Final Government Distribution                                      Chapter 6                                                IPCC WGIII AR6

1    Table of Contents
2    Chapter 6:          Energy Systems ............................................................................................................ 6-1

3      Executive Summary ......................................................................................................................... 6-3
4      6.1        Introduction .......................................................................................................................... 6-6
5      6.2        The Scope of the Energy System and its Possible Evolution ............................................... 6-8
6      6.3        Recent Energy System Trends and Developments .............................................................. 6-9
 7     6.4    Mitigation Options ............................................................................................................. 6-22
 8       6.4.1 Elements of Characterization ......................................................................................... 6-22
 9       6.4.2 Energy Sources and Energy Conversion ........................................................................ 6-22
10       6.4.3 Energy System Integration ............................................................................................. 6-48
11       6.4.4 Energy Storage for Low-Carbon Grids .......................................................................... 6-52
12       6.4.5 Energy Transport and Transmission .............................................................................. 6-58
13       6.4.6 Demand Side Mitigation Options from an Energy Systems Perspective ....................... 6-63
14       6.4.7 Summary of Mitigation Options .................................................................................... 6-65
15     6.5    Climate Change Impacts on the Energy System ................................................................ 6-69
16       6.5.1 Climate Impacts on the Energy System ......................................................................... 6-69
17       6.5.2 Impacts on Energy Supply ............................................................................................. 6-71
18       6.5.3 Impacts on Energy Consumption ................................................................................... 6-75
19       6.5.4 Impacts on Electricity System Vulnerability ................................................................. 6-77
20     6.6    Key Characteristics of Net Zero Energy Systems .............................................................. 6-78
21       6.6.1 What is a Net Zero Energy System? .............................................................................. 6-78
22       6.6.2 Configurations of Net-zero Energy Systems.................................................................. 6-80
23       6.6.3 The Institutional and Societal Characteristics of Net-zero Energy Systems .................. 6-94
24       6.6.4 Regional Circumstances and Net-zero Energy Systems ................................................ 6-95
25     6.7    Low-Carbon Energy System Transitions in the Near- and Medium-Term ........................ 6-98
26       6.7.1 Low-Carbon Energy System Transition Pathways ........................................................ 6-98
27       6.7.2.   Investments in Technology and Infrastructure ......................................................... 6-110
28       6.7.3 Energy System Lock-In and Path Dependence ............................................................ 6-112
29       6.7.4 Fossil fuels in a low-carbon transition ......................................................................... 6-117
30       6.7.5 Policy and Governance ................................................................................................ 6-118
31       6.7.6 Behaviour and Societal Integration .............................................................................. 6-120
32       6.7.7 The Costs and Benefits of Low-Carbon Energy System Transitions in the Context of
33       Sustainable Development ......................................................................................................... 6-123
34     Frequently Asked Questions ........................................................................................................ 6-127
35     References .................................................................................................................................... 6-129

     Do Not Cite, Quote or Distribute                                        6-2                                                    Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    Executive Summary
2    Warming cannot be limited to well below 2°C without rapid and deep reductions in energy system
3    CO2 and GHG emissions. In scenarios limiting likely warming to 1.5°C with limited overshoot (likely
4    below 2°C), net energy system CO2 emissions (interquartile range) fall by 87% to 97%% (60% to 79%)
5    in 2050. In 2030, in scenarios limiting warming to 1.5°C with no or limited overshoot, net CO2 and
6    GHG emissions fall by 35-51% and 38-52% respectively. In scenarios limiting warming to 1.5°C with
7    no or limited overshoot (likely below 2°C), net electricity sector CO2 emissions reach zero globally
8    between 2045 and 2055 (2050 and 2080) (high confidence) {6.7}
 9   Limiting warming to well below 2°C will require substantial energy system changes over the next
10   30 years. This includes reduced fossil fuel consumption, increased production from low- and zero-
11   carbon energy sources, and increased use of electricity and alternative energy carriers. Coal
12   consumption without CCS falls by 67% to 82% (interquartile range) in 2030 in scenarios limiting
13   warming to 1.5°C with no or limited overshoot. Oil and gas consumption fall more slowly. Low-carbon
14   sources produce 93% to 97% of global electricity by 2050 in scenarios that limit likely warming to 2°C
15   or below. In scenarios limiting warming to 1.5°C with no or limited overshoot (likely below 2°C),
16   electricity supplies 48% to 58% (36% to 47%) of final energy in 2050, up from 20% in 2019. (high
17   confidence) {6.7}
18   Net Zero energy systems will share common characteristics, but the approach in every country
19   will depend on national circumstances. Common characteristics of net zero energy systems will
20   include: (1) electricity systems that produce no net CO2 or remove CO2 from the atmosphere; (2)
21   widespread electrification of end uses, including light-duty transport, space heating, and cooking; (3)
22   substantially lower use of fossil fuels than today (4) use of alternative energy carriers such as hydrogen,
23   bioenergy, and ammonia to substitute for fossil fuels in sectors less amenable to electrification; (5) more
24   efficient use of energy than today; (6) greater energy system integration across regions and across
25   components of the energy system; and (7) use of CO2 removal (e.g., DACCS, BECCS) to offset any
26   residual emissions. (high confidence) {6.6}
27   Energy demands and energy sector emissions have continued to rise. From 2015 to 2019, global
28   final energy consumption grew by 6.6%, CO2 emissions from the global energy system grew by 4.6%,
29   and total GHG emissions from energy supply rose by 2.7%. Methane emissions, mainly fugitive
30   emissions from oil, gas, and coal, accounted for 18% of GHG emissions in 2019. Coal electricity
31   capacity grew by 7.6% between 2015 and 2019, as new builds in some countries offset declines in
32   others. Total consumption of oil and oil products increased by 5%, and natural gas consumption grew
33   by 15%. Declining energy intensity in almost all regions has been balanced by increased energy
34   consumption. (high confidence) {6.3}
35   Prices have dropped rapidly over the last five years for several key energy system mitigation
36   options, notably solar PV, wind power, and batteries. From 2015 to 2020, the prices of electricity
37   from PV and wind dropped 56% and 45%, respectively, and battery prices dropped by 64%. Electricity
38   from PV and wind is now cheaper than electricity from fossil sources in many regions, electric vehicles
39   are increasingly competitive with internal combustion engines, and large-scale battery storage on
40   electricity grids is increasingly viable. (high confidence) {6.3, 6.4}
41   Global wind and solar PV capacity and generation have increased rapidly. Solar PV grew by 170%
42   (to 680 TWh); wind grew by 70% (to 1420 TWh) from 2015 to 2019. Policy, societal pressure to limit
43   fossil generation, low interest rates, and cost reductions have all driven wind and solar PV deployment.
44   Solar PV and wind together accounted for 21% of total low-carbon electricity generation and 8% of
45   total electricity generation in 2019. Nuclear generation grew 9% between 2015 and 2019 and accounted
46   for 10% of total generation in 2019 (2790 TWh); hydroelectric power grew by 10% and accounted for

     Do Not Cite, Quote or Distribute                   6-3                                   Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    16% (4290 TWh) of total generation. In total, low- and zero-carbon electricity generation technologies
2    produced 37% of global electricity in 2019. (high confidence) {6.3, 6.4}
3    If investments in coal and other fossil infrastructure continue, energy systems will be locked-in to
4    higher emissions, making it harder to limit warming to well below 2°C. Many aspects of the energy
5    system – physical infrastructure; institutions, laws, and regulations; and behaviour – are resistant to
6    change or take many years to change. New investments in coal-fired electricity without CCS are
7    inconsistent with limiting warming to well below 2°C. (high confidence) {6.3, 6.7}
 8   Limiting warming to well below 2°C will strand fossil-related assets, including fossil
 9   infrastructure and unburned fossil fuel resources. The economic impacts of stranded assets could
10   amount to trillions of dollars. Coal assets are most vulnerable over the coming decade; oil and gas assets
11   are more vulnerable toward mid-century. CCS can allow fossil fuels to be used longer, reducing
12   potential stranded assets. (high confidence) {6.7}
13   A low-carbon energy transition will shift investment patterns and create new economic
14   opportunities. Total energy investment needs will rise, relative to today, over the next decades, if likely
15   warming is limited to 2°C or below. These increases will be far less pronounced, however, than the
16   reallocations of investment flows that are likely to be seen across sub-sectors, namely from fossil fuels
17   (extraction, conversion, and electricity generation) without CCS and toward renewables, nuclear power,
18   CCS, electricity networks and storage, and end-use energy efficiency. A significant and growing share
19   of investments between now and 2050 will be made in emerging economies, particularly in Asia. (high
20   confidence) {6.7}
21   Climate change will affect many future local and national low-carbon energy systems. The
22   impacts, however, are uncertain, particularly at the regional scale. Climate change will alter
23   hydropower production, bioenergy and agricultural yields, thermal power plant efficiencies, and
24   demands for heating and cooling, and it will directly impact power system infrastructure. Climate
25   change will not affect wind and solar resources to the extent that it would compromise their ability to
26   reduce emissions. (high confidence) {6.5}
27   Electricity systems powered predominantly by renewables will be increasingly viable over the
28   coming decades, but it will be challenging to supply the entire energy system with renewable
29   energy. Large shares of variable solar PV and wind power can be incorporated in electricity grids
30   through batteries, hydrogen, and other forms of storage; transmission; flexible non-renewable
31   generation; advanced controls; and greater demand-side responses. Because some applications (e.g., air
32   travel) are not currently amenable to electrification, 100% renewable energy systems would likely need
33   to include alternative fuels such as hydrogen or biofuels. Economic, regulatory, social, and operational
34   challenges increase with higher shares of renewable electricity and energy. The ability to overcome
35   these challenges in practice is not fully understood. {high confidence} {6.6}
36   Multiple energy supply options are available to reduce emissions over the next decade. Nuclear
37   power and hydropower are already established technologies. Solar PV and wind are now cheaper than
38   fossil-generated electricity in many locations. Bioenergy accounts for about a tenth of global primary
39   energy. Carbon capture is widely used in the oil and gas industry, with early applications in electricity
40   production and biofuels. It will not be possible to widely deploy all of these and other options without
41   efforts to address the geophysical, environmental-ecological, economic, technological, socio-cultural,
42   and institutional factors that can facilitate or hinder their implementation. (high confidence) {6.4}
43   Some mitigation options can provide more immediate and cost-effective emissions reductions
44   than others, but a comprehensive approach will be required over the next ten years to limit
45   warming to well below 2°C. There are substantial, cost-effective opportunities to reduce emissions
46   rapidly in several sectors, including electricity generation and light-duty transportation. But near-term
47   reductions in these sectors will not be sufficient to limit warming to well below 2°C. A broad-based
     Do Not Cite, Quote or Distribute                   6-4                                   Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    approach across the energy sector will be necessary to reduce emissions over the next ten years and to
2    set the stage for still deeper reductions beyond 2030. (high confidence) {6.4, 6.6, 6.7}
3    Enhanced integration across energy system sectors and across scales will lower costs and facilitate
4    low-carbon energy system transitions. Greater integration between the electricity sector and end use
5    sectors can facilitate integration of Variable Renewable Energy (VRE) options. Energy systems can be
6    integrated across district, regional, national, and international scales. (high confidence) {6.4, 6.6}
 7   The viable speed and scope of a low-carbon energy system transition will depend on how well it
 8   can support sustainable development goals (SDGs) and other societal objectives. Energy systems
 9   are linked to a range of societal objectives, including energy access, air and water pollution, health,
10   energy security, water security, food security, economic prosperity, international competitiveness,
11   employment. These linkages and their importance vary among regions. Energy sector mitigation and
12   efforts to achieve SDGs generally support one another, though there are important region-specific
13   exceptions. (high confidence) {6.1, 6.7}
14   The economic outcomes of low-carbon transitions in some sectors and regions may be on par with,
15   or superior to those of an emissions-intensive future. Cost reductions in key technologies,
16   particularly in electricity and light-duty transport, have increased the economic attractiveness of near-
17   term low-carbon transitions. Long-term mitigation costs are not well understood and depend on policy
18   design and implementation, and the future costs and availability of technologies. Advances in low-
19   carbon energy resources and carriers such as next-generation biofuels, hydrogen produced from
20   electrolysis, synthetic fuels, and carbon-neutral ammonia would substantially improve the economics
21   of net zero energy systems. (medium confidence) {6.4, 6.7}


     Do Not Cite, Quote or Distribute                  6-5                                  Total pages: 217
    Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

1   6.1 Introduction
2   The global energy system is the largest source of CO2 emissions (Chapter 2). Reducing energy sector
3   emissions is therefore essential to limit warming. The energy systems of the future will be very different
4   from those of today if the world successfully limits warming to well below 2°C. Energy will be
5   provided, converted, and used in different ways than it is today (Figure 6.1). Achieving and responding
6   to these changes presents an impressive range of challenges and opportunities.



    Do Not Cite, Quote or Distribute                   6-6                                  Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

2          Figure 6.1 Global energy flows within the 2019 global energy system (top panel) and within two
3      illustrative future, net zero CO2 emissions global energy system (bottom panels). Source: IEA, IPCC
4       Database. Flows below 1 EJ are not represented, rounded figures. The illustrative net zero scenarios
5     correspond to the years in which net energy system CO2 emissions reach zero – 2045 in IP1 and 2060 in
6                         IP2. Source: IP1: Luderer et al.(2021); IP2: Riahi, K. et al., 2021;IP2


 8   Within this context, this chapter has two main objectives. First, it aims to assess specific, individual
 9   mitigation options in energy supply, energy transformation, and energy transportation and transmission.
10   This assessment is complementary to a set of chapters that explore mitigation options in agriculture,
11   forestry, and other land uses (Chapter 7), urban systems and other settlements (Chapter 8), buildings
12   (Chapter 9), transport (Chapter 10), industry (Chapter 11), and cross-sectoral perspectives (Chapter 12).
13   Second, this chapter aims to assess system-level mitigation opportunities and challenges across the
14   entirety of energy systems. These systems include energy supply, transformation, transmission, storage,
15   transportation, and end-uses. They also include the societal systems that interact with the physical
16   energy system. As energy systems become increasingly integrated and interconnected, a system-wide
17   perspective is necessary for understanding mitigation opportunities and challenges.
18   Within this context, this chapter addresses six topics, each of which is addressed in a separate section.
19   First, Section 6.2 defines the scope of the energy system. Section 6.3 then discusses the recent trends in
20   energy systems that might exert the most significant influence on energy system evolution and options
21   for reducing emissions. Section 6.4 assesses the status and potential of individual energy supply,
22   transformation, storage, transportation and transmission, and integration mitigation options in the
23   energy sector. Section 6.5 explores how climate change might affect energy systems and alter potential
24   energy system mitigation options and strategies. Section 6.6 identifies key characteristics of net zero
25   energy systems – those that emit very little or no CO2. Section 6.7 explores transition pathways toward
26   and through net zero energy systems. Across all of these sections, the chapter aims to explore the ways
27   that energy sector mitigation options and strategies interact with SDGs and other societal and
28   environmental goals.

     Do Not Cite, Quote or Distribute                   6-7                                  Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    6.2 The Scope of the Energy System and its Possible Evolution
 2   For this chapter, energy systems are defined broadly to include both physical and societal elements. The
 3   physical infrastructure includes all the infrastructure and equipment used to extract, transform,
 4   transport, transmit, and convert energy to provide energy services. In addition to the physical system, a
 5   broad range of societal systems and dynamics are relevant to the energy system. Human societies use
 6   energy to transport themselves and the goods that they use and consume, to heat, cool, and light their
 7   homes, to cook their food, and to produce goods and services. Energy systems are therefore tied to the
 8   systems involved in the provision of these various goods and services. All energy users engage in the
 9   operation of energy systems by demanding energy at particular times and in particular forms. They can
10   adjust their behaviour and demands, for example, by using less energy or by changing when they use
11   energy. Consumers can invest in equipment that reduces their energy needs, and they can invest in
12   technologies that transform energy (e.g., rooftop solar) or store energy (e.g., batteries). Firms and
13   governments invest in equipment to produce, transform, and transport energy such as power plants,
14   refineries, electric transmission lines, and oil tankers. All aspects of energy systems are governed by
15   laws, regulations, and actual institutions that reside within businesses and governments at all levels.
16   This includes, for example, rules for trading emissions permits, deciding when particular electricity
17   generation technologies might come online, water management and related environmental rules that
18   define the availability of hydropower or influence water availability for cooling power plants,
19   regulations for injecting CO2 into underground reservoirs or disposing of nuclear waste, and even
20   company policies regarding work hours or teleworking, which can have important implications for
21   energy demand profiles. Many people are employed in the energy sector, and energy system mitigation
22   will eliminate some jobs while creating others.
23   This broader view of energy systems is essential for understanding energy system mitigation, as these
24   broader societal and institutional factors can have an important influence on energy system
25   transformations and the potential to rapidly reduce energy CO2 emissions. Energy system mitigation is
26   as much about the challenges of societal change as it is about the challenges of changes in physical
27   infrastructure, technologies, and operations. While this chapter does not attempt to draw a specific
28   boundary around all the different systems that interact with the energy system, it frequently explores
29   these broader system interactions when assessing different mitigation options and strategies.
30   There is no single spatial scale at which energy systems might be defined and assessed. They can be
31   assessed at the scales of homes, cities, states or provinces, countries, regions, or the entire world. These
32   different scales are frequently both distinct with their own internal dynamics yet also connected to one
33   another. This chapter most frequently assesses energy systems from the country and global perspective.
34   Because the energy system is so complex, it can be hard to define particular parts of it precisely, and
35   there may be competing definitions in the literature. For the purposes of this chapter, “energy supply”
36   encompasses all primary energy, conversion, and transmission processes with the exception of those
37   that use final energy to provide energy services in the end-use sectors (transport, buildings, industry
38   and agriculture). The “energy system” includes energy end uses sectors along with energy supply.
39   “Low-emissions” is used for energy technologies that produce little CO2 or no CO2 or that remove CO2
40   from the atmosphere. Similarly, “low-carbon” transitions is used to describe transitions that limit likely
41   warming to 2°C or below. “Net-zero” energy systems refer to those that produce very little or no CO2
42   or may even sequester CO2 from the atmosphere.

     Do Not Cite, Quote or Distribute                    6-8                                   Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    6.3 Recent Energy System Trends and Developments
2    Global energy sector emissions continue to grow but at a decreasing rate
3    Current energy sector emissions trends, if continued, will not limit global temperature change to well
4    below 2°C (high confidence). Global energy system fossil fuel CO2 emissions grew by 4.6% between
5    2015 and 2019 (1.1% yr-1), reaching 38 GtCO2 yr-1 and accounting for approximately two-thirds of
6    annual global anthropogenic GHG emissions. In 2020, with the worldwide COVID pandemic, energy
7    sector CO2 emissions dropped by roughly 2 GtCO2 yr-1 (Figure 6.2). However global energy-related
8    CO2 emissions are projected to rebound by nearly 5% in 2021, approaching the 2018-19 peak (IEA
9    2021d)

11     Figure 6.2 Global energy sector fossil fuel CO2 emissions and annual change 2000-2019 (MtCO2 yr-1).
12                         (Source: adapted from (Minx et al. 2021a; Crippa et al. 2021)

13   Coal was the single largest contributor to energy sector CO2 emissions between 2015 and 2019,
14   accounting for about 44% of energy sector CO2 emissions in 2019. Oil accounted for about 34% and
15   natural gas accounted for about 22% of energy sector CO2 emissions. Coal, oil and natural gas CO2
16   emissions grew respectively by 1.2%, 2% and 12.7% (annual rates of 0.31%, 0.5% and 3%) (Figure
17   6.3). The electricity sector remains the single largest source of energy sector CO2 emissions, accounting
18   for about 36% in 2019, followed by industry at 22% and transport (excluding international shipping and
19   aviation transport) at about 18% (Figure 6.3). Shipping and aviation accounted for a little over 3%.
20   These proportions have remained relatively unchanged over the last decade. Recent trends reinforce the
21   near-term challenges facing energy sector mitigation - electricity sector emissions continue to rise
22   despite rapid deployment of wind and solar power (see below); transportation emissions continue to
23   rise, and petroleum remains the dominant fuel, despite advances in batteries and electric cars (see
24   below). Some specific sectors, such as shipping and aviation, may present longer-term challenges.
25   Energy supply GHG emissions, including CO2 and non-CO2 greenhouse gases, reached 20 GtCO2-eq
26   yr-1 in 2019, rising by 2.7% between 2015 and 2019 (0.66% yr-1). Approximately 18% of energy supply
27   emissions were non-CO2 emissions. Electricity and heat contributed approximately 69% of total energy
28   supply GHG emissions in 2019 (Figure 6.3). This growth has occurred despite the high penetration of
29   solar PV and wind power, particularly in Asia and developed countries.
30   Fugitive emissions from fossil fuel production, primarily methane, accounted for about 18% of sector
31   supply emissions in 2019, with 2.6 Gt CO2eq yr-1 linked to oil and gas production and 1.3 Gt CO2eq yr-
     Do Not Cite, Quote or Distribute                  6-9                                  Total pages: 217
     Final Government Distribution                   Chapter 6                            IPCC WGIII AR6

1     to coal mining (Crippa et al. 2021). Oil and gas operations produced 2.9 GtCO2eq yr-1 in 2019 (82 Mt
2    yr-1 as methane), split roughly equally between the two (IEA 2020a). There remains a high degree of
3    uncertainty in methane emissions estimates from oil and gas operations despite the emergence of new
4    data from satellites and other measurement campaigns. According to a recent study (Hmiel et al. 2020),
5    methane emissions are underestimated by about 25 to 40%.
            Global Energy-Sector CO2 Emissions by Fuel           Global Energy Sector CO2 Emissions by

          Global Energy Supply GHG Emissions by Sector           Global Energy Supply GHG Emissions by

7        Figure 6.3 Global energy sector CO2 emissions and global energy supply GHG emission. Panel a: (IEA
8                                       2020a), other panels (Crippa et al. 2021)

 9   Increasing global energy sector GHG emissions have been driven by rising emissions in some large
10   developing and emerging countries; however, per capita emissions in these countries remain well below
11   those in developed countries (Yu et al. 2019). From 2015 to 2019, Eastern Asia, Southern Asia, and
12   South-East Asia energy sector CO2 emissions grew by 2.4% yr-1, 2.6% yr-1, and 5.1% yr-1, respectively.
13   The relative and absolute shares of Europe and North America have continued to decline, partly due to
14   the growth in other countries (Figure 6.3).

     Do Not Cite, Quote or Distribute                  6-10                                 Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

2            Figure 6.4 Drivers of greenhouse gas emissions across selected regions (Lamb et al. 2021).

 3   Despite the declining energy intensity, global energy system CO2 emissions have closely tracked GDP
 4   per capita (Figure 6.4). This is especially true in the Asian economies, which have experienced rapid
 5   GDP per capita growth in the past decades and a massive rise in energy demand. Similarly, emissions
 6   have declined in times of economic downturns – for example, in Eurasia in the 1990s and globally in
 7   2009 and 2020. Population growth has also contributed to emissions growth globally and in most
 8   regions, particularly Africa, but the effect of population growth has been less than that of economic
 9   growth. Since 2015, energy intensity has been declining (IEA 2020b), limiting the impact of economic
10   and population growth. However, there is no region where this factor alone would have been sufficient
11   to decrease CO2 emissions from the energy system. In Europe and North America, the only two regions
12   where emissions decreased meaningfully since 2010, a steady decrease in the carbon intensity of energy
13   was a significant downward driver. The reduction in carbon intensity in the EU is due primarily to the
14   increase of renewable electricity production coupled with the low levels of fossil fuel-based production
15   in the energy mix (Dyrstad et al. 2019).

     Do Not Cite, Quote or Distribute                  6-11                                   Total pages: 217
     Final Government Distribution                    Chapter 6                                IPCC WGIII AR6

1    Global energy production and demand continue to grow, but at a declining rate
 2   Recent changes in the energy system can be viewed within the context of longer-term trends in energy
 3   supply and use. Over the last decade, there has been a significant increase in the total primary energy
 4   supply (TPES) and major changes in energy sources. From 2015 to 2019, TPES grew by 6.6% (1.6%
 5   yr-1) from 569 EJ yr-1 to 606 EJ yr-1. Natural gas consumption grew most quickly during this period, at
 6   3.5% yr-1. Coal, oil and oil products grew at annual rates of 0.23% yr-1 and 0.83% yr-1, respectively. In
 7   2019, the shares of coal, oil, and natural gas in global TPES were 27%, 31% and 23%, representing
 8   only a modest shift from 2015, when the shares were 28%, 32% and 22%, respectively. Renewables,
 9   excluding hydropower, grew at an annual rate of 12% yr-1 during this period; however, their share
10   remains marginal in 2019 with just 2.2% of the TPES compared to 1.5% in 2015 (Figure 6.5). Bioenergy
11   (including traditional bioenergy) accounted for 9.4% of the TPES, a similar share compared with 2015.
12   The total final energy consumption (TFC) grew by 6.6% (1.6% yr-1) from 2015 to 2019, rising from
13   392 EJ yr-1 to 418 EJ yr-1. This is a slower growth rate than the previous decade (2.8 % yr-1) (Figure
14   6.5). In 2019, oil products used for transportation accounted for 41% of TFC. The penetration of non-
15   fossil fuels is still marginal despite the significant growth of electric vehicles in recent years. Coal still
16   accounted for 9.5% of TFC in 2019, dropping from 11.7% in 2015. Coal is mainly used as a primary
17   energy source in industry and to a lesser extent in the residential sector. The share of electricity
18   increased modestly, from 18.6% in 2015 to 20.2% in 2019, reflecting increasing access in developing
19   countries and increasing use of electricity for a wide variety of end uses in the residential sector (see
20   Box 6.1). Heat accounts for approximately 3% of TFC, used mainly in industry and the residential
21   sector. Biofuels and waste accounted for 10.4% of TFC in 2019, only modestly changed compared with
22   2015.

23    Figure 6.5 World Total Primary Energy Supply (TPES) (EJ) and total final energy consumption (TFC)
24         2000-2019 (adapted from IEA world energy balances (Minx et al. 2021b)database for IPCC)

25   There are important differences in fuel use across countries. While developed countries almost
26   exclusively use modern fuels, many countries still obtain a significant fraction of their energy from
27   traditional bioenergy (fuelwood and charcoal). Traditional bioenergy (fuelwood and charcoal) is
28   particularly important in sub-Saharan countries and some Asian countries such as India, particularly in
29   the residential sector for cooking. Africa is still characterized by a high share of traditional bioenergy
30   in TPES and TFC. In 2019, biomass and waste in Africa accounted for 44% of the TPES. The global
31   average was 9.4%.
32   Asia has been particularly important in TFC growth since 2015. In 2019, Eastern Asia accounted for
33   more 24% of TFC (1.52% annual growth from 2015). In contrast, TFC has increased by only 0.58% in
34   Europe and 1.24% in North America. Despite an increase of 2.05% over the same period, Africa’s TFC
35   remains relatively low (6.1% of global TFC), particularly in sub-Saharan countries. Approximately 860
36   million people, mostly in sub-Saharan Africa and some Asian countries, lacked access to electricity and

     Do Not Cite, Quote or Distribute                    6-12                                    Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    about 2.65 billion to clean-cooking facilities in 2018 (IEA 2019a). Achieving universal energy access
2    (SDG-7) will require energy transitions in the domestic sector, including new developments in off-grid
3    energy technologies, emphasis on rationalizing energy subsidies, and increasing efforts to address
4    health concerns related to the use of traditional fuels (Box 6.1).
5    Non-climate factors continue to drive energy systems changes
6    While energy system changes are vital to climate mitigation, recent energy system changes have arisen
7    in response to a much broader set of factors. Important factors include economic growth, energy access,
8    energy justice, energy security, air pollution, technological progress in low-emissions technologies,
9    local job creation. Several of these are discussed here.

10   Energy Access. Between 2000 and 2019, the proportion of the population with access to electricity
11   increased from 73% to 90% (IEA 2020c). Although most of those people gaining access to energy have
12   gained access to fossil fuel-based electricity, an increasing number are gaining access to electricity from
13   renewable sources. Low-emissions, decentralised systems are proving a cost-effective way to provide
14   electricity in rural areas (Scott et al. 2016; IEA 2019b; Muchunku et al. 2018), although the use of diesel
15   generators continues in some remote areas. Between 2000 and 2019 the proportion of the population
16   with access to clean cooking (modern fuels and/or improved biomass cookstoves) rose from 52% to
17   66%.

18   Energy Security. The ability of countries to maintain access to reliable and affordable energy resources
19   continues to shape energy policy. Energy security is perceived as a national security issue and often
20   prioritized over climate concerns (Nyman 2018). The linkage between climate and energy security is
21   now widely recognized (Toke and Vezirgiannidou 2013, Fu et al. 2021; La Viña et al. 2018; Blarke and
22   Lund 2007; World Energy Council 2020; United Nations 2021). Approaches to energy security are
23   frequently driven by the scope of domestic energy resources. For example, energy security concerns
24   have led to continued reliance on domestic coal production and consumption (Jakob et al. 2020) and
25   increased investment in domestic renewable generation (Konstantinos & Ioannidis, 2017). LNG
26   Importers have diversified their sources as reliance on LNG has increased (Vivoda 2019).

27   Air Pollution. The energy system is an important source of air pollution, including both indoor and
28   outdoor air pollution. Efforts to address air pollution in several countries and regions (the U. S., Mexico,
29   China, India, European Union, Africa, Southeast Asia, among others) have had an importance influence
30   on energy system changes (Bollen and Brink 2014, Fang et al. 2019). Policies aimed at controlling NOX
31   and SO2 emissions have driven emissions abatement efforts and coal fleet retirements (Drake & York,
32   2021) (Singh and Rao 2015). In some places, the prospect of reducing local air pollution remains more
33   salient to policymakers and the public than climate mitigation when deciding to tighten regulations on
34   coal use (Brauers & Oei, 2020).

35   Technology and costs. Costs for renewable technologies have fallen significantly in recent years,
36   driving significant changes in electricity production and transportation (see below). These advances are
37   not divorced from climate and other environmental concerns (Kuik, Branger, & Quirion, 2019;
38   Timilsina & Shah, 2020). Recent advances in PV cells, for example, can be traced in part to aggressive
39   deployment policies spurred by energy security, climate, and other environmental concerns (Kreuz and
40   Müsgens 2017, 6.3.5, 6.4.2). The falling costs of batteries, manly Li-ion batteries, has boosted the
41   competitiveness of electric vehicles (6.3.7. Nykvist et al. 2015).


43                        Box 6.1 Energy access, energy systems, and sustainability

     Do Not Cite, Quote or Distribute                   6-13                                   Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

1    Successful mitigation must work in tandem with fundamental development goals such as access to
2    modern forms of energy. In many developing countries, access to electricity, clean cooking fuels, and
3    modern and efficient energy remain an essential societal priority. This is particularly true in sub-Saharan
4    Africa and several Asian countries. SDG7 on universal access to modern energy includes targets on
5    modern energy services, renewable energy, and energy efficiency, which implies a profound
6    transformation of the current energy systems. Although there are different definitions of energy access,
7    the ultimate goal is universal access to clean and modern fuels.

9                    BOX 6.1 Figure 1 Measuring access to energy (ESMAP-World Bank, 2015)

10   Despite progress in some countries such as India, Bangladesh and Kenya, 860 million people were
11   without access to electricity in 2018, compared with 1.2 billion in 2010. About 2.65 billion households
12   were cooking with solid fuels, distributed across Asia and Africa (IEA et al. 2020). Around 850 million
13   people in sub-Saharan Africa relied on traditional biomass (firewood and charcoal) for cooking, and 60
14   million relied on kerosene and coal to meet their energy needs (IEA 2018a). Air pollution was likely
15   responsible for 1.1 million deaths across Africa in 2019 (Fisher et al. 2021). It has been estimated that
16   2.2 billion people will still be dependent on inefficient and polluting energy sources for cooking by
17   2030, mainly in Asia and Sub-Saharan Africa, and 650 million people are likely to remain without
18   access to electricity in 2030, 90% of whom will reside in Sub-Saharan Africa (IEA et al. 2020).
19   Research indicates that decentralised and on-grid renewables are likely the least cost options to provide
20   universal access to electricity by 2030 (Section 6.4.2). Natural gas, LPG, and improved biomass
21   cookstoves are the most important options for cooking. Universal access to electricity and clean cooking
22   requires a rapid shift from traditional biomass to cleaner fuels and/or clean cooking technologies (IEA
23   et al. 2020). It has been estimated that the provision of electricity and clean cooking for all would require
24   USD 786 billion in cumulative investment to 2030, equal to 3.4% of total energy sector investment over
25   the period (IEA 2017).
26   Even without universal access to modern energy, increased access will substantially affect energy
27   systems, particularly electricity systems through the deployment of renewable energy, LPG, and
28   biomass supply chains. Universal access for households, however, will have a minimal impact on global
29   energy demand; it has been estimated that universal access for household will increase energy demand
30   by 0.2% in 2030 (37 Mtoe yr-1) relative to a future without any change in access to modern energy
31   (citation)

32   [END BOX 6.1 HERE]

33   There have been initial efforts to phase out coal but only modest declines in use

     Do Not Cite, Quote or Distribute                    6-14                                   Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    Global coal consumption has been declining, with small fluctuations, since it peaked in 2013 (IEA
2    2020d). Coal is faring differently across regions. Coal use has been decreasing in the OECD regions,
3    particularly in the U.S. and the European Union, while remaining mostly flat in China after a period of
4    growth, and it is continuing to increase in other major developing Asian economies (IEA 2020d). Trends
5    in the electricity sector, where most coal is being consumed, are similar. Growth in coal-fired electricity
6    generation capacity in the Asia Pacific region has offset retirements in North America and Europe
7    (Jakob et al. 2020).
 8   Reductions in coal consumption have been driven in large part by non-climate factors, most notably
 9   environmental regulations to address air pollution, rapidly declining costs of renewables, and lower
10   natural gas prices, especially inexpensive unconventional gas in the U.S. (Diluiso et al 2021;
11   Vinichenko et al., 2021; Culver and Hong 2016). Older coal-fired power plants that cannot meet new
12   environmental regulations or have become unprofitable or uncompetitive have been closed in many
13   regions. Moreover, coal power expansion has slowed down in Asia, as countries have suspended and
14   cancelled new projects for reasons such as overcapacity, environmental constraints, and the
15   development of renewables (see Box 6.2).
16   Different regions have replaced retired coal with different energy sources. Old coal fleets have been
17   replaced approximately half by gas and half by renewables in the U.S., mainly by renewables in the
18   European Union, and by advanced coal plants and renewables in Asia (EMBER 2020). Replacing old
19   coal with new coal facilities is inconsistent with limiting warming to 2ºC or below {high confidence}
20   (Pfeiffer et al. 2016, 2018; Smith et al. 2019; Tong et al. 2019, Section 6.7.4).
21   Major coal consuming countries with abundant coal reserves remain far from phasing out coal (Spencer
22   et al. 2018; Edenhofer et al. 2018). In most developing countries with large coal reserves, coal use has
23   been increasing to support energy security and because it is perceived to have lower costs than
24   alternatives (Steckel et al. 2015; Kalkuhl et al 2019). However, coal faces increasing business risks
25   from the decreasing costs of alternative, low-emissions energy sources and increasing focus on air
26   pollution and other environmental impacts from coal mining and use (Garg et al. 2017; Sovacool et al
27   2021). Continued coal builds, mostly in developing countries, will increase the risks of stranded assets
28   (see Box 6.13) (Farfan Orozco 2017; Saygin et al. 2019; Cui et al. 2019).
29   Economic, social, and employment impacts of accelerated coal phaseouts tend to be significant in coal-
30   dependent regions. Tailored reemployment has been used to support coal transitions in some regions.
31   Although some estimates show higher employment opportunities from low-carbon energy (Garrett-
32   Peltier 2017), results vary across regions. Moreover, even with a net increase in total employment, in
33   the long run, renewable jobs are often located outside of coal regions and require different skill sets
34   from the coal industry (Spencer et al. 2018). In a broader sense, achieving a “just transition” also
35   requires managing the impacts on regional economic development for coal-dependent communities and
36   the effects of higher energy prices for consumers and energy-intensive industries through a
37   comprehensive policy package (see Box6.2) (Jakob et al. 2020; Green and Gambhir 2020).
38   [START BOX 6.2 HERE]

39                           Box 6.2 Status and Challenges of a Coal Phase-Out

40   Limiting global warming to 2ºC or below requires a rapid shift away from unabated coal consumption
41   (coal without CCS) in the energy system by 2050 (IPCC 2018a Section 6.7, Chapter 3). This will require
42   cancellation of new coal power projects and accelerated retirement of existing coal plants (Pfeiffer et
43   al. 2018; Edenhofer et al. 2018; Tong et al. 2019, Kriegler et al. 2018Smith et al. 2019). To limit
44   warming to 2ºC or below, and without new builds, existing coal plants will need to retire 10 to 25 years
45   earlier than the historical average operating lifetime. Completing all planned projects will further reduce
46   the viable lifetime of all plants by 5 to 10 years if warming is to be limited to 2ºC or below (Cui et al.

     Do Not Cite, Quote or Distribute                   6-15                                  Total pages: 217
     Final Government Distribution                     Chapter 6                                IPCC WGIII AR6

1    2019). Phasing-out coal in the next few decades will present economic, social, and security challenges.
2    These will vary across regions based on the characteristics of existing coal infrastructure, the availability
3    of alternatives, economic development, and technological and institutional lock-in (Jakob et al. 2020).


 6      Box 6.2 Figure 1 Retirement of coal-fired power plants to limit warming to 1.5°C and likely 2°C. (a)
 7     Historical facility age at retirement (b) the vintage year of existing units, (c) global coal capacity under
 8    different plant lifetimes, compared to capacity levels consistent with a well-below 2°C (green) and 1.5°C
 9    (blue) pathway assuming no new coal plants, and (d) and assuming plants currently under construction
10     come online as scheduled, but those in planning or permitting stages are not built. (Source: (Cui et al.
11                                                        2019)).

12   Between 2015 and 2019, global coal power capacity grew by 146 GW, or 7.6%, as new builds offset
13   retirements in some countries (Global Energy Monitor et al. 2021). Meanwhile, an increasing number
14   of countries and regions have committed to or operationalised coal phase-outs (Jewell et al. 2019; Watts
15   et al. 2019; Littlecott et al. 2021). Actions are being taken by various international and subnational
16   actors, including national and subnational governments, public and private power companies, and
17   financial institutions and pension funds that have committed not to fund new coal or coal-based
18   infrastructure (yan Nie et al. 2016; Buckley 2019; Auger et al. 2021). Although these initial efforts are
19   not yet sufficient in limiting warming to 1.5°C, and most have occurred in regions with older coal fleets,
20   these examples provide insight into possible coal phaseout strategies (Spencer et al. 2018) and help
21   identify the mechanisms driving the move away from coal, such as market, technology, policy, or other
22   societal objectives. They also enable better understanding of the possible character of oil and gas phase-
23   downs that would ultimately be needed to limit warming two well below 2°C (Section 6.7.4; (Raimi et
24   al. 2019).
25   Europe: Several European countries are part of the Powering Past Coal Alliance (PPCA) and have
26   committed to phase-out unabated coal on or before 2030 (Jewell et al. 2019). Because these countries
27   represent a small share of global coal generation capacity and have mostly ageing coal plants, they tend

     Do Not Cite, Quote or Distribute                     6-16                                    Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

 1   to face fewer changes in phasing out coal. The effectiveness of PPCA in countries with younger coal
 2   fleets has thus been questioned (Jewell et al. 2019; Blondeel et al. 2020). Germany recently joined the
 3   PPCA and has committed to phaseout unabated coal by 2038. As part of its commitment to phase out
 4   coal, Germany is implementing a set of measures that include compensation for power plant closures,
 5   labour market measures for coal workers, and substantial support of structural change in coal-mining
 6   regions. Poland, another coal-heavy country in Europe, has not indicated a coal phaseout target and
 7   faces substantial challenges (Whitley et al. 2017; Antosiewicz et al. 2020). European efforts to phase
 8   out coal indicate that appropriate financial instruments are needed (Rentier et al. 2019), and a just
 9   transition for workers are important to gain broad public support and help those most affected by the
10   phaseout (Johnstone and Hielscher 2017; Osička et al. 2020).
11   North America: Coal use has been declining in North America. In the U.S., the primary driver has
12   been the availability of cheap shale gas and ageing coal fleets. Coal use in the US has dropped by over
13   50% since 2008 (EIA 2019). The recently announced NDC by the Biden Administration sets a 100%
14   carbon-free electricity goal by 2035 (The White House 2021), indicating a phaseout not only of
15   unabated coal electricity generation, but also of natural gas generation. As one of the two founding
16   countries of the PPCA, Canada has committed to phasing out unabated coal power by 2030
17   (Government of Canada 2018). Declining coal use in both the U.S. and Canada has decreased GHG
18   emissions, local air pollutants, and cooling water use (Harris et al. 2015; Kondash et al. 2019). However,
19   there have been concerns about social and economic consequences, particularly at the local level. For
20   instance, the U.S. has lost about 50,000 coal mining jobs between 2011 and 2021 (U.S. Bureau of Labor
21   Statistics, 2021), with significant regional and economic inequities (Bodenhamer 2016; Abraham 2017;
22   Greenberg 2018). Comprehensive social programs, such as retirement compensation, training for
23   reemployment, and business support for economic diversification, have been suggested as means to
24   support a just transition (Patrizio et al. 2018; Homagain et al. 2015; Grubert 2020).
25   Asia: After a period of rapid growth, coal expansion has slowed in Asia, but it still the primary driver
26   of the global increase in coal demand (IEA 2020e). China's coal consumption reached a plateau under
27   policy efforts during the 13th Five-Year Plan (2016-2020), and new coal plants are being built at a slower
28   rate than previously. Both China and India have suspended and cancelled many new coal power projects
29   and retired a small set of old, dirty, inefficient coal plants (Global Energy Monitor et al. 2021; CEA
30   2019). These efforts are largely due to non-climate reasons, such air pollution and health (Singh and
31   Rao 2015; Gass et al. 2016; Peng et al. 2018; Malik et al. 2020), overcapacity (Blondeel and Van de
32   Graaf 2018), and rural electrification and renewable investments (Aklin et al. 2017; Thapar et al. 2018).
33   However, as new builds offset retirements, coal generation capacity has continued to grow in both
34   countries since 2015 (Global Energy Monitor et al. 2021). Other fast-growing Southeast Asian
35   countries, such as Indonesia, Vietnam, and Philippines have experienced strong growth in coal use (IEA
36   2020b), but an increasing number of new coal power projects are being cancelled (Littlecott et al. 2021).
37   Coal projects in these countries are decreasingly likely to proceed because they rely on international
38   financing, and China, Japan, United States, and other G7 countries have pledged to end overseas coal
39   financing (Schiermeier 2021).
40   Africa: New coal power projects in Africa have been declining since 2016, with only South Africa and
41   Zimbabwe currently building new coal plants and several others has planned projects (Littlecott et al.
42   2021). However, these projects also largely depend on international financing and are thus less likely
43   to be implemented (see above). In South Africa, employment in the coal mining sector has dropped by
44   almost half since the 1980s and is has been estimated to fall from 77,00 today to 22,000 to 42,000 by
45   2050 (Strambo et al. 2019; Cock 2019). Policy and financial support are essential to ensure a sustainable
46   transition for these workers (Swilling et al. 2016).
47   [END BOX 6.2 HERE]

     Do Not Cite, Quote or Distribute                  6-17                                  Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    Solar and wind energy have grown dramatically, but global shares remain low relative to other
2    sources
 3   Global PV and wind electric capacities grew 170% and 70%, respectively, between 2015 and 2019.
 4   Total solar and wind capacities in 2019 were 609 GW and 623 GW (Figure 6.6) and generation was
 5   680 TWh yr-1 and 1420 TWh yr-1. The combined share of solar and wind in the total global electricity
 6   generation in 2019 was around 8% (5.5% wind, 2.5% solar), up from around 5% in 2015 (IEA 2021a).
 7   Since 2015, the cost of solar PVs has declined by over 60%. Offshore wind costs have fallen by 32%,
 8   and onshore wind costs have fallen by 23% (Section 6.4). PV was around 99% of total solar capacity in
 9   2019; onshore wind was about 95% of total wind capacity. Concentrating solar power (CSP)
10   deployment has also continued to grow, but it remains far below PV. Prior to 2010, 50% of all wind
11   capacity was in Europe, but, since then, capacity growth in Asia, led by China, has surpassed the growth
12   in Europe. As a consequence, Europe’s share in global solar capacity has declined from 74% in 2010
13   to 24% in 2019. Asia’s share in wind and solar capacity in 2019 was 41% and 56%, followed by Europe
14   (31% and 24%) and North America (20% and 12%) (IRENA 2021a, 2020a).


16        Figure 6.6 Global solar and wind electricity installed capacities (GW) from 2015–2019 and their
17         combined share in total electricity generation Source: data from (IEA 2021a) IRENA, 2021).

18   Although the shares of wind and solar remains low in the global total electricity generation, recent
19   growth rates signal the potential for these technologies to support substantial mitigation. The prospects
20   for a continuation of recent growth rates will depend on meeting key challenges such as rapidly
21   integrating wind and solar into electricity grids (Section 6.6.2, Box 6.8) and retiring fossil power plants
22   (see above).
23   Low-carbon energy sources beyond wind and solar have continued to grow
24   Low-carbon energy sources such as nuclear, hydropower, bioenergy, geothermal, marine, and fossil or
25   bioenergy with CCUS have continued to grow since 2015 (IEA, 2017; IEA, 2021a). Hydroelectric
26   power grew from 3890 TWh yr-1 (14.0 EJ yr-1) in 2015 to 4290 TWh yr-1 (15.5 EJ yr-1) in 2019, or
27   10.3%; nuclear power grew from 2570 TWh yr-1 (9.3 EJ yr-1) to 2790 TWh yr-1 (10.1 EJ yr-1), or 8.6%.
28   Hydroelectric and nuclear shares in global total electricity generation remained around 16% and 10%,
29   respectively (IEA, 2017; IEA 2021a). Global biofuels production grew from 3.2 EJ yr-1 to 4.0 EJ yr-1
30   from 2015 to 2019 (IEA, 2017; IEA 2021a). Bioenergy accounted for 2.4% of electricity generation in
31   2019. Geothermal energy sources produced 92 TWh yr-1 (0.33 EJ yr-1) of electricity in 2019, up from
32   80 TWh yr-1 (0.28 EJ yr-1) in 2015 (IEA, 2017; IEA 2021a). At present, there are 28 commercially

     Do Not Cite, Quote or Distribute                   6-18                                  Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    operating CCUS facilities with a CO2 removal capacity of around 40 million tonnes yr-1 (Mtpa). Only
2    two of these are associated with electricity production; the majority are in industrial applications. 37
3    commercial projects, accounting for about 75 Mtpa, are in various stages of development and
4    construction (Global CCS Institute 2020). The share of marine energy in the global electricity
5    generation has remained at approximately 1 TWh yr-1 since 2015. In total, low- and zero-carbon
6    electricity generation technologies produced 37% of global electricity in 2019.
7    Battery prices have dropped substantially, spurring deployment in electricity and transportation
 8   Recent years have seen a rapid decline in the cost of energy storage, particularly batteries (see Section
 9   6.4.4). The price of lithium-ion batteries (LIBs) has declined by 97% in the past three decades and by
10   90% in the past decade alone (IEA 2021a; Ziegler and Trancik 2021)These declines have important
11   implications for the energy systems, most notably in supporting increased deployment of variable
12   renewable electricity (VRE) generation and electrification of the vehicle fleet.

13   Battery electricity storage has emerged as an important for supporting the flexibility of electricity
14   systems as they accommodate rising shares of VRE. Although pumped-storage hydropower systems
15   accounted for 160 GW, or over 90%, of total energy storage capacity in 2019 (IEA 2020c), battery
16   energy storage systems, led by Li-ion technology, have accounted for over 90% of new capacity
17   addition since 2015 (IRENA 2019a). In 2019, 10 GW of batteries were connected at the grid and
18   consumer level, rising from 0.6 GW in 2015 (IEA 2020c; IEA WEO 2019).

19   In California, in the U.S., legislation was passed to procure around 1.3 GW energy storage (excluding
20   pumped storage) by 2020. One of the largest utility-scale battery storage facilities (300 MW) recently
21   went online in California (Vistra Corp. 2021). Other major projects are in Florida in the US (409 MW),
22   London in the UK (320 MW), Lithuania (200 MW), Australia (150 MW), Chile (112 MW) and
23   Germany (90 MW), (Katz 2020; ARENA 2020; IRENA 2019a).

24   The drop in battery prices has also had important implications in the transportation sector. Automotive
25   LIB production rose from around 40 GWh in 2015 to 160 GWh in 2020 (32%). The stock of battery
26   electric vehicles (BEVs) grew from around 0.7 million in 2015 to 4.8 million in 2019 (IEA 2020d). The
27   number of publicly accessible vehicle chargers reached 1.3 million in 2020, 30% of which were fast
28   chargers. The average battery size of BEVs reached 67 kWh in 2019 due to consumer preferences and
29   government incentives for long-range vehicles (Agency 2020; IEA 2021b).

30   The energy policy landscape continues to evolve
31   The current energy sector policy landscape consists of policy mixes or policy packages, including
32   regulatory, market-based and other approaches. These mixes have evolved over time and include many
33   sectoral but also some economy-wide policy instruments, such as carbon pricing, subsidies.
34   Governments have chosen a mix of policies and institutional mechanisms that consists of regulatory
35   instruments, like efficiency and technology standards, economic instruments (e.g. carbon pricing,
36   subsidies) (Bertram et al. 2015; Martin and Saikawa 2017) and other policies, such as government
37   interventions to provide infrastructure, information policies, and voluntary actions by non-government
38   actors (Somanathan et al. 2014). In recent years, regulatory instruments to promote low-carbon
39   infrastructure have gained traction in developing countries (Finon 2019). The choice of policies has
40   depended on institutional capacities, technological maturity and other developmental priorities of
41   governments. For example, governments have favoured regulatory instruments over economic
42   instruments when there has been sufficient institutional capacity to implement and monitor the
43   regulations and standards (Hughes and Urpelainen 2015). Furthermore, institutional capacity has also
44   determined the extent of implemented measures (Adenle et al. 2017). Market conditions and
45   technological maturity are other important determinants of policy mixes being deployed in the energy
     Do Not Cite, Quote or Distribute                  6-19                                 Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    sector. For example, subsidies for mitigation like feed-in-tariffs (FIT) have worked best when the
2    technologies are in nascent stages of development (Gupta et al. 2019a).
 3   On the other hand, market-based instruments like emission trading schemes (ETS) and auctions coupled
 4   with a regulatory framework have been a favourable strategy for more mature technologies (Polzin et
 5   al. 2015; Kitzing et al. 2018). FIT, tax incentives, and renewable portfolio standards - despite potentially
 6   substantial program costs (Andor and Voss 2016; Abrell et al. 2019) - have played a significant role in
 7   attracting foreign direct investments in the renewable sector (Wall et al. 2019). Subsidies and carbon
 8   pricing have also played an important role in mainstreaming these renewable energy sources (Best and
 9   Burke 2018). Recently, subsidy-free investments in renewables, e.g. wind offshore (Jansen et al. 2020),
10   backed by power purchase agreements, have gained momentum (Frankfurt School-UNEP Centre and
11   BNEF 2020). Similar considerations apply for policy mixes targeted to other sectors, for example
12   transport and buildings.
13   The role of carbon pricing is still limited though increasing. Different measures have been suggested to
14   improve the performance of the ETS, such as “price floors and caps” and other carbon pricing schemes
15   (Bataille et al. 2018; Campiglio 2016; Goulder and Morgenstern 2018). In 2020, 61 regional, national
16   and sub-national carbon pricing instruments, representing 22% of the global GHG emissions, were in
17   action or scheduled for implementation (World Bank 2019). Over 51% of emissions covered are priced
18   at less than USD 10 per tCO2-e.. At present, however, only 5% of the global emissions covered under
19   carbon pricing initiatives are consistent with the range of carbon prices that have been suggested as
20   needed to limit warming to well below 2°C (Stiglitz and Stern 2017). Most of the carbon pricing
21   schemes have taken place in the OECD countries. The limited application of carbon pricing instruments
22   in developing, and emerging economies may be due to political economy constraints (Campiglio 2016;
23   Finon 2019). Carbon pricing had a sizeable impact on emissions , e.g. the EU ETS impacts emissions
24   from electricity in Germany (Schäfer 2019) and manufacturing in France (Colmer et al.), respectively.
25   Emission reductions could be increased with higher carbon prices and without free allocation of
26   allowances.
27   In the absence of a global comprehensive carbon price, regional regulatory policies for fossil fuels
28   supply and key demand sectors like transport, industry and buildings (Chapters 9-11), coupled with
29   regional carbon pricing instruments, were implemented to help initiating the climate actions consistent
30   with the Paris agreement (Kriegler et al. 2018). However, differences in the stringency of climate
31   regulation has triggered fear that regulation reduces the competitiveness of industries in regulated
32   countries and lead to industry re-location and “carbon leakage” (Schenker et al. 2018). In recent years,
33   however, there is little evidence of carbon leakage (Schäfer 2019; Naegele and Zaklan 2019), and even
34   positive effects of carbon pricing on efficiency have been observed (e.g., Löschel et al. 2019, for
35   German manufacturing firms, and (Germeshausen 2020) for German power plants). However, with
36   asymmetric rising carbon prices, discussions about specific policy mechanisms to address carbon
37   leakage like carbon border adjustments (Cosbey et al. 2019) were amplified. Furthermore, multiple
38   policies - often implemented by different governmental levels (national vs subnational) - interacted with
39   each other and thereby affected their environmental and economic effectiveness. Recent examples
40   include interactions of ETS with renewable support policies (e.g. Boehringer and Behrens 2015; Del
41   Rio 2017), energy efficiency policies (e.g. Wiese et al. 2018) or electricity market reform (e.g. Teng et
42   al. 2017), respectively.
43   Apart from explicit carbon pricing, various implicit carbon pricing mechanisms such as fossil fuel taxes
44   and removal of fossil fuel subsidies (see Box 6.3) and regulatory instruments are used by many countries
45   as part of their climate policies. In addition, public provision and procurement of low-carbon
46   infrastructure and technologies such as energy-efficient devices, renewable energy, and upgrades in
47   electricity grids through state-sponsored institutions and public-private partnerships have played an
48   important role in low-carbon development (e.g., (Baron 2016)).

     Do Not Cite, Quote or Distribute                   6-20                                   Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    [START BOX 6.3 HERE]

2                                           Box 6.3 Energy Subsidies

3    Energy subsidies continue to be widely applied. Global fossil fuel subsidies represent more than half of
4    total energy subsidies with predominantly adverse environmental, economic, and social effects (high
5    confidence).
 6   Energy subsidies can be defined as policy measures in the energy sector to lower the prices for
 7   consumers, raise the prices for producers, or reduce energy production costs (IEA 1999). There are
 8   subsidies for fossil fuels, renewables, and energy efficiency measures. The majority of the renewable
 9   subsidies are generation-based incentives for solar, wind or biomass in the form of feed-in-tariffs (FIT)
10   (Chapter 13), with total annual renewable subsidy estimates of about USD 150 billion yr-1 globally
11   (IEA 2018b). Estimates of fossil fuel subsidies can vary by an order of magnitude. For the year 2017,
12   the IEA estimated fossil fuel subsidies of USD 300 billion using IEA’s pre-tax, price-gap method (IEA
13   2018b), while the IMF included unpriced externalities in calculating subsidies of USD 5.2 trillion or
14   6.5% of global GDP (World Bank 2019; Coady et al. 2017, 2019). It has been estimated that the amount
15   spent on fossil fuel subsidies was around double the amount of subsidies spent on renewables (IEA
16   2018b). There are adverse environmental, economic and social consequences of fossil fuel subsidies
17   (Rentschler and Bazilian 2017). More than 75% of the distortions created by fossil fuel subsidies are
18   domestic, and studies indicate that reforming them can have substantial in-country benefits (Coady et
19   al. 2019, 2017). Some of the G-20 countries have implemented subsidy reforms based on low oil prices
20   (Jewell et al. 2018).
21   Fossil fuel subsidies most commonly pursue non-climate objectives, for example, enhanced access to
22   energy sources (high confidence). In some cases, these energy access subsidies have helped extend
23   modern energy sources to the poor (Kimemia and Annegarn 2016) and thereby contribute to SDG-7.
24   However, the subsidies have proven to be regressive in most cases, with little benefit reaching the poor
25   (Lockwood 2015). For example, Indonesia has introduced liquefied petroleum gas (LPG) subsidies for
26   cooking. The kerosene to LPG conversion program (“Zero Kero”) was launched in 2007 and provided
27   mainly households with free initial LPG equipment and LPG at a low subsidized price (Thoday et al.
28   2018; Imelda et al. 2018b). Besides the national government, provincial governments and industry
29   played a crucial role in implementation. Overall, the LPG conversion program in Indonesia reduced
30   cooking kerosene use (Andadari et al. 2014; Imelda et al. 2018b) and GHG emissions (Permadi et al.
31   2017) with positive health effects (Thoday et al. 2018; Imelda et al. 2018b). However, the program is
32   generally viewed as regressive and has failed to reduce traditional solid fuel use (Andadari et al. 2014;
33   Toft 2016; Thoday et al. 2018). Furthermore, even if the program decreased greenhouse gas emissions
34   relative to continued kerosene use, these subsidies are still targeted at fossil fuels and contribute to GHG
35   emissions.
36   India started a large LPG program in 2015 that provided a capital cost subsidy to poor households (e.g.
37   Kar et al. 2019; Jose et al. 2018; Gould 2018). While the program has increased adoption of LPG in
38   India (e.g. Sharma et al. 2019), it has not yet achieved a sustained use of LPG and replacement of solid
39   fuels for cooking, amplifying the need for complementary policy measures ( Gould 2018; Kar et al.
40   2019; Mani et al. 2020). The climate impacts of switching from biomass to LPG depend on the degree
41   of biomass combustion in stoves and the extent to which biomass originates from non-renewable
42   sources (Jose et al. 2018; Singh and Rao 2015). Barriers to increasing LPG use for cooking further
43   included abundance of solid fuels at zero (monetary) costs (Mani et al. 2020) as well as benefits of solid
44   fuels, such as maintaining the traditional taste of food and space heating in colder seasons (Gould 2018;
45   Sharma et al. 2020).
46   [END BOX 6.3 HERE]

     Do Not Cite, Quote or Distribute                   6-21                                   Total pages: 217
     Final Government Distribution                      Chapter 6                                  IPCC WGIII AR6


3    6.4 Mitigation Options
4    6.4.1 Elements of Characterization
 5   This section characterizes energy system mitigation options and discusses which factors enable and
 6   inhibit their implementation. We touch on a broad range of factors that may enable and inhibit the
 7   implementation of mitigation options by considering six dimensions that affect their feasibility (Table
 8   6.1, Annex II.11). The assessment aims to identify which mitigation options can be readily
 9   implemented and which face barriers that would need to be overcome before they can be deployed at
10   scale.
11      Table 6.1 Dimensions and indicators to assess the barriers and enablers of implementing mitigation
12                                   options in low carbon energy systems.

      Metric                           Indicators
      Geophysical: Are the             Physical potential: physical constraints to implementation
      required resources available?    Geophysical resources (including geological storage capacity): availability of
                                       resources needed for implementation
                                       Land use: claims on land where an option would be implemented
      Environmental-ecological:        Air pollution: increase or decrease in air pollutants, such as NH 4, CH4 and
      What are the wider               fine dust
      environmental and ecological     Toxic waste, mining, ecotoxicity and eutrophication
      impacts of the option?           Water quantity and quality: changes in the amount of water available for
                                       other uses
                                       Biodiversity: changes in conserved primary forest or grassland that affect
                                       biodiversity, and management to conserve and maintain land carbon stocks
      Technological: Can the           Simplicity: is the option technically simple to operate, maintain and integrate
      required option be upscaled      Technology scalability: can the option be scaled up, technically
      soon?                            Maturity and technology readiness: R&D and time needed to implement the
      Economic: What economic          Costs in 2030 and in the long term: investment costs, costs in USD tCO 2-eq-1
      conditions can support or        Employment effects and economic growth: decrease or increase in jobs and
      inhibit the implementation of    economic welfare
      the option?
      Socio-cultural: What social      Public acceptance: the extent to which the public supports the option and
      conditions could support or      will change their behaviour accordingly
      inhibit acceptance, adoption,    Effects on health and wellbeing
      and use of the option?           Distributional effects: equity and justice across groups, regions, and
                                       generations, including energy, water, and food security and poverty
      Institutional: What              Political acceptance: the extent to which politicians support the option
      institutional conditions could   Institutional capacity and governance, cross-sectoral coordination: capability
      support or inhibit the           of institutions to implement and handle the option
      implementation of the            Legal and administrative capacity
14   6.4.2 Energy Sources and Energy Conversion
15   Solar Energy
16   Solar PV is increasingly competitive with other forms of electricity generation and is the low-cost option
17   in many applications (high confidence). Costs have declined by 62% since 2015 (high confidence) and

     Do Not Cite, Quote or Distribute                      6-22                                     Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

1    are anticipated to decline by an additional 16% by 2030 if current trends continue (low confidence,
2    medium evidence). Key areas for continued improvement are grid integration and non-module costs for
3    rooftop systems (high confidence). Most deployment is now utility-scale (high confidence). Global
4    future potential is not limited by solar irradiation, but by grid integration needed to address its
5    variability, as well as access to finance, particularly in developing countries (high confidence).
 6   The global technical potential of direct solar energy far exceeds that of any other renewable energy
 7   resource and is well beyond the total amount of energy needed to support ambitious mitigation over the
 8   current century (high confidence). Estimates of global solar resources have not changed since AR5
 9   (Lewis 2007; Besharat et al. 2013) even as precision and near-term forecasting have improved (Abreu
10   et al. 2018; Diagne et al. 2013). Approximately 120,000 TW of sunlight reaches the Earth’s surface
11   continuously, almost 10,000 times average world energy consumption; factoring in competition for
12   land-use leaves a technical potential of about 300 PWh yr-1 (1080 EJ yr-1) for solar PV, roughly double
13   current consumption (Dupont et al. 2020). The technical potential for concentrating solar power (CSP)
14   is estimated to be 45–82 PWh yr-1 (162–295 EJ yr-1) (Dupont et al. 2020). Areas with the highest solar
15   irradiation are western South America; northern, eastern and southwestern Africa; and the Middle East
16   and Australia (Figure 6.7) (Prăvălie et al. 2019).
17   In many parts of the world, the cost of electricity from PV is below the cost of electricity generated
18   from fossil fuels, and in some, it is below the operating costs of electricity generated from fossil fuels
19   (high confidence). The weighted average cost of PV in 2019 was USD 68 MWh-1, near the bottom of
20   the range of fossil fuel prices (IRENA 2019b). The cost of electricity from PV has fallen by 89% since
21   2000 and 69% since AR5, at a rate of –16% per year. The 5:95 percentile range for PV in 2019 was
22   USD 52–190 MWh-1 (IRENA 2021b). Differences in solar insolation, financing costs, equipment
23   acquisition, installation labour, and other sources of price dispersion explain this range (Nemet et al.
24   2016; Vartiainen et al. 2020) and scale. For example, in India, rooftop installations cost 41% more than
25   utility-scale installations, and commercial-scale costs are 39% higher than utility-scale. Significant
26   differences in regional cost persist (Kazhamiaka et al. 2017; Vartiainen et al. 2020), with particularly
27   low prices in China, India, and parts of Europe. Globally, the range of global PV costs is quite similar
28   to the range of coal and natural gas prices.

30         Figure 6.7 Distribution of the daily mean global horizontal irradiation (GHI, kWh m-2 day-1).
31   Source: Global Solar Atlas (ESMAP 2019)
32   PV costs (Figure 6.8) have fallen for various reasons: lower silicon costs, automation, lower margins,
33   automation, higher efficiency, and a variety of incremental improvements (Fu et al. 2018; Green 2019)
34   (Chapter 16). Increasingly, the costs of PV electricity are concentrated in the installation and related
35   “soft costs” (marketing, permitting) associated with the technology rather than in the modules
36   themselves, which now account for only 30% of installed costs of rooftop systems (O’Shaughnessy et

     Do Not Cite, Quote or Distribute                  6-23                                  Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    al. 2019; IRENA 2021b). Financing costs are a significant barrier in developing countries (Ondraczek
2    et al. 2015) and growth there depends on access to low-cost finance (Creutzig et al. 2017).
3    CSP costs have also fallen, albeit at about half the rate of PV: –9% yr-1 since AR5. The lowest prices
4    for CSP are now competitive with more expensive fossil fuels, although the average CSP cost is above
5    the fossil range. Other data sources put recent CSP costs at USD 120 MWh-1, in the middle of the fossil
6    range (Lilliestam et al. 2020). Continuing the pace of change since AR5 will make CSP competitive
7    with fossil fuels in sunny locations, although it will be difficult for CSP to compete with PV and even
8    hybrid PV-battery systems. CSP electricity can be more valuable, however, because CSP systems can
9    store heat longer than PV battery systems.

12     Figure 6.8 Levelized costs of electricity (LCOE) of solar energy technologies 2000–2020. Range of fossil
13    fuel LCOE indicated as dashed lines USD 50–177 MWh-1. Linear fit lines were applied to data for AR4–
14    AR5 and post-AR5 (2012). Red dots are capacity-weighted global averages for utility-scale installations.
15    The blue area shows the range between the 5th and 95th percentile in each year. Data: ( IRENA 2021b).

16   The share of total costs of PV-intensive electricity systems attributed to integration costs has been
17   increasing but can be reduced by enhancing grid flexibility (Section 6.4.3, 6.6, Box 6.8) (high
18   confidence). The total costs of PV include grid integration, which varies tremendously depending on
19   PV's share of electricity, other supply sources like wind, availability of storage, transmission capacity,
20   and demand flexibility (Heptonstall and Gross 2020). Transmission costs can add USD 1–10 MWh-1 or
21   3–33% to the cost of utility-scale PV (Gorman et al. 2019). Distributed (rooftop) PV involves a broader
22   set of grid integration costs - including grid reinforcement, voltage balancing and control, and impacts
23   on other generations - and has a larger range of integration costs from USD 2–25 MWh-1, which is –3%
24   to +37% (Hirth et al. 2015; Wu et al. 2015; Gorman et al. 2019). Other meta-analyses put the range at
25   USD 1–7 MWh-1 in the USA (Luckow et al.; Wiser et al. 2017), while a comprehensive study put the
26   range at USD 12–18 MWh-1 for up to 35% renewables and USD 25–46 MWh-1 above 35% renewables
27   (Heptonstall and Gross 2020). Increased system flexibility can reduce integration costs of solar energy
28   (Wu et al. 2015) including storage, demand response, sector-coupling (Bogdanov et al. 2019; Brown et
29   al. 2018), and increase complementarity between wind and solar (Heide et al. 2010) (Sections 6.4.3 and
30   6.4.4).
31   Since solar PV panels have very low-operating costs, they can, at high penetrations and in the absence
32   of adequate incentives to shift demand, depress prices in wholesale electricity markets, making it
33   difficult to recoup investment and potentially reducing incentives for new installations (Hirth 2013;
34   Millstein et al. 2021). Continued cost reductions help address this issue of value deflation, but only
35   partially. Comprehensive solutions depend on adding transmission and storage (Das et al. 2020) and,
36   more fundamentally, adjustments to electricity market design (Bistline and Young 2019; Roques and
37   Finon 2017).

     Do Not Cite, Quote or Distribute                   6-24                                  Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

 1   The most important ways to minimize PVs impact on the environment lie in recycling materials at end
 2   of life and making smart land-use decisions (medium confidence). A comprehensive assessment of PVs
 3   environmental impacts requires life-cycle analysis (LCA) of resource depletion, land-use, ecotoxicity,
 4   eutrophication, acidification, ozone, and particulates, among other things (Mahmud et al. 2018). LCA
 5   studies show that solar PVs produce far less CO2 per unit of electricity than fossil generation, but PV
 6   CO2 emissions vary due to the carbon intensity of manufacturing energy and offset electricity (Grant
 7   and Hicks 2020). Concerns about systemic impacts, such as reducing the Earth’s albedo by covering
 8   surfaces with dark panels, have shown to be trivial compared to the mitigation benefits (Nemet 2009)
 9   (Box 6.7). Even though GHG LCA estimates span a considerable range of 9–250 gCO2 kWh-1 (de Wild-
10   Scholten 2013; Kommalapati et al. 2017), recent studies that reflect higher efficiencies and
11   manufacturing improvements find lower life-cycle emissions, including a range of 18–60 gCO2 kWh-1
12   (Wetzel and Borchers 2015) and central estimates of 80 gCO2 kWh-1 (Hou et al. 2016), 50 gCO2 kWh-
13     (Nugent and Sovacool 2014), and 20 gCO2 kWh-1 (Louwen et al. 2016). These recent values are an
14   order of magnitude lower than coal and natural gas and further decarbonization of the energy system
15   will make them lower still. Thin films and organics produce half the life-cycle emissions of silicon
16   wafer PV, mainly because they use less material (Lizin et al. 2013; Hou et al. 2016). Novel materials
17   promise even lower environmental impacts, especially with improvements to their performance ratios
18   and reliability (Gong et al. 2015; Muteri et al. 2020). Higher efficiencies, longer lifetimes, sunny
19   locations, less carbon intensive manufacturing inputs, and shifting to thin films could reduce future life-
20   cycle impacts.
21   Another environmental concern with large PV power plants is the conversion of land to collect solar
22   energy (Hernandez et al. 2015). Approximately 2 hectares of land are needed for 1 MW of solar
23   electricity capacity (Kabir et al. 2018; Perpiña Castillo et al. 2016); at 20% efficiency, a square of PV
24   panels of 550 km by 550 km, comprising 0.2% of Earth’s land area, could meet global energy demand.
25   Land conversion can have local impacts, especially near cities and where land used for solar competes
26   with alternative uses, such as agriculture. Large installations can also adversely impact biodiversity
27   (Hernandez et al. 2014), especially where the above ground vegetation is cleared and soils are typically
28   graded. Landscape fragmentation creates barriers to the movement of species. However, a variety of
29   means have emerged to mitigate land use issues. Substitution among renewables can reduce land
30   conversion (Tröndle and Trondle 2020). Solar can be integrated with other uses through ‘agrivoltaics’
31   (the use of land for both agriculture and solar production) (Dupraz et al. 2011) by, for example, using
32   shade-tolerant crops (Dinesh and Pearce 2016). Combining solar and agriculture can also create income
33   diversification, reduced drought stress, higher solar output due to radiative cooling, and other benefits
34   (Elamri et al. 2018; Hassanpour Adeh et al. 2018; Barron-Gafford et al. 2019). PV installations floating
35   on water also avoid land-use conflicts (Sahu et al. 2016; Lee et al. 2020), as does dual-use infrastructure,
36   such as landfills (Jäger-Waldau 2020) and reservoirs where evaporation can also be reduced (Farfan
37   and Breyer 2018).
38   Material demand for PV will likely increase substantially to limit warming to well below 2°C, but PV
39   materials are widely available, have possible substitutes, and can be recycled (medium confidence) (Box
40   6.4). The primary materials for PV are silicon, copper, glass, aluminium, and silver, the costliest being
41   silicon and glass being the most essential by mass, at 70%. None of these materials is considered to be
42   either critical or potentially scarce (IEA 2020e). Thin-film cells, such as amorphous silicon, cadmium
43   telluride and copper indium gallium diselenide (CIGS), use far less material (though they use more
44   glass), but account for less than 10% of the global solar market. Other thin-films, such as those based
45   on perovskites, organic solar cells, or earth-abundant, non-toxic materials such as kesterites, either on
46   their own, or layered on silicon, could further reduce material use per energy produced. (Box 6.4)
47   After a typical lifetime of 30 years of use, PV modules can be recycled to prevent environmental
48   contamination from the toxic materials within them, reusing valuable materials and avoiding waste

     Do Not Cite, Quote or Distribute                   6-25                                   Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

 1   accumulation. Recycling allows the reuse of nearly all - 83% in one study - of the components of PV
 2   modules, other than plastics (Ardente et al. 2019) and would add less than 1% to lifecycle GHG
 3   emissions (Latunussa et al. 2016). Glass accounts for 70% of the mass of a solar cell and is relatively
 4   easy to recycle. Recycling technology is advancing, but the scale and share of recycling is still small
 5   (Li et al. 2020d). By 2050, however, end-of-life PV could total 80 MT and comprise 10% of global
 6   electronic waste (Stolz and Frischknecht 2017), although most of it is glass. IEA runs a program to
 7   enable PV recycling by sharing best practices to minimise recycling life cycle impacts. Ensuring that a
 8   substantial amount of panels are recycled at end of life will likely require policy incentives, as the
 9   market value of the recovered materials, aside from aluminium and copper, is likely to be too low to
10   justify recycling on its own (Deng et al. 2019). A near-term priority is maximizing the recovery of
11   silver, silicon, and aluminium, the most valuable PV material components (Heath et al. 2020).
12   Many alternative PV materials are improving in efficiency and stability, providing longer-term
13   pathways for continued PV costs reductions and better performance (high confidence). While solar PV
14   based on semi-conductors constructed from wafers of silicon still captures 90% of the market, new
15   designs and materials have the potential to reduce costs further, increase efficiency, reduce resource
16   use, and open new applications. The most significant technological advance within silicon PV in the
17   past ten years has been the widespread adoption of the passivated emitter and rear cell (PERC) design
18   (Green 2015), which now accounts for the majority of production. This advance boosts efficiency over
19   traditional aluminium backing by increasing reflectivity within the cell and reducing electron hole
20   recombination (Blakers 2019). Bifacial modules increase efficiency by using reflected light from the
21   ground or roof on the backside of modules (Guerrero-Lemus et al. 2016). Integrating PV into buildings
22   can reduce overall costs and improve building energy performance (Shukla et al. 2016). Concentrating
23   PV uses lenses or mirrors that collect and concentrate light onto high efficiency PV cells (Li et al.
24   2020a). Beyond crystalline silicon, thin films of amorphous silicon, cadmium telluride, and copper
25   indium gallium selenide (among others) have the potential for much lower costs while their efficiencies
26   have increased (Green et al. 2019). Perovskites, inexpensive and easy to produce crystalline structures,
27   have increased in efficiency by a factor of six in the past decade; the biggest challenge is light-induced
28   degradation as well as finding lead-free efficient compounds or establish lead recycling at the end of
29   the life cycle of the device (Petrus et al. 2017; Chang et al. 2018; Wang et al. 2019b; Zhu et al. 2020).
30   Organic solar cells are made of carbon-based semiconductors like the ones found in the displays made
31   from organic light emitting diodes (OLEDs) and can be processed in thin films on large areas with
32   scalable and fast coating processes on plastic substrates. The main challenges are raising the efficiency
33   and improving their lifetime (Ma et al. 2020; Riede et al. 2021). Quantum dots, spherical semi-
34   conductor nano-crystals, can be tuned to absorb specific wavelengths of sunlight giving them the
35   potential for high efficiency with very little material use (Kramer et al. 2015). A common challenge for
36   all emerging solar cell technologies is developing the corresponding production equipment. Hybrids of
37   silicon with layers of quantum dots and perovskites have the potential to take advantage of the benefits
38   of all three, although those designs require that these new technologies have stability and scale that
39   match those of silicon (Palmstrom et al. 2019; Chang et al. 2017). This broad array of alternatives to
40   making PV from crystalline silicon offer realistic potential for lower costs, reduced material use, and
41   higher efficiencies in future years (Victoria et al. 2021).
42   Besides PV, alternative solar technologies exist, including CSP, which can provide special services in
43   high-temperature heat and diurnal storage, even if it is more costly than PV and its potential for
44   deployment is limited. CSP uses reflective surfaces, such as parabolic mirrors, to focus sunlight on a
45   receiver to heat a working fluid, which is subsequently transformed into electricity (Islam et al. 2018).
46   Solar heating and cooling are also well established technologies, and solar energy can be utilized
47   directly for domestic or commercial applications such as drying, heating, cooling, and cooking (Ge et
48   al. 2018). Solar chimneys, still purely conceptual, heat air using large transparent greenhouse-like
49   structures and channel the warm air to turbines in tall chimneys (Kasaeian et al. 2017). Solar energy
     Do Not Cite, Quote or Distribute                  6-26                                  Total pages: 217
     Final Government Distribution                    Chapter 6                                IPCC WGIII AR6

1    can also be used to produce solar fuels, for example, hydrogen or synthetic gas (syngas) (Nocera 2017;
2    Montoya et al. 2016; Detz et al. 2018). In addition, research proceeds on space-based solar PV, which
3    takes advantage of high insolation and a continuous solar resource (Kelzenberg et al. 2018), but faces
4    the formidable obstacle of developing safe, efficient, and inexpensive microwave or laser transmission
5    to the Earth’s surface (Yang et al. 2016). CSP is the most widely adopted of these alternative solar
6    technologies.
 7   Like PV, CSP facilities can deliver large amounts of power (up to 200 MW per unit) and maintain
 8   substantial thermal storage, which is valuable for load balancing over the diurnal cycle (McPherson et
 9   al. 2020). However, unlike PV, CSP can only use direct sunlight, constraining its cost-effectiveness to
10   North Africa, the Middle East, Southern Africa, Australia, the Western U.S., parts of South America
11   (Peru, Chile), the Western part of China, and Australia (Deng et al. 2015; Dupont et al. 2020). Parabolic
12   troughs, central towers and parabolic dishes are the three leading solar thermal technologies (Wang et
13   al. 2017d). Parabolic troughs represented approximately 70% of new capacity in 2018 with the balance
14   made up by central tower plants (Islam et al. 2018). Especially promising research directions are on
15   tower-based designs that can achieve high temperatures, useful for industrial heat and energy storage
16   (Mehos et al. 2017), and direct steam generation designs (Islam et al. 2018). Costs of CSP have fallen
17   by nearly half since AR5 (Figure 6.8) albeit at a slower rate than PV. Since AR5, almost all new CSP
18   plants have storage (Figure 6.9)(Thonig 2020).

21   Figure 6.9 CSP plants by storage capacity in hours (vertical), year of installation (horizontal), and size of
22      plant in MW (circle size). Since AR5, almost all new CSP plants have storage (Thonig 2020). Data
23                                    source:

24   Solar energy elicits favourable public responses in most countries (high confidence) (Bessette and Arvai
25   2018; Hanger et al. 2016; Jobin and Siegrist 2018; Ma et al. 2015; Mcgowan and Sauter 2005; Hazboun
26   and Boudet 2020; Roddis et al. 2019). Solar energy is perceived as clean and environmentally friendly
27   with few downsides (Faiers and Neame 2006; Whitmarsh et al. 2011b). Key motivations for
28   homeowners to adopt photovoltaic systems are expected financial gains, environmental benefits, the
29   desire to become more self-sufficient, and peer expectations (Korcaj et al. 2015; Palm 2017; Vasseur
30   and Kemp 2015). Hence, the observability of photovoltaic systems can facilitate adoption (Boudet
31   2019). The main barriers to the adoption of solar PV by households are its high upfront costs, aesthetics,
32   landlord-tenant incentives, and concerns about performance and reliability (Whitmarsh et al. 2011b;
33   Vasseur and Kemp 2015; Faiers and Neame 2006).

     Do Not Cite, Quote or Distribute                    6-27                                   Total pages: 217
     Final Government Distribution                   Chapter 6                              IPCC WGIII AR6

1   Wind Energy
2    Wind power is increasingly competitive with other forms of electricity generation and is the low-cost
3    option in many applications (high confidence). Costs have declined by 18% and 40% on land and
4    offshore since 2015 (high confidence), and further reductions can be expected by 2030 (medium
5    confidence). Critical areas for continued improvement are technology advancements and economies of
6    scale (high confidence). Global future potential is primarily limited by onshore land availability in wind
7    power-rich areas, lack of supporting infrastructure, grid integration, and access to finance (especially in
8    developing countries) (high confidence).
 9   Energy from wind is abundant, and the estimated technical potentials surpass the total amount of energy
10   needed to limit warming to well below 2°C (high confidence). Recent global estimates of potentially
11   exploitable wind energy resource are in the range of 557–717 PWh yr-1 (2005–2580 EJ yr-1) (Eurek et
12   al. 2017; Bosch et al. 2017, 2018; McKenna et al. 2022), or 20–30 times the 2017 global electricity
13   demand. Studies have suggested that ‘bottom-up’ approaches may overestimate technical potentials
14   (Miller et al. 2015; Kleidon and Miller 2020). But even in the most conservative ‘top-down’ approaches,
15   the technical wind potential surpasses the amount needed to limit warming to well below 2°C (Bosch
16   et al. 2017; Eurek et al. 2017; Volker et al. 2017). The projected climate change mitigation from wind
17   energy by 2100 ranges from 0.3°C–0.8°C depending on the precise socio-economic pathway and wind
18   energy expansion scenario followed (Barthelmie and Pryor 2021). Wind resources are unevenly
19   distributed over the globe and by time of the year (Petersen and Troen 2012), but potential hotspots
20   exist on every continent (Figure 6.10) as expressed by the wind power density (a quantitative measure
21   of wind energy available at any location). Technical potentials for onshore wind power vary
22   considerably, often because of inconsistent assessments of suitability factors (McKenna et al. 2020).
23   The potential for offshore wind power is larger than for onshore because offshore wind is stronger and
24   less variable (Bosch et al. 2018). Offshore wind is more expensive, however, because of higher costs
25   for construction, maintenance, and transmission. Wind power varies at a range of time scales, from
26   annual to sub-seconds; the effects of local short-term variability can be offset by power plant control,
27   flexible grid integration, and storage (Barra et al. 2021) (section 6.4.3). In some regions, interannual
28   variations in wind energy resources could be important for optimal power system design (Wohland et
29   al. 2019a; Coker et al. 2020).

32   Figure 6.10 Mean wind power density [W m-2] at 100 m above ground level over land and within 100 km
33                  of the coastline. Source: Global Wind Atlas

34   Wind power cost reductions (Figure 6.11) are driven mainly by larger capacity turbines, larger rotor
35   diameters and taller hub heights - larger swept areas increase the energy captured and the capacity

     Do Not Cite, Quote or Distribute                   6-28                                  Total pages: 217
     Final Government Distribution                   Chapter 6                              IPCC WGIII AR6

 1   factors for a given wind speed; taller towers provide access to higher wind speeds (Beiter et al. 2021).
 2   All major onshore wind markets have experienced rapid growth in both rotor diameter (from 81.2 m in
 3   2010 to 120 m in 2020) (IRENA 2021b), and average power ratings (from 1.9 MW in 2010 to 3 MW
 4   in 2020). The generation capacity of offshore wind turbines grew by a factor of 3.7 in less than two
 5   decades, from 1.6 MW in 2000 to 6 MW in 2020 (Wiser et al. 2021). Floating foundations could
 6   revolutionize offshore wind power by tapping into the abundant wind potential in deeper waters. This
 7   technology is particularly important for regions where coastal waters are too deep for fixed-bottom
 8   wind turbines. Floating wind farms potentially offer economic and environmental benefits compared
 9   with fixed-bottom designs due to less-invasive activity on the seabed during installation, but the long-
10   term ecological effects are unknown and meteorological conditions further offshore and in deeper
11   waters are harsher on wind turbine components (IRENA 2019c). A radical new class of wind energy
12   converters has also been conceived under the name of Airborne Wind Energy Systems that can harvest
13   strong, high-altitude winds (typically between 200–800m), which are inaccessible by traditional wind
14   turbines (Cherubini et al. 2015). This technology has seen development and testing of small devices
15   (Watson et al. 2019).
16   Wind capacity factors have increased over the last decade (Figure 6.11). The capacity factor for onshore
17   wind farms increased from 27% in 2010 to 36% in 2020 (IRENA 2021a). The global average offshore
18   capacity factor has decreased from a peak of 45% in 2017. This has been driven by the increased share
19   of offshore development in China, where projects are often near-shore and use smaller wind turbines
20   than in Europe (IRENA 2021b). Improvements in capacity factors also come from increased
21   functionality of wind turbines and wind farms. Manufactures can adapt the wind turbine generator to
22   the wind conditions. Turbines for windy sites have smaller generators and smaller specific capacity per
23   rotor area, and therefore operate more efficiently and reach full capacity for a longer time period (Rohrig
24   et al. 2019).
25   Electricity from onshore wind is less expensive than electricity generated from fossil fuels in a growing
26   number of markets (high confidence). The global average LCOE onshore declined by 38% from 2010
27   to 2020 (Figure 6.11), reaching USD 0.039 kWh-1. However, the decrease in cost varies substantially
28   by region. Since 2014, wind costs have declined more rapidly than the majority of experts predicted
29   (Wiser et al. 2021). New modelling projects onshore wind LCOE of USD .037 kWh-1 by 2030
30   (Junginger et al. 2020a), and additional reductions of 37–39% have been predicted by 2050 (Wiser et
31   al. 2021). The future cost of offshore wind is more uncertain because other aspects besides increases in
32   capacity factors influence the cost (Junginger et al. 2020b).

     Do Not Cite, Quote or Distribute                   6-29                                  Total pages: 217
     Final Government Distribution                   Chapter 6                              IPCC WGIII AR6


3     Figure 6.11 Global weighted average total installed costs, capacity factors, and LCOE for onshore (top)
4        and offshore (bottom) wind power of existing power plants per year (2010-2020). The shaded area
5       represents the 5th and 95th percentiles and the red dashed line represents the fossil fuel cost range.
6                                             Source: (IRENA 2021a)

 7   The cost of the turbine (including the towers) makes up the largest component of wind’s LCOE. Total
 8   installed costs for both onshore and offshore wind farms have decreased since 2015 (Figure 6.11), but
 9   the total installed costs for onshore wind projects are very site- and market-specific, as reflected in the
10   range of LCOEs. China, India, and the U.S. have experienced the largest declines in total installed costs.
11   In 2020, typical country-average total installed costs were around USD 1150 kW-1 in China and India,
12   and between USD 1403–2472 kW-1 elsewhere (IRENA 2021b). Total installed costs of offshore wind
13   farms declined by 12% between 2010 and 2020. But, because some of the new offshore wind projects
14   have moved to deeper waters and further offshore, there are considerable year-to-year variations in their
15   price (IRENA 2021b). Projects outside China in recent years have typically been built in deeper waters
16   (10–55 m) and up to 120 km offshore, compared to around 10 m in 2001–2006, when distances rarely
17   exceeded 20 km. With the shift to deeper waters and sites further from ports, the total installed costs of
18   offshore wind farms rose, from an average of around USD 2500 kW-1 in 2000 to around USD 5127 kW-
19     by 2011–2014, before falling to around USD 3185 kW-1 in 2020 (IRENA 2020a). The full cost of
20   wind power includes the transmission and system integration costs (Sections 6.4.3, 6.4.6. A new
21   technology in development is the co-location of wind and solar PV power farms, also known as hybrid
22   power plants. Co-locating wind, solar PV, and batteries can lead to synergies in electricity generation,
23   infrastructure, and land usage, which may lower the overall plant cost compared to single technology
24   systems (Lindberg et al. 2021).
25   Wind power plants pose relatively low environmental impact, but sometimes locally significant
26   ecological effects (high confidence). The environmental impact of wind technologies, including CO2
27   emissions, is concentrated in the manufacturing, transport, and building stage and in disposal as the
28   end-of-life of wind turbines is reached (Liu and Barlow 2017; Mishnaevsky 2021). The operation of
29   wind turbines produces no waste or pollutants. The LCA for wind turbines is strongly influenced by the
30   operating lifetime, quality of wind resources, conversion efficiency, and size of the wind turbines

     Do Not Cite, Quote or Distribute                   6-30                                  Total pages: 217
     Final Government Distribution                 Chapter 6                             IPCC WGIII AR6

1    (Laurent et al. 2018; Kaldellis and Apostolou 2017). But, all wind power technologies repay their
2    carbon footprint in less than a year (Bonou et al. 2016).
 3   Wind farms can cause local ecological impacts, including impacts on animal habitat and movements,
 4   biological concerns, bird and bat fatalities from collisions with rotating blades, and health concerns
 5   (Morrison and Sinclair 2004). The impacts on animal habitats and collisions can be resolved or reduced
 6   by selectively stopping some wind turbines in high risk locations, often without affecting the
 7   productivity of the wind farm (de Lucas et al. 2012). Many countries now require environmental studies
 8   of impacts of wind turbines on wildlife prior to project development, and, in some regions, shutdowns
 9   are required during active bird migration (de Lucas et al. 2012). Offshore wind farms can also impact
10   migratory birds and other sea species (Hooper et al. 2017). Floating foundations pose lower
11   environmental impacts at build stage (IRENA 2019c), but their cumulative long-term impacts are
12   unclear (Goodale and Milman 2016). Recent studies find weak associations between wind farm noise
13   and measures of long-term human health (Poulsen et al. 2018a,b, 2019a,b).
14   Public support for onshore and particularly offshore wind energy is generally high, although people
15   may oppose specific wind farm projects (high confidence) (e.g., Rand and Hoen 2017; Steg 2018; Bell
16   et al. 2005; Batel and Devine-Wright 2015). People generally believe that wind energy is associated
17   with environmental benefits and that it is relatively cheap. Yet, some people believe wind turbines can
18   cause noise and visual aesthetic pollution, threaten places of symbolic value (Russell et al. 2020;
19   Devine-Wright and Wiersma 2020), and have adverse effects on wildlife (Bates and Firestone 2015),
20   which challenges public acceptability (Rand and Hoen 2017). Support for local wind projects is higher
21   when people believe fair decision-making procedures have been implemented (Aitken 2010a; Dietz and
22   Stern 2008). Evidence is mixed whether distance from wind turbines or financial compensation
23   increases public acceptability of wind turbines (Hoen et al. 2019; Rand and Hoen 2017; Cass et al.
24   2010; Rudolph et al. 2018). Offshore wind farms projects have higher public support, but can also face
25   resistance (Rudolph et al. 2018; Bidwell 2017).
26   Common economic barriers to wind development are high initial cost of capital, long payback periods,
27   and inadequate access to capital. Optimal wind energy expansion is most likely to occur in the presence
28   of a political commitment to establish, maintain, and improve financial support instruments,
29   technological efforts to support a local supply chains, and grid investments integrate VRE electricity
30   (Diógenes et al. 2020).
31   [START BOX 6.4 HERE]
32             Box 6.4 Critical strategic minerals and a low-carbon energy system transition

33   The secure supply of many metals and minerals (e.g., cobalt, copper, lithium, and rare earth elements,
34   REEs) is critical to supporting a low-emissions energy system transition (Sovacool et al. 2020). A low-
35   carbon energy system transition will increase the demand for these minerals to be used in technologies
36   like wind turbines, PV cells, and batteries (World Bank 2020). Reliance on these minerals has raised
37   questions about possible constraints to a low-carbon energy system transition, including supply chain
38   disruptions (Chapter 10.6). Concerns have also been raised about mining for these materials, which
39   frequently results in severe environmental impacts (Sonter et al. 2020), and metal production itself is
40   energy-intensive and difficult to decarbonize (Sovacool et al. 2020).
41   Wind energy depends on two critical REEs - neodymium and dysprosium - used in magnets in high-
42   performance generators (Pavel et al. 2017; Li et al. 2020b). Silicon-wafer-based solar PV, which
43   accounted for 95% of PV production in 2020, does not use REEs but utilizes aluminium, copper, and
44   silver (IEA 2021a). Lithium, nickel, cobalt, and phosphorous are used in batteries. Many critical
45   minerals are used in EVs, including aluminium and copper in manufacturing the necessary EV charging
46   infrastructure, and neodymium in permanent magnet motors.

     Do Not Cite, Quote or Distribute                 6-31                                 Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    These strategic minerals are found in a limited number of countries, and concerns have been raised that
2    geopolitical factors could disrupt the supply chain necessary for a low-carbon energy system transition.
3    However, excluding cobalt and lithium, no single country holds more than a third of the world reserves.
4    The known supply of some strategic minerals is still close to 600 years at current levels of demand (BP
5    2020), but increased demand would cut more quickly into supplies.
 6   There are alternatives to the strategic minerals currently used to support a low-carbon transition. Wind
 7   turbines can be manufactured without permanent magnets to reduce the need for strategic minerals, but
 8   the production costs are higher, and their efficiency is reduced (Månberger and Stenqvist 2018).
 9   Alternatives to silicon, such as thin films, could be used to produce PVs. Thin-films use much less
10   material than silicon-based PV, but they contain other potentially critical metals like tellurium,
11   cadmium, and gallium. Alternatives to lithium-ion batteries, such as sodium-ion batteries, are becoming
12   more practical and feasible (Sovacool et al. 2020).
13   [END BOX 6.4HERE]
14 Hydroelectric Power
15   Hydropower is technically mature, proved worldwide as a primary source of renewable electricity, and
16   may be used to balance electricity supply by providing flexibility and storage. The LCOE of
17   hydropower is lower than the cheapest new fossil fuel-fired option. However, the future mitigation
18   potential of hydropower depends on minimizing environmental and social impacts during the planning
19   stages, reducing the risks of dam failures, and modernising the aging hydropower fleet to increase
20   generation capacity and flexibility (high confidence).
21   Estimates of global gross theoretical available hydropower potential varies from 31–128 PWh yr-1 (112–
22   460 EJ yr-1), exceeding total electricity production in 2018 (Banerjee et al. 2017; IEA 2021d; BP 2020).
23   This potential is distributed over 11.8 million locations (Figure 6.12), but many of the locations cannot
24   be developed for (current) technical, economic, or political reasons. The estimated technical potential
25   of hydropower is 8–30 PWh yr-1 (29–108 EJ yr-1), and its estimated economic potential is 8–15 PWh
26   yr-1 (29–54 EJ yr-1) (van Vliet et al. 2016c; Zhou et al. 2015). Actual hydropower generation in 2019
27   was 4.2 PWh (15.3 EJ), providing about 16% of global electricity and 43% of global electricity from
28   renewables (BP 2020; Killingtveit 2020; IEA 2020f). Asia holds the largest hydropower potential
29   (48%), followed by S. America (19%) (Hoes et al. 2017).

31    Figure 6.12 Global map of gross hydropower potential distribution [GWh yr -1], Data: (Hoes et al. 2017)

32   Hydropower is a mature technology with locally adapted solutions (high confidence) (Zhou et al. 2015;
33   Killingtveit 2020). The peak efficiency of hydroelectric plants is greater than 85%. Hydropower plants
34   without storage or with small storage typically produce a few kWs to 10 MWs (examples of such plants
35   producing higher amounts do exist), and are useful for providing electricity at a scale from households
     Do Not Cite, Quote or Distribute                  6-32                                  Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    to small communities (El Bassam et al. 2013; Towler 2014). However, hydropower plants without or
2    with small storage may be susceptible to climate variability, especially droughts, when the amount of
3    water may not be sufficient to generate electricity (see Section 6.5, Premalatha et al. 2014).
4    Hydropower plants with storage may produce 10 GW, reaching over 100 TWh yr -1 (0.36 EJ yr-1), but
5    generally require large areas. Pumped storage hydropower stores energy by pumping water to higher
6    reservoirs during low-demand periods (Killingtveit 2020). The storage in hydropower systems provides
7    flexibility to compensate for rapid variations in electricity loads and supplies. The regulating
8    characteristics of the storage play an important role in assuring continuity of energy supply from
9    renewable sources (Yang et al. 2018b).
10   Hydropower is one of the lowest-cost electricity technologies (Mukheibir 2013; IRENA 2021b). Its
11   operation and maintenance costs are typically 2–2.5% of the investment costs per kW yr-1 for a lifetime
12   of 40–80 years (Killingtveit 2020). Construction costs are site specific. The total cost for an installed
13   large hydropower project varies from USD 10,600–804,500 kW-1 if the site is located far away from
14   transmission lines, roads, and infrastructure. Investment costs increase for small hydropower plants and
15   may be as high as USD 100,000 kW-1 or more for the installation of plants of less than 1 MW - 20% to
16   80% more than for large hydropower plants (IRENA 2015). During the past 100 years, total installed
17   costs and LCOE have risen by a few percent, but the LCOE of hydropower remains lower than the
18   cheapest new fossil fuel-fired option (IRENA 2019b, 2021).
19   Hydroelectric power plants may pose serious environmental and societal impacts (high confidence)
20   (Mccartney 2009). Dams may lead to fragmentation of ecological habitats because they act as barriers
21   for migration of fish and other land and water-borne fauna, sediments, and water flow. These barriers
22   can be mitigated by sediment passes and fish migration aids, and with provision of environmental flows.
23   Below dams, there can be considerable alterations to vegetation, natural river flows, retention of
24   sediments and nutrients, and water quality and temperature. Construction of large reservoirs leads to
25   loss of land, which may result in social and environmental consequences. Minimizing societal and
26   environmental impacts requires taking into account local physical, environmental, climatological,
27   social, economic, and political aspects during the planning stage (Killingtveit 2020). Moreover, when
28   large areas of land are flooded by dam construction, they generate GHGs (Phyoe and Wang 2019;
29   Maavara et al. 2020; Prairie et al. 2018). On the other hand, hydropower provides flexible, competitive
30   low-emission electricity, local economic benefits (e.g., by increasing irrigation and electricity
31   production in developing countries), and ancillary services such as municipal water supply, irrigation
32   and drought management, navigation and recreation, and flood control (IRENA 2021b). However, the
33   long term economic benefits to communities affected by reservoirs are a subject of debate (de Faria et
34   al. 2017; Catolico et al. 2021).
35   Public support for hydroelectric energy is generally high (Steg 2018), and higher than support for coal,
36   gas, and nuclear. Yet, public support for hydro seems to differ for existing and new projects (high
37   confidence). Public support is generally high for small and medium scale hydropower in regions where
38   hydropower was historically used (Gormally et al. 2014). Additionally, there is high support for existing
39   large hydropower projects in Switzerland (Plum et al. 2019; Rudolf et al. 2014), Canada (Boyd et al.
40   2019), and Norway (Karlstrøm and Ryghaug 2014), where it is a trusted and common energy source.
41   Public support seems lower for new hydropower projects (Hazboun and Boudet 2020), and the
42   construction of new large hydropower plants has been met with strong resistance in some areas
43   (Bronfman et al., 2015; Vince, 2010). People generally perceive hydroelectric energy as clean and a
44   non-contributor to climate change and environmental pollution (Kaldellis et al. 2013). For example, in
45   Sweden, people believed that existing hydropower projects have as few negative environmental impacts
46   as solar, and even less than wind (Ek 2005). However, in areas where the construction of new large-
47   scale hydroelectric energy is met with resistance, people believe that electricity generation from hydro
48   can cause environmental, social, and personal risks (Bronfman et al., 2012; Kaldellis et al., 2013).

     Do Not Cite, Quote or Distribute                  6-33                                 Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

 1   The construction time of hydroelectric power plants is longer than many other renewable technologies,
 2   and that construction time may be extended by the additional time it takes to fill the reservoir. This
 3   extended timeline can create uncertainty in the completion of the project. The uncertainty is due to
 4   insecurity in year-to-year variations in precipitation and the water inflows required to fill reservoirs.
 5   This is especially critical in the case of trans-boundary hydroelectric power plants, where filling up the
 6   reservoirs can have large implications on downstream users in other nations. As a result of social and
 7   environmental constraints, only a small fraction of potential economic hydropower projects can be
 8   developed, especially in developed countries. Many developing countries have major undeveloped
 9   hydropower potential, and there are opportunities to develop hydropower combined with other
10   economic activities such as irrigation (Lacombe et al. 2014). Competition for hydropower across
11   country borders can lead to conflict, which could be exacerbated if climate alters rainfall and streamflow
12   (Ito et al. 2016).
13 Nuclear Energy
14   Nuclear power can deliver low-carbon energy at scale (high confidence). Doing so will require
15   improvements in managing construction of reactor designs that hold the promise of lower costs and
16   broader use (medium confidence). At the same time, nuclear power continues to be affected by cost
17   overruns, high up-front investment needs, challenges with final disposal of radioactive waste, and
18   varying public acceptance and political support levels (high confidence).
19   There are sufficient resources for substantially increasing nuclear deployment (medium confidence).
20   Estimates for identified uranium resources have been increasing steadily over the years. Conventional
21   uranium resources have been estimated to be sufficient for over 130 years of supply at current levels of
22   use; 100 years were estimated in 2009 (Hahn 1983; NEA/IAEA 2021). In the case of future uranium
23   resource scarcity, thorium or recycling of spent fuel might be used as alternatives. Interest in these
24   alternatives has waned with better understanding of uranium deposits, their availability, and low prices
25   (OECD NEA 2015; IAEA 2005).
26   There are several possible nuclear technology options for the period from 2030 to 2050 (medium
27   confidence). In addition to electricity, nuclear can also be used to produce low-carbon hydrogen and
28   freshwater (Kayfeci et al. 2019; Kavvadias and Khamis 2014)
29   •   Large reactors. The nuclear industry has entered a new phase of reactor construction, based on
30       evolutionary designs. These reactors achieve improvements over previous designs through small to
31       moderate modifications, including improved redundancy, increased application of passive safety
32       features, and significant improvements to containment design to reduce the risk of a major accident
33       (MIT 2018). Examples include European - EPR, Korean - APR1400, U.S. - AP1000, Chinese -
34       HPR1000 or Russian - VVER-1200.
35   •   Long-term operation (LTO) of the current fleet. Continued production from nuclear power will
36       depend in part on life extensions of the existing fleet. By the end of 2020, two-thirds of nuclear
37       power reactors will have been operational for over 30 years. The design lifetime of most of existing
38       reactors is 30–40 years. Engineering assessments have established that reactors can operate safely
39       for longer if key replaceable components (e.g., steam generator, mechanical and electrical
40       equipment, instrumentation and control parts) are changed or refurbished (IAEA 2018). The first
41       lifetime extension considered in most of the countries typically is 10–20 years (OECD IEA NEA
42       2020).
43   •   Small Modular Reactors. There are more than 70 SMR designs at different stages of consideration
44       and development, from the conceptual phase to licensing and construction of first-of-a-kind
45       facilities (IAEA 2020). Due to smaller unit sizes, the SMRs are expected to have lower total
46       investment costs, although the cost per unit of generation might be higher than conventional large
47       reactors (Mignacca and Locatelli 2020). Modularity and off-site pre-production may allow greater
48       efficiency in construction, shorter delivery times, and overall cost optimization (IEA 2019c). SMR
     Do Not Cite, Quote or Distribute                  6-34                                  Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1        designs aim to offer an increased load-following capability that makes them suitable to operate in
2        smaller systems and in systems with increasing shares of VRE sources. Their market development
3        by the early 2030s will strongly depend on the successful deployment of prototypes during the
4        2020s.
 5   Nuclear power costs vary substantially across countries (high confidence). First-of-a-kind projects
 6   under construction in Northern America and Europe have been marked by delays and costs overruns
 7   (Berthelemy and Rangel 2015). Construction times have exceeded 13–15 years and cost has surpassed
 8   3–4 times initial budget estimates (OECD IEA NEA 2020). In contrast, most of the recent projects in
 9   Eastern Asia (with construction starts from 2012) were implemented within 5–6 years (IAEA PRIS
10   2021). In addition to region-specific factors, future nuclear costs will depend on the ability to benefit
11   from the accumulated experience in controlling the main drivers of cost. These cost drivers fall into
12   four categories: design maturity, project management, regulatory stability and predictability, and multi-
13   unit and series effects (NEA 2020). With lessons learned from first-of-a-kind projects, the cost of
14   electricity for new builds are expected to be in the range of USD 42–102 MWh-1 depending on the
15   region (OECD IEA NEA 2020).
16   Lifetime extensions are significantly cheaper than new builds and cost competitive with other low-
17   carbon technologies. The overnight cost of lifetime extensions is estimated in the range of USD 390–
18   630 kWe-1 for Europe and North America, and the LCOE in the range of USD 30–36 MWh-1 for
19   extensions of 10–20 years (OECD IEA NEA 2020).
20   Cost-cutting opportunities, such as design standardization and innovations in construction approaches,
21   are expected to make SMRs competitive against large reactors by 2040 (Rubio and Tricot 2016)
22   (medium confidence). As SMRs are under development, there is substantial uncertainty regarding the
23   construction costs. Vendors have estimated first-of-a-kind LCOEs at USD 131–190 MWh-1. Effects of
24   learning for nth-of-a-kind SMR are anticipated to reduce the first-of-a-kind LCOE by 19–32%.
25   Despite low probabilities, the potential for major nuclear accidents exists, and the radiation exposure
26   impacts could be large and long-lasting (Steinhauser et al. 2014). However, new reactor designs with
27   passive and enhanced safety systems reduce the risk of such accidents significantly (high confidence).
28   The (normal) activity of a nuclear reactor results in low volumes of radioactive waste, which requires
29   strictly controlled and regulated disposal. On a global scale, roughly 421 ktons of spent nuclear fuel
30   have been produced since 1971 (IEA 2014). Out of this volume, 2–3% is high-level radioactive waste,
31   which presents challenges in terms of radiotoxicity and decay longevity, and ultimately entails
32   permanent disposal.
33   Nuclear energy is found to be favourable regarding land occupation (Cheng and Hammond 2017;
34   Luderer et al. 2019) and ecological impacts (Brook and Bradshaw 2015; Gibon et al. 2017). Similarly,
35   bulk material requirements per unit of energy produced are low (e.g. aluminum, copper, iron, rare earth
36   metals) (Vidal et al. 2013; Luderer et al. 2019). Water-intensive inland nuclear power plants may
37   contribute to localized water stress and competition for water uses. The choice of cooling systems
38   (closed-loop instead of once-through) can significantly moderate withdrawal rates of the freshwater (Jin
39   et al. 2019; Fricko et al. 2016; Mouratiadou et al. 2016; Meldrum et al. 2013). Reactors situated on the
40   seashore are not affected by water scarcity issues (JRC EU 2021). Life cycle assessment (LCA) studies
41   suggest that the overall impacts on human health (in terms of disability adjusted life years (DALYs)
42   from the normal operation of nuclear power plants are substantially lower than those caused by fossil
43   fuel technologies and are comparable to renewable energy sources (Treyer et al. 2014; Gibon et al.
44   2017).
45   Nuclear power continues to suffer from limited public and political support in some countries (high
46   confidence). Public support for nuclear energy is consistently lower than for renewable energy and
47   natural gas, and in many countries as low as support for energy from coal and oil (Hobman and

     Do Not Cite, Quote or Distribute                  6-35                                 Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

 1   Ashworth 2013; Corner et al. 2011; Pampel 2011). The major nuclear accidents (i.e. Three Mile Island,
 2   Chernobyl, and Fukushima) decreased public support (Poortinga et al. 2013; Bird et al. 2014). The
 3   public remains concerned about the safety risks of nuclear power plants and radioactive materials
 4   (Tsujikawa et al. 2016; Bird et al. 2014; Pampel 2011). At the same time, some groups see nuclear
 5   energy as a reliable energy source, beneficial for the economy and helpful in climate change mitigation.
 6   Public support for nuclear energy is higher when people are concerned about energy security, including
 7   concerns about the availability of energy and high energy prices (Gupta et al. 2019b; Groot et al. 2013),
 8   and when they expect local benefit (Wang et al. 2020c). Public support also increases when trust in
 9   managing bodies is higher (de Groot and Steg 2011). Similarly, transparent and participative decision-
10   making processes enhance perceived procedural fairness and public support (Sjoberg 2004).
11   Because of the sheer scale of the investment required (individual projects can exceed USD 10 billion in
12   value), nearly 90% of nuclear power plants under construction are run by state-owned or controlled
13   companies with governments assuming significant part of the risks and costs. For countries that choose
14   nuclear power in their energy portfolio, stable political conditions and support, clear regulatory regimes,
15   and adequate financial framework are crucial for successful and efficient implementation.
16   Many countries have adopted technology-specific policies for low-carbon energy courses, and these
17   policies influence the competitiveness of nuclear power. For example, feed-in-tariffs and feed-in
18   premiums for renewables widely applied in the EU (Kitzing et al. 2012) or renewable portfolio
19   standards in the U.S. (Barbose et al. 2016) impact wholesale electricity price (leading occasionally to
20   low or even negative prices), which affects the revenues of existing nuclear and other plants (Bruninx
21   et al. 2013; Newbery et al. 2018; Lesser 2019).
22   Nuclear power’s long-term viability may hinge on demonstrating to the public and investors that there
23   is a long-term solution to spent nuclear fuel. Evidence from countries steadily progressing towards first
24   final disposals - Finland, Sweden and France - suggests that broad political support, coherent nuclear
25   waste policies, and a well-managed, consensus-based decision-making process are critical for
26   accelerating this process (Metlay 2016). Proliferation concerns surrounding nuclear power are related
27   to fuel cycle (i.e., uranium enrichment and spent fuel processing). These processes are implemented in
28   a very limited number of countries following strict national and internationals norms and rules, such as
29   IAEA guidelines, treaties, and conventions. Most of the countries which might introduce nuclear power
30   in the future for their climate change mitigation benefits do not envision developing their own full fuel
31   cycle, significantly reducing any risks that might be linked to proliferation (IAEA 2014, 2019).
32 Carbon Dioxide Capture, Utilization, and Storage
33   Since AR5, there have been increased efforts to develop novel platforms that reduce the energy penalty
34   associated with CO2 capture, develop CO2 utilization pathways as a substitute to geologic storage, and
35   establish global policies to support CCS (high confidence). CCS can be used within electricity and other
36   sectors. While it increases the costs of electricity, CCS has the potential to contribute significantly to
37   low-carbon energy system transitions (IPCC 2018).
38   The theoretical global geologic storage potential is about 10,000 Gt-CO2, with more than 80% of this
39   capacity existing in saline aquifers (medium confidence). Not all the storage capacity is usable because
40   geologic and engineering factors limit the actual storage capacity to an order of magnitude below the
41   theoretical potential, which is still more than the CO2 storage requirement through 2100 to limit
42   temperature change to 1.5C (Martin-Roberts et al. 2021) (high confidence). One of the key limiting
43   factors associated with geologic CO2 storage is the global distribution of storage capacity (Table 6.2).
44   Most of the available storage capacity exists in saline aquifers. Capacity in oil and gas reservoirs and
45   coalbed methane fields is limited. Storage potential in the U.S. alone is >1,000 Gt-CO2, which is more
46   than 10% of the world total (NETL 2015). The Middle East has more than 50% of global enhanced oil
47   recovery potential (Selosse and Ricci 2017). It is likely that oil and gas reservoirs will be developed

     Do Not Cite, Quote or Distribute                   6-36                                  Total pages: 217
     Final Government Distribution                         Chapter 6                                  IPCC WGIII AR6

 1   before saline aquifers because of existing infrastructure and extensive subsurface data (Alcalde et al.
 2   2019; Hastings and Smith 2020). Notably, not all geologic storage is utilizable. In places with limited
 3   geologic storage, international CCS chains are being considered, where sources and sinks of CO2 are
 4   located in two or more countries (Sharma and Xu 2021). For economic long-term storage, the desirable
 5   conditions are a depth of 800-3000 m, thickness of greater than 50 m and permeability greater than 500
 6   mD (Singh et al. 2020; Chadwick et al. 2008). Even in reservoirs with large storage potential, the rate
 7   of injection might be limited by the subsurface pressure of the reservoir (Baik et al. 2018a). It is
 8   estimated that geologic sequestration is safe with overall leakage rates at <0.001% yr -1 (Alcalde et al.
 9   2018). In many cases, geological storage resources are not located close to CO2 sources, increasing
10   costs and reduces viability (Garg et al. 2017a).
11    Table 6.2 Geologic storage potential across underground formations globally. These represent order-of-
12                             magnitude estimates. Data: (Selosse and Ricci 2017)

      Reservoir Type         Africa   Australia   Canada   China   CSA    EEU   FSU    India   MEA    Mexico   ODA    USA     WEU

      Enhanced Oil                3          0         3       1      8     2    15       0     38         0      1       8      0
      Depleted oil and gas       20          8        19       1     33     2   191       0     252       22     47    32      37
      Enhanced Coalbed            8         30        16      16      0     2    26       8       0        0     24    90      12
      Methane Recovery
      Deep saline aquifers     1000        500       667     500   1000   250   1000    500     500      250   1015   1000     250

13   CSA: Central and South America, EEU: Eastern Europe, FSU: Former Soviet Union, MEA: Middle East, ODA:
14   Other Asia (except China and India), WEU: Western Europe.

15   CO2 utilization (CCU) - instead of geologic storage - could present an alternative method of
16   decarbonization (high confidence). The global CO2 utilization potential, however, is currently limited
17   to 1–2 GtCO2 yr-1 for use of CO2 as a feedstock (Hepburn et al. 2019; Kätelhön et al. 2019) but could
18   increase to 20 GtCO2 by the mid-century (medium confidence). CCU involves using CO2 as a feedstock
19   to synthesize products of economic value and as substitute to fossil feedstock. However, several CO 2
20   utilization avenues might be limited by energy availability. Depending on the utilization pathway, the
21   CO2 may be considered sequestered for centuries (e.g., cement curing, aggregates), decades (plastics),
22   or only a few days or months (e.g. fuels) (Hepburn et al. 2019). Moreover, when carbon-rich fuel end-
23   products are combusted, CO2 is emitted back into the atmosphere. Because of presence of several
24   industrial clusters (regions with high density of industrial infrastructure) globally, a number of regions
25   demonstrate locations where CO2 utilization potential could be matched with large point sources of CO2
26   (Wei et al. 2020).
27   The technological development for several CO2 utilization pathways is still in the laboratory, prototype,
28   and pilot phases, while others have been fully commercialized (such as urea manufacturing).
29   Technology development in some end-uses is limited by purity requirements for CO2 as a feedstock.
30   The efficacy of CCU processes depends on additional technological constraints such as CO2 purity and
31   pressure requirements. For instance, urea production requires CO2 pressurized to 122 bar and purified
32   to 99.9%. While most utilization pathways require purity levels of 95-99%, algae production may be
33   carried out with atmospheric CO2 (Ho et al. 2019; Voldsund et al. 2016).
34   Existing post-combustion approaches relying on absorption are technologically ready for full-scale
35   deployment (high confidence). More novel approaches using membranes and chemical looping that
36   might reduce the energy penalty associated with absorption are in different stages of development -
37   ranging from laboratory phase to prototype phase (Abanades et al. 2015) (high confidence). There has
38   been significant progress in post-combustion capture technologies that used absorption in solvents such
39   as monoethanol amine (MEA). There are commercial-scale application of solvent-based absorption at
40   two facilities – Boundary Dam since 2015 and Petra Nova (temporarily suspended) since 2017, with
41   capacities of 1 and 1.6 MtCO2 yr-1 respectively (Mantripragada et al. 2019; Giannaris et al. 2020a).
     Do Not Cite, Quote or Distribute                         6-37                                     Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

1    Several 2nd and 3rd generation capture technologies are being developed with the aim of not just lowering
2    costs but also enhancing other performance characteristics such as improved ramp-up and lower water
3    consumption. These include processes such as chemical looping, which also has the advantage of being
4    capable of co-firing with biomass (Bhave et al. 2017; Yang et al. 2019). Another important
5    technological development is the Allam cycle, which utilizes CO2 as a working fluid and operates based
6    on oxy-combustion capture. Applications using the Allam Cycle can deliver net energy efficiency
7    greater than 50% and 100% CO2 capture, but they are quite sensitive to oxygen and CO2 purity needs
8    (Scaccabarozzi et al. 2016; Ferrari et al. 2017).
 9   CO2 capture costs present a key challenge, remaining higher than USD 50 tCO2-1 for most technologies
10   and regions; novel technologies could help reduce some costs (high confidence). The capital cost of a
11   coal or gas electricity generation facility with CCS is almost double one without CCS (Zhai and Rubin
12   2016; Rubin et al. 2015; Bui et al. 2018). Additionally, the energy penalty increases the fuel requirement
13   for electricity generation by 13–44%, leading to further cost increases (Table 6.3).
14    Table 6.3 Costs and efficiency parameters of CCS in electric power plants. Data: (Muratori et al. 2017a)

                                        Capital      Efficiency      CO2 Capture         CO2 Avoided Cost
                                        Cost [USD    [%]             Cost [USD ton-      [USD ton-CO2-1]
                                        kW-1]                        CO2-1]
      Coal (steam plant) + CCS          5800         28%             63                  88
      Coal (IGCC) + CCS                 6600         32%             61                  106
      Natural Gas (CC) + CCS            2100         42%             91                  33
      Oil (CC) + CCS                    2600         39%             105                 95
      Biomass (steam plant) + CCS       7700         18%             72                  244
      Biomass (IGCC) + CCS              8850         25%             66                  242
16   In addition to reductions in capture costs, other approaches to reduce CCS costs rely on utilizing the
17   revenues from co-products such as oil, gas, or methanol, and on clustering of large-point sources to
18   reduce infrastructure costs. The potential for such reductions is limited in several regions due to low
19   sink availability, but it could jumpstart initial investments (medium confidence). Injecting CO2 into
20   hydrocarbon formations for enhanced oil or gas recovery can produce revenues and lower costs
21   (Edwards and Celia 2018). While enhanced oil recovery potential is <5% of the actual CCS needs, they
22   can enable early pilot and demonstration projects (Núñez-López and Moskal 2019; Núñez-López et al.
23   2019). Substantial portions of CO2 are effectively stored during enhanced oil recovery (Sminchak et al.
24   2020; Menefee and Ellis 2020). By clustering together of several CO2 sources, overall costs may be
25   reduced by USD 10 tCO2-1 (Abotalib et al. 2016; Garg et al. 2017a), but geographical circumstances
26   determine the prospects of these cost reductions via economies-of-scale. The major pathways for
27   methanol, methane, liquid fuel production, and cement curing have costs greater than USD 500 tCO2-1
28   (Hepburn et al. 2019). The success of these pathways therefore depends on the value of such fuels and
29   on the values of other alternatives.
30   The public is largely unfamiliar with carbon capture, utilization, and storage technologies (Tcvetkov et
31   al. 2019; L’Orange Seigo et al. 2014) (high confidence), and many people may not have formed stable
32   attitudes and risk perceptions regarding these technologies (Daamen et al. 2006; Jones et al. 2015; Van
33   Heek et al. 2017) (medium confidence). In general, low support has been reported for CCS technologies
34   (Allen and Chatterton 2013; Demski et al. 2017). When presented with neutral information on CCS,
35   people favour other mitigation options such as renewable energy and energy efficiency (de Best-
36   Waldhober et al. 2009; Scheer et al. 2013; Karlstrøm and Ryghaug 2014). Although few totally reject
37   CCS, specific CCS projects have faced strong local resistance, which has contributed to the cancellation
38   of CCS projects (Terwel et al. 2012; L’Orange Seigo et al. 2014). Communities may also consider CCU
39   to be lower-risk and view it more favourably than CCS (Arning et al. 2019).

     Do Not Cite, Quote or Distribute                  6-38                                    Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    CCS requires considerable increases in some resources and chemicals, most notably water. Power plants
2    with CCS could shutdown periodically due to water scarcity. In several cases, water withdrawals for
3    CCS are 25–200% higher than plants without CCS (Yang et al. 2020; Rosa et al. 2020b) due to energy
4    penalty and cooling duty. The increase is slightly lower for non-absorption technologies. In regions
5    prone to water scarcity such as the Southwestern U.S. or Southeast Asia, this may limit deployment and
6    result in power plant shutdowns during summer months (Liu et al. 2019b; Wang et al. 2019c). The water
7    use could be managed by changing heat integration strategies and implementing reuse of wastewater
8    (Magneschi et al. 2017; Giannaris et al. 2020b).

10          Figure 6.13 Costs and potential for different CO2 utilization pathways (Hepburn et al. 2019)

11   Because CCS always adds cost, policy instruments are required for it to be widely deployed (high
12   confidence). Relevant policy instruments include financial instruments such as emission certification
13   and trading, legally enforced emission restraints, and carbon pricing (Haszeldine 2016; Kang et al.
14   2020). There are some recent examples of policy instruments specifically focused on promoting CCS.
15   The recent U.S. 45Q tax credits offers nationwide tax credits for CO2 capture projects above USD 35–
16   50 tCO2-1 which offset CO2 capture costs at some efficient plants (Esposito et al. 2019). Similarly,
17   California’s low-carbon fuel standard offers benefits for CO2 capture at some industrial facilities such
18   as biorefineries and refineries (Von Wald et al. 2020).
19 Bioenergy
20   Bioenergy has the potential to be a high-value and large-scale mitigation option to support many
21   different parts of the energy system. Bioenergy could be particularly valuable for sectors with limited
22   alternatives to fossil fuels (e.g., aviation, heavy industry), production of chemicals and products, and,
23   potentially, in carbon dioxide removal (CDR) via BECCS or biochar. While traditional biomass and
24   first-generation biofuels are widely used today, the technology for large-scale production from
25   advanced processes is not competitive, and growing dedicated bioenergy crops raises a broad set of
26   sustainability concerns. Its long-term role in low-carbon energy systems is therefore uncertain (high
27   confidence). [Note that this section focuses on the key technological developments for deployment of
28   commercial bioenergy.]
29   Bioenergy is versatile: technology pathways exist to produce multiple energy carriers from biomass -
30   electricity, liquid fuels, gaseous fuels, hydrogen, and solid fuels - as well as other value-added products
31   (high confidence). Different chemical and biological conversion pathways exist to convert diverse
32   biomass feedstocks into multiple final energy carriers (Figure 6.14). Currently, biomass is mostly used
33   to produce heat or for cooking purposes (traditional biomass), electricity, or first-generation sugar-based

     Do Not Cite, Quote or Distribute                   6-39                                  Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

1    biofuels (e.g., ethanol produced via fermentation), as well as biodiesel produced from vegetable oils
2    and animal fats. Electricity generated from biomass contributes about 3% of global generation. Tens of
3    billions of gallons of first-generation biofuels are produced per year. The processing requirements
4    (drying, dewatering, pelletizing) of different feedstocks for producing electricity from biomass are
5    energy-intensive, and when utilizing current power plants, the efficiency is around 22%, with an
6    increase up to 28% with advanced technologies (Zhang et al. 2020).
 7   Scaling up bioenergy use will require advanced technologies such as gasification, Fischer-Tropsch
 8   processing, hydrothermal liquefaction (HTL), and pyrolysis. These pathways could deliver several final
 9   energy carriers starting from multiple feedstocks, including forest biomass, dedicated cellulosic
10   feedstocks, crop residues, and wastes (Figure 6.14). While potentially cost-competitive in the future,
11   pyrolysis, Fischer-Tropsch, and HTL are not currently cost-competitive (IEA 2018c; Molino et al. 2018;
12   Prussi et al. 2019), and scaling-up these processes will require robust business strategies and optimized
13   use of co-products (Lee and Lavoie 2013). Advanced biofuels production processes are at the pilot or
14   demonstration stage and will require substantial breakthroughs or market changes to become
15   competitive. Moreover, fuels produced from these processes require upgrading to reach “drop-in”
16   conditions – that is, conditions in which they may be used directly consistent with current standards in
17   existing technologies (van Dyk et al. 2019). Additional opportunities exist to co-optimize second
18   generation biofuels and engines (Ostadi et al. 2019; Salman et al. 2020). In addition, gaseous wastes,
19   or high-moisture biomass, such as dairy manure, wastewater sludge and organic MSW could be utilized
20   to produce renewable natural gas. Technologies for producing biogas (e.g. digestion) tend to be less
21   efficient than thermochemical approaches and often produce large amounts of CO2, requiring the
22   produced fuels to undergo significant upgrading (Melara et al. 2020).

24   Figure 6.14 Range of advanced bioenergy conversion pathways (excluding traditional biomass, direct heat
25   generation, first-generation biofuels, and non-energy products) based on feedstock, targeted end product,
26     and compatibility with CDR via CCS and soil carbon sequestration (Modified from Baker et al, 2020)

27   A major scale-up of bioenergy production will require dedicated production of advanced biofuels. First
28   generation biofuels produced directly from food crops or animal fats both have limited potential and
29   lower yield per land area than advanced biofuels. Wastes and residues (e.g., from agricultural, forestry,
30   animal manure processing) or biomass grown on degraded, surplus, and marginal land can provide
31   opportunities for cost-effective and sustainable bioenergy at significant but limited scale (Saha and
32   Eckelman 2018; Fajardy and Mac Dowell 2020; Spagnolo et al. 2020; Morris et al. 2013). Assessing
33   the potential for a major scale-up of purpose-grown bioenergy is challenging due to its far-reaching
34   linkages to issues beyond the energy sector, including competition with land for food production and
35   forestry, water use, impacts on ecosystems, and land-use change) (IPCC 2020; Chapter 12; (Roe et al.
36   2021)). These factors, rather than geophysical characteristics, largely define the potential for bioenergy
     Do Not Cite, Quote or Distribute                  6-40                                  Total pages: 217
     Final Government Distribution                       Chapter 6                               IPCC WGIII AR6

1    and explain the difference in estimates of potential in the literature. Biomass resources are not always
2    in close proximity to energy demand, necessitating additional infrastructure or means to transport
3    biomass or final bioenergy over larger distances and incur additional energy use (Baik et al. 2018b;
4    Singh et al. 2020).
 5   An important feature of bioenergy is that it can be used to remove carbon from the atmosphere by
 6   capturing CO2 in different parts of the conversion process and then permanently storing the CO2
 7   (BECCS or biochar) (Chapter 3, Chapter 7; Chapter 12.5; Smith et al. 2016; Fuss et al. 2018). Some
 8   early opportunities for low-cost BECCS are being utilized in the ethanol sector but these are applicable
 9   only in the near-term at the scale of ≤100 Mt-CO2 yr-1 (Sanchez et al. 2018). Several technological and
10   institutional barriers exist for large-scale BECCS implementation, including large energy requirements
11   for CCS, limit and cost of biomass supply and geologic sinks for CO2 in several regions, and cost of
12   CO2 capture technologies (high confidence). Besides BECCS, biofuels production through pyrolysis
13   and hydrothermal liquefaction creates biochar, which could also be used to store carbon as 80% of the
14   carbon sequestered in biochar will remain in the biochar permanently (Chapter 7). In addition to its
15   ability to sequester carbon, biochar can be used as a soil amendment (Wang et al. 2014b).
16   First-generation bioenergy is currently competitive in some markets, though on average its costs are
17   higher than other forms of final energy. Bioenergy from waste and residues from forestry and
18   agriculture is also currently competitive, but the supply is limited (Aguilar et al. 2020). These costs are
19   context-dependent, and regions having large waste resources are already producing low-cost bioenergy
20   (Jin and Sutherland 2018). In the future, technology costs are anticipated to decrease, but bioenergy
21   produced through cellulosic feedstocks may remain more expensive than fossil alternatives. Large-scale
22   deployment of early opportunities especially in the liquid fuel sector may reduce the technological costs
23   associated with biomass conversion (IEA 2020g). At the same time, the cost of feedstocks may rise as
24   bioenergy requirements increase, especially in scenarios with large bioenergy deployment (Muratori et
25   al. 2020). The costs of bioenergy production pathways are highly uncertain (Table 6.4).
27        Table 6.4 The costs of electricity generation, hydrogen production, and second-generation liquid fuels
28       production from biomass in 2020. These costs are adapted from (Daioglou et al. 2020), (Bhave et al. 2017),
29                   (NREL 2020a), (Lepage et al. 2021), (Witcover and Williams 2020), (NREL 2020b)

                                                          Unit              Low      Median           High
            Bioelectricity with CCS                  USD/MWh                 74         86             160
            Bioelectricity without CCS               USD/MWh                 66         84             112
            Biohydrogen with CCS*                      USD/kg               1.63       2.37            2.41
            Biohydrogen without CCS*                   USD/kg               1.59       1.79            2.37
            Liquid biofuels with CCS                   USD/gge              1.34       4.20            7.85
            Liquid biofuels without CCS                USD/gge              1.15       4.00            7.60
30                                                * Using cellulosic feedstocks

31   •     Electricity. The costs of baseload electricity production with biomass are higher than corresponding
32         fossil electricity production with and without CCS, and are likely to remain as such without carbon
33         pricing (Bhave et al. 2017). The additional cost associated with CO2 capture are high for
34         conventional solvent-based technologies. However, upcoming technologies such as chemical
35         looping are well-suited to biomass and could reduce CCS costs.
36   •     Hydrogen. The costs of hydrogen production from biomass are somewhat higher than, but
37         comparable, to that produced by natural gas reforming with CCS. Further, the incremental costs for
38         incorporating CCS in this process are less than 5% of the levelized costs in some cases, since the

     Do Not Cite, Quote or Distribute                        6-41                                 Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

 1       gasification route creates a high-purity stream of CO2 (Muratori et al. 2017a; Sunny et al. 2020).
 2       While these processes have fewer ongoing prototypes/demonstrations, the costs of biomass-based
 3       hydrogen (with or without CCS) are substantially cheaper than that produced from electrolysis
 4       utilizing solar/wind resources (Kayfeci et al. 2019; Newborough and Cooley 2020), even though
 5       electrolysis costs are dropping.
 6   •   Liquid Biofuels. First-generation sugar-based biofuels (e.g., ethanol produced via fermentation) or
 7       biodiesel produced from vegetable oils and animal fats are produced in several countries at large
 8       scale and costs competitive with fossil fuels. However, supply is limited. The costs for second
 9       generation processes (Fischer-Tropsch and cellulosic ethanol) are higher in most regions (Li et al.
10       2019). Technological learning is projected to reduce these costs by half (IEA 2020g).
11   Large-scale bioenergy production will require more than wastes/residues and cultivation on marginal
12   lands, which may raise conflicts with SDGs relevant to environmental and societal priorities (Gerten et
13   al. 2020; Heck et al. 2018) (Chapter 12). These include competition with food crops, implications for
14   biodiversity, potential deforestation to support bioenergy crop production, energy security implications
15   from bioenergy trade, point-of-use emissions and associated effects on air quality, and water use and
16   fertilizer use (Fajardy and Mac Dowell 2018; Tanzer and Ramírez 2019; Fuss et al. 2018; Brack and
17   King 2020). Overall, the environmental impact of bioenergy production at scale remains uncertain and
18   varies by region and application.
19   Alleviating these issues would require some combination of increasing crop yields, improving
20   conversion efficiencies, and developing advanced biotechnologies for increasing the fuel yield per
21   tonne of feedstock (Henry et al. 2018). Policy structures would be necessary to retain biodiversity,
22   manage water use, limit deforestation and land-use change emissions, and ultimately optimally integrate
23   bioenergy with transforming ecosystems. Large-scale international trade of biomass might be required
24   to support a global bioeconomy, raising questions about infrastructure, logistics, financing options, and
25   global standards for bioenergy production and trade (Box 6.10). Additional institutional and economic
26   barriers are associated with accounting of carbon dioxide removal, including BECCS (Fuss et al. 2014;
27   Muratori et al. 2016; Fridahl and Lehtveer 2018).
28   Life-cycle emissions impacts from bioenergy are subject to large uncertainties and could be
29   incompatible with net zero emissions in some contexts. Due to the potentially large energy conversion
30   requirements and associated GHG emissions (Chapter 7, Chapter 12), bioenergy systems may fail to
31   deliver near-zero emissions depending on operating conditions and regional contexts (Staples et al.
32   2017; Lade et al. 2020; Daioglou et al. 2017; Hanssen et al. 2020; Elshout et al. 2015). As a result,
33   bioenergy carbon neutrality is debated and depends on factors such as the source of biomass, conversion
34   pathways and energy used for production and transport of biomass, and land use changes, as well as
35   assumed analysis boundary and considered timescale (Fan et al. 2021; Wiloso et al. 2016; Zanchi et al.
36   2012; Booth 2018). Similarly, the lifecycle emissions of BECCS remain uncertain and will depend on
37   how effectively bioenergy conversion processes are optimized (Fajardy and Mac Dowell 2017; Tanzer
38   and Ramírez 2019).
39   Acceptability of bioenergy is relatively low compared to other renewable energy sources like solar and
40   wind (Poortinga et al. 2013; EPCC 2017; Peterson et al. 2015; Ma et al. 2015) and comparable to
41   natural gas (Scheer et al. 2013). People also know relatively little about bioenergy compared to other
42   energy sources (Whitmarsh et al. 2011a; EPCC 2017) and tend be be more ambivalent towards
43   bioenergy compared to other mitigation options (Allen and Chatterton 2013). People evaluate biomass
44   from waste products (e.g., food waste) more favourably than grown-for-purpose energy crops, which
45   are more controversial (Demski et al. 2015; Plate et al. 2010). The most pressing concerns for use of
46   woody biomass are air pollution and loss of local forests (Plate et al. 2010). Various types of bioenergy
47   additionally raise concerns about landscape impacts (Whitmarsh et al. 2011a) and biodiversity

     Do Not Cite, Quote or Distribute                  6-42                                 Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    (Immerzeel et al. 2014). Moreover, many people do not see biomass as a renewable energy source,
2    possibly because it involves burning of material.
4                         Box 6.5 Methane mitigation options for coal, oil, and gas

 5   Methane emissions mainly from coal, oil, and gas currently represent in 2019 about 18% of energy
 6   supply sector GHG emissions and 90% of global energy supply non-CO2 emissions in 2019 (Minx et
 7   al. 2021b). While approximately 80% of the life-cycle methane emissions in the coal sector occur during
 8   underground mining, oil and gas emissions are spread throughout upstream, midstream, and
 9   downstream stages (IPCC, 2019) (Alvarez et al. 2018). For this reason, methane reductions from coal
10   mining can be accomplished through coal mine methane recovery (where methane and coal are
11   recovered simultaneously) and from the ventilation air, which can reduce methane emissions by 50–
12   75% (Singh and Hajra 2018; Zhou et al. 2016). Governments incentivize such operations through a
13   number of emissions trading and offset programs (Haya et al. 2020). Methane emissions in the oil and
14   gas sector can be reduced by leak detection and repair, relevant across varying time scales (hours to
15   decades) and regional scopes (component/facility level to continental) (Fox et al. 2019). Around 50%
16   of the methane emitted from oil and gas infrastructure can be mitigated at net-negative costs; that is,
17   the market price of the recovered methane is higher than the mitigation costs (IEA 2021e). As CO2
18   emissions are reduced and fossil fuel consumption decreases, methane emissions associated with these
19   supply chains are anticipated to decline (section 6.7). That said, substantial ‘legacy’ methane emissions
20   – methane leaks after abandonment – will remain even if a complete fossil fuel phase-out takes place.
21   These legacy emissions are estimated to be less than 1–4% of overall methane emissions across all
22   fossil fuel sources (Kholod et al. 2020; Williams et al. 2021b). Even without a complete phase-out, 50-
23   80% of methane emissions from coal, oil and gas could be avoided with currently available technologies
24   at less than USD 50 tCO2-eq-1 (Höglund-Isaksson et al. 2020; Harmsen et al. 2019). Methane recovery
25   from abandoned coal mines could offset most project costs (Singh and Sahu 2018). For abandoned oil
26   and gas wells, low plugging costs could be offset through methane recovery, while high plugging costs
27   would likely require some market or policy support (Kang et al. 2019).
28   [END BOX 6.5 HERE]
29 Fossil Energy
30   Fossil fuels could play a role in climate change mitigation if strategically deployed with CCS (high
31   confidence). On the one hand, the primary mechanism for reducing emissions is to eliminate the
32   unabated fossil fuel use. On the other hand, fossil energy combined with CCS provides a means of
33   producing low-carbon energy while still utilizing the available base of fossil energy worldwide and
34   limiting stranded assets. While Section discusses the important aspects of CCS with fossil fuels,
35   this section aims to elucidate the feasibility criteria around these fuels itself.
36   Fossil fuel reserves have continued to rise because of advanced exploration and utilization techniques
37   (high confidence). A fraction of these available reserves can be used consistent with mitigation goals
38   when paired with CCS opportunities in close geographical proximity (high confidence). Based on
39   continued exploration, the fossil fuel resource base has increased significantly; for example, a 9%
40   increase in gas reserves and 12% in oil reserves was observed in the U.S. between 2017 and 2018. This
41   increase is a result of advanced exploration techniques, which are often subsidized (Lazarus and van
42   Asselt 2018; MA et al. 2018). Fossil reserves are distributed unevenly throughout the globe. Coal
43   represents the largest remaining resource (close to 500 ZJ). Conventional oil and gas resources are an
44   order of magnitude smaller (15–20 ZJ each). Technological advances have increased the reserves of
45   unconventional fossil in the last decade. Discovered ultimate recoverable resources of unconventional
46   oil and gas are comparable to conventional oil and gas (Fizaine et al. 2017).

     Do Not Cite, Quote or Distribute                  6-43                                 Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    It is unlikely that resource constraints will lead to a phaseout of fossil fuels, and instead, such a phase-
2    out would require policy action. Around 80% of coal, 50% of gas, and 20% of oil reserves are likely to
3    remain unextractable under 2C constraints (McGlade and Ekins 2015; Pellegrini et al. 2020). Reserves
4    are more likely to be utilized in a low-carbon transition if they can be paired with CCS. Availability of
5    CCS technology not only allows continued use of fossil fuels as a capital resource for countries but also
6    paves the way for CDR through BECCS (Pye et al. 2020; Haszeldine 2016). While the theoretical
7    geologic CO2 sequestration potential is vast, there are limits on how much resource base could be
8    utilized based on geologic, engineering, and source-sink mapping criteria (Budinis et al. 2017).
 9   Technological changes have continued to drive down fossil fuel extraction costs. Significant
10   decarbonization potential also exists via diversification of the fossil fuel uses beyond combustion (high
11   evidence). The costs of extracting oil and gas globally have gone down by utilizing hydraulic fracturing
12   and directional drilling for resources in unconventional reservoirs (Wachtmeister and Höök 2020).
13   Although the extraction of these resources is still more expensive than those derived from conventional
14   reservoirs, the large availability of unconventional resources has significantly reduced global prices.
15   The emergence of liquefied natural gas (LNG) markets has also provided opportunities to export natural
16   gas significant distances from the place of production (Avraam et al. 2020). The increase in availability
17   of natural gas has been accompanied by an increase in the production of natural gas liquids as a co-
18   product to oil and gas. Over the period from 2014 to 2019, exports of natural gas liquids increased by
19   160%. Natural gas liquids could potentially be a lower-carbon alternative to liquid fuels and
20   hydrocarbons. On the demand side, natural gas can be used to produce hydrogen using steam methane
21   reforming, which is a technologically mature process (Sections 6.4.4, 6.4.5). When combined with 90%
22   CO2 capture, the costs of producing hydrogen are around USD 1.5–2 kg(H2)-1 (Newborough and Cooley
23   2020; Collodi et al. 2017), considerably less than hydrogen produced via electrolysis.
24   Significant potential exists for gasifying deep-seated coal deposits in situ to produce hydrogen. Doing
25   so reduces fugitive methane emissions from underground coal mining. The integration costs of this
26   process with CCS are less than with natural gas reforming. The extent to which coal gasification could
27   be compatible with low-carbon energy would depend on the rate of CO2 capture and the ultimate use of
28   the gas (Verma and Kumar 2015). Similarly, for ongoing underground mining projects, coal mine
29   methane recovery can be economic for major coal producers such as China and India. Coal mine
30   methane and ventilation air methane recovery can reduce the fugitive methane emissions by 50–75%
31   (Zhou et al. 2016; Singh and Sahu 2018).
32   The cost of producing electricity from fossil sources has remained roughly the same with some regional
33   exceptions while the costs of transport fuels has gone down significantly (high confidence). The cost of
34   producing electricity from fossil fuels has remained largely static, with the exception of some regional
35   changes, for example, a 40% cost reduction in the U.S. for natural gas (Rai et al. 2019), where the gas
36   wellhead price has declined by almost two-thirds due to large reserves. Similarly, the global price of
37   crude oil has declined from almost USD 100–55 bbl-1 in the last five years.
38   The energy return of investment (EROI) is a useful indicator of full fossil lifecycle costs. Fossil fuels
39   create significantly more energy per unit energy invested – or in other words have much larger EROI –
40   than most cleaner fuels such as biomass or electrolysis-derived hydrogen, where intensive processing
41   reduces EROI (Hall et al. 2014). That said, recent years have seen a decrease in fossil EROI, especially
42   as underground coal mining has continued in China. Exploitation of unconventional gas reservoirs is
43   also energy intensive and has led to a reduction in EROI. The primary energy EROI of fossil fuels has
44   converged at about 30, which represents a 20-point decrease from the 1995 value for coal (Brockway
45   et al. 2019). When processing and refining stages are considered, these EROI values further decrease.
46   Several countries have large reserves of fossil fuels. Owing to climate constraints, these may become
47   stranded causing considerable economic impacts (6.7.3, 6.7.4, Box 6.13) (high confidence). While
48   global fossil energy resources are greater than 600 ZJ, more than half of these resources would likely
     Do Not Cite, Quote or Distribute                  6-44                                 Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    be unburnable even in the presence of CCS (Pye et al. 2020; McGlade and Ekins 2015). This would
2    entail a significant capital loss for the countries with large reserves. The total amount of stranded assets
3    in such a case would amount to USD 1–4 trillion at present value (Box 6.13).
 4   Apart from CO2 emissions and air pollutants from fossil fuel combustion, other environmental impacts
 5   include fugitive methane leakages and implications to water systems. While the rate of methane leakage
 6   from unconventional gas systems is uncertain, their overall GHG impact is less than coal (Deetjen and
 7   Azevedo 2020; Tanaka et al. 2019). The stated rate of leakage in such systems ranges from 1-8%, and
 8   reconciling different estimates requires a combination of top-down and bottom-up approaches (Grubert
 9   and Brandt 2019; Zavala-Araiza et al. 2015). Similarly, for coal mining, fugitive methane emissions
10   have grown despite some regulations on the degree to which emission controls must be deployed.
11   Recent IPCC inventory guidance also notes considerable CO2 emissions resulting from spontaneous
12   combustion of the coal surface, and accounting for these emissions will likely increase the overall life-
13   cycle emissions by 1–5% (Fiehn et al. 2020; Singh 2019; IPCC 2019).
14   Another key issue consistently noted with unconventional wells (both oil and gas, and coalbed methane)
15   is the large water requirements (Qin et al. 2018). The overall water footprint of unconventional
16   reservoirs is higher than conventional reservoirs because of higher lateral length and fracturing
17   requirements (Scanlon et al. 2017; Kondash et al. 2018). Moreover, produced water from such
18   formations is moderately to highly brackish, and treating such waters has large energy consumption
19   (Singh and Colosi 2019; Bartholomew and Mauter 2016).
20   Oil and coal consistently rank among the least preferred energy sources in many countries (high
21   confidence). The main perceived advantage of fossil energy is the relatively low costs, and emphasizing
22   these costs might increase acceptability somewhat (Pohjolainen et al. 2018; Hazboun and Boudet 2020;
23   Boyd et al. 2019). Acceptability of fossil fuels is on average similar to acceptability of nuclear energy,
24   although evaluations are less polarized. People evaluate natural gas as somewhat more acceptable than
25   other fossil fuels, although they generally oppose hydraulic fracturing (Clarke et al. 2016). Yet, natural
26   gas is evaluated as less acceptable than renewable energy sources, although evaluations of natural gas
27   and biogas are similar (Liebe and Dobers 2019; Plum et al. 2019). Acceptability of fossil energy tends
28   to be higher in countries and regions that strongly rely on them for their energy production (Boyd et al.
29   2019; Pohjolainen et al. 2018). Combining fossil fuels with CCS can increase their acceptability (Van
30   Rijnsoever et al. 2015; Bessette and Arvai 2018). Some people seem ambivalent about natural gas, as
31   they perceive both benefits (e.g., affordability, less carbon emissions than coal) and disadvantages (e.g.,
32   finite resource, contributing to climate change) (Blumer et al. 2018).
33   Fossil fuel subsidies have been valued of the order of USD 0.5–5 trillion annually by various estimates
34   which have the tendency to introduce economic inefficiency within systems (Merrill et al. 2015; Jakob
35   et al. 2015) (high confidence). Subsequent reforms have been suggested by different researchers who
36   have estimated reductions in CO2 emissions may take place if these subsidies are removed (Mundaca
37   2017). Such reforms could create the necessary framework for enhanced investments in social welfare
38   – through sanitation, water, clean energy - with differentiating impacts (Edenhofer 2015).
39 Geothermal Energy
40   Geothermal energy is heat stored in the Earth’s subsurface and is a renewable resource that can be
41   sustainably exploited. The geophysical potential of geothermal resources is 1.3 to 13 times the global
42   electricity demand in 2019 (medium confidence). Geothermal energy can be used directly for various
43   thermal applications, including space heating and industrial heat input, or converted to electricity
44   depending on the source temperature (Moya et al. 2018; REN21 2019; Limberger et al. 2018).
45   Suitable aquifers underlay 16% of the Earth's land surface and store an estimated 110,000–1,400,000
46   PWh (400,000–1,450,000 EJ) that could theoretically be used for direct heat applications. For electricity
47   generation, the technical potential of geothermal energy is estimated to be between 30 PWh yr -1 (108
     Do Not Cite, Quote or Distribute                   6-45                                   Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

 1   EJ yr-1) (to 3 km depth) and 300 PWh yr-1 (1080 EJ yr-1) (to 10 km depth). For direct thermal uses, the
 2   technical potential is estimated to range from 2.7–86 PWh yr-1 (9.7–310 EJ yr-1)(IPCC 2011). Despite
 3   the potential, geothermal direct heat supplies only 0.15% of the annual global final energy consumption.
 4   The technical potential for electricity generation, depending on the depth, can meet one third to almost
 5   three times the global final consumption (based on IEA database for IPCC). The mismatch between
 6   potential and developed geothermal resources is caused by high up-front costs, decentralized
 7   geothermal heat production, lack of uniformity among geothermal projects, geological uncertainties,
 8   and geotechnical risks (IRENA 2017a; Limberger et al. 2018). A limited number of countries have a
 9   long history in geothermal. At least in two countries (Iceland and New Zealand), geothermal accounts
10   for 20–25% of electricity generation (Spittler et al. 2020; Pan et al. 2019). Furthermore, in Iceland
11   approximately 90% of the households are heated with geothermal energy. In Kenya, as of July 2019,
12   geothermal accounted for 734 MW effective capacity spread over 10 power plants and approximately
13   one third of the total installed capacity (Kahlen 2019).

15     Figure 6.15 Global weighted average total installed costs, capacity factors and LCOE for geothermal
16    power per year (2010-2020). The shaded area represents the 5% and 95% percentiles. Source: (IRENA
17                                                    2021a)

18   There are two main types of geothermal resources: convective hydrothermal resources, in which the
19   Earth’s heat is carried by natural hot water or steam to the surface, and hot, dry rock resources, in which
20   heat cannot be extracted using water or steam, and other methods must be developed. There are three
21   basic types of geothermal power plants: (1) dry steam plants use steam directly from a geothermal
22   reservoir to turn generator turbines; (2) flash steam plants take high-pressure hot water from deep inside
23   the Earth and convert it to steam to drive generator turbines and (3) binary cycle power plants transfer
24   the heat from geothermal hot water to another liquid. Many of the power plants in operation today are
25   dry steam plants or flash plants (single, double and triple) harnessing temperatures of more than 180°C.
26   However, medium temperature fields are increasingly used for electricity generation or combined heat
27   and power. The use of medium temperature fields has been enabled through the development of binary
28   cycle technology, in which a geothermal fluid is used via heat exchangers. Increasing binary generation
29   technologies are now being utilized instead of flash steam power plants. This will result in almost 100%
30   injection and essentially zero GHG emissions, although GHG emissions from geothermal power
31   production are generally small compared to traditional baseload thermal energy power generation
32   facilities (Fridriksson et al. 2016).
33   Additionally, new technologies are being developed like Enhanced Geothermal Systems (EGS), which
34   is in the demonstration stage (IRENA 2018), deep geothermal technology, which may increase the
35   prospects for harnessing the geothermal potential in a large number of countries, or shallow-geothermal
36   energy, which represents a promising supply source for heating and cooling buildings (Narsilio and Aye
37   2018). Successful large-scale deployment of shallow geothermal energy will depend not only on site-
38   specific economic performance but also on developing suitable governance frameworks (Bloemendal
     Do Not Cite, Quote or Distribute                   6-46                                  Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

1    et al. 2018; García-Gil et al. 2020). Technologies for direct uses like district heating, geothermal heat
2    pumps, greenhouses, and other applications are widely used and considered mature. Given the limited
3    number of plants commissioned, economic indicators (Figure 6.15) vary considerably depending on site
4    characteristics.
 5   Public awareness and knowledge of geothermal energy is relatively low (high confidence). Geothermal
 6   energy is evaluated as less acceptable than other renewable energy sources such as solar and wind, but
 7   is preferred over fossil and nuclear energy, and in some studies, over hydroelectric energy (Karytsas et
 8   al. 2019; Pellizzone et al. 2015; Steel et al. 2015; Hazboun and Boudet 2020) (high confidence). Some
 9   people are concerned about the installation of geothermal facilities close to their homes, similar to solar
10   and wind projects (Pellizzone et al. 2015). The main concerns about geothermal energy, particularly for
11   large scale, high-temperature geothermal power generation plants, involve water usage, water scarcity,
12   and seismic risks of drilling (Dowd et al. 2011). Moreover, noise, smell and damages to the landscape
13   have been reasons for protests against specific projects (Walker 1995). However, with the
14   implementation of modern technologies, geothermal presents fewer adverse environmental impacts. At
15   the same time, people perceive geothermal energy as relatively environmentally friendly (Tampakis et
16   al. 2013).
17 Marine Energy
18   The ocean is a vast source of energy (Hoegh-Guldberg et al. 2019). Ocean energy can be extracted from
19   tides, waves, ocean thermal energy conversion (OTEC), currents, and salinity gradients (Bindoff et al.
20   2019). Their technical potentials, without considering possible exclusion zones, are explored below.
21   Tidal energy, which uses elevation differences between high and low tides, appears in two forms:
22   potential energy (rise and fall of the tide) and current energy (from tidal currents). The global technically
23   harvestable tidal power from areas close to the coast is estimated as ~1.2 PWh yr-1 (4.3 EJ yr-1) (IRENA
24   2020b). The potential for tidal current energy is estimated to be larger than that for tidal range or barrage
25   (Melikoglu 2018). Ocean wave energy is abundant and predictable and can be extracted directly from
26   surface waves or pressure fluctuations below the surface (Melikoglu 2018). Its global theoretical
27   potential is 29.5 PWh yr-1 (106 EJ yr-1) ,which means that wave energy alone could meet all global
28   energy demand (Mørk et al. 2010; IRENA 2020b). The temperature gradients in the ocean can be
29   exploited to produce energy, and its total estimated available resource could be up to 44.0 PWh yr -1
30   (158 EJ yr-1) (Rajagopalan and Nihous 2013). Salinity gradient energy, also known as osmotic power,
31   has a global theoretical potential of over 1.6 PWh yr-1 (6.0 EJ yr-1) (IRENA 2020b). The greatest
32   advantage of most marine energy, excluding wave energy, is that their sources are highly regular and
33   predictable, and energy can be furthermore generated both day and night. An additional use of sea water
34   is to develop lower-cost district cooling systems near the sea (Hunt et al. 2019). The greatest barrier to
35   most marine technology advances is the relatively high upfront costs, uncertainty on environmental
36   regulation and impact, need for investments and insufficient infrastructure (Kempener and Neumann
37   2014a,b). There are also concerns about technology maturity and performance; thus, not all have the
38   potential to become economically viable (IRENA 2020b).
39 Waste-to-Energy
40   Waste-to-energy (WTE) is a strategy to recover energy from waste in a form of consumable heat,
41   electricity, or fuel (Zhao et al. 2016). Thermal (incineration, gasification, and pyrolysis) and biological
42   (anaerobic digestion and landfill gas to energy) technologies are commonly used (Ahmad et al.
43   2020).When WTE technologies are equipped with proper air pollution reduction facilities they can
44   contribute to clean electricity production and reduction of GHG emissions. However, if not properly
45   operated, they can exacerbate air quality issues.
46   In 2019, there were more than 1,200 WTE incineration facilities worldwide, with estimated capacity of
47   310 million tons per year (UNECE 2020). It is estimated that treatment of a minimum of 261 million

     Do Not Cite, Quote or Distribute                    6-47                                   Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

 1   tons/year of waste could produce 283 TWh (1 EJ) of power and heat by 2022 (Awasthi et al., 2019).
 2   Incineration plants can reduce the mass of waste by 70%-80% and the volume of waste by 80%-90%
 3   (Haraguchi et al. 2019). Incineration technology can reduce water and soil pollution (Gu et al., 2019).
 4   However, if not properly handled, dust, and gases such as SO2, HCL, HF, NO2, and dioxins in the flue
 5   gases can harm the environment (Mutz et al. 2017). Anaerobic digestion technology has a positive
 6   environmental impact and the ability to reduce GHG emissions (Ayodele et al. 2018; Cudjoe et al.
 7   2020). The by-product of the anaerobic digestion process could be used as a nutrient-rich fertilizer for
 8   enhancing soil richness for agricultural purposes (Wainaina et al. 2020). Due to the potential negative
 9   impacts on domestic environment and residents’ health, WTE projects such as incineration encounter
10   substantial opposition from the local communities in which they are located (Ren et al., 2016; Baxter
11   et al., 2016). Therefore, for WTE to be deployed more widely, policies would need to be tailored with
12   specific guidelines focused on mitigating emissions, which may have adverse effect on the environment.
13   Depending on the origin of the waste used, the integration of WTE and carbon capture and storage
14   (CCS) could enable waste to be a net zero or even net negative emissions energy source (Kearns 2019;
15   Wienchol et al. 2020). For example, in Europe only, the integration of CCS with WTE facilities has the
16   potential to capture about 60 to 70 million tons of carbon dioxide annually (Tota et al. 2021).
17   Waste-to-energy is an expensive process compared to other energy sources such as fossil fuels and
18   natural gas (Mohammadi and Harjunkoski 2020). However, the environmental and economic benefits
19   make its high financial costs justifiable. In 2019, the global WTE market size was valued at USD 31
20   billion, and it is predicted to experience 7.4% annual growth until 2027 (UNECE 2020).
22   6.4.3 Energy System Integration
23   Greenhouse gases are emitted across all economic activities. Therefore, cost-effective decarbonization
24   requires a “system of systems” approach that considers the interaction between different energy sectors
25   and systems. Flexibility technologies and advanced control of integrated energy systems (e.g.,
26   considering the interaction between electricity, heating/cooling, gas/hydrogen, transport sectors) could
27   reduce energy infrastructure investments substantially in future low-carbon energy systems (Strbac et
28   al. 2015b; Jacobson et al. 2019)
29   The electricity grid will serve as a backbone of future low-carbon energy systems. Integration of large
30   amounts of VRE generation (Hansen et al. 2019), particularly wind and solar generation (Perez et al.
31   2019; Bistline and Young 2019), presents economic and technical challenges to electricity system
32   management across different timescales from sub-seconds, hours, days, seasons, to multiple years.
33   Furthermore, electrification of segments of the transport and heat sectors could disproportionately
34   increase peak demand relative to supply (Bistline et al. 2021). Increases in peak demand may require
35   reinforcing network infrastructures and generation in the historical passive system operation paradigm
36   (Strbac et al. 2020).
37   These challenges to electricity system management can be addressed through system integration and a
38   digitalized control paradigm involving advanced information and communication technologies. Real-
39   time maintenance of supply-demand balance and sufficient flexibility technologies such as electricity
40   storage, flexible demand, and grid forming converters (Strbac et al. 2015a; López Prol and Schill 2021)
41   would be increasingly valuable for incorporating larger amounts of VRE generation. This flexibility
42   will be particularly important to deal with sudden losses of supply, for example, due to a failure of a
43   large generator or interconnector or a rapid increase in demand (Teng et al. 2017; Chamorro et al. 2020).
44   The transition to a digitalized-based electricity system control paradigm would facilitate radical changes
45   in the security of supply, moving from the traditional approach of redundancy in assets to a smart control

     Do Not Cite, Quote or Distribute                  6-48                                  Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    paradigm. Advanced control and communication systems can significantly reduce the electricity system
2    investment and operation costs (2020; Münster et al. 2020; Harper et al. 2018).
3    Importance of cross-sector coupling for cost-effective energy system decarbonization
 4   Integrated whole-system approaches can reduce the costs of low-carbon energy system transitions (high
 5   confidence). A lack of flexibility in the electricity system may limit the cost-effective integration of
 6   technologies as part of broader net zero energy systems. At the same time, the enormous latent
 7   flexibility hidden in heating and cooling, hydrogen, transport, gas systems, and other energy systems
 8   provides opportunities to take advantage of synergies and to coordinate operations across systems
 9   (Martinez Cesena and Mancarella 2019; Zhang et al. 2018; Bogdanov et al. 2021; Pavičević et al. 2020;
10   Martin et al. 2017) (Figure 6.16).
11   Sector coupling can significantly increase system flexibility, driven by the application of advanced
12   technologies (Bogdanov et al. 2019; Solomon et al. 2019; Clegg and Mancarella 2016; Zhang and
13   Fujimori 2020; Zhao et al. 2021; Heinen et al. 2016; Zhang et al. 2019b). For example, district heating
14   infrastructure can generate both heat and power. Cooling systems and electrified heating systems in
15   buildings can provide flexibility through preheating and precooling via thermal energy storage Li, G.
16   et al. 2017; Li, Z. et al. 2016).. System balancing services can be provided by electric vehicles (EVs)
17   based on vehicle-to-grid concepts and deferred charging through smart control of EV batteries without
18   compromising customers’ requirements for transport (Aunedi and Strbac 2020).
19   Hydrogen production processes (power-to-gas and vice versa) and hydrogen storage can support short-
20   term and long-term balancing in the energy systems and enhance resilience (Stephen and Pierluigi 2016;
21   Strbac et al. 2020). However, the economic benefits of flexible power-to-gas plants, energy storage,
22   and other flexibility technological and options will depend on the locations of VRE sources, storage
23   sites, gas, hydrogen, and electricity networks (Jentsch et al. 2014; Heymann and Bessa 2015; Ameli et
24   al. 2020). Coordinated operation of gas and electricity systems can bring significant benefits in
25   supplying heat demands. For example, hybrid heating can eliminate investment in electricity
26   infrastructure reinforcement by switching to heat pumps in off-peak hours and gas boilers in peak hours
27   ( Dengiz et al. 2019; Fischer et al. 2017; Bistline et al. 2021). The heat required by direct air carbon
28   capture and storage (DACCS) could be effectively supplied by inherent heat energy in nuclear plants,
29   enhancing overall system efficiency (Realmonte et al. 2019).

31         Figure 6.16 Interaction between different energy sectors (extracted from Münster et al. 2020)

32   Rather than incremental planning, strategic energy system planning can help minimize long-term
33   mitigation costs (high confidence). With a whole-system perspective, integrated planning can consider
34   both short-term operation and long-term investment decisions, covering infrastructure from local to
35   national and international, while meeting security of supply requirements and incorporating the

     Do Not Cite, Quote or Distribute                  6-49                                  Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    flexibility provided by different technologies and advanced control strategies (Zhang et al. 2018;
2    O’Malley et al. 2020; Strbac et al. 2020). Management of conflicts and synergies between local district
3    and national level energy system objectives, including strategic investment in local hydrogen and heat
4    infrastructure, can drive significant whole-system cost savings (Fu et al. 2020; Zhang et al. 2019b). For
5    example, long-term planning of the offshore grid infrastructure to support offshore wind development,
6    including interconnection between different countries and regions, can provide significant savings
7    compared to a short-term incremental approach in which every offshore wind farm is individually
8    connected to the onshore grid (E3G 2021).
9   Role of flexibility technologies
10   Flexibility technologies - including energy storage, demand-side response, flexible/dispatchable
11   generation, grid forming converters, and transmission interconnection - as well as advanced control
12   systems, can facilitate cost-effective and secure low-carbon energy systems (high confidence).
13   Flexibility technologies have already been implemented, but they can be enhanced and deployed more
14   widely. Due to their interdependencies and similarities, there can be both synergies and conflicts for
15   utilizing these flexibility options (Bistline et al. 2021). It will therefore be important to coordinate the
16   deployment of the potential flexibility technologies and smart control strategies. Important electricity
17   system flexibility options include the following:
18   •   Flexible/dispatchable generation: Advances in generation technologies, for example, gas/hydrogen
19       plants and nuclear plants, can enable them to provide flexibility services. These technologies would
20       start more quickly, operate at lower power output, and make faster output changes, enabling more
21       secure and cost-effective integration of VRE generation and end-use electrification. There are
22       already important developments in increasing nuclear plants flexibility (e.g., in France (Office of
23       Nuclear Energy 2021)) and the development of small modular reactors, which could support system
24       balancing (FTI Consulting 2018).
25   •   Grid-forming converters (inverters): The transition from conventional electricity generation,
26       applying mainly synchronous machines to inverter-dominated renewable generation, creates
27       significant operating challenges. These challenges are mainly associated with reduced synchronous
28       inertia, system stability, and black start capability. Grid-forming converters will be a cornerstone
29       for the control of future electricity systems dominated by VRE generation. These converters will
30       address critical stability challenges, including the lack of system inertia, frequency and voltage
31       regulation, and black-start services while reducing or eliminating the need to operate conventional
32       generation (Tayyebi et al. 2019).
33   •   Interconnection: Electricity interconnections between different regions can facilitate more cost-
34       effective renewable electricity deployment. Interconnection can enable large-scale sharing of
35       energy and provide balancing services. Back-up energy carriers beyond electricity, such as
36       ammonia, can be shared through gas/ammonia/hydrogen-based interconnections, strengthening
37       temporal coupling of multiple sectors in different regions (Bhagwat et al. 2017; Brown et al. 2018)
38       (Section 6.4.5).
39   •   Demand-side response: Demand-side schemes – including, for example, smart appliances, EVs,
40       and building-based thermal energy storage (Heleno et al. 2014) – can provide flexibility services
41       across multiple time frames and systems. Through differentiation between essential and non-
42       essential needs during emergency conditions, smart control of demands can significantly enhance
43       system resilience (Chaffey 2016).
44   •   Energy storage: Energy storage technologies (Section 6.4.4) can act as both demand and generation
45       sources. They can provide services such as system balancing, various ancillary services, and
46       network management. Long-duration energy storage can significantly enhance the utilization of
47       renewable energy sources and reduce the need for firm low-carbon generation (Sepulveda et al.
48       2021).

     Do Not Cite, Quote or Distribute                   6-50                                   Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1   Role of digitalization and advanced control systems
 2   A digitalized energy system can significantly reduce energy infrastructure investments while enhancing
 3   supply security and resilience (high confidence) (Andoni et al. 2019; Strbac et al. 2020). Significant
 4   progress has been made in the development of technologies essential for the transition to a digitalized
 5   energy control paradigm, although the full implementation is still under development. Electrification
 6   and the increased integration of the electricity system with other systems will fundamentally transform
 7   the operational and planning paradigm of future energy infrastructure. A fully intelligent and
 8   sophisticated coordination of the multiple systems through smart control will support this paradigm
 9   shift. This shift will provide significant savings through better utilization of existing infrastructure
10   locally, regionally, nationally, and internationally. Supply system reliability will be enhanced through
11   advanced control of local infrastructure (Strbac et al. 2015a). Furthermore, this paradigm shift offers
12   the potential to increase energy efficiency through a combination of technologies that gather and analyse
13   data and consequently optimize energy use in real-time.
14   The transition to advanced data-driven control of energy system operations (Sun et al. 2019a; Cremer
15   et al. 2019) will require advanced information and communication technologies and infrastructure,
16   including the internet, wireless networks, computers, software, middleware, smart sensors, internet of
17   things components, and dedicated technological developments (Hossein Motlagh et al. 2020). The
18   transition will raise standardization and cybersecurity issues, given that digitalization can become a
19   single point of failure for the complete system (Unsal et al. 2021; Ustun and Hussain 2019).
20   Implementing peer-to-peer energy trading based on blockchain is expected to be one of the key elements
21   of next-generation electricity systems (Qiu et al. 2021). This trading will enable consumers to drive
22   system operation and future design, increasing overall system efficiency and security of supply while
23   reducing emissions without sacrificing users’ privacy (Andoni et al. 2019; Ahl et al. 2020). When
24   deployed with smart contracts, this concept will be suitable for energy systems involving many
25   participants, where a prerequisite is digitalization (e.g., smart meters, end-use demand control systems)
26   (Teufel et al. 2019; Juhar and Khaled 2018).
27   System benefits of flexibility technologies and advanced control systems
28   New sources of flexibility and advanced control systems provide a significant opportunity to reduce
29   low-carbon energy system costs by enhancing operating efficiency and reducing energy infrastructure
30   and low-carbon generation investments, while continuing to meet security requirements (high
31   confidence). In the U.S, for example, one study found that flexibility in buildings alone could reduce
32   U.S. CO2 emissions by 80 MT yr-1 and save USD 18 bn yr-1 in electricity system costs by 2030
33   (Satchwell et al. 2021). Key means for creating savings are associated with the following:
34   •   Efficient energy system operation: Flexibility technologies such as storage, demand-side response,
35       interconnection, and cross-system control will enable more efficient, real-time demand and supply
36       balancing. This balancing has historically been provided by conventional fossil-fuel generation
37       (Nuytten et al. 2013).
38   •   Savings in investment in low carbon/renewable generation capacity: System flexibility sources can
39       absorb or export surplus electricity, thus reducing or avoiding energy curtailment and reducing the
40       need for firm low-carbon capacity such as nuclear and fossil-fuel plants with CCS (Newbery et al.
41       2013; Solomon et al. 2019). For example, one study found that flexibility technologies and
42       advanced control systems could reduce the need for nuclear power by 14 GW and offshore wind by
43       20 GW in the UK’s low-carbon transition (Strbac et al. 2015b).
44   •   Reduced need for backup capacity: System flexibility can reduce energy demand peaks, reducing
45       the required generation capacity to maintain the security of supply, producing significant savings
46       in generation investments (Strbac et al. 2020).
47   •   Deferral or avoidance of electricity network reinforcement/addition: Flexibility technologies
48       supported by advanced control systems can provide significant savings in investment in electricity
     Do Not Cite, Quote or Distribute                  6-51                                 Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

1        network reinforcement that might emerge from increased demand, for example, driven by
2        electrification of transport and heat sectors. Historical network planning and operation standards
3        are being revised considering alternative flexibility technologies, which would further support cost-
4        effective integration of decarbonized transport and heat sectors (Strbac et al. 2020).

5    6.4.4 Energy Storage for Low-Carbon Grids
 6   Energy storage technologies make low carbon electricity systems more cost-effective, allowing VRE
 7   technologies to replace more expensive firm low carbon generation technologies (Carbon Trust 2016)
 8   and reducing investment costs in backup generation, interconnection, transmission, and distribution
 9   network upgrades (high confidence). Energy system decarbonization relies on increased electrification
10   (Section Meeting increasing demands with variable renewable sources presents challenges
11   and could lead to costly infrastructure reinforcements. Energy storage enables electricity from variable
12   renewables to be matched against evolving demands across both time and space, using short-, medium-
13   and long-term storage of excess energy for delivery later or different location. In 2017, an estimated
14   4.67 TWh (0.017 EJ) of electricity storage was in operation globally (IRENA 2017). If the integration
15   of renewables is doubled from 2014 levels by 2030, the total capacity of global electricity storage could
16   triple, reaching 11.89–15.27 TWh (0.043–0.055 EJ)(IRENA 2017b).
18       Table 6.5 Suitability of low carbon energy storage technologies, in terms of the grid services they can
19           provide, and overall features such as technology maturity, where Low represents an emerging
20   technology; Med represents a maturing technology and High a fully mature technology. The opportunity
21    for the cost of a technology to reduce over the next decade is represented by Low, Med and High and the
22      lifetime of installations by: Long, for projects lasting more than 25 years; Med for those lasting 15–25
23          years; Short, for those lasting less than 15 years. (PHS - Pumped Hydroelectric Storage, CAES -
24     Compressed Air Energy Storage, LAES - Liquid Air Energy Storage, TES - Thermal Energy Storage,
25        FES - Flywheel Energy Storage, LiB – Li-ion Batteries, Scap – Supercapacitors, RFB - Redox Flow
26    Batteries, RHFC - Reversible Hydrogen Fuel Cells, PtX – Power to fuels). [Footnote: References: PHS –
27       IRENA 2017, Barbour et al. 2016, Yang 2016; CAES – Brandon et al. 2015, IRENA 2017, Luo et al.
28   2014; LAES – Luo et al. 2014, Highview 2019; TES – Brandon et al. 2015, Smallbone et al. 2017, Gallo et
29      al. 2016; FES – Yulong et al. 2017, IRENA 2017; LiB – IRENA 2017, Hammond and Hazeldine 2015,
30     Staffell, I. and Rustomji, M. et al. 2016, Schmidt et al. 2017c,Nykvist and Nilsson 2015, May et al. 2018,
31   IRENA 2015b; Scap – Brandon et al. 2015, Gur 2018; RFB – IRENA 2017; RHFC – Gur 2018, IEA 2015]

      Suitability factor     PHS      CAES      LAES      TES      FES     LiB      Scap    RFB      PtX      RHF
      Upgrade deferral
      Energy Arbitrage
      Capacity firming
      Seasonal storage
      Frequency regulation
      Voltage support
      Black start
      Short term reserve
      Fast reserve
      power supply
      Maturity               High     High      Med       Low      High    Med      Low     Low      Low      Low

     Do Not Cite, Quote or Distribute                   6-52                                   Total pages: 217
     Final Government Distribution                     Chapter 6                             IPCC WGIII AR6

      Opportunity to reduce   Low     Low       Low        Med     Med     High    High    High     Med     High
      Lifetime                Long    Long      Long       Long    Med     Short   Med     Med      Med     Short
      Roundtrip Efficiency    60–     30–       55–        70–     90%     >95%    >95     80–      35–     <30%
                              80%     60%       90%        80%                     %       90%      60%
 2   Energy storage technologies can provide a range of different grid services (Table 6.5). Energy storage
 3   enhances security of supply by providing real time system regulation services (voltage support,
 4   frequency regulation, fast reserve, and short-term reserve). A greater proportion of variable renewable
 5   sources reduces system inertia, requiring more urgent responses to changes in system frequency, which
 6   rapid response storage technologies can provide (stability requires responses within sub second
 7   timescale for provision of frequency and voltage control services). Energy storage also provides
 8   intermittent renewable sources with flexibility, allowing them to contribute a greater proportion of
 9   electrical energy and avoiding curtailment (capacity firming). Investment costs in backup generation,
10   interconnection, transmission, and distribution network upgrades can thus be reduced (upgrade
11   deferral), meaning that less low carbon generation will need to be built while still reducing emissions.
12   In the event of an outage, energy storage reserves can keep critical services running (islanding) and
13   restart the grid (black start). The ability to store and release energy as required provides a range of
14   market opportunities for buying and selling of energy (arbitrage).
15   No single, sufficiently mature energy storage technology can provide all the required grid services - a
16   portfolio of complementary technologies working together can provide the optimum solution (high
17   confidence). Different energy storage technologies can provide these services and support cost-effective
18   energy system decarbonization (Carbon Trust 2016). To achieve very low carbon systems, significant
19   volumes of storage will be required (Strbac et al. 2015a; Section 0). There are few mature global supply
20   chains for many of the less-developed energy storage technologies. This means that although costs
21   today may be relatively high, there are significant opportunities for future cost reductions, both through
22   technology innovation and through manufacturing scale. Adding significant amounts of storage will
23   reduce the price variation and, therefore, the profitability of additional and existing storage, increasing
24   investment risk.
25   Energy storage extends beyond electricity storage and includes technologies that can store energy as
26   heat, cold, and both liquid and gaseous fuels. Energy storage is a conversion technology, enabling
27   energy to be converted from one form to another. This diversification improves the overall resilience
28   of energy systems, with each system being able to cover supply shortfalls in the others. For example,
29   storage can support the electrification of heating or cooling, as well as transport through electric
30   vehicles, powered by batteries or by fuel cells. Storage significantly reduces the need for costly
31   reinforcement of local distribution networks through smart charging schemes and the ability to flow
32   electricity back to the grid (e.g., through vehicle-to-grid). By capturing otherwise wasted energy
33   streams, such as heat or cold, energy storage improves the efficiency of many systems, such as
34   buildings, data centres and industrial processes.
35   Energy Storage Technologies
36   Pumped Hydroelectric Storage (PHS). PHS makes use of gravitational potential energy, using water as
37   the medium. Water is pumped into an elevated reservoir using off-peak electricity and stored for later
38   release when electricity is needed. These closed-loop hydropower plants have been in use for decades
39   and account for 97% of worldwide electricity storage capacity (IEA, 2018b; IRENA, 2017). PHS is best
40   suited to balancing daily energy needs at a large scale, and advances in the technology now allow both
41   rapid response and power regulation in both generating and pumping mode (Valavi and Nysveen 2018;
42   Kougias et al. 2019; Dong et al. 2019). The construction itself can cause disruption to the local
43   community and environment (Hayes et al. 2019), the initial investment is costly, and extended
     Do Not Cite, Quote or Distribute                  6-53                                 Total pages: 217
     Final Government Distribution                     Chapter 6                               IPCC WGIII AR6

1    construction periods delay return on investment (Section In addition, locations for large-scale
2    PHS plants are limited.
 3   Advanced pump-turbines are being developed, allowing both reversible and variable-speed operation,
 4   supporting frequency control and grid stability with improved round-trip efficiencies (Ardizzon et al.
 5   2014). New possibilities are being explored for small-scale PHS installations and expanding the
 6   potential for siting (Kougias et al. 2019). For example, in underwater PHS, the upper reservoir is the
 7   sea, and the lower is a hollow deposit at the seabed. Seawater is pumped out of the deposit to store off-
 8   peak energy and re-enters through turbines to recharge it (Kougias et al. 2019). Using a similar concept,
 9   underground siting in abandoned mines and caverns could be developed reasonably quickly (IEA
10   2020h). Storage of energy as gravitational potential can also be implemented using materials other than
11   water, such as rocks and sand. Pumped technology is a mature technology (Barbour et al. 2016; Rehman
12   et al. 2015) and can be important in supporting the transition to future low carbon electricity grids (IHA
13   2021).
14   Batteries. There are many types of batteries, all having unique features and suitability (c), but their key
15   feature is their rapid response times. A rechargeable battery cell is charged by using electricity to drive
16   ions from one electrode to another, with the reverse occurring on discharge, producing a usable electric
17   current (Crabtree et al. 2015). While lead-acid batteries (LABs) have been widely used for automotive
18   and grid applications for decades (May et al. 2018), li-ion batteries (LIBs) are increasingly being used
19   in grid-scale projects (Crabtree et al. 2015), displacing LABs. The rapid response time of batteries
20   makes them suitable for enhanced frequency regulation and voltage support, enabling the integration of
21   variable renewables into electricity grids (Strbac and Aunedi 2016). Batteries can provide almost all
22   electricity services, except for seasonal storage. Lithium-ion batteries, in particular, can store energy
23   and power in small volumes and with low weight, making them the default choice for EVs (Placke et
24   al. 2017). EV batteries are expected to form a distributed storage resource as this market grows, both
25   impacting and supporting the grid (Staffell and Rustomji 2016).
27   Table 6.6 Technical characteristics of a selected range of battery chemistries, categorized as those which
28   precede LIBs (white background), LIBs (yellow background) and post LIBs (blue background). With the
29    exception of the All Solid-State batteries, all use liquid electrolytes. (1 =Mahmoudzadeh et al. 2017; 2 =
30   Manzetti and Mariasiu 2015; 3 =Placke et al. 2017; 4 = Nykvist and Nilsson 2015; 5 =Cano et al. 2018; 6 =
31   BloombergENF 2019; 7 = You and Manthiram 2017; 8 = Fotouhi et al. 2017; 9 = IRENA 2017; 10 = Yang
32                                                     et al., 2020)

      Battery Type               Technology Maturity        Life     Span   Energy         Specific     Price (USD
                                                            (Cycles)        Density        Energy       kWh-1)   in
                                                                            (Wh L-1)       (Wh kg-1)    2017
      Lead Acid                  High                       300–800 5       102–106 5      38–60 5      70–160 5
      Ni MH                      High                       600–1200 5      220–250 5      42–110 5     210–365 5
      Ni Cd                      High                       1350 2          100 2          60 2         700
      High-temperature      Na   High                       1000 5          150–280 8      80–120 1     315–490 8
      LIB state of the art       High                       1000–6000 5     200–680 3      110–250 3    176 6
      LIB energy-optimized       Under Development                          600–850 3      300-440 3
      Classic Li Metal (CLIM)    Under Development                          800–1050 3     420–530 3
      Metal Sulfur (Li S)        Near Commercialization     100–500 5       350–680 3, 8   360–560 3,   36–130 5

      Metal Sulfur (Na S)        Under Development          5000–10,000

      Metal Air (Li/air)         Under Development          20–100 5                       470–900 4    70–200 5

     Do Not Cite, Quote or Distribute                    6-54                                   Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

      Metal Air (Zn/air)         Under Development          150–450 5                      200–410 4    70–160 5
      Na ion                     Under Development          500 7                          600 7
      All-Solid-State            Under Development                                         278–479 3
      Redox                      Under Development          >12,000–        15–2510        10–2010      6610
                                                            14,000 10
2    Drawbacks of batteries include relatively short lifespans and the use of hazardous or costly materials in
3    some variants. While LIB costs are decreasing (Schmidt et al. 2017; Vartiainen et al. 2020), the risk of
4    thermal runaway, which could ignite a fire (Gur 2018; Wang et al. 2019a), concerns about long-term
5    resource availability (Sun et al. 2017; Olivetti et al. 2017), and concerns about global cradle-to-grave
6    impacts (Peters et al. 2017; Kallitsis et al. 2020) need to be addressed.
 7   The superior characteristics of LIBs will keep them the dominant choice for EV and grid applications
 8   in the medium-term (high confidence). There are, however, several next-generation battery chemistries
 9   (Placke et al. 2017), which show promise (high confidence). Cost reductions through economies of scale
10   are a key area for development. Extending the life of the battery can bring down overall costs and
11   mitigate the environmental impacts (Peters et al. 2017). Understanding and controlling battery
12   degradation is therefore important. The liquid, air-reactive electrolytes of conventional LIBs are the
13   main source of their safety issues (Gur 2018; Janek and Zeier 2016), so All-Solid-State Batteries, in
14   which the electrolyte is a solid, stable material, are being developed. They are expected to be safe, be
15   durable, and have higher energy densities (Janek and Zeier 2016). New chemistries and concepts are
16   being explored, such as lithium-sulfur batteries to achieve even higher energy densities (Van Noorden
17   2014; Blomgren 2017) and sodium chemistries because sodium is more abundant than lithium (Hwang
18   et al. 2017). Cost-effective recycling of batteries will address many sustainability issues and prevent
19   hazardous and wasteful disposal of used batteries (Harper et al. 2019). Post-LIB chemistries include
20   metal sulfur, metal-air, metal ion (besides Li) and All-Solid-State Batteries.
21   Compressed Air Energy Storage (CAES). With CAES, off-peak electricity is used to compress air in a
22   reservoir – either in salt caverns for large scale or in high-pressure tanks for smaller-scale installations.
23   The air is later released to generate electricity. While conventional CAES has used natural gas to power
24   compression, new low carbon CAES technologies, such as isothermal or adiabatic CAES, control
25   thermal losses during compression and expansion (Wang et al. 2017c). Fast responses and higher
26   efficiencies occur in small-scale CAES installations, scalable to suit the application as a distributed
27   energy store, offering a flexible, low maintenance alternative (Luo et al. 2014; Venkataramani et al.
28   2016).
29   CAES is a mature technology in use since the 1970s. Although CAES technologies have been
30   developed, there are not many installations at present (Blanc et al. 2020; Wang et al. 2017b). While the
31   opportunities for CAES are significant, with a global geological storage potential of about 6.5 PW
32   (Aghahosseini and Breyer 2018), a significant amount of initial investment is required. Higher
33   efficiencies and energy densities can be achieved by exploiting the hydrostatic pressure of deep water
34   to compress air within submersible reservoirs (Pimm et al. 2014). CAES is best suited to bulk diurnal
35   electricity storage for buffering VRE sources and services, which do not need a very rapid response. In
36   contrast to PHS, CAES has far more siting options and poses few environmental impacts.
37   Liquid Air Energy Storage (LAES). Liquid air energy storage uses electricity to liquefy air by cooling
38   it to –196°C and storing it in this condensed form (largely liquid nitrogen) in large, insulated tanks. To
39   release electricity, the ‘liquid air’ is evaporated through heating, expanding to drive gas turbines. Low-
40   grade waste heat can be utilized, providing opportunities for integrating with industrial processes to
41   increase system efficiency. There are clear, exploitable synergies with the existing liquid gas
42   infrastructure (Peters and Sievert 2016).

     Do Not Cite, Quote or Distribute                    6-55                                   Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

1    LAES provides bulk daily storage of electricity, with the additional advantage of being able to capture
2    waste heat from industrial processes. This technology is in the early commercial stage (Regen 2017;
3    Brandon et al. 2015). Advances in whole systems integration can be developed to integrate LAES with
4    industrial processes, making use of their waste heat streams. LAES uniquely removes contaminants in
5    the air and could potentially incorporate CO2 capture (Taylor et al. 2012).
 6   Thermal Energy Storage (TES). Thermal energy storage refers to a range of technologies exploiting the
 7   ability of materials to absorb and store heat or cold, either within the same phase (sensible TES), through
 8   phase changes (latent TES), or through reversible chemical reactions (thermochemical TES). Pumped
 9   Thermal Energy Storage (PTES), a hybrid form of TES, is an air-driven electricity storage technology
10   storing both heat and cold in gravel beds, using a reversible heat-pump system to maintain the
11   temperature difference between the two beds and gas compression to generate and transfer heat (Regen
12   2017). TES technologies can store both heat and cold energy for long periods, for example in
13   underground water reservoirs for balancing between seasons (Tian et al. 2019; Dahash et al. 2019),
14   storing heat and cold to balance daily and seasonal temperatures in buildings and reducing heat build-
15   up in applications generating excessive waste heat, such as data centres and underground operations.
16   TES can be much cheaper than batteries and has the unique ability to capture and reuse waste heat and
17   cold, enabling the efficiency of many industrial, buildings, and domestic processes to be greatly
18   improved (high confidence). Integration of TES into energy systems is particularly important, as the
19   global demand for cooling is expected to grow (Peters and Sievert 2016; Elzinga et al. 2014). Sensible
20   TES is well developed and widely used; latent TES is less developed with few applications.
21   Thermochemical TES is the least developed, with no application yet (Prieto et al. 2016; Clark et al.
22   2020). The potential for high-density storage of industrial heat for long periods in thermochemical TES
23   (Brandon et al. 2015) is high, with energy densities comparable to that of batteries (Taylor et al. 2012),
24   but material costs are currently prohibitive, ranging from hundreds to thousands of dollars per tonne.
25   Flywheel Energy Storage (FES). Flywheels are charged by accelerating a rotor/flywheel. Energy is
26   stored in the spinning rotor’s inertia which is only decelerated by friction (minimized by magnetic
27   bearings in a vacuum), or by contact with a mechanical, electric motor. They can reach full charge very
28   rapidly, their state of charge can be easily determined (Amiryar and Pullen 2017), and they operate over
29   a wide range of temperatures. While they are more expensive to install than batteries and
30   supercapacitors, they last a long time and are best suited to stationary grid storage, providing high power
31   for short periods (minutes). Flywheels can be used in vehicles, but not as the primary energy source.
32   Flywheels are a relatively mature storage technology but not widely used, despite their many advantages
33   over electrochemical storage (Dragoni 2017). Conventional flywheels require costly, high tensile
34   strength materials, but high-energy flywheels, using lightweight rotor materials, are being developed
35   (Amiryar and Pullen 2017; Hedlund et al. 2015).
36   Supercapacitors, aka Ultracapacitors or Double Layer Capacitors (Scap). Supercapacitors consist of a
37   porous separator sandwiched between two electrodes, immersed in a liquid electrolyte (Gur 2018).
38   When a voltage is applied across the electrodes, ions in the electrolyte form electric double layers at the
39   electrode surfaces, held by electrostatic forces. This structure forms a capacitor, storing electrical charge
40   (Lin et al. 2017; Brandon et al. 2015) and can operate from –40°C to 65°C.
41   Supercapacitors can supply high peaks of power very rapidly for short periods (seconds up to minutes)
42   and are able to fulfil the grid requirements for frequency regulation, but they would need to be
43   hybridized with batteries for automotive applications. Their commercial status is limited by costly
44   materials and additional power electronics required to stabilize their output (Brandon et al. 2015).
45   Progress in this area includes the development of high energy supercapacitors, LIB-supercapacitor
46   devices (Gonzalez et al. 2016), and cheaper materials (Wang et al. 2017a), all providing the potential

     Do Not Cite, Quote or Distribute                    6-56                                   Total pages: 217
     Final Government Distribution                    Chapter 6                               IPCC WGIII AR6

1    to improve the economic case for supercapacitors, either by reducing manufacturing costs or extending
2    their service portfolio.
 3   Redox Flow Batteries (RFB). Redox flow batteries use two separate electrolyte solutions, usually
 4   liquids, but solid or gaseous forms may also be involved, stored in separate tanks, and pumped over or
 5   through electrode stacks during charge and discharge, with an ion-conducting membrane separating the
 6   liquids. The larger the tank, the greater the energy storage capacity, whereas more and larger cells in
 7   the stack increase the power of the flow battery. This decoupling of energy from power enables RFB
 8   installations to be uniquely tailored to suit the requirements of any given application. There are two
 9   commercially available types today: vanadium and zinc bromide, and both operate at near ambient
10   temperatures, incurring minimal operational costs.
11   RFBs respond rapidly and can perform all the same services as LIBs, except for onboard electricity for
12   EVs. Lower cost chemistries are emerging, to enable cost-effective bulk energy storage (Brandon et al.
13   2015). A new membrane-free design eliminates the need for a separator and also halves the system
14   requirements, as the chemical reactions can coexist in a single electrolyte solution (Navalpotro et al.
15   2017; Arenas et al. 2018).
16   Power to fuels (PtX). (see also Section The process of using electricity to generate a gaseous
17   fuel, such as hydrogen or ammonia, is termed power-to-gas (PtG/P2G) (IEA 2020h). When injected
18   into the existing gas infrastructure (section 6.4.5), it has the added benefit of decarbonizing gas
19   (Brandon et al. 2015). Electricity can be used to generate hydrogen, which is then converted back into
20   electricity using combined-cycle gas turbines that have been converted to run on hydrogen. For greater
21   compatibility with existing gas systems and appliances, the hydrogen can combined with captured
22   carbon dioxide to form methane and other synthetic fuels (Thema et al. 2019), however methane has
23   high global warming potential and its supply chain emissions have been found to be significant
24   (Balcombe et al. 2013).
25   PtX can provide all required grid services, depending on how it is integrated. However, a significant
26   amount of PtX is required for storage to produce electricity again (Bogdanov et al. 2019) due to the low
27   roundtrip efficiency of converting electricity to fuel and back again. However, portable fuels (hydrogen,
28   methane, ammonia, synthetic hydrocarbons) are useful in certain applications, for example in energy
29   systems lacking the potential for renewables, and the high energy density of chemical storage is
30   essential for more demanding applications, such as transporting heavy goods and heating or cooling
31   buildings (IEA 2020h). Research into more efficient and flexible electrolyzers which last longer and
32   cost less is needed (Brandon et al. 2015).
33   Hydrogen and Reversible Hydrogen Fuel Cells (H/RHFC). Hydrogen is a flexible fuel with diverse
34   uses, capable of providing electricity, heat, and long-term energy storage for grids, industry, and
35   transport, and has been widely used industrially for decades (Section 0). Hydrogen can be produced in
36   various ways and stored in significant quantities in geological formations at moderate pressures, often
37   for long periods, providing seasonal storage (Gabrielli et al. 2020). A core and emerging
38   implementation of PtX is hydrogen production through electrolyzers. Hydrogen is a carbon-free fuel
39   holding three times the energy held by an equivalent mass of gasoline but occupying a larger volume.
40   An electrolyzer uses excess electricity to split water into hydrogen and oxygen through the process of
41   electrolysis. A fuel cell performs the reverse process of recombining hydrogen and oxygen back into
42   water, converting chemical energy into electricity (Elzinga et al. 2014). Reversible hydrogen fuel cells
43   (RHFCs) can perform both functions in a single device, however they are still in the pre-commercial
44   stage, due to prohibitive production costs.
45   Hydrogen can play an important role in reducing emissions and has been shown to be the most cost-
46   effective option in some cases, as it builds on existing systems (Staffell et al. 2018). Fuel cell costs need
47   to be reduced and the harmonies between hydrogen and complementary technologies, such as batteries,

     Do Not Cite, Quote or Distribute                    6-57                                   Total pages: 217
     Final Government Distribution                    Chapter 6                                IPCC WGIII AR6

1    for specific applications need to be explored further. Hydrogen can provide long duration storage to
2    deal with prolonged extreme events, such as very low output of wind generation, to support resilience
3    of future low carbon energy systems. Research in this technology focuses on improving roundtrip
4    efficiencies, which can be as high as 80% with recycled waste heat and in high-pressure electrolyzers,
5    incorporating more efficient compression (Matos et al. 2019). Photo-electrolysis uses solar energy to
6    directly generate hydrogen from water (Amirante et al. 2017).
7   Societal Dimensions of Energy Storage
 8   Public awareness knowledge about electricity storage technologies, their current state, and their
 9   potential role in future energy systems is limited (Jones et al. 2018). For instance, people do not perceive
10   energy system flexibility and storage as a significant issue, or assume storage is already taking place.
11   Public perceptions differ across storage technologies. Hydrogen is considered a modern and clean
12   technology, but people also have safety concerns. Moreover, the public is uncertain about hydrogen
13   storage size and the possibility of storing hydrogen in or near residential areas (Eitan and Fischhendler
14   2021). Battery storage both on the household and community level was perceived as slightly positive
15   in one study in the UK (Ambrosio-Albala et al. 2020). However, financial costs are seen as a main
16   barrier. The potential of electric vehicle batteries to function as flexible storage is limited by the current
17   numbers of EV owners and concerns that one’s car battery might not be fully loaded when needed.
18   6.4.5 Energy Transport and Transmission
19   The linkage between energy supply and distribution, on the one hand, and energy use on the other is
20   facilitated by various mechanisms for transporting energy. As the energy system evolves, the way that
21   energy is transported will also evolve.
22   Hydrogen: Low-Carbon Energy Fuel
23   Hydrogen is a promising energy carrier for a decarbonized world (Box 6.9). It can be utilized for
24   electricity, heat, transport, industrial demand, and energy storage (Abdin et al. 2020). In low-carbon
25   energy systems, hydrogen is expected to be utilized in applications that are not as amenable to
26   electrification, such as a fuel for heavy-duty road transport and shipping, or as a chemical feedstock
27   (Griffiths et al. 2021; Schemme et al. 2017). Hydrogen could also provide low-carbon heat for industrial
28   processes or be utilized for direct reduction of iron ore (Vogl et al. 2018). Hydrogen could replace
29   natural gas-based electricity generation (do Sacramento et al. 2013) in certain regions and support the
30   integration of variable renewables into electricity systems by providing a means of long-term electricity
31   storage. Hydrogen-based carriers, such as ammonia and synthetic hydrocarbons, can likewise be used
32   in energy-intensive industries and the transport sector (Schemme et al. 2017; IRENA 2019b) (e.g.,
33   synthetic fuels for aviation). These hydrogen-based energy carriers are easier to store than hydrogen.
34   At present hydrogen has limited applications – mainly being produced onsite for the creation of
35   methanol and ammonia (IEA 2019c), as well as in refineries.
36   Low- or zero-carbon produced hydrogen is not currently competitive for large-scale applications, but it
37   is likely to have a significant role in future energy systems, due to its wide-range of applications (high
38   confidence). Key challenges for hydrogen are: (a) cost-effective low/zero carbon production, (b)
39   delivery infrastructure cost, (c) land area (i.e., ‘footprint’) requirements of hydrogen pipelines,
40   compressor stations, and other infrastructure, (d) challenges in using existing pipeline infrastructure,
41   (e) maintaining hydrogen purity, (e) minimizing hydrogen leakage, and (f) the cost and performance of
42   end-uses. Furthermore, it is necessary to consider the public perception and social acceptance of
43   hydrogen technologies and their related infrastructure requirements (Scott and Powells 2020; Iribarren
44   et al. 2016)
45   Hydrogen Production. Low- or zero-carbon hydrogen can be produced from multiple sources. While
46   there is no consensus on the hydrogen production spectrum, “blue” hydrogen (Goldmann and

     Do Not Cite, Quote or Distribute                    6-58                                    Total pages: 217
     Final Government Distribution                                         Chapter 6                                          IPCC WGIII AR6

 1   Dinkelacker 2018) generally refers to hydrogen produced from natural gas combined with CCS through
 2   processes such as steam methane reforming (SMR)(Sanusi and Mokheimer 2019) and advanced gas
 3   reforming (Zhou et al. 2020). Low-carbon hydrogen could also be produced from coal coupled with
 4   CCS (Hu et al. 2020) (Table 6.7). Current estimates are that adding CCS to produce hydrogen from
 5   SMR will add on average 50% on the capital cost, 10% to fuel, and 100% to operating costs. For coal
 6   gasification, CCS will add 5% to the capital and fuel costs and 130% to operating costs (IEA 2019d;
 7   Staffell et al. 2018). Further, biomass gasification could produce renewable hydrogen, and when joined
 8   with CCS could provide negative carbon emissions. “Green” hydrogen (Jaszczur et al. 2016) most often
 9   is referred to as hydrogen produced from zero-carbon electricity sources such as solar power and wind
10   power (Schmidt et al. 2017) (Table 6.8). Nuclear power could also provide clean hydrogen, via
11   electrolysis or thermochemical water splitting (EERE 2020). Hydrogen can even be produced pyrolysis
12   of methane (Sánchez-Bastardo et al. 2020), sometimes called as “turquoise” hydrogen, solar
13   thermochemical water splitting, biological hydrogen production (cyanobacteria) (Velazquez Abad and
14   Dodds 2017), and microbes that use light to make hydrogen (under research)(EIA 2020).
16        Table 6.7 Key performance and cost characteristics of different non-electric hydrogen production
17                                        technologies (including CCS)
18    (1) CSIRO 2021; (2) IEA 2020; (3) IRENA 2019; (4) Hydrogen Council 2020; (5) CCC 2018; (6) BEIS 2021; (7) Ishaq et al. 2021; (8) Al-Mahtani et al.
19                                                                2021; (9) IEA 2019

         Technology                                LHV Efficiency (%)               Carbon Intensity              Cost Estimates* (USD
                                                                                    (kgCO2 (kgH2)-1)                         (kgH2)-1)
                                                  Current        Long-term                                     Current             Long-term
         SMR                                     65(5)           74(5,6)            1.0–3.6(5,9)             1.0–                1.5–2.6(5)
         Advanced gas reforming                  -               81–84(5,6)         0.9–2.9(5)               1.3–2.1(5)          1.2–3.4(5,6)
         Hydrogen     from    coal               54(5)           54(5)              2.1–5.5(5,9)             1.8–                2.4–3.3(5)
         gasification                                                                                        3.1
         Hydrogen from biomass                   53.6(7)         40–60(5)              Potential to          4.9(5)              2.9–5.9(5,6)
         gasification                                                                   achieve-
20   *USD per GBP exchange rate: 0.72 (August 2021); LHV: Lower Heating Values; Long-term refers to 2040 and 2050 according to different
21   references

22         Table 6.8 Efficiency and cost characteristics of electrolysis technologies for hydrogen production
23      (1) CSIRO 2021; (2) IEA 2020; (3) IRENA 2019; (4) Hydrogen Council 2020; (5) CCC 2018; (6) BEIS 2021; (7) IEA 2019; (8) Christensen 2020

      Technology                              LHV Efficiency (%)                    CAPEX (USD kWe-1)                Cost Estimates*,+ (USD
                                        Current                Long-term            Current (7)     Long-           Current          Long-
                                                                                                    term (7)                         term
      Alkaline Electrolysers            58–77(1,2,5,6,8)       70–82                500–1400        200–700         2.3–6.9(1,2,3,5) 0.9–
      PEM                               54–72(1,2,5,6,8)       67–82                1100–           200–900         3.5–9.3(1,4,5,6) 2.2–
                                                                                    1800                                             7.2(5,6)
      SOEC                              74–81(2,6,8)           77–92                2800–           500–            4.2 (5)
                                                                                    5600            1000
24   *USD per GBP exchange rate: 0.72 (August 2021); + The cost of hydrogen production from electrolysers is highly dependent on the
25   technology, source of electricity, and operating hours, and here some values based on the assumptions made in the references are provided.

26   Hydrogen energy carriers. Hydrogen can be both an energy carrier itself, be converted further for into
27   other energy carriers (such as synthetic fuels) and be a means of transporting other sources of energy.
28   For example, hydrogen could be transported in its native gaseous form or liquified. Hydrogen can also
     Do Not Cite, Quote or Distribute                                        6-59                                               Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    be combined with carbon and transported as a synthetic hydrocarbons (Gumber and Gurumoorthy 2018)
2    (IRENA 2019d) as well as be transported via liquid organic hydrogen carriers (LOHCs) or ammonia
3    (IRENA 2019d). For synthetic hydrocarbons such as methane or methanol to be considered zero carbon,
4    the CO2 used to produce them would need to come from the atmosphere either directly through DACCS
5    or indirectly through BECCS (IRENA 2019b). LOHCs are organic substances in liquid or semi-solid
6    states, which store hydrogen based on reversible catalytic hydrogenation and de-hydrogenation of
7    carbon double bounds (Rao and Yoon 2020; Niermann et al. 2019). Hydrogen produced from
8    electrolysis could also be seen as an electricity energy carrier. This is an example of the PtX processes
9    (section 6.4.4), entailing the conversion of electricity to other energy carriers for subsequent use.
10   Ammonia is a promising cost-effective hydrogen carrier (Creutzig et al. 2019). Onsite generation of
11   hydrogen for the production of ammonia already occurs today, and the NH3 could be subsequently
12   “cracked” (with a 15–25% energy loss) to reproduce hydrogen (Bell and Torrente-Murciano 2016;
13   Hansgen et al. 2010; Montoya et al. 2015). Because the energy density of ammonia is 38% higher than
14   liquid hydrogen (Osman and Sgouridis 2018), it is potentially a suitable energy carrier for long-distance
15   transport and storage (Salmon et al. 2021) Moreover, ammonia is more easily condensable (liquefied at
16   0.8 MPa, 20°C), which provides economically viable hydrogen storage and supply systems. Ammonia
17   production and transport are also established industrial processes (∼180 MMT yr-1 (Valera-Medina et
18   al. 2017), and hence ammonia is considered to be a scalable and cost-effective hydrogen-based energy
19   carrier. At present, most ammonia is used in fertilizers (~80%), followed by many industrial processes,
20   such as the manufacturing of mining explosives and petrochemicals (Jiao and Xu 2018). In contrast to
21   hydrogen, ammonia can be used directly as a fuel without any phase change for internal combustion
22   engines, gas turbines, and industrial furnaces (Kobayashi et al. 2019). Ammonia can also be used in
23   low and high temperature fuel cells (Lan and Tao 2014), whereby both electricity and hydrogen can be
24   produced without any NOx emissions. Furthermore, ammonia provides the flexibility to be
25   dehydrogenated for hydrogen-use purposes. Ammonia is considered a carbon-free sustainable fuel for
26   electricity generation, since in a complete combustion, only water and nitrogen are produced (Valera-
27   Medina et al. 2017). Like hydrogen, ammonia could facilitate management of variable RES, due to its
28   cost-effective grid-scale energy storage capabilities. In this regard, production of ammonia via hydrogen
29   from low- or zero-carbon generation technologies along with ammonia energy recovery technologies
30   (Afif et al. 2016) could play a major role in forming a hydrogen and/or ammonia economy to support
31   decarbonization. However, there are serious concerns regarding the ability to safely use ammonia for
32   all these purposes, given its toxicity - whereas hydrogen is not considered toxic.
33   In general, challenges around hydrogen-based energy carriers - including safety issues around
34   flammability, toxicity, storage, and consumption - require new devices and techniques to facilitate their
35   large-scale use. Relatively high capital costs and large electricity requirements are also challenges for
36   technologies that produce hydrogen energy carriers. Yet, these energy carriers could become
37   economically viable through the availability of low-cost electricity generation and excess of renewable
38   energy production (Daiyan et al. 2020) A key challenge in use of ammonia is related to significant
39   amount of NOx emissions, which is released from nitrogen and oxygen combustion, and unburned
40   ammonia. Both have substantial air pollution risks, which can result in lung and other injuries, and can
41   reduce visibility (EPA 2001). Due to the low flammability of hydrogen energy carriers such as liquified
42   hydrogen (Nilsson et al. 2016) and ammonia (Li et al. 2018), a stable combustion (Zengel et al. 2020;
43   Lamas and Rodriguez 2019) in the existing gas turbines is not currently feasible. In recent
44   developments, however, the proportion of hydrogen in gas turbines has been successfully increased,
45   and further development of gas turbines may enable them to operate on 100% hydrogen by 2030 (Pflug
46   et al. 2019)
47   Long-Distance Hydrogen Transport. Hydrogen can allow regional integration and better utilization of
48   low- or zero-carbon energy sources (Box 6.9 and Box 6.10). Hydrogen produced from renewables or

     Do Not Cite, Quote or Distribute                  6-60                                 Total pages: 217
     Final Government Distribution                  Chapter 6                             IPCC WGIII AR6

1    other low-carbon sources in one location could be transported for use elsewhere (Philibert 2017; Ameli
2    et al. 2020). Depending on the distance to the user and specific energy carrier utilized (e.g., gaseous
3    hydrogen or LOHC), various hydrogen transport infrastructures, distribution systems, and storage
4    facilities would be required (Hansen 2020; Schönauer and Glanz 2021) (Figure 6.17).
 5   Hydrogen can be liquefied and transported at volume over the ocean without pressurization. This
 6   requires a temperature of –253ºC and is therefore energy-intensive and costly (Niermann et al. 2021).
 7   Once it reaches its destination, the hydrogen needs to be re-gasified, adding further cost. A
 8   demonstration project is under development exporting liquid hydrogen from Australia to Japan
 9   (Yamashita et al. 2019). Hydrogen could also be transported as ammonia by ocean in liquid form.
10   Ammonia is advantageous because it is easier to store than hydrogen (Zamfirescu and Dincer 2008;
11   Nam et al. 2018; Soloveichik 2016). Liquid ammonia requires temperatures below –33ºC and is
12   therefore more straightforward and less costly to transport than liquified hydrogen and even liquified
13   natural gas (Singh and Sahu 2018). A project exporting ammonia from Saudi Arabia to Japan is under
14   consideration (Nagashima 2018). LOHCs could also be used to transport hydrogen at ambient
15   temperature and pressure. This advantageous property of LOHCs makes them similar to oil products,
16   meaning they can be transported in existing oil infrastructure including oil tankers and tanks (Niermann
17   et al. 2019; IEA 2019). A project is under development to export hydrogen from Brunei to Japan using
18   LOHCs (Kurosaki 2018).

20      Figure 6.17 Hydrogen value chain. Hydrogen can be produced by various means and input and fuel
21    sources. These processes have different emissions implications. Hydrogen can be transported by various
22       means and in various forms, and it can be stored in bulk for longer-term use. It also has multiple
23                              potential end uses. CHP: Combined heat and power

24   Intra-Regional Hydrogen Transportation. Within a country or region, hydrogen would likely be
25   pressurized and delivered as compressed gas. About three times as much compressed hydrogen by
26   volume is required to supply the same amount of energy as natural gas. Security of supply is therefore
27   more challenging in hydrogen networks than in natural gas networks. Storing hydrogen in pipelines
28   (linepack) would be important to maintaining security of supply (Ameli et al. 2019, 2017). Due to the
29   physics of hydrogen, in most cases exiting gas infrastructure would need to be upgraded to transport
30   hydrogen. Transporting hydrogen in medium- or high-pressure networks most often would require
31   reinforcements in compressor stations and pipeline construction routes (Dohi et al. 2016). There are
32   several recent examples of efforts to transport hydrogen by pipeline. For example, in the Iron Mains
33   Replacement Programme in the UK, the existing low pressure gas distribution pipes are being converted

     Do Not Cite, Quote or Distribute                 6-61                                  Total pages: 217
     Final Government Distribution                  Chapter 6                               IPCC WGIII AR6

1    from iron to plastic (Committee on Climate Change 2018). In the Netherlands, an existing low-pressure
2    12 km natural gas pipeline has been used for transporting hydrogen (Dohi et al. 2016).
3    To bypass gas infrastructure in transporting hydrogen, methane can be transported using the existing
4    gas infrastructure, while hydrogen can be produced close to the demand centres. This approach will
5    only make sense if the methane is produced in a manner that captures carbon from the atmosphere
6    and/or if CCS is used when the methane is used to produce hydrogen.
 7   Bulk Hydrogen Storage. Currently, hydrogen is stored in bulk in chemical processes such as metal and
 8   chemical hydrides as well as in geologic caverns (Andersson and Grönkvist 2019; Caglayan et al. 2019)
 9   (e.g., salt caverns operate in Sweden) (Elberry et al. 2021). There are still many challenges, however,
10   due to salt or hard rock geologies, large size, and minimum pressure requirements of the sites (IEA
11   2019c). Consequently, alternative carbon-free energy carriers, which store hydrogen, may become more
12   attractive (Kobayashi et al. 2019; Lan et al. 2012).
13   Electricity Transmission
14   Given the significant geographical variations in the efficiency of renewable resources across different
15   regions and continents, electricity transmission could facilitate cost effective deployment of renewable
16   generation, enhance resilience and security of supply, and increase operational efficiency (high
17   confidence). The diurnal and seasonal characteristics of different renewable energy sources such as
18   wind, solar, and hydropower can vary significantly by location. Through enhanced electricity
19   transmission infrastructure, more wind turbines can be deployed in areas with high wind potential and
20   more solar panels in areas with larger solar irradiation. Increases in electricity transmission and trade
21   can also enhance operational efficiency and reduce or defer the need for investment in peaking plants,
22   storage, or other load management techniques needed to meet security of supply requirements
23   associated with localized use of VRE sources. Increased interconnectivity of large-scale grids also
24   allows the aggregation of ‘smart grid’ solutions such as flexible heating and cooling devices for flexible
25   demand in industrial, commercial, and domestic sectors (Hakimi et al. 2020) and EVs (Li et al. 2021;
26   Muratori and Mai 2020). In general, interconnection is more cost-optimal for countries that are
27   geographically close to each other and can benefit from the diversity of their energy mixes and usage
28   (Schlachtberger et al. 2017). Such developments are not without price, however, and amongst other
29   concerns, raise issues surrounding land use, public acceptance, and resource acquisition for materials
30   necessary for renewable developments (Vakulchuk et al. 2020; Capellán-Pérez et al. 2017).
31   A number of studies have demonstrated the cost benefits of interconnected grids in a range of
32   geographical settings, including across the United States (Bloom et al. 2020), across Europe (2020;
33   Newbery et al. 2013; Cluet et al. 2020), between Australia and parts of Asia (Halawa et al. 2018), and
34   broader global regions, for example between the Middle East and Europe or North Africa and Europe
35   (Tsoutsos et al. 2015). While there is growing interest in interconnection among different regions or
36   continents, a broad range of geopolitical and socio-techno-economic challenges would need to be
37   overcome to support this level of international co-operation and large-scale network expansion (Bertsch
38   et al. 2017; Palle 2021).
39   Status of electricity transmission technology. Long-distance electricity transmission technologies are
40   already available. High voltage alternating current (HVAC), high-voltage direct current (HVDC), and
41   ultra HVDC (UHVDC) technologies are well-established and widely used for bulk electricity
42   transmission (Alassi et al. 2019). HVDC is used with underground cables or long-distance overhead
43   lines (typically voltages between 100–800 kV (Alassi et al. 2019) where HVAC is infeasible or not
44   economic. A ~USD 17 bn project development agreement was signed in January 2021 that would
45   connect 10 GW of PVs in the north of Australia via a 4500 km 3 GW HVDC cable to Singapore,
46   suggesting that this would be cost effective (Sun Cable 2021). In September 2019, the Changji-Guquan
47   ±1,100 kV ultra-high voltage direct current (UHVDC) transmission project built by State Grid

     Do Not Cite, Quote or Distribute                  6-62                                  Total pages: 217
     Final Government Distribution                  Chapter 6                              IPCC WGIII AR6

1    Corporation of China was officially completed and put into operation. The transmission line is able to
2    transmit up to 12 GW over 3341 km (Pei et al. 2020). This is the UHVDC transmission project with the
3    highest voltage level, the largest transmission capacity, and the longest transmission distance in the
4    world (Liu 2015).
 5   Other technologies that could expand the size of transmission corridors and/or improve the operational
 6   characteristics include low-frequency AC transmission (LFAC) (Xiang et al. 2021; Tang et al. 2021b)
 7   and half-wave AC transmission (HWACT) (Song et al. 2018; Xu et al. 2019). LFAC is technically
 8   feasible, but the circumstances in which it is the best economic choice compared to HVDC or HVAC
 9   still needs to be established (Xiang et al. 2016). HWACT is restricted to very long distances, and it has
10   not been demonstrated in practice, so its feasibility is unproven. There are still a number of
11   technological challenges for long-distance transmission networks such as protection systems for DC or
12   hybrid AC-DC networks (Franck C. et al. 2017; Chaffey 2016), improvement in cabling technology,
13   and including the use of superconductors and nanocomposites (Ballarino et al. 2016; Doukas 2019),
14   which require advanced solutions.
15   Challenges, barriers, and recommendations. The main challenge to inter-regional transmission is the
16   absence of appropriate market designs and regulatory and policy frameworks. In addition, there are
17   commercial barriers for further enhancement of cross-border transmission. The differing impacts of
18   cross-border interconnections on costs and revenues for generation companies in different regions could
19   delay the development of these interconnectors. It is not yet clear how the investment cost of
20   interconnections should be allocated and recovered, although there is growing support for allocating
21   costs in accordance with the benefits delivered to the market participants. Increased cross-border
22   interconnection may also require new business models which provide incentives for investment and
23   efficient operation, manage risks and uncertainties, and facilitate coordinated planning and governance
24   (Poudineh and Rubino 2017).
25   Optimizing the design and operation of the interconnected transmission system, both onshore and
26   offshore grids, also requires more integrated economic and reliability approaches (Moreno et al. 2012)
27   to ensure the optimal balance between the economics and the provision of system security while
28   maximizing the benefits of smart network technologies.
29   A wide range of factors, including generation profiles, demand profiles circuit losses, reliability
30   characteristics, and maintenance, as well as the uncertainties around them will need to be considered in
31   designing and operating long-distance transmission systems if they are to be widely deployed (De Sa
32   and Al Zubaidy 2011; Du 2009; Djapic et al. 2008; E3G 2021). Public support for extending
33   transmission systems will also be crucial, and studies indicate that such support is frequently low
34   (Perlaviciute et al. 2018; Vince 2010).
35   6.4.6 Demand Side Mitigation Options from an Energy Systems Perspective
36   Demand-side measures are fundamental to an integrated approach to low carbon energy systems (high
37   confidence). Mitigation options, such as wind parks, CCS, and nuclear power plants, may not be
38   implemented when actors oppose these options. Further, end users, including consumers, governments,
39   businesses and industry, would need to adopt the relevant options, and then use these as intended; user
40   adoption can be a key driver to scale up markets for low carbon technologies. This section discusses
41   which factors shape the likelihood that end users engage in relevant mitigation actions, focusing on
42   consumers; strategies to promote mitigation actions are discussed in Section
43   A wide range of actions of end users would reduce carbon emissions in energy systems (Abrahamse et
44   al. 2007; Dietz 2013; Creutzig et al. 2018; Hackmann et al. 2014; Grubler et al. 2018), including:
45   • use of low carbon energy sources and carriers. Actors can produce and use their own renewable
46        energy (e.g., install solar PV, solar water heaters, heat pumps), buy shares in a renewable energy
47        project (e.g., wind shares), or select a renewable energy provider.
     Do Not Cite, Quote or Distribute                  6-63                                 Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

 1   •   adoption of technologies that support flexibility in energy use and sector coupling, thereby
 2       providing flexibility services by balancing demand and renewable energy supply. This would
 3       reduce the need to use fossil fuels to meet demand when renewable energy production is low and
 4       put less pressure on deployment of low-emission energy supply systems. Examples are technologies
 5       to store energy (e.g., batteries and electric vehicles) or that automatically shift appliances on or off
 6       (e.g., fridges, washing machines).
 7   • adoption of energy-efficient appliances and systems and increase of resource efficiency of end uses
 8       so that less energy is required to provide the same service. Examples are insulating buildings, and
 9       passive or energy positive buildings.
10   • change behaviour to reduce overall energy demand or to match energy demand to available energy
11       supplies. Examples include adjusting indoor temperature settings, reducing showering time,
12       reducing car use or flying, and operating appliances when renewable energy production is high.
13   • purchase and use products and services that are associated with low GHG emissions during their
14       production (e.g., reduce dairy and meat consumption) or for transporting products (e.g., local
15       products). Also, end users can engage in behaviour supporting a circular economy, by reducing
16       waste (e.g., of food), sharing products (e.g., cars, equipment), and refurbishing products (e.g., repair
17       rather than buying new products) so that less new products are used.
18   Various factors shape whether such mitigation actions are feasible and considered by end users,
19   including contextual factors, individual abilities, and motivational factors. Mitigation actions can be
20   facilitated and encouraged by targeting relevant barriers and enablers (section
21   Contextual factors, such as physical and climate conditions, infrastructure, available technology,
22   regulations, institutions, culture, and financial conditions define the costs and benefits of mitigation
23   options that enable or inhibit their adoption (high confidence). Geographic location and climate factors
24   may make some technologies, such as solar PV or solar water heaters, impractical (Chang et al. 2009).
25   Culture can inhibit efficient use of home heating or PV (Sovacool and Griffiths 2020), low carbon diets
26   (Dubois et al. 2019), and advanced fuel choices (Van Der Kroon et al. 2013). Also, favourable financial
27   conditions promote the uptake of PV (Wolske and Stern 2018), good facilities increase recycling
28   (Geiger et al. 2019), and vegetarian meal sales increase when more vegetarian options are offered..
29   Mitigation actions are more likely when individuals feel capable to adopt them (Pisano and Lubell 2017;
30   Geiger et al. 2019), which may depend on income and knowledge. Low-income groups may lack
31   resources to invest in refurbishments and energy-efficient technology with high upfront costs (Andrews-
32   Speed and Ma 2016; Chang et al. 2009; Wolske and Stern 2018). Yet, higher income groups can afford
33   more carbon-intensive lifestyles (Golley and Meng 2012; Namazkhan et al. 2019; Frederiks et al. 2015;
34   Santillán Vera and de la Vega Navarro 2019; Mi et al. 2020; Wiedenhofer et al. 2017). Knowledge of
35   the causes and consequences of climate change and of ways to reduce GHG emissions is not always
36   accurate, but lack of knowledge is not a main barrier of mitigation actions (Boudet 2019).
37   Motivation to engage in mitigation action, reflecting individuals’ reasons for actions, depends on
38   general goals that people strive for in their life (i.e., values). People who strongly value protecting the
39   environment and other people are more likely to consider climate impacts and to engage in a wide range
40   of mitigation actions than those who strongly value individual consequences of actions, such as pleasure
41   and money (Taylor et al. 2014; Steg 2016). Values affect which types of costs and benefits people
42   consider and prioritize when making choices, including individual, affective, social, and environmental
43   costs and benefits (Gowdy 2008; Steg 2016).
44   First, people are more likely to engage in mitigation behaviour (i.e., energy savings, energy efficiency,
45   resource efficiency in buildings, low-carbon energy generation) when they believe such behaviour has
46   more individual benefits than costs (Harland et al. 1999; Steg and Vlek 2009; Kastner and Matthies
47   2016; Kastner and Stern 2015; Kardooni et al. 2016; Wolske et al. 2017; Korcaj et al. 2015), including
48   financial benefits, convenience, comfort, autonomy, and independence in energy supply (Wolske and
     Do Not Cite, Quote or Distribute                   6-64                                   Total pages: 217
     Final Government Distribution                   Chapter 6                               IPCC WGIII AR6

1    Stern 2018). Yet, financial consequences seem less important for decisions to invest in energy-
2    efficiency and renewable energy production than people indicate (Zhao et al. 2012).
3    Second, people are less likely to engage in mitigation behaviours that are unpleasurable or inconvenient
4    (Steg 2016), and more likely to do so when they expect to derive positive feelings from such actions
5    (Smith et al. 1994; Pelletier et al. 1998; Steg 2005; Carrus et al. 2008; Brosch et al. 2014; Taufik et al.
6    2016). Positive feelings may be elicited when behaviour is pleasurable, but also when it is perceived as
7    meaningful (Bolderdijk et al. 2013; Taufik et al. 2015).
 8   Third, social costs and benefits can affect climate action (Farrow et al. 2017), although people do not
 9   always recognize this (Nolan et al. 2008; Noppers et al. 2014). People engage more in mitigation actions
10   when they think others expect them to do so and when others act as well (Rai et al. 2016; Harland et al.
11   1999; Nolan et al. 2008). Being part of a group that advocates mitigation actions encourages such
12   actions (Biddau et al. 2016; Fielding and Hornsey 2016; Jans et al. 2018). Talking with peers can reduce
13   uncertainties and confirm benefits about adoption of renewable energy technology (Palm 2017), and
14   peers can provide social support (Wolske et al. 2017). People may engage in mitigation actions when
15   they think this would signal something positive about them (Griskevicius et al. 2010; Milinski et al.
16   2006; Kastner and Stern 2015; Noppers et al. 2014). Social influence can also originate from political
17   and business leaders (Bouman and Steg 2019); GHG emissions are lower when legislators have strong
18   environmental records (Jensen and Spoon 2011; Dietz et al. 2015).
19   Fourth, mitigation actions, including saving energy and hot water, limiting meat consumption, and
20   investing in energy efficiency, resource efficiency in buildings, and renewable energy generation are
21   more likely when people more strongly care about others and the environment (Van Der Werff and Steg
22   2015; Steg et al. 2015; Wolske et al. 2017). People across the world generally strongly value the
23   environment (Bouman and Steg 2019; Steg 2016), suggesting that they are motivated to mitigate climate
24   change. The more individuals are aware of the climate impact of their behaviour, the more they think
25   their actions can help reduce such impacts, which strengthens their moral norms to act accordingly, and
26   promotes mitigation actions (Steg and de Groot 2010; Jakovcevic and Steg 2013; Chen 2015; Wolske
27   et al. 2017).
28   Initial mitigation actions can encourage engagement in other mitigation actions when people experience
29   that such actions are easy and effective (Lauren et al. 2016), and when initial actions make them realize
30   they are a pro-environmental person, motivating them to engage in more mitigation actions so as to be
31   consistent (van der Werff et al. 2014; Lacasse 2015, 2016; Peters et al. 2018). This implies it would be
32   important to create conditions that make it likely that initial mitigation actions motivate further actions.
33   6.4.7 Summary of Mitigation Options
34   Designing feasible, desirable, and cost-effective energy sector mitigation strategies requires comparison
35   between the different mitigation options. One such metric is the cost of delivering one unit of energy,
36   for example, the levelized cost, or USD MWh-1, of electricity produced from different sources. LCOEs
37   are useful because they normalize the costs per unit of service provided. While useful in characterizing
38   options in broad strokes, it is important to acknowledge and understand several caveats associated with
39   these metrics, particularly when applied globally. They may be constructed with different discount
40   rates; they require information on energy input costs for options that require energy inputs (e.g., fossil
41   electricity generation, biofuels); they depend on local resource availability, for example solar insolation
42   for solar power, wind classes for wind power, and rainfall and streamflow for hydropower; and actual
43   implementation costs may include additional elements, for example, the costs of managing electricity
44   grids heavily dependent on VRE electricity sources. These complicating factors vary across regions,
45   some depend strongly on the policy environment in which mitigation options are deployed, and some
46   depend on how technologies are constructed and operated.

     Do Not Cite, Quote or Distribute                   6-65                                   Total pages: 217
     Final Government Distribution                                                                                                                                                                                                                                                                            Chapter 6                                                                                                                                                                                                                   IPCC WGIII AR6

 1   The literature provides multiple LCOE estimates for mitigation options today and in the future (see
 2   Table 6.9 for electricity generation options). LCOE ranges for low- and zero-carbon electricity
 3   technologies overlap with LCOE’s of fossil generation without CCS. For example, LCOEs for utility
 4   solar and wind today and in the future overlap with those of new coal and gas without CCS (Figure
 5   6.18, NREL 2021; Lazard, 2020; IEA WEO 2020, IEA WEO 2020). Some of the overlap stems from
 6   differences in assumptions or regional conditions that apply to all technologies (e.g., variations in
 7   assumed discount rates), but the overlap also reflects the fact that low- and zero-carbon electricity
 8   generation options are, and will be, less expensive than emitting options in many regions. Future cost
 9   projections also illustrate that several technologies are anticipated to experience further cost declines
10   over the coming decades, reinforcing the increasingly competitiveness of low- and zero-carbon
11   electricity. For example, IEA’s LCOEs estimates for offshore wind halve between 2020 and 2040 in

12   several regions (IEA WEO 2020).
       Figure with new IEA & Lazard values :

       2020 USD / kWh
       2020 Cents/ kWh







                                                                >2020 to 2040

                                                                                                                >2020 to 2040

                                                                                                                                                   >2020 to 2040

                                                                                                                                                                                                  >2020 to 2040

                                                                                                                                                                                                                                                    >2020 to 2040

                                                                                                                                                                                                                                                                                              >2020 to 2040

                                                                                                                                                                                                                                                                                                                                       >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                  >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                               >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                                                                   >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 >2020 to 2040

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       >2020 to 2040
                              circa 2020

                                                                                circa 2020

                                                                                                                                circa 2020

                                                                                                                                                                   circa 2020

                                                                                                                                                                                                                    circa 2020

                                                                                                                                                                                                                                                                       circa 2020

                                                                                                                                                                                                                                                                                                                  circa 2020

                                                                                                                                                                                                                                                                                                                                                       circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                  circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                                                               circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         circa 2020

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 circa 2020
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      Bio−electricity with CCS

                                                                                                                                                                                Wind − On−shore

                                                                                                                                                                                                                                 Wind − Off−shore
                                                                                             Solar PV − Other
                                           Solar PV − Utility

                                                                                                                                                                                                                                                                                                                                                                                                               NGCC with CCS

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   Coal with CCS

                                                                                                                                                                                                                                                                                                                                                                    Natural gas



                                                                                             Authors                                         EIA                   IEA                                            IRENA                                             NREL                                  Other                 Region                                     Global                              North America                                                       Europe                                          Asia                                     South America

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   Fossil fuel LCOE range in circa 2020
14                       Figure 6.18 Range of LCOEs (in USD cents kWh-1) from recent studies for different electricity generating
15                                            technologies circa 2020 and in the future between 2020-2040. LCOEs are primarily taken from recent
16                                           studies, because the costs of some technologies are changing rapidly. To make the figure more tractable
17                                         across the studies, we highlight the data from IEA WEO 2020 STEPS scenario in red (EIA, 2020), the EIA
18                                          AEO 2021 in blue (EIA, 2021), NREL ATB 2021 in green, (NREL, 2021), and IRENA Renewable Power
19                                             Generation Costs in 2020 in yellow (IRENA, 2021). All other studies are shown in light grey markers.
20                                         Marker shapes identify the regions included in the studies. Studies that included several regions are labelled
21                                         as global. Only sources that provided LCOEs are included. Ranges for studies frequently reflect variations
22                                            among regional estimates. Studies that are shown as a mid-point and a solid line represent studies that
23                                            reported either a median or an average, and that had either a confidence interval or a minimum and a
24                                         maximum reported. Dashed lines with markers at the end represent the range of values reported in studies
25                                           that had several point estimates for either different regions or used different assumptions. All estimates
26                                          were converted to 2020 USD. The publication year was used if no USD year was provided. Some studies
27                                           included transmissions costs, and some the CCS studies included storage and sequestration costs, while
28                                          others did not. Vertical axis is capped at USD2020 0.30 kWh-1, but some estimates for hydro, geothermal,
29                                          natural gas and bioelectricity were higher than 0.30. The grey horizontal band denotes the range of fossil
30                                                                              fuel electricity LCOEs in circa 2020.

31   A more direct metric of mitigation options is the cost to reduce one tonne of CO2 or equivalent GHGs,
32   or USD tCO2-eq-1 avoided. In addition to the comparison challenges noted above, this metric must
33   account for the costs and emissions of the emitting options that is being displaced by the low-carbon
34   option. Assumptions about the displaced option can lead to very different mitigation cost estimates

     Do Not Cite, Quote or Distribute                                                                                                                                                                                                                                                                             6-66                                                                                                                                                                                                                                 Total pages: 217
     Final Government Distribution                        Chapter 6                                  IPCC WGIII AR6

1    (Table 6.9). Despite these challenges, these metrics are useful for identifying broad trends and making
2    broad comparisons, even from the global perspective in this assessment. But local information will
3    always be critical to determine which options are most cost-effective in any specific applications.
5      Table 6.9 Examples of cost of mitigation for selected electricity options. Results represent variations in
6       mitigation options and displaced fossil generation. LCOEs are illustrative, but consistent with recent
7         estimates. Negative values mean that the mitigation option is cheaper than the displaced option,
8                                         irrespective of emissions benefits.

                                                     New coal         Existing coal                      Existing NGCC
                              Baseline emissions
                              rate (tonCO2 MWh-         0.8               0.9                0.34             0.42

                                LCOE (USD2020
                                                       0.065             0.041               0.044           0.028
     Utility scale solar PV                        44 USD tCO2-       66 USD tCO2-     165 USD tCO2-     171 USD tCO2-
     (poor resource site)                               eq-1              eq-1              eq-1              eq-1
     Utility scale solar PV                        –38 USD tCO2-      –7 USD tCO2-     –26 USD tCO2-      17 USD tCO2-
     (good resource site)                               eq-1              eq-1              eq-1              eq-1
10   The feasibility and desirability of mitigation options extends well beyond the market economic costs of
11   installation and operation (Section 6.4.1). Figure 6.19 summarizes the barriers and enablers for
12   implementing different mitigation options in energy systems. The feasibility of different options can be
13   enhanced by removing barriers and/or strengthening enablers of the implementation of the options. The
14   feasibility of options may differ across context (e.g., region), time (e.g., 2030 versus 2050), scale (e.g.,
15   small versus large) and the long-term warming goal (e.g., 1.5°C versus 2°C).


     Do Not Cite, Quote or Distribute                          6-67                                    Total pages: 217
    Final Government Distribution                     Chapter 6                                IPCC WGIII AR6

1   Figure 6.19 Summary of the extent to which different factors would enable or inhibit the deployment of mitigation options in energy systems. Blue bars indicate the
2         extent to which the indicator enables the implementation of the option (E) and orange bars indicate the extent to which an indicator is a barrier (B) to the
3       deployment of the option, relative to the maximum possible barriers and enablers assessed. An X signifies the indicator is not applicable or does not affect the
4     feasibility of the option, while a forward slash indicates that there is no or limited evidence whether the indicator affects the feasibility of the option. The shading
5   indicates the level of confidence, with darker shading signifying higher levels of confidence. Appendix II provides an overview of the factors affecting the feasibility
6   of options and how they differ across context (e.g., region), time (e.g., 2030 versus 2050), and scale (e.g., small versus large), and includes a line of sight on which the
7                                                    assessment is based. The assessment method is explained in Annex II.11.


    Do Not Cite, Quote or Distribute                     6-68                                    Total pages: 217
    Final Government Distribution                    Chapter 6                        IPCC WGIII AR6

1   6.5 Climate Change Impacts on the Energy System
2   6.5.1 Climate Impacts on the Energy System
3   Many components of the energy system are affected by individual weather events and climate
4   conditions (Table 6.10). In addition, a range of compounding effects can be anticipated, as the complex,
5   interconnected climate and energy system are influenced by multiple weather and climate conditions.
6   This raises the question of whether the energy system transformation needed to limit warming will be
7   impacted by climate change.

    Do Not Cite, Quote or Distribute                 6-69                                  Total pages: 217
    Final Government Distribution                                                      Chapter 6                                                                                                                                                                                               IPCC WGIII AR6

1      Table 6.10 Relevance of the key climatic impact-drivers (and their respective changes in intensity, frequency, duration, timing, and spatial extent) for major
2       categories of activities in the energy sector. The climate impact-drivers (CIDs) are identified in AR6/GWI/Chapter 12 (ref). The relevance is assessed as:
3                                         positive/negative (+ or –), or both (±). D&O: Design and Operation, CF: Capacity Factor.

                                                                                                                                                                                                                                                                                                                                               Climate Impact Driver
                                                                           Heat and cold                                                                                                 Wet and dry                                                                                                                            Wind                                                                                      Snow and ice                                                                                          Coastal                                                                     Oceanic                                                                              Other

                                                                                                                                                              Heavy precipitation and pluvial flood

                                                                                                                                                                                                                                                   Agricultural and ecological drought

                                                                                                                                                                                                                                                                                                                                                                                                                                                            Heavy snowfall and ice storm
                                                                                                                                                                                                                                                                                                                                                                                       Snow, glacier and ice sheet

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  Atmospheric CO2 at surface
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          Mean ocean temperature
                                                                                                                                                                                                                                                                                                                                                                                                                                  Lake, river and sea ice
                                                                Mean air temperature

                                                                                                                                                                                                                            Hydrological drought

                                                                                                                                                                                                                                                                                                                                                                 Sand and dust storm

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         Air pollution weather

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               Radiation at surface
                                                                                                                                                                                                                                                                                                                          Severe wind storm
                                                                                                                           Mean precipitation

                                                                                                                                                                                                                                                                                                        Mean wind speed

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   Relative sea level

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   Marine heatwave

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      Dissolved oxygen
                                                                                                                                                                                                                                                                                                                                              Tropical cyclone

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  Snow avalanche

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        Coastal erosion

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     Ocean salinity
                                                                                       Extreme heat

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     Ocean acidity
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        Coastal flood
                                                                                                                                                                                                                                                                                         Fire weather

                                                                                                                                                River flood
                                                                                                      Cold spell


                  Energy sector         Energy activity
                                       Resources (dammed)                                                                       +                  +                                                                –            –                           –                                                                                                                                ±                                         –
                                             D&O (dammed)                                                                                                               ±                               –                                                                                                                                                                                                                                                           –
                                     Resources (undammed)                                                                       +                  +                    +                                                        –                           –
                                          D&O (undammed)                                                                                                                ±                               –                                                                                                                                                                                     ±                                         –                           –
                                            Capacity factors –                                                      –                                                                                                                                                                                       +                 –                   –
                  Wind power                 D&O (onshore)                                                          –                                                                                                                                                                                                         –                   –                   –                                                                                             –                       –                                              –                –
                                              D&O (offshore)                                                                                                                                                                                                                                                                  –                   –                                                                                                                                         –                           –
                                                     CF (PV) –                            –                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    –                                                    +
                  Solar power                      CF (CSP)                                             –                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      –                                                    +
                                                          D&O                                                                                                                                                                                                                               _                                                                         –                       _                                                                                             –
                 Ocean energy                     Resources                                                                                                                                                                                                                                                 +                 ±                   +                                                                                                                                         –                           ±                  ±                                    –                                                        ±
                   Bio-energy                     Resources ±                             –                                     ±                                                                                                –                           –                              –                                                                                                                                                                                                                                                                                                                                                                                                            +
              Thermal power plants                 Efficiency ±                           –                                                                                                                                      –
                  (incl nuclear)                Vulnerability                             –             –           –                              –                    –                               –                        –                                                          –                                                                                                                                                                       –                                                   –                  –
                      CCS                          Efficiency ±                                                                                                                                                     –            –                           –
                                                     Heating +                                          –                                                                                                                                                                                                   –                                                                                                                                                                                                                                                                                                                                                                                                                       +
               Energy consumption
                                                     Cooling –                            –                                                                                                                                                                                                                 ±                                                                                                                                                                                                                                                                                                                                                                                                                       –
                  Electric power                          D&O –                           –                                                                                                                                                                                                 –
               transmission system              Vulnerability                             –                                                                                                                                                                                                 –                                 –                   –                                                                                                                 –                                 –

    Do Not Cite, Quote or Distribute                                                   6-70                                                                                                                                                                                                                               Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    The impacts of climate change on the energy system can be divided into three areas: impacts on the
2    energy supply, impacts on energy consumption, and impacts on energy infrastructure. The rest of this
3    section focuses on how the future changes in climate drivers might affect the ability of the energy
4    system transformation needed to mitigate climate change. The discussion of energy infrastructure in
5    this section is limited to electric electricity system vulnerability.
6    6.5.2 Impacts on Energy Supply
 7   The increased weather-dependency of future low-carbon electricity systems amplifies the possible
 8   impacts of climate change (Staffell and Pfenninger 2018). However, globally climate change impacts
 9   on electricity generation – including hydro, wind and solar power potentials – should not compromise
10   climate mitigation strategies (high confidence). Many of the changes in the climate system will be
11   geographically complex at the regional and local levels. Thus, regionally climate change impacts on
12   electricity generation could be significant. Climate change impacts on bioenergy potentials are more
13   uncertain because of uncertainties associated with the crop response to climate change, future water
14   availability and crop deployment. Climate change can reduce the efficiency of thermal power generation
15   and increase the risk of power plant shutdowns during droughts. The potential additional cooling water
16   needs of CCS can increase these risks.
17   Hydropower
18   The impacts of climate change on hydropower will vary by region. High latitudes in the northern
19   hemisphere are anticipated to experience increased runoff and hydropower potential. For other regions,
20   studies find both increasing and decreasing runoff and hydropower potential. Areas with decreased
21   runoff are anticipated to experience reduced hydropower production and increased water conflict among
22   different economic activities. (high confidence)
23   Hydropower production is directly related to the availability of water. Changes in runoff and its
24   seasonality and changes in temperature and precipitation intensity will influence hydro electricity
25   production (IHA 2019). In general, increased precipitation will increase water availability and
26   hydropower production. Increased precipitation intensity, however, may impact the integrity of dam
27   structures and affect power production by increasing debris accumulation and vegetation growth.
28   Additionally, increased precipitation intensity results in the silting of the reservoirs or increases the
29   amount of water spilt, resulting in erosion (Schaeffer et al. 2012; IHA 2019). Climate change will likely
30   lead to higher air temperatures, resulting in more surface evaporation, less water storage, and loss of
31   equipment efficiency (Ebinger and Vergara 2011; Fluixá-Sanmartín et al. 2018; Hock et al. 2019;
32   Mukheibir 2013). Climate change may alter the demands for water use by other sectors that often rely
33   on stored water in multi-purpose reservoirs and may therefore generate conflicts over water use. The
34   increased need for water for irrigation and/or industry can affect the availability of water for hydropower
35   generation (Solaun and Cerdá 2017; Spalding-Fecher et al. 2016). Higher temperatures increase glacier
36   melt, increasing water availability for hydropower while the glaciers exist. Changes in the timing of
37   snow and ice melt may require upgrading in storage capacity and adaptation of the hydropower plant
38   management for fully exploiting the increase in water availability.

     Do Not Cite, Quote or Distribute                   6-71                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

2                                        Global Solar Atlas (ESMAP 2019)


5      Figure 6.20 Global spatial patterns of changes in gross hydropower potential based on climate forcing
6      from five climate models. Changes are shown for the 2050s (upper) and the 2080s (lower) for the low
7     emission scenario (RCP2.6; left) and highest emission scenario (RCP8.5; right) scenarios relative to the
8                       control period (1971–2000). [Data source: (van Vliet et al. 2016a)].

 9   The conclusions regarding climate change impacts on hydropower vary due to differences in modelling
10   assumptions and methodology, such as choice of the climate and hydrological models, choice of metrics
11   (e.g., projected production vs hydropower potential), level of modelling details between local and global
12   studies, reservoir operation assumptions. Also important is how hydropower production matches up
13   with other reservoir purposes, accounting for other water and energy users, and how the competing uses
14   are impacted by climate change (Turner et al. 2017; van Vliet et al. 2016b). Nonetheless, analyses
15   consistently demonstrate that the global impact of climate change on hydropower will be small, but the
16   regional impacts will be larger, and will be both positive and negative (Figure 6.20) Gross global
17   hydropower potential in the 2050s has been estimated to slightly decrease (Hamududu and Killingtveit
18   2012) between 0.4% (for the low emission scenario) and 6.1% (for the highest emission scenario) for
19   the 2080s compared to 1971–2000 (van Vliet et al. 2016a).
20   Regional changes in hydropower are estimated from 5–20% increases for most areas in high latitudes
21   (van Vliet et al. 2016b; Turner et al. 2017) to decreases of 5–20% in areas with increased drought
22   conditions (Cronin et al. 2018). Models show a consistent increase in streamflow and hydropower
23   production by 2080 in high latitudes of the northern hemisphere and parts of the tropics (Figure 6.20)
24   (e.g., central Africa and southern Asia) while decreasing in the U.S., southern and central Europe,
25   Southeast Asia and southern South America, Africa and Australia (van Vliet et al. 2016c,a). Decreases
26   in hydropower production are indicated for parts of North America, central and southern Europe, the
27   Middle East, central Asia and Southern South America. Studies disagree on the changes in hydropower
28   production in China, central South America, and partially in southern Africa (Solaun and Cerdá 2019;
29   Hamududu and Killingtveit 2012; van Vliet et al. 2016b; Fan et al. 2020).

     Do Not Cite, Quote or Distribute                  6-72                                   Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1   Wind Energy
2    Climate change will not substantially impact future wind resources and will not compromise the ability
3    of wind energy to support low-carbon transitions (high confidence). Changing wind variability may
4    have a small to modest impact on backup energy and storage needs (low confidence); however, current
5    evidence is largely from studies focused on Europe.
 6   Long-term global wind energy resources are not expected to substantially change in future climate
 7   scenarios (Karnauskas et al. 2018; Yalew et al. 2020; Pryor et al. 2020). However, recent research has
 8   indicated consistent shifts in the geographic position of atmospheric jets in the high emission scenarios
 9   (Harvey et al. 2014), which would decrease wind power potentials across the Northern Hemisphere
10   mid-latitudes and increase wind potentials across the tropics and the Southern Hemisphere. However,
11   the climate models used to make these assessments differ in how well they can reproduce the historical
12   wind resources and wind extremes, which raises questions about the robustness of their predictions of
13   future wind resources (Pryor et al. 2020).
14   There are many regional studies on changes in wind resources from climate change. For Europe, there
15   is medium evidence and moderate agreement that wind resources are already increasing and will
16   continue to increase in Northern Europe and decrease in Southern Europe (Moemken et al. 2018;
17   Carvalho et al. 2017; Devis et al. 2018). For North America, the various studies have low agreement
18   for the changes in future wind resources in part because the year-to-year variations in wind resources
19   are often larger than the future change due to climate change (Johnson and Erhardt 2016; Wang et al.
20   2020b; Costoya et al. 2020; Chen 2020). Studies show increases in future wind resources in windy areas
21   in South America (Ruffato-Ferreira et al. 2017; de Jong et al. 2019). No robust future changes in wind
22   resources have been identified in China (Xiong et al. 2019). However, none of the global or regional
23   studies of the effects of climate change on wind resources considers the fine-scale dependence of wind
24   resources on the topography and wind direction (Sanz Rodrigo et al. 2016; Dörenkämper et al. 2020)
25   or the effect of expanding wind energy exploitation (Lundquist et al. 2019; Volker et al. 2017). There
26   is limited evidence that extreme wind speeds, which can damage wind turbines, will increase due to
27   climate change (Pes et al. 2017; Pryor et al. 2020). Nevertheless, projected changes in Europe and North
28   America – regions where the most extensive analysis has been undertaken – are expected to be within
29   the estimates embedded in the design standards of wind turbines (Pryor and Barthelmie 2013).
30   Future wind generation in Europe could decrease in summer and autumn, increasing in winter in
31   northern-central Europe but decreasing in southernmost Europe (Carvalho et al. 2017). Towards 2100,
32   intra-annual variations increase in most of Europe, except around the Mediterranean area (Reyers et al.
33   2016), but this may reflect natural multidecadal variability (Wohland et al. 2019b). Wind speeds may
34   become more homogeneous over large geographical regions in Europe due to climate change,
35   increasing the likelihood of large areas experiencing high or low wind speeds simultaneously (Wohland
36   et al. 2017). These changes could result in fewer benefits in the transmission of wind generation between
37   countries and increased system integration costs. Europe could require a modest increase (up to 7%) in
38   backup energy towards the end of the 21st century due to more homogeneous wind conditions over
39   Europe (Wohland et al. 2017; Weber et al. 2018). However, other studies report that impact of climate
40   change is substantially smaller than interannual variability, with no significant impact on the occurrence
41   of extreme low wind production events in Europe (Van Der Wiel et al. 2019). If European electricity
42   systems are designed to manage the effects of existing weather variability on wind power, they can
43   likely also cope with climate change impacts on wind power (Ravestein et al. 2018). Changes in wind
44   generation variability caused by climate change are also reported for North America (Haupt et al. 2016;
45   Losada Carreño et al. 2018), with modest impacts on electricity system operation (Craig et al. 2019).

     Do Not Cite, Quote or Distribute                  6-73                                  Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1   Solar Energy
 2   Climate change is not expected to substantially impact global solar insolation and will not compromise
 3   the ability of solar energy to support low-carbon transitions (high confidence). Models show dimming
 4   and brightening in certain regions, driven by cloud, aerosol and water vapour trends [WGI,ch12,p31].
 5   The increase in surface temperature, which affects all regions, decreases solar power output by reducing
 6   the PV panel efficiency. In some models and climate scenarios, the increases in solar insolation are
 7   counterbalanced by reducing efficiency due to rising surface air temperatures, which increase
 8   significantly in all models and scenarios (Jerez et al. 2015; Bartók et al. 2017; Emodi et al. 2019).
 9   Increases in aerosols would reduce the solar resource available and add to maintenance costs
10   [AR6,WGI,ch12].
11   In many emission scenarios, the effect on solar PV from temperature-induced efficiency losses is
12   smaller than the effect expected from changes on solar insolation due to variations in water vapour and
13   clouds in most regions. Also, future PV technologies will likely have higher efficiency, which would
14   offset temperature-related declines (Müller et al. 2019). Cloud cover is projected to decrease in the
15   subtropics (around –0.05% per year), including parts of North America, vast parts of Europe and China,
16   South America, South Africa and Australia (medium agreement, medium evidence). Thus, models
17   project modest (< 3%) increases in solar PV by the end of the century for southern Europe, northern
18   and southern Africa, Central America, and the Caribbean (Emodi et al. 2019). There are several studies
19   projecting decreasing solar production, but these are generally influenced by other factors, for example,
20   increasing air pollution (Ruosteenoja et al. 2019). The multi-model means for solar insolation in
21   regional models decrease 0.60 W m-2 per decade from 2006 to 2100 over most of Europe (Bartók et
22   al. 2017), with the most significant decreases in the Northern countries (Jerez et al. 2015).
23   Bioenergy
24   Climate change can affect biomass resource potential directly, via changes in the suitable range (i.e.,
25   the area where bioenergy crops can grow) and/or changes in yield, and indirectly, through changes in
26   land availability. Increases in CO2 concentration increase biomass yield; climate changes (e.g.,
27   temperature, precipitation, etc.) can either increase or decrease the yield and suitable range.
28   Climate change will shift the suitable range for bioenergy towards higher latitudes, but the net change
29   in the total suitable area is uncertain (high confidence). Several studies show northward shifts in the
30   suitable range for bioenergy in the northern hemisphere (Tuck et al. 2006; Bellarby et al. 2010; Preston
31   et al. 2016; Barney and DiTomaso 2010; Hager et al. 2014; Conant et al. 2018; Cronin et al. 2018;
32   Wang et al. 2014a), but the net effect of climate change on total suitable area varies by region, species,
33   and climate model (Barney and DiTomaso 2010; Hager et al. 2014; Wang et al. 2014a).
34   The effect of climate change on bioenergy crop yields will vary across region and feedstock (high
35   confidence); however, in general, yields will decline in low latitudes (medium confidence) and increase
36   in high latitudes (low confidence) (Haberl et al. 2010; Cosentino et al. 2012; Mbow et al. 2019; Cronin
37   et al. 2018; Preston et al. 2016). However, the average change in yield varies significantly across studies,
38   depending on the feedstock, region, and other factors (Dolan et al. 2020; Kyle et al. 2014) Mbow et al.
39   (2019); Beringer et al. (2011). Only a few studies extend the modelling of climate change impacts on
40   bioenergy to quantify the effect on bioenergy deployment or its implications on the energy system
41   (Calvin et al. 2013, 2019; Thornton et al. 2017; Kyle et al. 2014). These studies find that changes in
42   deployment are of the same sign as changes in yield; that is, if yields increase, then deployment
43   increases.
44   Some of the uncertainty in the sign and magnitude of the impacts of climate change on bioenergy
45   potential is due to uncertainties in CO2 fertilization (the increase in photosynthesis due to increases in
46   atmospheric CO2 concentration) (Bonjean Stanton et al. 2016; Haberl et al. 2011; Cronin et al. 2018;
47   Solaun and Cerdá 2019; Yalew et al. 2020). For example, earlier studies found that without CO2
     Do Not Cite, Quote or Distribute                   6-74                                   Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    fertilization, climate change will reduce global bioenergy potential by about 16%; with CO2
2    fertilization, however, climate change increases this potential by 45% (Haberl et al. 2011). However,
3    newer studies in the U.S. find little effect of CO2 fertilization on switchgrass yield (Dolan et al. 2020).
4    There is also a considerable uncertainty across climate and crop models in estimating bioenergy
5    potential (Hager et al. 2014).
6   Thermal power plants
 7   The operation of thermal power plants will be affected by climate change, deriving from changes in the
 8   ambient conditions like temperature, humidity and water availability (Schaeffer et al. 2012) (high
 9   confidence). Changes in ambient temperature have relatively small impacts on coal-fired and nuclear
10   power plants (Rankine cycle); however, gas-fired power plants (Brayton or combined-cycle) may have
11   their thermal efficiency and power output significantly decreased (De Sa and Al Zubaidy 2011;
12   Schaeffer et al. 2012). Droughts decrease potential cooling water for thermal power plants and increase
13   the probability of water outlet temperatures exceeding regulatory limits, leading to lower production or
14   even shutdowns. Thermal power utilization has been reported to be on average 3.8% lower during
15   drought years globally (van Vliet et al. 2016c). and further significant decreases in available thermal
16   power plant capacity due to climate change are projected (Koch et al. 2014; van Vliet et al. 2016b;
17   Yalew et al. 2020) . An increase in climate-related nuclear power disruptions has been reported in the
18   past decades globally (Ahmad 2021).
19   Carbon capture may increase cooling water usage significantly, especially in retrofits, with up to 50%
20   increase in water usage for coal-fired power plants globally, depending on the CCS technology (Rosa
21   et al. 2020, Section 6.4). In Asia, planned coal capacity is expected to be vulnerable to droughts, sea
22   level rise, and rising air temperatures, and this may be exacerbated by incorporating carbon capture
23   (Wang et al. 2019c). Recently, however, studies have proposed designs of CCS with a minimal increase
24   in water requirements (Mikunda et al. 2021; Magneschi et al. 2017).
25   Older thermal power plants can be retrofitted to mitigate climate impacts by altering and redesigning
26   the cooling systems (Westlén 2018), although the costs for these solutions may be high. For example,
27   dry cooling may be used instead of once-through cooling; however, it lowers thermal efficiency and
28   would leave plants vulnerable to ambient temperature increase (Ahmad 2021). Closed-circuit cooling
29   is much less sensitive to water temperature than once-through cooling (Bonjean Stanton et al. 2016).
30   Modifying policies and regulation of water and heat emissions from power plants may also be used to
31   mitigate plant reliability problems induced by climate change (Eisenack 2016; Mu et al. 2020), albeit
32   with potential impacts for other water users and ecology. Improvements in water use and thermal
33   efficiencies and the use of transmission capabilities over large geographical regions to mitigate risks on
34   individual plants are also possible mitigation options (Miara et al. 2017).
35   6.5.3 Impacts on Energy Consumption
36   Heating demand will decrease, and cooling demand will increase in response to climate change. Peak
37   load may increase more than energy consumption, and the changing spatial and temporal load patterns
38   can impact transmission and needs for storage, demands-side management, and peak-generating
39   capacity. (high confidence)
40   Climate change will decrease heating demands, especially in cold regions, and it will increase cooling
41   demands, especially in warm regions (Yalew et al. 2020). Recent studies report significant net impacts,
42   with the commercial and industrial sectors and substantial air condition penetration driving an increase
43   in energy demand (De Cian and Sue Wing 2019; Levesque et al. 2018; van Ruijven et al. 2019; Davis
44   and Gertler 2015; Yalew et al. 2020). For example, globally, De Cian and Sue Wing (2019) found a 7–
45   17% increase in energy consumption due to climate change in 2050, with the range depending on the
46   climate change scenario. The overall effects of climate change on building energy consumption are

     Do Not Cite, Quote or Distribute                   6-75                                  Total pages: 217
     Final Government Distribution                     Chapter 6                        IPCC WGIII AR6

1    regionally dependent. For example, Zhang et al. (2019) find that reduced heating will outweigh
2    increased cooling in the residential buildings in Europe, but the reverse will be true in China.
 3   While many studies have focused on energy consumption, climate extremes are expected to alter peak
 4   energy demands, with the potential for blackouts, brownouts, and other short-term energy system
 5   impacts (Yalew et al. 2020). For example, peak energy demand during heatwaves can coincide with
 6   reduced transmission and distribution capacity at higher temperatures. In large cities, extreme heat
 7   events increase cooling degree days significantly, with the urban heat island effect compounding the
 8   impact (Morakinyo et al. 2019). One study found that total electricity consumption at the end of the
 9   century in the U.S. could increase on average by 20% during summer months and decrease on average
10   by 6% in the winder (Ralston Fonseca et al. 2019). While the average increase in consumption is
11   modest, climate change is projected to have severe impacts on the frequency and intensity of peak
12   electricity loads. (Auffhammer et al. 2017). Bartos et al. (2016) find that peak per-capita summertime
13   load in the U.S. may rise by 4.2%–15% by mid-century. Efficient cooling technologies and other
14   demand side measures can limit cooling energy loads during periods of particularly high
15   demand(Dreyfus et al. 2020; International Energy Agency (IEA) 2018).
16   [START BOX 6.6 HERE]
17                                         Box 6.6 Energy Resilience

18   In February 2021, the state of Texas was hit by three major storms and suffered significant scale power
19   outages. More than 4.5 million homes and businesses on the Texas electric grid were left without
20   electricity for days, limiting the ability to heat homes during dangerously low temperatures and leading
21   to food and clean water shortages (Busby et al. 2021). The Texas and other events – e.g., during
22   Typhoon Haiyan that affected Southeast Asia in 2013; the Australian bush fires in 2019–2020 and forest
23   fires in 2018 in California; water shortages in Cape Town, South Africa in 2018 and the western United
24   States during 2021 – raise the question of whether future low-carbon energy systems will be more or
25   less resilient than those of today.
26   Some characteristics of low-carbon energy systems will make them less resilient. Droughts reduce
27   hydroelectric electricity generation (Gleick 2016; van Vliet et al. 2016c); wind farms do not produce
28   electricity in calm conditions or shut down in very strong winds (Petersen and Troen 2012); solar PV
29   generation is reduced by clouds and is less efficient under extreme heat, dust storms, and wildfires
30   (Perry and Troccoli 2015; Jackson and Gunda 2021). In addition, the electrification of heating will
31   increase the weather dependence of electricity consumption (Staffell and Pfenninger 2018; Gea-
32   Bermúdez et al. 2021). Non-renewable generation, for example from nuclear and fossil power plants,
33   are also vulnerable to high temperatures and droughts as they depend on water for cooling (Cronin et
34   al. 2018; Ahmad 2021).
35   But some aspects of low-carbon energy systems will make them more resilient. Wind and solar farms
36   are often spread geographically, which reduces the chances of being affected by the same extreme
37   weather event (Perera et al. 2020). The diversification of energy sources, in which each component has
38   different vulnerabilities, increases resilience. Less reliance on thermal electricity generation
39   technologies will reduce the risks of curtailment or efficiency losses from droughts and heat waves.
40   (Lohrmann et al. 2019). More generally, increased electricity system integration and flexibility (Section
41   6.4.3) and weatherization of generators increases electricity system resilience (Heffron et al. 2021;
42   Busby et al. 2021). Likewise, local district micro-grids with appropriate enabling technologies (e.g.,
43   distributed generation, energy storage, greater demand-side participation, electric vehicles) may ensure
44   access to electricity during major long-duration power outage events and radically enhance the
45   resilience of supply of essential demand (Stout et al. 2019).
46   [END BOX 6.6 HERE]

     Do Not Cite, Quote or Distribute                  6-76                                 Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

2    6.5.4 Impacts on Electricity System Vulnerability
3    While long-term trends are important for electricity system planning, short-term effects associated with
4    loss of power can be disruptive and lead to significant economic losses along with cascading impacts
5    on health and safety. Extreme weather and storms threaten the electricity system in different ways,
6    affecting system resilience, reliability, and adequacy (Moreno-Mateos et al. 2020). The implications of
7    climate change for electricity system vulnerability will depend on the degree to which climate change
8    alters the frequency and intensity of extreme weather events. The complex compounding effects of
9    simultaneous events (e.g., high winds and lightning occurring at the same time) are not well understood.
10   High wind speeds can shear lines through mechanical failure or cause lines to collide, causing transient
11   events (Panteli and Mancarella 2015; Yalew et al. 2020). Hurricane conditions can damage electricity
12   system infrastructures, including utility-scale wind and solar PV plants. Electricity systems may
13   experience high demand when lines are particularly at risk from mechanical failure from wind and
14   storm-related effects. However, except for medium evidence of increases in heavy precipitation
15   associated with tropical cyclones, there is limited evidence that extreme wind events will increase in
16   frequency or intensity in the future (Kumar et al. 2015; Pryor et al. 2020).
17   Wildfires pose a significant threat to electricity systems in dry conditions and arid regions (Dian et al.
18   2019). With climate change, wildfires will probably become more frequent (Flannigan et al. 2013) and
19   more difficult to address given that they frequently coincide with dry air and can be exacerbated by
20   high winds (Mitchell 2013).
21   Lightning can cause wildfires or common-mode faults on electricity systems associated with vegetation
22   falling on power substations or overhead lines but is more generally associated with flashovers and
23   overloads (Balijepalli et al. 2005). Climate change may change the probability of lightning-related
24   events (Romps et al. 2014).
25   Snow and icing can impact overhead power lines by weighing them down beyond their mechanical
26   limits, leading to collapse and cascading outages (Feng et al. 2015). Snow can also lead to flashovers
27   on lines due to wet snow accumulation on insulators (Croce et al. 2018; Yaji et al. 2014) and snow and
28   ice can impact wind turbines (Davis et al. 2016). Climate change will lower risk of snow and ice
29   conditions (McColl et al. 2012), but there is still an underlying risk of sporadic acute cold conditions
30   such as those associated with winter storms in Texas in 2021 (Box 6.).
31   Flooding poses a threat to the transmission and distribution systems by inundating low-lying substations
32   and underground cables. Coastal flooding also poses a threat to electricity system infrastructure. Rising
33   sea levels from climate change and associated storm surge may also pose a significant risk for coastal
34   electricity systems (Entriken and Lordan 2012).
35   Temperature increases influence electricity load profiles and electricity generation, as well as
36   potentially impact supporting information and communication infrastructure. Heat can pose direct
37   impacts to electricity system equipment such as transformers. Referred to as solar heat faults, they occur
38   under high temperatures and low wind speeds and can be exacerbated by the urban heat island effect
39   (McColl et al. 2012). Increasing temperatures affect system adequacy by reducing electric transmission
40   capacity, simultaneously increasing peak load due to increased air conditioning needs (Bartos et al.
41   2016).
42   [START BOX 6.7 HERE]
43                     Box 6.7 Impacts of Renewable Energy Production on Climate

44   While climate change will affect energy systems (Section 6.5), the reverse is potentially also true:
45   increasing the use of renewable energy sources could affect local climate. Large solar PV arrays and
     Do Not Cite, Quote or Distribute                  6-77                                  Total pages: 217
     Final Government Distribution                       Chapter 6                          IPCC WGIII AR6

1    hydroelectric dams darken the land surface, and wind turbines extract the wind’s kinetic energy near
2    the Earth’s surface. Their environmental impacts of renewable energy production are mostly confined
3    to areas close to the production sources and have shown to be trivial compared to the mitigation benefits
4    of renewable energy (high confidence).
 5   Solar Energy. Observations and model simulations have addressed whether large-scale solar PV power
 6   plants can alter the local and regional climate. In rural areas at the local scale, large-scale solar PV farms
 7   change the surface characteristics and affect air temperatures (Taha 2013). Measurements in rural
 8   Arizona, U.S. show local nighttime temperatures 3–4°C warmer at the PV farm than surroundings
 9   (Barron-Gafford et al. 2016). In contrast, measurements in urban settings show that solar PV panels on
10   roofs provide a cooling effect (Ma et al. 2017; Taha 2013). On the regional scale, modelling studies
11   suggest cooling in urban areas (0.11–0.53°C) and warming in rural areas (up to 0.27°C) (Millstein and
12   Menon 2011). Global climate model simulations show that solar panels induce regional cooling by
13   converting part of the incoming solar energy to electricity (Hu et al. 2016). However, converting the
14   generated electricity to heat in urban areas increases regional and local temperatures, compensating for
15   the cooling effect.
16   Wind Energy. Surface temperature changes in the vicinity of wind farms have been detected (Xia et al.
17   2019; Smith et al. 2013; Lee and Lundquist 2017; Takle et al. 2019) in the form of nighttime warming.
18   Data from field campaigns suggest that a “suppression of cooling” can explain the observed warming
19   (Takle et al. 2019). Regional and climate models have been used to describe the interactions between
20   turbines and the atmosphere and find minor impacts (Vautard et al. 2014). More sophisticated models
21   confirm the local warming effect of wind farms but report that the impact on the regional area is slight
22   and occasional (Wang et al. 2019d). Wind turbines alter the transport and dissipation of momentum
23   near the surface but do not directly impact the Earth's energy balance (Fischereit et al. 2021). However,
24   the secondary modifications to the energy and water exchanges have added implications for the climate
25   system (Jacobson and Archer 2012).
26   Hydropower. The potential climate impacts of hydropower concentrate on the GHG emissions from
27   organic matter decomposition when the carbon cycle is altered by the flooding of the hydroelectric
28   power plant reservoir (Ocko and Hamburg 2019), but emissions from organic matter decomposition
29   decrease over time. The darker surface of the reservoir, compared to the lighter surrounding land may
30   counterbalance part of the reduced GHG emissions by hydropower production (Wohlfahrt et al. 2021).
31   However, these impacts vary significantly among facilities due to the surrounding land properties and
32   the area inundated by the reservoir.
33   [END BOX 6.7 HERE

36   6.6    Key Characteristics of Net Zero Energy Systems
37   6.6.1 What is a Net Zero Energy System?
38   Limiting warming to well below 2°C requires that CO2 emissions from the energy sector be reduced to
39   near zero or even below zero (Chapter 3, 6.7). Policies, technologies, behaviours, investments, and other
40   factors will determine the speed at which countries transition to net zero energy systems – those that
41   emit very little or no emissions. An understanding of these future energy systems can help to chart a
42   course toward them over the coming decades.
43   This section synthesizes current understanding of net zero energy systems. Discussions surrounding
44   efforts to limit warming are frequently communicated in terms of the point in time at which net
45   anthropogenic CO2 emissions reach zero, accompanied by substantial reductions in non-CO2 emissions
     Do Not Cite, Quote or Distribute                    6-78                                    Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    (IPCC 2018, Chapter 3). Net-zero GHG goals are also common, and they require net-negative CO2
2    emissions to compensate for residual non-CO2 emissions. Economy-wide CO2 and GHG goals appear
3    in many government and corporate decarbonization strategies, and they are used in a variety of ways.
4    Most existing carbon-neutrality commitments from countries and subnational jurisdictions aim for
5    economies with very low emissions rather than zero emissions. Offsets, carbon dioxide removal (CDR)
6    methods, and/or land sink assumptions are used to achieve net zero goals (Kelly Levin et al. 2020).
 7   Precisely describing a net zero energy system is complicated by the fact that different scenarios attribute
 8   different future CO2 emissions to the energy system, even under scenarios where economy-wide CO2
 9   emissions reach net zero. It is also complicated by the dependence of energy system configurations on
10   unknown future conditions like population and economic growth, and technological change. The energy
11   system is not the only source or sink of CO2 emissions. Terrestrial systems may store or emit carbon,
12   and CDR options like BECCS or DACCS can be used to store CO2, relieving pressure on the energy
13   system (Chapter 3). The location of such CDR options is ambiguous, as it might be deployed within or
14   outside of the energy sector (Figure 6.21), and many CDR options, such as DACCS, would be important
15   energy consumers (Bistline and Blanford 2021a, 6.6.2). If CDR methods are deployed outside of the
16   energy system (e.g., net negative agriculture, forestry, and land use CO2 emissions), it is possible for
17   the energy system to still emit CO2 but have economy-wide emissions of zero or below. When global
18   energy and industrial CO2 emissions reach net zero, the space remaining for fossil energy emissions is
19   determined by deployment of CDR options (Figure 6.21).
20   This section focuses on energy systems that produce very little or no CO2 emissions, referred to in this
21   chapter as net zero energy systems. While energy systems may not reach net zero concurrently with
22   economy-wide CO2 or GHG emissions, they are a useful benchmark for planning a path to net zero.
23   Note that the focus here is on energy systems with net zero CO2 emissions from fossil fuel combustion
24   and industrial processes, but the lessons will be broadly applicable to net zero GHG energy systems as
25   well. Net-zero GHG energy systems would incorporate the major efforts made to reduce non-CO2
26   emissions (e.g., CH4 from oil, gas and coal as discussed in Section 6.4) and would also need to
27   incorporate more CDR to compensate for remaining non-CO2 GHG emissions. Energy sector emissions
28   in many countries may not reach net zero at the same time as global energy system emissions (Figure
29   6.25, Cross-Chapter Box 3 in Chapter 3).


     Do Not Cite, Quote or Distribute                   6-79                                  Total pages: 217
     Final Government Distribution                     Chapter 6                        IPCC WGIII AR6

2       Figure 6.21 Residual emissions and CDR when global energy and industrial CO2 emissions reach net
3      zero. Residual emissions and CDR in net zero scenarios show global differences across warming levels
4     (blue = <1.5˚C, green = <2.0˚C, orange = >2.0˚C). Points represent different models and scenarios from
5    the AR6 database. In each case, the boxes show the 25th to 75th percentile ranges, and whiskers show the
6         5th and 95th percentiles. Lines and circles within the boxes denote the median and mean values,
7                                                    respectively.

8    6.6.2 Configurations of Net-zero Energy Systems
 9   Net-zero energy systems entail trade-offs across economic, environmental, and social dimensions
10   (Davis et al. 2018). Many socioeconomic, policy, and market uncertainties will also influence the
11   configuration of net zero energy systems (van Vuuren et al. 2018; Krey et al. 2019; Bistline et al. 2019;
12   Smith et al. 2015, Azevedo et al. 2021, Pye et al, 2021). There are reasons that countries might focus
13   on one system configuration versus another, including cost, resource endowments, related industrial
14   bases, existing infrastructure, geography, governance, public acceptance, and other policy priorities
15   (Section 6.6.4 and Chapter 18 of WGII).
16   Explorations of net zero energy systems have been emerging in the detailed systems modelling literature
17   (Azevedo et al. 2021; Bistline 2021b). Reports associated with net zero economy-wide targets for
18   countries and subnational entities typically do not provide detailed roadmaps or modelling but discuss
19   high-level guiding principles, though more detailed studies are emerging at national levels (Williams et
20   al. 2021a; Duan et al. 2021; Capros et al. 2019; Wei et al. 2020). Most analysis has focused on
21   identifying potential decarbonization technologies and pathways for different sectors, enumerating
22   opportunities and barriers for each, their costs, highlighting robust insights, and characterizing key
23   uncertainties (Hepburn et al. 2019; Davis et al. 2018).
24   The literature on the configuration of net zero energy systems is limited in a few respects. On the one
25   hand, there is a robust integrated assessment literature that provides characterizations of these systems
26   in broad strokes (AR6 database), offering internally consistent global scenarios to link global warming
27   targets to regional/national goals. All integrated assessment scenarios that discuss net zero energy
28   system CO2 emissions provide high-level characterizations of net zero systems. Because these
29   characterizations have less temporal, spatial, technological, regulatory, and societal detail, however,
30   they may not consider the complexities that could ultimately influence regional, national, or local

     Do Not Cite, Quote or Distribute                  6-80                                 Total pages: 217
    Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1   pathways. High-fidelity models and analyses are needed to assess the economic and environmental
2   characteristics and the feasibility of many aspects of net zero or net negative emissions energy systems
3   (high confidence) (Bistline and Blanford 2020; Blanford et al. 2018). For example, evaluating the
4   competitiveness of electricity sector technologies requires temporal, spatial, and technological detail to
5   accurately represent system investments and operations (Bistline 2021c; Victoria et al. 2021; Helistoe
6   et al. 2019; Collins et al. 2017; Santen et al. 2017).
7   Configurations of net zero energy systems will vary by region but are likely to share several common
8   characteristics (high confidence) (Figure 6.22). We focus on seven of those common characteristics in
9   the remainder of this subsection.

    Do Not Cite, Quote or Distribute                  6-81                                  Total pages: 217
    Final Government Distribution                    Chapter 6                         IPCC WGIII AR6


3     Figure 6.22 Characteristics of global net zero energy systems when global energy and industrial CO2
4    emissions reach net zero. Scenarios reaching net zero emissions show differences in residual emissions

    Do Not Cite, Quote or Distribute                 6-82                                  Total pages: 217
     Final Government Distribution                       Chapter 6                         IPCC WGIII AR6

1     and carbon removal (a), energy resources (b), electrification (c), energy intensity (as measured here by
2       energy GDP-1) (d), and emissions trajectory (e), particularly with respect to warming levels (blue =
3    <1.5˚C, green = <2.0˚C, orange = >2.0˚C, gray = unspecified). Points represent individual scenarios, with
4    probability density distributions shown along each axis for each warming level (colours corresponding to
5     warming levels) and for all scenarios (black). Points represent different models and scenarios from the
6                                                 AR6 database.

7   Limited and/or Targeted Use of Fossil Fuels
 8   Net-zero energy systems will use far less fossil fuel than today (high confidence). The precise
 9   quantity of fossil fuels will largely depend upon the relative costs of such fuels, electrification,
10   alternative fuels, and CDR (see Section in the energy system (high confidence). All of these are
11   affected by regional differences in resources (McGlade and Ekins 2015), existing energy infrastructure
12   (Tong et al. 2019), demand for energy services, and climate and energy policies. Fossil fuel use may
13   persist, for example, if and where the costs of such fuels and the compensating carbon management
14   (e.g., CDR, CCS) are less than non-fossil energy. For most applications, however, it is likely that
15   electrification (McCollum et al. 2014; Madeddu et al. 2020; Zhang and Fujimori 2020) or use of non-
16   fossil alternative fuels (Zeman and Keith 2008; Graves et al. 2011; Hänggi et al. 2019; Ueckerdt et al.
17   2021) will prove to be the cheapest options. Most residual demand for fossil fuels is likely to
18   predominantly be petroleum and natural gas given their high energy density (Davis et al. 2018), while
19   demand for coal in net zero energy systems is likely to be very low (Luderer et al. 2018; Jakob et al.
20   2020, Section 6.7.4) (high confidence).
21   There is considerable flexibility regarding the overall quantity of liquid and gaseous fuels that will be
22   required in net zero energy systems (high confidence) (Figure 6.22, Section 6.7.4). This will be
23   determined by the relative value of such fuels as compared to systems which rely more or less heavily
24   on zero-emissions electricity. In turn, the share of any fuels that are fossil or fossil-derived is uncertain
25   and will depend on the feasibility of CCS and CDR technologies and long-term sequestration as
26   compared to alternative, carbon-neutral fuels. Moreover, to the extent that physical, biological, and/or
27   socio-political factors limit the availability of CDR (Smith et al. 2015; Field and Mach 2017), carbon
28   management efforts may prioritize residual emissions related to land use and other non-energy sources.
29   Zero or Negative CO2 Emissions from Electricity
30   Net-zero energy systems will rely on decarbonized or net-negative CO2 emissions electricity systems,
31   due to the many lower-cost options for producing zero-carbon electricity and the important role of end-
32   use electrification in decarbonizing other sectors (high confidence).
33   There are many possible configurations and technologies for zero- or net-negative-emissions electricity
34   systems (high confidence). These systems could entail a mix of variable renewables, dispatchable
35   renewables (e.g., biomass, hydropower), other firm, dispatchable (“on-demand”) low-carbon generation
36   (e.g., nuclear, CCS-equipped capacity), energy storage, transmission, carbon removal options (e.g.,
37   BECCS, DACCS), and demand management (Bistline and Blanford 2021b; Bistline et al. 2018; Jenkins
38   et al. 2018b; Luderer et al. 2017). The marginal cost of deploying electricity sector mitigation options
39   increases as electricity emissions approach zero; in addition, the most cost-effective mix of system
40   resources changes as emissions approach zero and, therefore, so do the implications of electricity sector
41   mitigation for sustainability and other societal goals (Cole et al. 2021; Jayadev et al. 2020; Bistline et
42   al. 2018; Mileva et al. 2016; Sepulveda et al. 2018). Key factors influencing the electricity mix include
43   relative costs and system benefits, local resource bases, infrastructure availability, regional integration
44   and trade, co-benefits, societal preferences and other policy priorities, all of which vary by country and
45   region (Section 6.6.4). Many of these factors depend on when the net zero point is reached (Figure
46   6.22).

     Do Not Cite, Quote or Distribute                    6-83                                   Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

 1   Based on their increasing economic competitiveness, VRE technologies, especially wind and solar
 2   power, will likely comprise large shares of many regional generation mixes (high confidence) (Figure
 3   6.22). While wind and solar will likely be prominent electricity resources, this does not imply that 100%
 4   renewable energy systems will be pursued under all circumstances, since economic and operational
 5   challenges increase nonlinearly as shares approach 100% (Box 6.8) (Bistline and Blanford 2021a; Cole
 6   et al. 2021; Shaner et al. 2018; Frew et al. 2016; Imelda et al. 2018b). Real-world experience planning
 7   and operating regional electricity systems with high instantaneous and annual shares of renewable
 8   generation is accumulating, but debates continue about how much wind and solar should be included in
 9   different systems, and the cost-effectiveness of mechanisms for managing variability (Box 6.8). Either
10   firm, dispatchable generation (including nuclear, CCS-equipped capacity, dispatchable renewables such
11   as geothermal, and fossil units run with low capacity factors and CDR to balance emissions) or seasonal
12   energy storage (alongside other balancing resources discussed in Box 6.8) will be needed to ensure
13   reliability and resource adequacy with high percentages of wind and solar (Jenkins et al. 2018b;
14   Dowling et al. 2020; Denholm et al. 2021) though each option involves uncertainty about costs, timing,
15   and public acceptance (Albertus et al. 2020).
16   Electricity systems require a range of different functional roles – for example, providing energy,
17   capacity, or ancillary services. As a result, a range of different types of generation, energy storage, and
18   transmission resources may be deployed in these systems (Baik et al. 2021). There are many options
19   for each of these roles, each with their strengths and weaknesses (Sections 6.4.3 and 6.4.4), and
20   deployment of these options will be influenced by the evolution of technological costs, system benefits,
21   and local resources (Veers et al. 2019; Mai et al. 2018; Bistline et al. 2018; Hirth 2015; Fell and Linn
22   2013).
23   System management is critical for zero- or negative-emissions electricity systems. Maintaining
24   reliability will increasingly entail system planning and operations that account for characteristics of
25   supply- and demand-side resources (Hu et al. 2018). Coordinated planning and operations will likely
26   become more prevalent across portions of the electricity system (e.g., integrated generation,
27   transmission, and distribution planning), across sectors, and across geographies (Bistline and Young
28   2019; Chan et al. 2018; Konstantelos et al. 2017; EPRI 2017, Section 6.4.3).
29   Energy storage will be increasingly important in net zero energy systems, especially in systems with
30   shares of VRE (high confidence). Deployment of energy storage will vary based on the system benefits
31   and values of different options (Arbabzadeh et al. 2019; Denholm and Mai 2019). Diurnal storage
32   options like lithium-ion batteries have different value than storing and discharging electricity over
33   longer periods through long-duration energy storage with less frequent cycling, which require different
34   technologies, supporting policies, and business models (Sepulveda et al. 2021; Dowling et al. 2020;
35   Gallo et al. 2016; Albertus et al. 2020; Blanco and Faaij 2017) (Section 6.4.4). The value of energy
36   storage varies with the level of deployment and on the competitiveness of economic complements such
37   as VRE options (Bistline and Young 2020; Mileva et al. 2016) and substitutes such as flexible demand
38   (Brown et al. 2018; Merrick et al. 2018), transmission (Merrick et al. 2018; Brown et al. 2018;
39   Schlachtberger et al. 2017; Bistline and Young 2019), trade (Bistline et al. 2020b), dispatchable
40   generators (Hittinger and Lueken 2015; Gils et al. 2017; Arbabzadeh et al. 2019), DAC (Daggash et al.
41   2019), and efficiencies in system operations (Tuohy et al. 2015).
42   The approach to other sectors could impact electricity sector planning, and the role of some technologies
43   (e.g., hydrogen, batteries, CCS) could depend on deployment in other sectors. CCS offers opportunities
44   for CO2 removal when fuelled with syngas or biomass containing carbon captured from the atmosphere
45   (Hepburn et al. 2019); however, concerns about lifecycle environmental impacts, uncertain costs, and
46   public acceptance are potential barriers to widespread deployment (Section 6.4.2). It is unclear whether
47   CDR options like BECCS will be included in the electricity mix to offset continued emissions in other
48   parts of the energy system or beyond (Mac Dowell et al. 2017; Luderer et al. 2018; Bauer et al. 2018a).

     Do Not Cite, Quote or Distribute                   6-84                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    Some applications may also rely on PtX electricity conversion to create low-emissions synthetic fuels
2    (Sections, 6.4.4, and 6.4.5), which could impact electricity system planning and operations.
3    Additionally, if DAC technologies are used, electricity and heat requirements to operate DAC could
4    impact electricity system investments and operations (Bistline and Blanford 2021a; Realmonte et al.
5    2019).
6    [START BOX 6.8 HERE]
7                          Box 6.8 100% Renewables in Net Zero Energy Systems

 8   The decreasing cost and increasing performance of renewable energy has generated interest in the
 9   feasibility of providing nearly all energy services with renewables. Renewable energy includes wind
10   power, solar power, hydroelectric power, bioenergy, geothermal energy, tidal power, and ocean power.
11   There are two primary frames around which 100% renewable energy systems are discussed: 100%
12   renewable electricity systems and 100% renewable energy systems, considering not only electricity but
13   all aspects of the energy system.
14   It is technically feasible to use very high renewable shares (e.g., above 75% of annual regional
15   generation) to meet hourly electricity demand under a range of conditions, especially when VRE
16   options, notably wind and solar, are complemented by other resources (high confidence). There are
17   currently many grids with high renewable shares and large anticipated roles for VRE sources, in
18   particular wind and solar (see Section 6.4), in future low-carbon electricity systems. An increasingly
19   large set of studies examines the feasibility of high renewable penetration and economic drivers under
20   different policy, technology, and market scenarios (Denholm et al. 2021; Blanford et al. 2021; Bistline
21   et al. 2019; Hansen et al. 2019; Jenkins et al. 2018b; Cochran et al. 2014; Dowling et al. 2020; Deason
22   2018). High wind and solar penetration involves technical and economic challenges due to their unique
23   characteristics such as spatial and temporal variability, short- and long-term uncertainty, and non-
24   synchronous generation (Cole et al. 2017). These challenges become increasingly important as
25   renewable shares approach 100% (Sections and 6.4.3).
26   There are many balancing options in systems with very high renewables (Denholm et al. 2021; Bistline
27   2021a; Mai et al. 2018; Milligan et al. 2015; Jenkins et al. 2018b).
28   ●   Energy storage: Energy storage technologies like batteries, pumped hydro, and hydrogen can
29       provide a range of system services (Balducci et al. 2018; Bistline et al. 2020a; Section 6.4.4).
30       Lithium-ion batteries have received attention as costs fall and installations increase, but very high
31       renewable shares typically entail either dispatchable generation or long-duration storage in addition
32       to short-duration options (Schill 2020; Arbabzadeh et al. 2019; Jenkins et al. 2018b). Energy storage
33       technologies are part of a broad set of options (including synchronous condensers, demand-side
34       measures, and even inverter-based technologies themselves) for providing grid services (Castillo
35       and Gayme 2014; EPRI 2019a).
36   ●   Transmission and trade: To balance differences in resource availability, high renewable systems
37       will very likely entail investments in transmission capacity (Zappa et al. 2019; Pleßmann and
38       Blechinger 2017; Macdonald et al. 2016; Mai and Et al 2014; Section 6.4.5) and changes in trade
39       (Abrell and Rausch 2016; Bistline et al. 2019). These increases will likely be accompanied by
40       expanded balancing regions to take advantage of geographical smoothing.
41   ●   Dispatchable (“on-demand”) generation: Dispatchable generation could include flexible fossil
42       units or low-carbon fuels such as hydrogen with lower minimum load levels (Bistline 2019;
43       Denholm et al. 2018), renewables like hydropower, geothermal, or biomass (Hansen et al. 2019;
44       Hirth 2016), or flexible nuclear (Jenkins et al. 2018a). The composition depends on costs and other
45       policy goals, though in all cases, capacity factors are low for these resources (Mills et al. 2020).
46   ●   Demand management: Many low-emitting and high-renewables systems also utilize increased
47       load flexibility in the forms of energy efficiency, demand response, and demand flexibility, utilizing
     Do Not Cite, Quote or Distribute                  6-85                                  Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1        newly electrified end uses such as electric vehicles to shape demand profiles to better match supply
2        (Bistline 2021a; Imelda et al. 2018a; Hale 2017; Brown et al. 2018; Ameli et al. 2017).
3    ●   Sector coupling: Sector coupling includes increased end-use electrification and PtX electricity
4        conversion pathways, which may entail using electricity to create synthetic fuels such as hydrogen
5        (Ueckerdt et al. 2021; Davis et al. 2018) (see Sections 6.4.3, 6.4., 6.4.5,, and
 6   Deployment of integration options depends on their relative costs and value, regulations, and electricity
 7   market design. There is considerable uncertainty about future technology costs, performance,
 8   availability, scalability, and public acceptance (Kondziella and Bruckner 2016; Bistline et al. 2019).
 9   Deploying balanced resources likely requires operational, market design, and other institutional
10   changes, as well as technological changes in some cases (Denholm et al. 2021; Cochran et al. 2014).
11   Mixes will differ based on resources, system size, flexibility, and whether grids are isolated or
12   interconnected.
13   Although there are no technical upper bounds on renewable electricity penetration, the economic value
14   of additional wind and solar capacity typically decreases as their penetration rises, creating economic
15   challenges at higher deployment levels (Denholm et al. 2021; Millstein et al. 2021; Cole et al. 2021;
16   Gowrisankaran et al. 2016; Hirth 2013). The integration options above, as well as changes to market
17   design, can mitigate these challenges but likely will not solve them, especially since these options can
18   exhibit declining value themselves (Denholm and Mai 2019; Bistline 2017; De Sisternes et al. 2016)
19   and may be complements or substitutes to each other.
20   Energy systems that are 100% renewable (including all parts of the energy sector, and not only
21   electricity generation) raise a range of technological, regulatory, market, and operational challenges that
22   make their competitiveness uncertain (high confidence). These systems require decarbonizing all
23   electricity, using this zero-carbon electricity broadly, and then utilizing zero-carbon energy carriers for
24   all end uses not served by electricity, for example, air travel, long-distance transport, and high-
25   temperature process heat. Broader questions emerge regarding the attractiveness of supplying all
26   energy, and not just electricity, with renewables (Figure 6.22). Integrated assessment and energy
27   systems research suggest large roles for renewables, but energy and electricity shares are far from 100%,
28   even with stringent emissions reductions targets and optimistic assumptions about future cost reductions
29   (Huntington et al. 2020; Jenkins et al. 2018b; Bauer et al. 2018; Bistline et al. 2018, Section 6.7.1).
30   Scenarios with 100% renewable energy systems are an emerging subset in the decarbonization
31   literature, especially at regional levels (Denholm et al. 2021; Hansen et al. 2019). Many 100%
32   renewables studies focus more heavily on electrification for decarbonizing end uses, and include less
33   biofuels and hydrogen than the broader literature on deep decarbonization (Bauer et al. 2018a). These
34   studies typically assume a constrained set of available technologies to demonstrate the technical
35   feasibility of very high renewable systems and do not optimize to find least-cost, technology-neutral
36   decarbonization pathways, and many 100% renewables studies focus on the electricity sector or a
37   limited number of sectors (Hansen et al. 2019; Jenkins et al. 2018a). In addition to renewables, studies
38   broadly agree that including additional low-carbon options – including not only low-carbon electricity
39   but also targeted use of fossil fuels with and without CCS (Section and alternative fuels for
40   sectors that are difficult to electrify (Section – can lower the cost of decarbonization even with
41   very high shares of renewables (Figure 6.22). However, there is disagreement about the magnitude of
42   cost savings from larger portfolios, which depend on context- and scenario-specific assumptions about
43   technologies, markets, and policies.
44   [END BOX 6.8 HERE]
45   Widespread Electrification of End Uses
46   Net-zero energy systems will rely more heavily on increased use of electricity (electrification) in end
47   uses (high confidence). The literature on net zero energy systems almost universally calls for increased

     Do Not Cite, Quote or Distribute                   6-86                                  Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1    electrification (Williams et al. 2012; Sugiyama 2012; Williams et al. 2014; Rogelj et al. 2015a; Sachs
2    et al. 2016; Kriegler et al. 2014a; Sven et al. 2018; Luderer et al. 2018; Schreyer et al. 2020). At least
3    30% of the global final energy needs are expected to be served by electricity, with some estimates
4    suggesting upwards of 80% of total energy use being electrified (Figure 6.22, panel c). Increased
5    electrification is especially valuable in net zero energy systems in tandem with decarbonized electricity
6    generation or net-negative emissions electricity generation (Section Flexible electric loads
7    (electric vehicles, smart appliances) can in turn facilitate incorporation of VRE electricity options,
8    increase system flexibility, and reduce needs for grid storage (Section 6.4.3) (Mathiesen et al. 2015);
9    Lund et al., 2018).
10   Several end-uses such as passenger transportation (light-duty electric vehicles, two and three wheelers,
11   buses, rail) as well as building energy uses (lighting, cooling) are likely to be electrified in net zero
12   energy systems (high confidence). Variations in projections of electrification largely result from
13   differences in expectations about the ability and cost-competitiveness of electricity to serve other end
14   uses such as non-rail freight transport, aviation, and heavy industry (McCollum et al. 2014; Breyer et
15   al. 2019; Bataille et al. 2016; EPRI 2018) (Section, especially relative to biofuels and hydrogen
16   (‘low carbon fuels’) (Sachs et al. 2016; Rockström et al. 2017; McCollum et al. 2014), the prospects for
17   which are still quite uncertain (Section 6.4). The emergence of CDR technologies and the extent to
18   which they allow for residual emissions as an alternative to electrification will also affect the overall
19   share of energy served by electricity (Section
20   Regions endowed with cheap and plentiful low-carbon electricity resources (wind, solar, hydropower)
21   are likely to emphasize electrification, while those with substantial bioenergy resources or availability
22   of other liquid fuels might put less emphasis on electrification, particularly in hard-to-electrify end-uses
23   (medium confidence). For example, among a group of Latin American countries, relative assumptions
24   about liquid fuels and electricity result in an electrification range of 28–82% for achieving a net zero
25   energy system (Bataille et al. 2020). Similarly, the level of penetration of biofuels that can substitute
26   for electrification will depend on regional circumstances such as land-use constraints, competition with
27   food, and sustainability of biomass production (Section
28   Electrification of most buildings services, with the possible exception of space heating in extreme
29   climates, is expected in net zero energy systems (high confidence) (Chapter 9). Space cooling and water
30   heating are expected to be largely electrified. Building electrification is expected to rely substantially
31   on heat pumps, which will help lower emissions both through reduced thermal requirements and higher
32   efficiencies (Mathiesen et al. 2015; Rissman et al.; Sven et al. 2018). The level of electrification for
33   heating will depend on the tradeoffs between building or household level heat pumps versus more
34   centralized district heating network options (Mathiesen et al. 2015; Brown et al. 2018), as well as the
35   cost and performance of heat pumps in more extreme climates and grid infrastructure (EPRI 2018;
36   Waite and Modi 2020).
37   A significant share of transportation, especially road transportation, is expected to be electrified in net
38   zero energy systems (high confidence). In road transportation, two-three wheelers, light-duty vehicles
39   (LDVs), and buses, are especially amenable to electrification, with more than half of passenger LDVs
40   expected to be electrified globally in net zero energy systems (medium confidence) ((Bataille et al. 2020;
41   Sven et al. 2018; Khalili et al. 2019; Fulton et al. 2015). Long-haul trucks, large ships, and aircraft are
42   expected to be harder to electrify absent technological breakthroughs (Mathiesen et al. 2015; Fulton et
43   al. 2015), although continued improvements in battery technology may enable electrification of long-
44   haul trucks (Nykvist and Olsson 2021; Chapter 10). Due to the relative ease of rail electrification, near
45   complete electrification of rail and a shift of air and truck freight to rail is expected in net zero energy
46   systems (Sven et al. 2018; Khalili et al. 2019; Rockström et al. 2017; Fulton et al. 2015). The degree of
47   modal shifts and electrification will depend on local factors such as infrastructure availability and
48   location accessibility. Due to the challenges associated with electrification of some transport modes,

     Do Not Cite, Quote or Distribute                   6-87                                   Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1    net zero energy systems may include some residual emissions associated with the freight sector that are
2    offset through CDR technologies (Muratori et al. 2017b), or reliance on low and zero-carbon fuels
3    instead of electrification.
 4   A non-trivial number of industry applications could be electrified as a part of a net zero energy system,
 5   but direct electrification of heavy industry applications such as cement, primary steel manufacturing,
 6   and chemical feedstocks is expected to be challenging (medium confidence) (Davis et al. 2018;
 7   Madeddu et al. 2020; Philibert 2019; van Sluisveld et al. 2021). Process and boiler heating in industrial
 8   facilities are anticipated to be electrified in net zero energy systems. Emissions intensity reductions for
 9   cement and concrete production can be achieved through the use of electrified cement kilns, while
10   emissions associated with steel production can be reduced through the use of an electric arc furnace
11   (EAF) powered by decarbonized electricity (Rissman et al.). Electricity can also be used to replace
12   thermal heat such as resistive heating, electric arc furnaces, and laser sintering (Rissman et al.; Madeddu
13   et al. 2020). One study found that as much as 60% of the energy end-use in European industry could be
14   met with direct electrification using existing and emerging technologies (Madeddu et al. 2020). Industry
15   electrification for different regions will depend on the economics and availability of alternative
16   emissions mitigation strategies such as carbon neutral fuels and CCS (Davis et al. 2018; Madeddu et al.
17   2020).
18   Alternative Fuels in Sectors not Amenable to Electrification
19   Net-zero energy systems will need to rely on alternative fuels – notably hydrogen or biofuels – in several
20   sectors that are not amenable to electricity and otherwise hard to decarbonize (medium confidence).
21   Useful carbon-based fuels (e.g., methane, petroleum, methanol), hydrogen, ammonia, or alcohols can
22   be produced with net zero CO2 emissions and without fossil fuel inputs (Sections 6.4.4 and 6.4.5). For
23   example, liquid hydrocarbons can be synthesized via hydrogenation of non-fossil carbon by processes
24   such as Fischer-Tropsch (Mac Dowell et al. 2017) or by conversion of biomass (Tilman et al. 2009).
25   The resulting energy-dense fuels can serve applications that are difficult to electrify, but it is not clear
26   if and when the combined costs of obtaining necessary feedstocks and producing these fuels without
27   fossil inputs will be less than continuing to use fossil fuels and managing the related carbon through,
28   for example, CCS or CDR (Ueckerdt et al. 2021)
29   CO2 emissions from some energy services are expected to be particularly difficult to cost-effectively
30   avoid, among them aviation; long-distance freight by ships; process emissions from cement and steel
31   production; high-temperature heat (e.g., >1000°C); and electricity reliability in systems with high
32   penetration of variable renewable energy sources (NAS; Davis et al. 2018; Luderer et al. 2018;
33   Chiaramonti 2019; Sepulveda et al. 2018; Bataille 2020; Rissman et al.; Thiel and Stark 2021; Madeddu
34   et al. 2020). The literature focused on these services and sectors is growing, but remains limited, and
35   provides minimal guidance on the most promising or attractive technological options and systems for
36   avoiding these sectors’ emissions. Technological solutions do exist, but those mentioned in the literature
37   are prohibitively expensive, exist only at an early stage, and/or are subject to much broader concerns
38   about sustainability (e.g., biofuels) (Davis et al. 2018).
39   Liquid biofuels today supply about 4% of transportation energy worldwide, mostly as ethanol from
40   grain and sugar cane and biodiesel from oil seeds and waste oils (Davis et al. 2018). These biofuels
41   could conceivably be targeted to difficult-to-electrify sectors, but face substantial challenges related to
42   their life-cycle carbon emissions, cost, and further scalability (Tilman et al. 2009; Staples et al. 2018),
43   (Section 6.4.2). The extent to which biomass will supply liquid fuels or high temperature heat for
44   industry in a future net zero energy system will thus depend on advances in conversion technology that
45   enable use of feedstocks such as woody crops, agricultural residues, algae, and wastes, as well as
46   competing demands for bioenergy and land, the feasibility of other sources of carbon-neutral fuels, and
47   integration of bioenergy production with other objectives, including CDR, economic development, food
48   security, ecological conservation, and air quality (Lynd 2017; Laurens 2017; Williams and Laurens
     Do Not Cite, Quote or Distribute                   6-88                                   Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    2010; Strefler et al. 2018; Bauer et al. 2018a; Fargione 2010; Creutzig et al. 2015; Bauer et al. 2018b;
2    Muratori et al. 2020b; Chatziaras et al. 2016; Fennell et al. 2021) (Section
 3   Costs are the main barrier to synthesis of net zero emissions fuels (high confidence), particularly costs
 4   of hydrogen (a constituent of hydrocarbons, ammonia, and alcohols) (Section 6.4.5). Today, most
 5   hydrogen is supplied by steam reformation of fossil methane (CH4 into CO2 and H2) at a cost of 1.30-
 6   1.50 USD kg-1 (Sherwin 2021). Non-fossil hydrogen can be obtained by electrolysis of water, at current
 7   costs of 5-7 USD kg-1 H2-1 (assuming relatively low electricity costs and high utilization rates) (Graves
 8   et al. 2011; Newborough and Cooley 2020; Peterson et al. 2020; DOE 2020a). At these costs for
 9   electrolytic hydrogen, synthesized net zero emissions fuels would cost at least 1.6 USD per liter of
10   diesel equivalent (or 6 USD gallon-1 and 46 USD GJ-1, assuming non-fossil carbon feedstock costs of
11   100 USD per ton of CO2 and low process costs of 0.05 USD liter-1 or 1.5 USD GJ-1). Similar calculations
12   suggest that synthetic hydrocarbon fuels could currently avoid CO2 emissions at a cost of 936-1404
13   USD ton-1 (Ueckerdt et al. 2021). However, economies of scale are expected to bring these costs down
14   substantially in the future (Ueckerdt et al. 2021; IRENA 2020c), and R&D efforts are targeting 60-80%
15   reductions in costs (to less than 2 USD kg-1 (H2)-1) possibly by use of less mature but promising
16   technologies such as high-temperature electrolysis and thermochemical water splitting (Schmidt et al.
17   2017; Pes et al. 2017; DOE 2018; Saba et al. 2018; Kuckshinrichs et al. 2017; DOE 2020b).
18   Technologies capable of producing hydrogen directly from water and sunlight (photoelectrochemical
19   cells or photocatalysts) are also under development, but still at an early stage (DOE 2020a; Nielander
20   et al. 2015). High hydrogen production efficiencies have been demonstrated, but costs, capacity factors,
21   and lifetimes need to be improved in order to make such technologies feasible for net zero emissions
22   fuel production at scale (DOE 2020a; Newborough and Cooley 2020; McKone et al. 2014).
23   The carbon contained in net zero emissions hydrocarbons must have been removed from the atmosphere
24   either through DAC, or, in the case of biofuels, by photosynthesis (which could include CO 2 captured
25   from the exhaust of biomass or biogas combustion) (Zeman and Keith 2008; Graves et al. 2011). A
26   number of different groups are now developing DAC technologies, targeting costs of 100 USD per ton
27   of CO2 or less (Darton and Yang 2018; Keith et al. 2018; Fasihi et al. 2019).

29     Figure 6.23 Schematic of net zero emissions energy system, including methods to address difficult-to-
30                                  electrify sectors. (Source: Davis et al. 2018)

     Do Not Cite, Quote or Distribute                  6-89                                  Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    [START BOX 6.9 HERE]
2                                       Box 6.9 The Hydrogen Economy

 3   The phrase “hydrogen economy” is often used to describe future energy systems in which hydrogen
 4   plays a prominent role. These future energy systems would not use hydrogen for all end uses; they
 5   would use hydrogen to complement other energy carriers, mainly electricity, where hydrogen might
 6   have advantages. Hydrogen could provide long-term electricity storage to support high-penetration of
 7   intermittent renewables and could enable trading and storage of electricity between different regions to
 8   overcome seasonal or production capability differences (Dowling et al. 2020; Sepulveda et al. 2021). It
 9   could also be used in lieu of natural gas for peaking generation, provide process heat for industrial
10   needs, or be used in the metal sector via direct reduction of iron ore (Chapter 11). Clean hydrogen could
11   be used as a feedstock in the production of various chemicals and synthetic hydrocarbons. Finally,
12   hydrogen-based fuel cells could power vehicles. Recent advances in battery storage make electric
13   vehicles the most attractive alternative for light-duty transport. However, fuel cell technology could
14   complement electric vehicles in supporting the decarbonisation of heavy-duty transport segments (e.g.,
15   trucks, buses, ships, and trains) (Chapter 10).
16   Hydrogen production costs have historically been prohibitive, but recent technological developments
17   are bringing costs down. These developments include improvements in hydrogen production
18   technologies in terms of efficiency and capital costs (e.g., SMR) (Alrashed and Zahid 2021; Boretti and
19   Banik 2021) and the emergence of alternative production technologies such as electrolysers (Dawood
20   et al. 2020). These technological changes, along with decreasing costs of renewable power, are
21   increasing the viability of hydrogen. Other improvements in hydrogen-based technologies are also
22   emerging quickly. Gas turbines now run on blended fuels containing 5-95% hydrogen by volume (GE
23   2020) and could operate entirely on hydrogen by 2030 (Pflug et al. 2019). Fuel cell costs have decreased
24   by 80-95% since the early 2000s, while power density and durability have improved (Kurtz et al. 2019;
25   IEA 2019e; Jouin et al. 2016).
26   For hydrogen to support decarbonisation, it will need to be produced from zero-carbon or extremely
27   low-carbon energy sources. One such production category is “green hydrogen.” While there is no
28   unified definition for green hydrogen, it can be produced by the electrolysis of water using electricity
29   generated without carbon emissions (such as renewables). Hydrogen can also be produced through
30   biomass gasification with CCS (BECCS), leading to negative carbon emissions (del Pozo et al. 2021).
31   Additionally, “blue hydrogen” can be produced from natural gas through the process of auto-thermal
32   reforming (ATR) or steam methane reforming (SMR), combined with CCS technology that would
33   absorb most of the resulting CO2 (80-90%).
34   However, the potential role of hydrogen in future energy systems depends on more than just production
35   methods and costs. For some applications, the competitiveness of hydrogen also depends on the
36   availability of the infrastructure needed to transport and deliver it at relevant scales (Lee et al. 2021).
37   Transporting hydrogen through existing gas pipelines is generally not feasible without changes to the
38   infrastructure itself (Muratori et al. 2018; Gumber and Gurumoorthy 2018). Existing physical barriers,
39   such as steel embrittlement and degradation of seals, reinforcements in compressor stations, and valves,
40   require retrofitting during the conversion to H2 distribution or new H2 dedicated pipelines to be
41   constructed (Dohi et al. 2016). The capacity to leverage and convert existing gas infrastructure to
42   transport hydrogen will vary regionally, but in many cases could be the most economically viable
43   pathway (Brändle et al. 2021; Cerniauskas et al. 2020; Brooks 2021; Wettengel 2021). Hydrogen could
44   also be transported as liquid gas or as liquid organic hydrogen carriers such as ammonia, for which
45   industry knowledge exists (Hong et al. 2021; Wulf et al. 2018; Demir et al. 2018). Additionally,
46   improvements in fuel cell technologies are needed to make hydrogen-based transport economically

     Do Not Cite, Quote or Distribute                   6-90                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    viable. There are also safety concerns associated with the flammability (Nilsson et al. 2017) and storage
2    (Andersson and Grönkvist 2019; Caglayan et al. 2019) of hydrogen which will need to be considered.
3    [END BOX 6.9 HERE]
4   Using Less Energy and Using It More Efficiently
5    Demand-side or demand reduction strategies include technology efficiency improvements, strategies
6    that reduce energy consumption or demand for energy services (such as reducing the use of personal
7    transportation, often called “conservation”) (Creutzig et al. 2018), and strategies such as load
8    curtailment.
 9   Net-zero energy systems will use energy more efficiently than those of today (high confidence). Energy
10   efficiency and energy use reduction strategies are generally identified as being flexible and cost-
11   effective, with the potential for large scale deployment (Chapters 5, 9, 10, and 11). For this reason,
12   existing studies find that energy efficiency and demand reduction strategies will be important
13   contributors to net zero energy systems (Creutzig et al. 2018; Davis et al. 2018; DeAngelo et al. 2021).
14   Lower demand reduces the need for low-carbon energy or alternative fuel sources.
15   Characterizing efficiency of net zero energy systems is problematic due to measurement challenges
16   (high confidence). Efficiency itself is difficult to define and measure across full economies (Saunders
17   et al. 2021). There is no single definition of energy efficiency and the definition understandably depends
18   on the context used (Patterson 1996), which ranges from device level efficiency all the way to the
19   efficient use of energy throughout an economy. Broadly, energy efficient strategies allow for the same
20   level of services or output while using less energy. At the level of the entire economy, measures such
21   as primary or final energy per capita or per GDP are often used as a proxy for energy efficiency, but
22   these measures reflect not only efficiency, but also many other factors such as industrial structure,
23   endowed natural resources, consumer preferences, policies, and regulations. Energy efficiency and
24   other demand-side strategies represent such a large set of technologies, strategies, policies, market and
25   consumers’ responses and policies that aggregate measures can be difficult to define (Saunders et al.
26   2021).
27   Measurement issues notwithstanding, virtually all studies that address net zero energy systems assume
28   improved energy intensity in the future (high confidence). The overall efficiency outcomes and the
29   access to such improvements across different nations, however, is not clear. Energy consumption will
30   increase over time despite energy efficiency improvements due to population growth and development
31   (DeAngelo et al. 2021).
32   A study (DeAngelo et al. 2021) reviewed 153 IAM scenarios that attain net zero energy sector CO2
33   emissions and found that, under a scenario with net zero emissions: global final energy per capita lies
34   between 21–109 GJ per person (median: 57), in comparison to 2018 global final energy use of 55 GJ
35   per person; many countries use far more energy per capita than today as their incomes increase; global
36   final energy use per unit of economic output ranges from 0.7–2.2 EJ per trillion USD (median: 1.5), in
37   comparison to 5 EJ per trillion USD in 2018; and the median final energy consumption is 529 EJ. By
38   comparison, final energy consumption would be 550 EJ if current energy consumption per capita
39   continued under a future population of 10 billion people. Across all scenarios, total final energy
40   consumption is higher today than in the year in which net zero emissions are attained, and regionally,
41   only the OECD+EU and Eurasia have lower median total final energy than in 2010.
42   Net-zero energy systems will be characterized by greater efficiency and more efficient use of energy
43   across all sectors (high confidence). Road transportation efficiency improvements will require a shift
44   from liquid fuels (Chapters 5 and 10). Emissions reductions will come from a transition to electricity,
45   hydrogen, or synthetic fuels produced with low carbon energy sources or processes. Vehicle
46   automation, ride-hailing services, online shopping with door delivery services, and new solutions like

     Do Not Cite, Quote or Distribute                  6-91                                  Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    last mile delivery with drones may result in increased service share. Lighter vehicles, a shift to public
2    transit, and incorporation of 2- and 3-wheelers will be features of a net zero energy system (Chapter
3    10). Teleworking and automation of work may provide reductions in driving needs. Other sectors, such
4    as air travel and marine transportation may rely on alternative fuels such as biofuels, synthetic fuels,
5    ammonia, produced with zero carbon energy source (Section
 6   Under net zero energy systems, buildings would by characterized by improved construction materials,
 7   an increase in multi-family dwellings, early retirement of inefficient buildings, smaller floor areas, and
 8   smart controls to optimize energy use in the building, namely for heating, cooling, LED lighting, and
 9   water heating (Chapter 9). End-uses would utilize electricity, or potentially hydrogen, produced from
10   zero carbon sources. The use of electricity for heating and cooking may often be a less efficient process
11   at converting primary energy to energy services than using natural gas, but using natural gas would
12   require CDR in order to be considered net zero emissions. Changes in behaviour may modestly lower
13   demand. Most economies would have buildings with more efficient technologies powered by zero
14   carbon electricity, and developing economics would shift from biomass to electricity, raising their
15   energy consumption as population and wealth increase under net zero energy systems.
16   Industry has seen major efficiency improvements in the past, but many processes are now close to their
17   thermodynamic limits. Electrification and breakthrough processes (such as producing steel with
18   electricity and H2), using recycled materials, using heat more efficiently by improving thermal
19   insulation, and using waste heat for heat pumps, as well using advanced sensors, monitoring, and
20   visualization and communication technologies may provide further efficiency improvements. (Chapter
21   11)
22   Greater Reliance on Integrated Energy System Approaches
23   Energy systems integration refers to connected planning and operations across energy carriers,
24   including electricity, fuels, and thermal resources. Coordinated planning could be important in lowering
25   system costs, increasing reliability, minimizing environmental impacts, and ensuring that costs of R&D
26   and infrastructure account for not just current needs but also for those of future energy systems (Section
27   6.4.3). Integration includes not only the physical energy systems themselves but also simultaneous
28   societal objectives (e.g., sustainable development goals), innovation processes (e.g., coordinating R&D
29   to increase the likelihood of beneficial technological spillovers), and other institutional and
30   infrastructural transformations (Sachs et al. 2019). Given system variability and differences in regional
31   resources, there are economic and technical advantages to greater coordination of investments and
32   policies across jurisdictions, sectors, and levels of government (Schmalensee and Stavins 2017).
33   Coordinated planning and operations can improve system economics by sharing resources, increasing
34   the utilization of capital-intensive assets, enhancing the geographical diversity of resource bases, and
35   smoothing demand. But integration could require regulatory and market frameworks to facilitate and
36   appropriate price signals to align incentives and to coordinate investments and operations.
37   Carbon-neutral energy systems are likely to be more interconnected than those of today (high
38   confidence). The many possible feedstocks, energy carriers, and interconversion processes imply a
39   greater need for the integration of production, transport, storage, and consumption of different fuels
40   (Davis et al. 2018). For instance, electrification is expected to play an important role in decarbonizing
41   light-duty vehicles (Chapter 10, Section 6.4.3), yet the electricity and transport sectors have few direct
42   interactions today. Systems integration and sectoral coupling are increasingly relevant to ensure that net
43   zero energy systems are reliable, resilient, and affordable (EPRI 2017; O’Malley et al. 2020; Buttler
44   and Spliethoff 2018; Martin et al. 2017). Deep decarbonization offers new opportunities and challenges
45   for integrating different sectors as well as supply- and demand-side options. For instance, increasing
46   electrification will change daily and seasonal load shapes, and end-use flexibilities and constraints could
47   impact the desirability of different supply-side technologies (EPRI 2019b; Brown et al. 2018). The
48   feasibility of net zero energy system configurations could depend on demonstrating cross-sector
     Do Not Cite, Quote or Distribute                   6-92                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    benefits like balancing VRE sources in the electricity sector, and on offering the flexibility to produce
2    multiple products. For instance, low-emissions synthetic fuels could help to bridge stationary and
3    mobile applications, since fuel markets have more flexibility than instantaneously balanced electricity
4    markets due to the comparative ease and cost of large-scale, long-term storage of chemical fuels (Davis
5    et al. 2018).
 6   There are few detailed archetypes of integrated energy systems that provide services with zero- or net-
 7   negative CO2 emissions (such as (Jacobson et al. 2019)), so there is considerable uncertainty about
 8   integration and interactions across parts of the system. Although alternate configurations, tradeoffs, and
 9   pathways are still being identified, common elements include fuels and processes like zero- or negative-
10   CO2 electricity generation and transmission, hydrogen production and transport, synthetic hydrocarbon
11   production and transport, ammonia production and transport, and carbon management, where linkages
12   across pathways could include the use of electricity to produce hydrogen via electrolysis (Davis et al.
13   2018; Jenkins et al. 2018b; van Vuuren et al. 2018; Shih et al. 2018; Moore 2017; Smith et al. 2016).
14   Linked analytical frameworks are increasing being used to understand the potential role for system
15   coupling with greater temporal resolution, spatial resolution, and heterogeneity of consumer and firm
16   decisions (Pye et al. 2021; Gerboni et al. 2017; Santen et al. 2017; Collins et al. 2017; Bistline and de
17   la Chesnaye 2017; Bohringer and Rutherford 2008).
18   Challenges associated with integrating net zero energy systems include rapid technological change, the
19   importance of behavioural dimensions in domains with limited experience and data, policy changes and
20   interactions, and path dependence. Technological cost and public acceptance will influence the degree
21   of integration. Sectoral pathways will likely be adaptive and adjust based on the resolution of
22   uncertainties over time, and the relative competitiveness will evolve as the technological frontier
23   evolves, which is a complex and path-dependent function of deployment, R&D, and inter-industry
24   spillovers. Supply-side options interact with demand-side measures in increasingly integrated energy
25   systems (van Vuuren et al. 2018; Sorrell 2015).
26 Carbon Dioxide Removal
27   While CDR is likely necessary for net zero energy systems, the scale and mix of strategies is unclear –
28   nonetheless some combination of BECCS and DACCS are likely to be part of net zero energy systems
29   (high confidence). Studies indicate that energy-sector CDR may potentially remove 5–12 GtCO2
30   annually globally in net zero energy systems (Fuss et al. 2018) (Figure 6.22; Section 6.7; Chapter 12).
31   CDR is not intended as a replacement for emissions reduction, but rather as a complementary effort to
32   offset residual emissions from sectors that are not decarbonized and from other low-carbon technologies
33   such as fossil CCS (McLaren et al. 2019; Gaffney et al. 2020; Iyer et al. 2021).
34   CDR covers a broad set of methods and implementation options (Chapters 7 and 12). The two CDR
35   methods most relevant to the energy sector are BECCS, which is used to produce energy carriers, and
36   DACCS which is an energy user (Smith et al. 2016; Singh and Colosi 2021). BECCS has value as an
37   electricity generation technology, providing firm, dispatchable power to support electricity grids with
38   large amounts of VRE sources, and reducing the reliance on other means to manage these grids,
39   including electricity storage (Bistline and Blanford 2021a; Mac Dowell et al. 2017). BECCS may also
40   be used to produce liquid fuels or gaseous fuels, including hydrogen (Section (Muratori et al.
41   2020b). For instance, CO2 from bio-refineries could be captured at <45 USD tCO2-1 (Sanchez et al.
42   2018). Similarly, while CO2 capture is expensive, its integration with hydrogen via biomass gasification
43   can be achieved at an incremental capital cost of 3-35% ((Muratori et al. 2020b); Section 6.4). As with
44   all uses of bioenergy, linkages to broad sustainability concerns may limit the viable development, as
45   will the presence of high-quality geologic sinks in close proximity (Melara et al. 2020).
46   DACCS offers a modular approach to CDR (Creutzig et al. 2019), but it could be a significant consumer
47   of energy. DAC could also interact with other elements of the energy systems as the captured CO2 could

     Do Not Cite, Quote or Distribute                  6-93                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    be reused to produce low-carbon methanol and other fuels (Realmonte et al. 2019; Hoppe et al. 2018;
2    Zhang and Fujimori 2020). DACCS might also offer an alternative for use of excess electricity produced
3    by variable renewables (Wohland et al. 2018), though there are uncertainties about the economic
4    performance of this integrated approach.
6    6.6.3 The Institutional and Societal Characteristics of Net-zero Energy Systems
 7   The transition to net zero energy systems is not just technological; it requires shifts in institutions,
 8   organizations, and society more generally. As such, it involves institutional changes alongside changes
 9   in supply, technology, or markets (Andrews-Speed 2016, Pai et al. 2021). Institutional relationships
10   between governments and energy sector actors (e.g., consumers, electricity companies) affect the nature
11   of net zero systems, as these entities may collaborate on or dispute net zero goals and measures to
12   achieve them. For example, following the Fukushima disaster, Japan placed emphasis on government-
13   utility-public cooperation on use of nuclear power as a means of reducing carbon emissions (Sklarew
14   2018). Institutions are instrumental in shaping net zero energy systems in multiple ways, complemented
15   by and interacting with the behaviours of actors and policy regimes in these systems (Figure 6.24).

17        Figure 6.24 A four-level framework for institutional change. The diagram depicts three levels of
18         institutions (1-3) which collectively govern actor behaviours (4). Source: Andrews-Speed 2016

19   One level of institutional interactions reflects embedded institutions, norms, beliefs, and ideas that
20   would need to change to support net zero energy systems. This applies, for example, to the objectives
21   of modern economies and the potentially contradictory dynamics embedded in the concept of “green
22   growth” (Stegemann and Ossewaarde 2018; Stoknes and Rockström 2018). The institutional
23   environment – the political and legal systems that govern exchanges and protect property rights – would
24   also need to be different in net zero energy systems. In this setting, changing regulations or subsidies
25   that continue to favour carbon-intensive systems over the technologies of a net zero energy system
26   might prove difficult (Sovacool 2017). More generally, net zero energy systems will need new
27   regulatory frameworks to undertake new challenges, from managing a more interconnected grid to
28   adequately governing underground storage of CO2. Institutions may also govern specific transactions,
29   such as firms or networks that supply energy fuels or services. Current actors are typically resistant to
30   disruptions, even if such disruptions may broadly benefit society (Kungl 2015, Mori 2018, Schmid et
31   al. 2017).

     Do Not Cite, Quote or Distribute                  6-94                                  Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

 1   For example, one energy system characterized by differentiated institutional interactions is the United
 2   States, where delivery of liquid fuels is lightly regulated, while electricity delivery is closely regulated
 3   (Dworkin et al. 2013). Reforming this two-pronged system for decarbonization would require four types
 4   of institutional change: (1) changes to the control systems that coordinate generation and transmission
 5   through a pyramidal architecture for the operational control, dispatch, and delivery of electricity with a
 6   primary emphasis on reliability; (2) changes to the financing of central-station power plants through
 7   long-term bonds, as valued by Wall Street ratings analysts; (3) changes to the structure of investor-
 8   owned utilities that attract private investors who expected decades of technological stability to yield
 9   long-term, low-risk revenues; and (4) changes to regulations to restructure and limit excessive returns
10   and easy entry of new retail competitors, all recognizing local and national concerns through state and
11   federal regulatory agencies. The example shows how decision-making and the infrastructures involved
12   are layered, and can create “nested hierarchies” where institutions fulfill multiple roles for energy
13   governance or regulation simultaneously (Stern et al. 2016b). Internationally and across different parts
14   of the energy system, institutional challenges such as these could become even more stark and complex
15   (Van de Graaf 2013).
16   6.6.4 Regional Circumstances and Net-zero Energy Systems
17   Countries have flexibility to pursue options that make the most sense for their national circumstances
18   (Figure 6.25). They may emphasize supply transformation over demand reduction; deploy different
19   resources; engage at different levels in international energy trade; support different energy industries;
20   focus on different energy carriers (e.g., electricity, hydrogen); or focus more on distributed or integrated
21   systems, among others. Many factors may influence the long-term net zero energy systems that are
22   appropriate for any country’s national circumstances, including the following.
23   Future Technology. Technological transitions have often been driven by the relative merits of different
24   technology options. Recent trends in the use of PV cells, wind power, and in batteries, for example,
25   have been spurred by their increasing economic competitiveness (Section 6.3). Yet future technology
26   cannot be fully predicted, so it provides only a partial guide today for charting a path toward future
27   systems.
28   Indigenous Energy Resources. Countries may emphasize approaches that take advantage of indigenous
29   energy resources such as solar power, wind, hydroelectric resources, land for bioenergy crops, CO2
30   storage capability, or fossil resources to be used with CCS. Countries with less abundant resources may
31   put greater emphasis on demand reductions and regional integration. Countries with resource bases that
32   are easily tradeable, like low-carbon electricity or bioenergy, may choose to trade those resources rather
33   than use them domestically (Box 6.10, Section 6.4.3, 6.4.5).
34   Regional Climate. Climate influences heating and cooling demand, both of which influence countries’
35   energy demands and energy infrastructure to meet those demands (Section 6.5). In addition to daily
36   demand profiles, heating and cooling are seasonal, influencing which energy sources may serve these
37   loads and the seasonal storage they require. Cooling is almost entirely served by electricity today, and
38   heating has commonly been served by non-electric fuels. In low-carbon energy systems, heating may
39   be increasingly served by electricity (Section 6.6.4), meaning that the influence of regional climate may
40   be strongest on countries’ electricity systems.
41   Current Energy System Configuration. Future sectoral energy demands and the potential for demand-
42   side transformation are partially determined by existing infrastructure (e.g., building stocks, transport
43   infrastructure). Countries with less developed or growing energy systems will have more flexibility to
44   create the systems that best match their long-term goals, but there may be substantial challenges in
45   transitioning directly to the most advanced low-carbon technology options, and countries may have
46   different capacities to absorb technology from other countries.

     Do Not Cite, Quote or Distribute                   6-95                                   Total pages: 217
     Final Government Distribution                    Chapter 6                         IPCC WGIII AR6

1    Regional Integration. Regional integration will allow countries to bridge energy gaps using external
2    linkages, including regional electricity integration and trade in hydrogen, biomass, and other fuels.
3    Countries with greater integration can rely more heavily on imports and may therefore rely less on
4    indigenous resources (Box 6.10).


 8    Figure 6.25 Characteristics of regional energy systems and emissions when energy and industrial CO 2
 9     emissions reach net zero. Regional differences are shown for residual emissions and carbon removal
10         (panel a), energy resources (panel b), electrification (panel c), and energy intensity (panel d).
11   Distributions of scenarios are shown along each axis for each region. Colour scheme is shown in panel a.
12                    Points represent different models and scenarios from the AR6 database.

13   Societal Preferences. Citizens in every country have preferences for certain technological options or
14   mitigation approaches over others that will influence energy system choices. The public generally
15   prefers a future energy system based largely on renewables. Preferences for non-renewable energy
16   differ across regions and groups. For example, studies have found that people in the U.K., Germany,
17   the Netherlands, and Switzerland prefer renewable energy and personal energy efficiency and savings

     Do Not Cite, Quote or Distribute                 6-96                                  Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    to nuclear, fossil fuels and CCS (Demski et al. 2017; Jones et al. 2012; Scheer et al. 2013; Volken et al.
2    2018; Bessette and Arvai 2018; Steg 2018). Studies have found that people with higher education levels,
3    higher incomes, females, and liberals prefer renewables to fossil fuels and nuclear (Van Rijnsoever et
4    al. 2015; Bertsch et al. 2016; Blumer et al. 2018; Jobin et al. 2019). The willingness to pay for renewable
5    electricity differs by source (Ma et al. 2015; Sundt and Rehdanz 2015).
 6   Technological Leadership, Economic Opportunities, and Growth. Countries may emphasize
 7   technologies in which they intend to have technological leadership and a competitive advantage. These
 8   could emerge over time or be based on current areas of opportunity or leadership. Industrial policy will
 9   influence future energy system as technological choices can benefit or hamper incumbents or new
10   market actors.
11   Energy Security. Countries emphasizing import security will tend to rely more heavily on indigenous
12   resources (Section 6.3). Some indigenous resources may raise security of supply issues that will
13   influence energy system configurations. Bioenergy and hydropower, for example, can be subject to
14   import climate risks (6.5), and significant integration of VRE technologies will influence electricity
15   system infrastructure and management (6.6.2, Box 6.8).
16   Other Factors. Countries will consider a wide range of other factors in building toward low-carbon
17   energy systems. Population density, for example, will influence building and transportation energy
18   demands; economic transitions will influence industrial energy demands. Societal priorities beyond
19   climate, notably SDGs may influence technology choices and types of energy systems (Sections 6.3
20   and 6.7.7).
21   [START BOX 6.10 HERE]
22                            Box 6.10 Regional Integration of Energy Systems

23   Energy systems are linked across countries in many ways: countries transport crude oil across the ocean
24   in supertankers, pipelines carry oil and natural gas across country boundaries, electric power lines cross
25   country boundaries, and countries trade industrial commodities that carry embodied energy or that are
26   essential inputs to mitigation technologies. Future systems will generate electricity using different
27   mixes of technologies, produce and transport different carriers (e.g., hydrogen or biofuels), and use far
28   less fossil fuel, among other major changes. Important examples include electricity, hydrogen, and
29   biomass.

30   Electricity System Integration. Net-zero energy systems will rely more heavily on electricity
31   generated from low-emissions technologies. Given the significant variations in the location of low-
32   carbon electricity resources and the temporal variability of some renewable electricity sources, notably
33   solar and wind power, regional electricity grids could reduce overall costs of net zero energy systems
34   (Section 6.4.5). Furthermore, electricity transmission interconnections could significantly reduce local
35   energy balancing costs and investment in peaking plants needed to meet security of supply
36   requirements, and it could increase system resilience, especially in the case of extreme events such as
37   heat waves or cold spells (Fasihi and Bogdanov 2016). Important challenges to regional electricity
38   integration include geopolitical concerns from cross-border trade and societal and technological
39   challenges associated with building new transmission lines.

40   Hydrogen Trade. Hydrogen may play an important role in future net zero energy systems, particularly
41   in applications where electricity is not economically advantageous (see Box 6.9). Hydrogen can be used
42   to decarbonize regions in which it is produced, and it can also be transported long distances to facilitate
43   decarbonization of sectors distant from sources of low-cost supply. Methods of long-distance, high-
44   volume hydrogen transport could include liquid storage, chemical carriers, and gaseous delivery via
45   pipelines (Section 6.4.5). In net zero systems with substantial wind and solar power generation,

     Do Not Cite, Quote or Distribute                   6-97                                  Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1    hydrogen can be generated through electrolysis and then shipped to other locations. Important
2    challenges to hydrogen trade include cost-effective low carbon production, cost of delivery
3    infrastructure, storage, and end-use technology costs and safety.
 4   Trade in Biomass. Biomass may also play an important role in net zero energy systems (Section 6.6.4,
 5   Chapter 3). Large-scale bioenergy production and consumption is likely to trigger global biomass trade.
 6   Global bioenergy trade volumes presently exceed 1 EJ yr-1, of which 60% is directly traded for energy
 7   purposes (Proskurina et al. 2019b). Established trade mechanisms include wood pellet transport,
 8   ethanol, and biodiesel (Proskurina et al. 2019a). In a net zero global energy system, bioenergy trade
 9   could be greater than current trade of coal or natural gas, but less than that of petroleum (Sharmina et
10   al. 2017) (Mandley et al, 2020). Some studies indicate that Latin America and Africa could become key
11   exporting regions, with the EU, the USA, and East Asia emerging as key importers (Rentizelas et al.
12   2019; Alsaleh and Abdul-Rahim 2018). Studies have found that net bioenergy exports could be as high
13   as 10% of GDP for some Latin American countries, while other regions like the EU may be faced with
14   burgeoning import reliance (Mahlknecht et al. 2020; Daioglou et al. 2020b). In addition to challenges
15   associated with bioenergy production (Section 6.4, Chapter 7), important challenges to biomass trade
16   include differences in sustainability criteria and land/biomass definitions in different jurisdictions, and
17   difficulties in establishing consistent monitoring and auditing systems (Lamers et al, 2016).
18   [END BOX 6.10 HERE]

21   6.7   Low-Carbon Energy System Transitions in the Near- and Medium-
22         Term
23   6.7.1 Low-Carbon Energy System Transition Pathways
24   Energy System Emissions
25   Without additional efforts to reduce emissions, it is very unlikely that energy system CO 2 emissions
26   will decrease sufficiently to limit warming to well below 2°C (high confidence). Scenarios assuming
27   improvements in technology but no additional climate policies beyond those in place today provide a
28   benchmark for comparison against energy-related CO2 emissions in mitigation scenarios (Figure 6.26).
29   Emissions in these reference scenarios increase through 2050 but span a broad range (Chapter 3 Figure
30   3.16; Riahi et al. 2017; Wei et al. 2018). The highest emissions levels are about four times current
31   emissions; the lowest are modestly below today’s emissions. Emissions in these scenarios increase in
32   most regions, but they diverge significantly across regions (Bauer et al. 2017). Asia and the Middle East
33   and Africa account for the majority of increased emissions across these scenarios (Figure 6.27). While
34   it is unlikely that there will be no new climate policies in the future, these scenarios nonetheless support
35   the conclusion that the energy sector will not be decarbonized without explicit policy actions to reduce
36   emissions.
37   Warming cannot be limited to well below 2°C without rapid and deep reductions in energy system GHG
38   emissions (high confidence). Energy sector CO2 emissions fall by 87-97% (interquartile range) by 2050
39   in scenarios limiting warming to 1.5°C with no or limited overshoot and 60-79% in scenarios limiting
40   likely warming to 2°C (Figure 6.26). Energy sector GHG emissions fall by 85-95% (interquartile range)
41   in scenarios limiting warming to 1.5°C with no or limited overshoot and 62-78% in scenarios limiting
42   likely warming to 2°C (Figure 6.26). In 2030, in scenarios limiting warming to 1.5°C with no or limited
43   overshoot, net CO2 and GHG emissions fall by 35-51% and 38-52% respectively. Key characteristics
44   of emissions pathways – the year of peak emissions, the year when net emissions reach zero, and the
45   pace of emissions reductions – vary widely across countries and regions. These differences arise from
     Do Not Cite, Quote or Distribute                   6-98                                   Total pages: 217
     Final Government Distribution                                                                                                                        Chapter 6                                                                             IPCC WGIII AR6

1    differences in economic development, demographics, resource endowments, land use, and potential
2    carbon sinks (Schreyer et al. 2020)(Schaeffer, et al.2020; Schreyer, et al., 2020; van Soest, Heleen, et
3    al., 2021;Figure 6.27, Figure 6.28, Box 6.11). If countries do not move quickly to reduce emissions – if
4    reductions are delayed – a more rapid energy transition will subsequently be required to limit warming
5    to 2°C or below (Rogelj et al. 2015a, 2018a; IPCC 2018).

                                                                              GHG                                                                          COCH4
                                                                                                                                                              2                                                                                  CH4
                                      120                       8000                                                                     120                                                                                       100

                                                                7000                                                                                                                                                                90
                                      100                                                                                                100
     Emissions relative to 2020 (%)

                                                                                                        Emissions relative to 2020 (%)

                                                                                                                                                                                                  Emissions relative to 2020 (%)
                                      Emissions (Mt-CO2/year)

                                       80                                                                                                 80                                                                                        70
                                       60                                                                                                 60
                                                                4000                                                                                                                                                                50

                                       40                                                                                                 40

                                       20                                                                                                 20                                                                                        30
                                            0                                                                                              0
                                                                1000                                                                                                                                                                10

                                      -20                          0                                                                     -20                                                                                         0
                                                                       2030     20402020   2050                                                 2030
                                                                                                                                                2030             2040              2050    2040                                          2030       2040 2050   2050

                                                                                            1.5°C without overshoot                            1.5°C with overshoot     Likely below 2°C   Likely below 2°C with delay
 8   Figure 6.26 Projected energy sector GHG emissions for the 1.5°C scenarios (without and with overshoot),
 9    and likely below 2°C scenarios (without and with delayed policy action) during 2020-2050 (Source: AR6
10        Scenario Database). Boxes indicate 25th and 75th percentiles while whiskers indicate 5th and 95th
11     percentiles. GHG emissions are inclusive of energy sector CO 2, CH4, N2O emissions and 80% of global
12   HFC emissions. Number of model-scenario combinations in AR6 database: 1.5C without overshoot: 170,
13                   1.5C with overshoot: 177, 2C without delay: 297, 2C without delay: 124.


     Do Not Cite, Quote or Distribute                                                                                                                     6-99                                                                                     Total pages: 217
    Final Government Distribution                    Chapter 6                        IPCC WGIII AR6


3      Figure 6.27 Net regional (R5) CO2 emissions from energy across scenarios. for the 1.5°C scenarios
4    (without and with overshoot), and likely below 2°C scenarios (without and with delayed policy action)
5      during 2020-2050 (Source: AR6 Scenario Database). Boxes indicate 25th and 75th percentiles while
6     whiskers indicate 5th and 95th percentiles. Most mitigation scenarios are based on a cost-minimizing
7           framework that does not consider historical responsibility or other equity approaches.

    Do Not Cite, Quote or Distribute                6-100                                  Total pages: 217
     Final Government Distribution                       Chapter 6                         IPCC WGIII AR6

1     Figure 6.28 The timing of net zero emissions for full economy GHGs, energy sector CO 2, and electricity
2       sector CO2. Boxes indicate 25th and 75th percentiles, centre black line is the median, while whiskers
3      indicate 1.5x the inter-quartile range. The vertical dashed lines represent the median point at which
4    emissions in the scenarios have dropped by 95% (pink) and 97.5% (purple), respectively. Dots represent
5    individual scenarios. The fraction indicates the number of scenarios reaching net zero by 2100 out of the
6                                 total sample. (Source: AR6 Scenario Database)

 7   The timing of net zero energy system emissions varies substantially across scenarios. In scenarios
 8   limiting warming to 1.5°C with no or limited overshoot (likely below 2°C), the energy system reaches
 9   net zero CO2 emissions (interquartile range) from 2060 onwards (2080-). (Figure 6.28). However, net
10   emissions reach near-zero more quickly. For example, in scenarios limiting warming to 1.5C with no
11   or limited overshoot (likely below 2C) net energy system CO2 emissions drop by 95% between 2056
12   and 2075 (2073 and 2093). Net full economy GHG emissions reach zero more slowly than net CO2
13   emissions. In some scenarios, net energy system CO2 and total GHG emissions do not reach zero this
14   century, offset by CDR in other sectors.
15   The timing of emissions reductions will vary across the different parts of the energy sector (Figure
16   6.28). To decarbonize most cost-effectively, global net CO2 emissions from electricity generation will
17   likely reach zero before the rest of the energy sector (medium confidence). In scenarios limiting
18   warming to 1.5C with no or limited overshoot (likely below 2C), net electricity sector CO2 emissions
19   (interquartile range) reach zero globally between 2044 and 2055 (2052 and 2078) (Figure 6.28). It is
20   likely to be less-costly to reduce net CO2 emissions close to or below zero in the electricity sector than
21   in other sectors, because there are relatively more low-emissions options in electricity. Sectors such as
22   long-distance transport, air transport, and process heat are anticipated to face greater challenges to
23   decarbonization than the electricity sector (Rogelj et al. 2018b, 2015b; Clark and Herzog 2014; IPCC
24   2018; Luderer et al. 2018).
25   In addition, there are potential options to remove CO2 from the atmosphere in the electricity sector,
26   notably BECCS, which would allow electricity sector emissions to drop below zero. Without CDR
27   options, electricity sector emissions may not fall all the way to zero. If CDR is accomplished in other
28   sectors and not in electricity, some fossil fuel plants may still lead to positive net electricity sector CO2
29   emissions even in net zero economies (Williams et al. 2021a; Bistline and Blanford 2021b)

     Do Not Cite, Quote or Distribute                   6-101                                   Total pages: 217
                                      Final Government Distribution                                                                                Chapter 6                                                                   IPCC WGIII AR6

                                                                                                            Electricity                                                                                         Transport
                                                                                 120                                                                                                      140

                                       Relative to 2020 levels (%)

                                                                                                                                                           Relative to 2020 levels (%)

                                                                                  40                                                                                                       80

                                                                                  20                                                                                                       60


                                                                                 -60                                                                                                        0
                                                                                                  2030               2040              2050                                                            2030                2040             2050
                                                                                                              Industry                                                                               Residential and commercial
                          7000                                                    200                                                                                                     140

Emissions (Mt-CO2/year)

                                                   Relative to 2020 levels (%)

                                                                                                                                                            Relative to 2020 levels (%)

                          5000                                                    140                                                                                                     100

                          4000                                                                                                                                                             80

                          3000                                                                                                                                                             60

                          2000                                                                                                                                                             40
                          1000                                                                                                                                                             20

                             0                                                         0                                                                                                    0
                                                                                           2020    2030               2040           20302050                                                          2030   2040         2040             2050   2050

                                                                                                          1.5°C without overshoot   1.5°C with overshoot                                  Likely below 2°C    Likely below 2°C with delay
                                 2    Figure 6.29 Reductions in CO2 emissions relative to 2020 levels for the 1.5°C scenarios (without and with
                                 3      overshoot), and likely below 2°C scenarios (without and with delayed policy action) during 2030-2050
                                 4    (Source: AR6 Scenario Database). Boxes indicate 25th and 75th percentiles while whiskers indicate 5th and
                                 5                                                 95th percentiles.

                                  6   We lack sufficient understanding to pin down precise dates at which energy system CO2 emissions in
                                  7   individual countries, regions, or sectors will reach net zero. Net-zero timing is based on many factors
                                  8   that are not known today or are bound up in development of key technologies, such as energy storage,
                                  9   bioenergy, or hydrogen. Some countries have low-carbon resource bases that could support deep
                                 10   emissions reductions, while others do not. Timing is also affected by the availability of CDR options,
                                 11   whether these options are in the energy sector or elsewhere, and the discount rate used to assess
                                 12   strategies (Bednar et al. 2019; Emmerling et al. 2019). Moreover, while many scenarios are designed
                                 13   to minimize global mitigation costs, many other frameworks exist for allocating mitigation effort across
                                 14   countries (Chapter 4; van den Berg, N. J., et al., 2019).

                                      Do Not Cite, Quote or Distribute                                                                            6-102                                                                              Total pages: 217
     Final Government Distribution                       Chapter 6                          IPCC WGIII AR6

1   Low-carbon energy transition strategies
2    There are multiple technological routes to reduce energy system emissions (see Section 6.6). Here we
3    discuss three of these: (1) decarbonizing primary energy and electricity generation, (2) switching to
4    electricity, bioenergy, hydrogen, and other fuels produced from low-carbon sources, and (3) limiting
5    energy use through improvement of efficiency and conservation. CDR is discussed in Section 6.7.1.X;
6    Fossil fuel transitions are discussed in Section 6.7.4.
 7   Decarbonizing Primary Energy and Electricity Generation. Limiting warming to well below 2°C
 8   requires a rapid and dramatic increase in energy produced from low- or zero-carbon sources (high
 9   confidence). Low- and zero-carbon technologies produce 74-82% (interquartile range) of primary
10   energy in 2050 in scenarios limiting warming to 1.5°C with no or limited overshoot and 55-68% in
11   likely below 2°C scenarios (Figure 6.29). The share of low-carbon technologies in global primary
12   energy supply today is below 20% (Section 6.3, Chapter 3,Figure 6.29). The percentage of low- and
13   zero-carbon energy will depend in part on the evolution of energy demand – the more that energy
14   demand grows, the more energy from low- and zero-carbon sources will be needed and the higher the
15   percentage of total primary energy these sources will represent.
16   Low- and zero-carbon sources produce 97-99% of global electricity in 2050 in scenarios limiting
17   warming to 1.5°C with no or limited overshoot and 93-97% in likely below 2°C scenarios (Figure 6.29)
18   (medium confidence). Decarbonizing electricity generation, in tandem with increasing use of electricity
19   (see below), is an essential near-term strategy for limiting warming. The increase in low- and zero-
20   carbon electricity will occur while electricity demand grows substantially. Studies have projected that
21   global electricity demand will roughly double by 2050 and quadruple to quintuple by 2100 irrespective
22   of efforts to reduce emissions (Bauer et al. 2017; Luderer et al. 2017; IEA 2019a).
23   Renewable energy, especially generation from solar and wind, is likely to have an important role in
24   many low-carbon electricity systems. The contributions of wind and solar electricity will depend on
25   their levelized costs relative to other options, integration costs, system value, and the ability to integrate
26   variable resources into the grid (Section 6.6). Electric sector technology mixes will vary by region but
27   will typically include additional resources such as hydropower, nuclear power, fossil generation with
28   CCS, energy storage resources, and geothermal energy, among others. Contributions of different
29   options vary widely across scenarios based on different assumptions about these factors (Figure 6.30).
30   Nonetheless, it is likely that wind and solar will dominate low-carbon generation and capacity growth
31   over the next couple decades due to supporting policies in many countries and due to their significant
32   roles in early electric sector decarbonization, alongside reductions in coal generation (Bistline and
33   Blanford 2021b; Pan et al. 2021). Clean firm technologies play important roles in providing flexibility
34   and on-demand generation for longer durations, though deployment of these technologies is typically
35   associated with deeper decarbonization levels (e.g., beyond 70-80% reductions), which are likely to be
36   more important after 2030 in many regions, and with more limited CDR deployment (Baik et al. 2021;
37   Williams et al. 2021a; Bistline and Blanford 2021a).

     Do Not Cite, Quote or Distribute                   6-103                                    Total pages: 217
     Final Government Distribution                                                                                   Chapter 6                                                           IPCC WGIII AR6

           Low-carbon energy in primary energy                                                 Bioenergy in primary energy                                                   Low-carbon energy in electricity
                 100                                                               40                                                                                    100

                    90                                                                                                                                                       90
                    80                                                                                                                                                       80
                    70                                                                                                                                                       70
     Share (%)


                                                                       Share (%)

                                                                                                                                                             Share (%)
                                                                                   20                                                                                        50
                    20                                                             10
                    10                                                               5
                                 2020     2030     2040    2050                      0                                       CH4                                              0
                                                                                                    2020           2030          2040          2050                               2020     2030       2040          2050

                               Solar and wind in electricity                                                  CCS in electricity                                                    Nuclear in electricity
                  90                                                                           40                                                                            30

     Emissions (Mt-CO2/year)

                               6000                                                            35
                  60                                                                                                                                                         20
          Share (%)

                                                                                   Share (%)

                                                                                                                                                                 Share (%)
                  50           4000
                                                                                               20                                                                            15
                               3000                                                            15
                  30                                                                                                                                                         10

                               2000                                                            10
                  10                                                                            5
                     0                                                                          0                                                                             0
                                  0       2030      2040       2050                                        2020           2030          2040          2050                        2020      2030       2040          2050
                                                   2020                                                     2030                                             2040                                            2050

                                                               1.5°C without overshoot                     1.5°C with overshoot         Likely below 2°C     Likely below 2°C with delay

3         Figure 6.30 Shares of low carbon energy (all sources except unabated fossil fuels) and bioenergy
4     (including both traditional and commercial biomass) in total primary energy, and solar+wind, CCS and
5    nuclear in electricity for the 1.5°C scenarios (without and with overshoot), and likely below 2°C scenarios
6       (without and with delayed policy action) during 2020-2050 (Source: AR6 Scenario Database). Boxes
7                 indicate 25th and 75th percentiles while whiskers indicate 5th and 95th percentiles.

8    [START BOX 6.11 HERE]

9                                                Box 6.11 Illustrative Low-Carbon Energy System Transitions

10   There are multiple possible strategies to transform the energy system to reach net zero CO2 emissions
11   and to limit likely warming to 2°C or below. All pathways rely on the strategies for net zero CO2 energy
12   systems highlighted in Section 6.6.2, but they vary in the emphasis that they put on different aspects of
13   these strategies and the pace at which they approach net zero emissions. The pathway that any country
14   or region might follow will depend on a wide variety of factors (Section 6.6.4), including, for example,
15   resource endowments, trade and integration with other countries and regions, carbon sequestration
16   potential, public acceptability of various technologies, climate, the nature of domestic industries, the
17   degree of urbanization, and the relationship with other societal priorities such as energy access, energy
18   security, air pollution, and economic competitiveness. The illustrative mitigation pathways presented
19   in this box demonstrate four distinct strategies for energy system transformations and how each plays
20   out for a different region, aligned with global strategies that would contribute to limiting warming to
21   1.5°C. Each pathway represents a very different vision of a net zero energy system. Yet, all these
22   pathways share the common characteristic of a dramatic system-wide transformation over the coming

     Do Not Cite, Quote or Distribute                                                                              6-104                                                                           Total pages: 217
     Final Government Distribution                    Chapter 6                         IPCC WGIII AR6

1    decades.

5       Box 6.11, Figure 1 Illustrative Mitigation Pathway 2.0-Neg: Latin America & Caribbean (LAM) in a
6    likely below 2°C scenario (LAM net-zero economy 2040-2045, net zero energy system 2045-2050). Supply
7    side focus with growing dependency on carbon dioxide removal and AFOLU, thus achieves net-zero CO2
8                                                   relatively early.

12    Box 6.11, Figure 2 Illustrative Mitigation Pathway 1.5-Renewables: Africa (AF) in a 1.5°C scenario (AF
13    net-zero economy, 2055-2060, AF net zero energy system 2055-2060). Rapid expansion of non-biomass
14                           renewables, high electrification, and a fossil fuel phaseout.

     Do Not Cite, Quote or Distribute                 6-105                                 Total pages: 217
     Final Government Distribution                       Chapter 6                      IPCC WGIII AR6

3     Box 6.11, Figure 3 Illustrative Mitigation Pathway 1.5-Low Demand: Developed Countries (DEV) in a
4    1.5°C scenario (DEV net-zero economy, 2055-2060, net zero energy system 2075-2080). Major reduction
5                    of energy demand, high electrification, and gradual fossil fuel phaseout.

 9    Box 6.11, Figure 4 Illustrative Mitigation Pathway 1.5-Shifting Pathways: Asia and Developing Pacific
10      (APC) in a 1.5°C scenario (APC net-zero economy, 2075-2080, net-zero energy system 2090-2095).
11    Renewables, high electrification, fossil fuel phaseout and low AFOLU emissions. Reaches net-zero CO2
12                                                   relatively late.

14   Box 6.11, Table 1. Summary of selected Illustrative Mitigation Pathways energy system characteristics in
15                                        2050 for the chosen regions.


                       CO2                                   Low          CO2
                      Reduc                 Variable        carbon      Removal
                       tion                renewable      electricity   BECCS,
                      2020-     Energy     electricity     capacity     AFOLU,                       Year net-zero CO2
                      2050     intensity   generation     additions      Total    GDP per capita         emissions

     Do Not Cite, Quote or Distribute                    6-106                               Total pages: 217
     Final Government Distribution                        Chapter 6                            IPCC WGIII AR6

                                MJ / PPP                                                    PPP                           Electri
                         %                   EJ/yr (%)      GW/yr       Gt CO2 yr-1                       Full   Energ
                                USD2010                                                 USD2010/person                     city
                                                                                                         econo     y
              Region            202    20          205     202    205
                                                                                                          my     sector
                        2050     0     50   2020     0      0      0        2050        2020    2050
                                      2.     0.5    7.7    15.    21.                   1295             2040-   2045-    2025-
      Neg     LAM       124      3                                      1.1, 0.2, 1.9           24860
                                      1      (9)   (53)     4      5                     2               2045    2050     2030
      Ren     AF                       1.    0.1    18                                                   2055-   2055-    2025-
                         85     7.6    9     (5)   (84)    5      217    0.1, 0, 0.1    2965    8521     2060    2060     2030
                                      0.     4.6    37                                  4294             2055-   2075-    2045-
      LD      DEV        92     3.1                        52     188    0, 0.6, 0.6            61291
                                      9     (13)   (72)                                  5               2060    2080     2050
                                      1.    3 (7    91                                  1051             2075-   2085-    2085-
      SP      APC        76     3.8                        123    603   0.1, 0.4, 0.4           37180
                                      1      )     (79)                                  4               2080    2090     2090

3    [END BOX 6.11 HERE]
 5   Switching to Low-Carbon Energy Carriers. Switching to energy carriers produced from low-carbon
 6   sources will be an important strategy for energy sector decarbonization. Accelerated electrification of
 7   end uses such as light duty transport, space heating, and cooking is a critical near-term mitigation
 8   strategy (Waisman et al. 2019; Sugiyama 2012; Zou et al. 2015; Rockström et al. 2017; IEA 2019f;
 9   Tang et al. 2021a). Electricity supplies 48-58% (interquartile range) of the global final energy demand
10   by 2050 in scenarios limiting warming to 1.5°C with no or limited overshoot and 36-47% in likely
11   below 2°C scenarios (Figure 6.29). Globally, the current level of electrification is about 20%.
12   Indirect electrification encompasses the use of electricity to produce hydrogen and synthetic fuels
13   (efuels or power fuels). The extent of indirect electrification of final energy will depend on resource
14   endowments and other regionally specific circumstances. Although indirect electrification is less
15   efficient compared to direct electrification, it allows low-carbon fuels to be imported from regions with
16   abundant low carbon electricity generation resources (Fasihi and Breyer 2020; Fasihi and Bogdanov
17   2016; Lehtveer et al. 2019, Box 6.10 on regional integration).
18   While electrifying end uses is a key decarbonization strategy, some end uses such as long-distance
19   transport (freight, aviation, and shipping) and energy-intensive industries will be harder to electrify. For
20   these sectors, alternative fuels or energy carriers such as biofuels, hydrogen, ammonia or synthetic
21   methane, may be needed (see Section 6.6 and Box 6.9). Most scenarios find that hydrogen consumption
22   will grow gradually, becoming more valuable when the energy system has become predominantly low-
23   carbon (Figure 6.31).

     Do Not Cite, Quote or Distribute                     6-107                                   Total pages: 217
     Final Government Distribution                                                                                    Chapter 6                                                                   IPCC WGIII AR6

                         Electricity in final energy                             Electricity in final energy                             Electricity in final energy                               Electricity in final energy
                 70                                                                     (transport)                                              (industry)                                       (residential & commercial)
                                                                            50                                                      80                                                     100
                                                                            45                                                      70                                                      90

                                                                            40                                                                                                              80
                                                                            35                                                                                                              70
     Share (%)

                 40                                                                                                                 50

                                                              Share (%)

                                                                                                                        Share (%)

                                                                                                                                                                              Share (%)
                                                                            30                                                                                                              60

                                                                            25                                                      40                                                      50
                                                                            20                                                      30                                                      40
                 20                                                         15                                                                                                              30
                                                                            10                                                                                                              20
                                                                             5                                                                                                              10

                     0                                                       0                                                       0                                                       0
                         2020      2030      2040      2050                      2020      2030      2040      2050                      2020      2030      2040      2050                         2020      2030    2040       2050

                         Hydrogen in final energy                                Hydrogen in final energy                                Hydrogen in final energy                                  Hydrogen in final energy
                 9                                                                    (transport)                                              (industry)                                         (residential & commercial)
                                                                            30                                                      16                                                      4.5
                                                                                                                                    14                                                       4
                 7                                                          25
     Share (%)

                                                                Share (%)

                                                                                                                        Share (%)

                                                                                                                                                                               Share (%)
                 5                                                                                                                                                                          2.5

                 4                                                          15                                                       8                                                       2

                 3                                                                                                                   6
                 2                                                                                                                   4
                 1                                                                                                                   2

                 0                                                           0                                                       0                                                     -0.5
                         2020     2030      2040       2050                       2020      2030      2040     2050                      2020      2030      2040      2050                          2020     2030     2040      2050
2       Figure 6.31 Shares of electricity and hydrogen in final energy for scenarios limiting warming to 1.5°C
3      scenarios (without and with overshoot) and scenarios limiting likely warming to below 2°C (without and
4      with delayed policy action) during 2020-2050 (Source: AR6 Scenario Database). Boxes indicate 25th and
5                          75th percentiles while whiskers indicate 5th and 95th percentiles.

 6   Reducing Energy Demand. Energy service demand is expected to continue to increase with growth of
 7   the economy, but there is great uncertainty about how much it will increase (Riahi et al. 2017; Bauer et
 8   al. 2017; Yu et al. 2018). Given the need to produce low-carbon energy, the scale of energy demand is
 9   a critical determinant of the mitigation challenge (Riahi et al. 2012). Higher energy demand calls for
10   more low-carbon energy and increases the challenge; lower energy demand reduces the need for low-
11   carbon sources and therefore can ease a low-carbon transtion. Recent studies have shown that tempering
12   the growth of energy demand, while ensuring services and needs are still satisfied, can materially affect
13   the need for technological CDR (see below) (Grubler et al. 2018; van Vuuren et al. 2018). Two of the
14   Illustrative Mitigation Pathways (IMP-SP, IMP-LD) feature substantially lower final energy demand
15   across buildings, transport, and industry than most other pathways in the literature. In some cases,
16   energy demand levels are lower in 2050 (and later) than in 2019. These lower demands result in less
17   reliance on bioenergy and a more limited role for CDR. [Ch. 3, Figure 3.18]
18 Technology options to offset residual emissions
19   CDR technologies can offset emissions from sectors that are difficult to decarbonize (Section 6.6),
20   altering the timeline and character of energy sector transitions. A number of studies suggest that CDR
21   is no longer a choice but rather a necessity to limit warming to 1.5°C (Luderer et al. 2018; Rogelj et al.
22   2015a; van Vuuren et al. 2018; Detz et al. 2018; Strefler et al. 2018). The reliance on CDR varies across
23   scenarios and is tightly linked future energy demand and the rate of emission reductions in the next two
24   decades: deeper near-term emissions reductions will reduce the need to rely on CDR to constrain
25   cumulative CO2 emissions. Some studies have argued that only with a transition to lower energy
26   demands will it be possible to largely eliminate the need for engineered CDR options (Grubler et al.
27   2018; van Vuuren et al. 2018). Overall, the amount of CDR will depend on CO2 capture costs, lifestyle
28   changes, reduction in non-CO2 GHGs, and utilization of zero-emission end-use fuels (van Vuuren et al.
29   2018)(Muratori et al, 2017; van Vuuren et al, 2018).
30   There is substantial uncertainty about the amount of CDR that might ultimately be deployed. In most
31   scenarios that limit warming to 1.5°C, CDR deployment is fairly limited through 2030 at less than 1
32   Gt-CO2 yr-1. The key projected increase in CDR deployment (BECCS and DAC only) occurs between
     Do Not Cite, Quote or Distribute                                                                                 6-108                                                                                 Total pages: 217
     Final Government Distribution                     Chapter 6                        IPCC WGIII AR6

 1   2030 and 2050 with annual CDR in 2050 projected at 2.5-7.5 Gt-CO2 yr-1 in 2050 (interquartile range)
 2   in scenarios limiting warming to 1.5C with limited or no overshoot and 0.7-1.4 Gt-CO2 yr-1 in 2050 in
 3   scenarios limiting warming to 2C with limited or no overshoot. This characteristic of scenarios largely
 4   reflects substantial capacity addition of BECCS power plants. BECCS is also deployed in multiple ways
 5   across sectors. For instance, the contribution (interquartile range) of BECCS to electricity is 1-5% in
 6   2050 in scenarios limiting warming to 1.5C with limited or no overshoot and 0-5% in scenarios limiting
 7   likely warming to below 2C . The contribution (interquartile range) of BECCS to liquid fuels is 9-21%
 8   in 2050 in scenarios limiting warming to 1.5C with limited or no overshoot and 2-11% in scenarios
 9   limiting likely warming to below 2C. Large-scale deployment of CDR allows flexibility in timing of
10   emissions reduction in hard-to-decarbonize sectors.
11   CDR will influence the potential fossil-related stranded assets (Box 6.13). Availability of low-cost CDR
12   can help reduce premature retirement for some fossil fuel infrastructure. CDR can allow countries to
13   reach net zero emissions without phasing out all fossil fuels. Specific infrastructure could also be
14   extended if it is used to burn biomass or other non-emitting sources. For example, existing coal-fired
15   power plants, particularly those with CCS, could be co-fired with biomass (Pradhan et al, 2021; Woolf
16   et al, 2016; Lu et al, 2019). In many scenarios, energy sector CDR is deployed to such an extent that
17   energy sector CO2 emissions become negative in the second half of the century (Chapter 3).
18   [START BOX 6.12 HERE]

19                        Box 6.12 Taking Stock of the Energy System Transition

20   The Global Stocktake is a regularly occurring process under the UNFCCC in which efforts will be made
21   to understand progress on, among other things, global mitigation. Collective progress of countries
22   towards the Paris Agreement goal will be assessed and its outcome will inform Parties in updating and
23   enhancing their NDCs. This box explores potential indicators to understand energy system mitigation
24   progress.
25   CO2 emissions from fuel combustion are the bottom line on energy system progress. Beyond CO 2
26   emissions, primary energy demand by energy sources, final energy consumption by sectors, and total
27   electricity demand provide a first order assessment of energy system transitions. The year at which CO2
28   emissions peak is also important. The Kaya Identity can be used to decompose energy system CO2
29   emissions into carbon intensity of the energy system (CO2 emissions from fossil-fuel combustion and
30   industry divided by energy use), energy intensity (energy use divided by economic output), and
31   economic output. The impacts of energy and climate policy are reflected in the changes of carbon
32   intensity and energy intensity. Carbon intensity captures decarbonization of energy supply systems, for
33   example through fuel switching from fossil fuels to non-fossil fuels, upscaling of low carbon energy
34   sources, and deploying carbon dioxide removal technologies. The carbon intensity of electricity is
35   specifically important, given the role of the electricity sector in near-term mitigation. Economy wide
36   energy intensity represents efforts of demand-side energy, such as energy conservation, increase of
37   energy performance of technologies, structural change of economy, and development of efficient urban
38   infrastructure.
39   Beyond these aggregate indicators, a second order assessment would capture more details, such as the
40   electrification rate, share of renewables, nuclear, CCS or other low carbon technologies in electricity
41   generation, land area used for energy production, and numbers of EV or PHEV. Consumption of coal,
42   oil and gas captures the underlying factors of CO2 emissions. The emphasis of these indicators could
43   differ across countries in the context of national specific circumstances. Technology- or project-based
44   statistics are also useful to check the progress of the low-carbon transition, for example, the number of
45   CCS facilities.

     Do Not Cite, Quote or Distribute                 6-109                                 Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    A critical challenge in the assessment of energy sector progress is how to measure societal, institutional,
2    and political progress. These factors are difficult to quantify, yet they are fundamental determinants of
3    the ability to reduce emissions. Public opinion, special interest politics, implications of mitigation for
4    employment, energy subsidies, and energy policies are all critical indicators of progress. In addition,
5    while much of the literature focuses on national level action, mitigation is increasingly being led by
6    cities, states, provinces, businesses, and other subnational or non-national actors. Understanding the
7    progress of these actors will be critical to assess energy system mitigation progress. New research is
8    needed to better assess these “societal” indicators and the role of non-national actors.
9    [END BOX 6.12 HERE]
10   6.7.2. Investments in Technology and Infrastructure
11   Total global energy investment was roughly USD 1940 billion yr-1 in 2019 (IEA 2021f). This total can
12   be broken down into the following main categories: fossil-related energy supply, including oil, gas, and
13   coal extraction and fossil electricity generation (USD 990 billion yr-1); renewable electricity, primarily
14   solar and wind (USD 340 billion yr-1); nuclear energy (USD 40 billion yr-1); electricity networks (USD
15   270 billion yr-1); and end-use energy efficiency (USD 270 billion yr-1).
16   Energy investment needs are projected to rise going forward, according to investment-focused scenario
17   studies found in the literature (Bertram et al. 2021; McCollum et al. 2018a; Zhou et al. 2019). While
18   these increases are projected to occur in emissions-intensive pathways as well as low-carbon pathways,
19   they are projected to be largest in low-carbon pathways. Average annual global energy investments over
20   the 2016-2050 period range (across six models) from USD 2100 to 4100 billion yr-1 in pathways likely
21   limiting warming to 2°C and from USD 2400 to 4700 billion yr-1 in pathways limiting warming to 1.5°C
22   with no or limited overshoot (McCollum et al., 2018). Whatever the scenario, a significant and growing
23   share of investments between now and 2050 will be channelled toward infrastructure build-out in
24   emerging economies, particularly in Asia (Zhou et al. 2019).
25   More widespread electrification of buildings, transport, and industry means particularly substantial
26   investment in the electricity system. According to C1-C3 pathways in the AR6 scenario database, such
27   investments could be at the following average annual levels (inter-quartile range, USD2015) over the
28   2023-2052 timeframe: USD 1670 to 3070 billion yr-1 (C1), 1600 to 2780 billion yr-1 (C2), and 1330-
29   2680 billion yr-1 (C3). (See also
30   Beyond these sector-wide numbers, a key feature of stringent mitigation pathways is a pronounced
31   reallocation of investment flows across sub-sectors, namely from unabated fossil fuels (extraction,
32   conversion, and electricity generation) and toward renewables, nuclear power, CCS, electricity
33   networks and storage, and end-use energy efficiency (McCollum et al. 2018a; Bertram et al. 2021; IEA
34   2021f) (Figure 6.32). Investments in solar, wind, and electricity transmission, distribution, and storage
35   increase the most in mitigation scenarios. Up to 2050, the bulk of these investments are made in OECD
36   and Asian countries (Figure 6.33). While fossil fuel extraction investments exhibit a marked
37   downscaling across all regions, compared to reference scenarios, the declines are especially strong in
38   the Middle East, REF, and OECD.

     Do Not Cite, Quote or Distribute                  6-110                                  Total pages: 217
     Final Government Distribution                     Chapter 6                        IPCC WGIII AR6

2    Figure 6.32 Global average annual investments from 2023 to 2052 (undiscounted, in USD billion yr-1) for
3    electricity supply sub-sectors and for extraction of fossil fuels in C1-C3 pathways (Source: AR6 Scenario
4      Database and Chapter 3). Historical investments are also shown for comparison (Source: IEA, 2021;
5     approximations are made for hydro and geothermal based on available data; solar and wind values are
6     for 2020). ‘T&D’: transmission and distribution of electricity. Bars show median values across models-
7     scenarios, and whiskers the inter-quartile ranges. See Chapters 3 and 15 for additional information on
8                                             investments and finance.


     Do Not Cite, Quote or Distribute                 6-111                                  Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1      Figure 6.33 Regional average annual investments from 2023 to 2052 (undiscounted, in USD billion yr-1)
2        for four of the largest sub-sectors of the energy system in C1-C3 pathways (Source: AR6 Scenario
3      Database and Chapter 3). Historical investments are also shown for comparison (Source: IEA, 2016).
4     ‘T&D’: transmission and distribution of electricity. ‘Extr.’: extraction of fossil fuels. Bars show median
5        values across models-scenarios, and whiskers the inter-quartile ranges. See Chapters 3 and 15 for
6                                 additional information on investments and finance.

 7   Investments into end-use energy efficiency are projected to also be substantial in mitigation pathways,
 8   potentially upwards of several hundred USD billion yr-1 on average to 2050, compared to USD 270
 9   billion yr-1 in 2019 (McCollum et al. 2018a; IEA 2021f). However, the literature is inconsistent in how
10   demand-side investments are calculated, as boundary conditions are less clear than for energy supply
11   investments. Taking a broader definition can result in estimates that are an order-of-magnitude higher,
12   meaning as large or larger than supply-side investments (IEA 2021f; Grubler et al. 2012).
13   Increasing low-carbon investment primarily requires shifting existing capital investment through
14   regulation and incentives as well as removing existing investment barriers (McCollum et al.
15   2018a)(Ameli, N. et al., 2021: Hafner et al. 2020: McCollum et al. 2018). While there is a considerable
16   amount of capital in the world, it is not always available to those wishing to invest in certain projects.
17   Total annual global investment in fixed capital was USD 22.4 trillion in 2021, over an order-of-
18   magnitude larger than energy sector investment (World Bank, 2021).
19   Future investment patterns will vary by region, as they do now, due to differences in risk profiles,
20   resource endowments and economic and governance structures (Zhou et al. 2019)(6.6; Ameli, N. et al.,
21   2021; Fizaine et al. 2016; Zhou et al., 2019). In rapidly growing countries, investments to support a
22   low-carbon energy system transition will be integrated with those needed to meet rapidly increasing
23   energy demands, irrespective of whether efforts are made to reduce emissions. In less-rapidly-growing
24   countries (Sun et al. 2019), investments will focus on transitioning current energy systems to low-
25   carbon configurations. Most current energy investments are concentrated in high- and upper-middle
26   income countries (IEA 2021f), but this will change as investment needs continue to grow in today’s
27   lower-middle and low-income countries (Bertram et al. 2021; McCollum et al. 2018a; Zhou et al. 2019;
28   IEA 2021f).
29   6.7.3 Energy System Lock-In and Path Dependence
30   Path dependence refers to resistance to change due to favourable socio-economic conditions with
31   existing systems; decisions made in the past unduly shape future trajectories. Carbon lock-in is a
32   specific type of path dependence (Seto et al. 2016). Given that energy system mitigation will require a
33   major course change from recent history, lock-in is an important issue for emission reductions in the
34   energy sector. While lock-in is typically expressed in terms of physical infrastructure that would need
35   to be retired early to reach mitigation goals, it involves a much broader set of issues that go beyond
36   physical systems and into societal and institutional systems (Table 6.11).
38     Table 6.11 Lock-in types and typical mechanisms (Kotilainen et al. 2020). Reproduced under Creative
39                                     Commons 4.0 International License.

     Do Not Cite, Quote or Distribute                  6-112                                   Total pages: 217
     Final Government Distribution                    Chapter 6                        IPCC WGIII AR6

2   Societal and Institutional Inertia
3    A combination of factors - user, business, cultural, regulatory, and transnational - will hinder low-
4    carbon energy transitions. Strong path dependencies, even in early formative stages, can have lasting
5    impacts on energy systems, producing inertia that cuts across technological, economic, institutional and
6    political dimensions (high confidence) (Vadén et al. 2019; Rickards et al. 2014) Chapter 5).
7    Energy systems exemplify the ways in which massive volumes of labor, capital, and effort become sunk
8    into particular institutional configurations (Bridge et al. 2013, 2018). Several embedded factors affect
9    large-scale transformation of these systems and make technological diffusion a complex process:
10   ●   User environments affect purchase activities and can involve the integration of new technologies
11       into user practices and the development of new preferences, routines, habits and even values
12       (Kanger et al. 2019).
13   ●   Business environments can shape the development of industries, business models, supply and
14       distribution chains, instrument constituencies and repair facilities (Béland and Howlett 2016).
15   ●   Culture can encompass the articulation of positive discourses, narratives, and visions that enhance
16       cultural legitimacy and societal acceptance of new technologies. Regulatory embedding can capture
17       the variety of policies that shape production, markets and use of new technologies.
18   ●   Transnational community can reflect a shared understanding in a community of global experts
19       related to new technologies that transcends the borders of a single place, often a country.
20   While low-carbon innovation involves systemic change (Geels et al. 2018), these are typically less
21   popular than energy supply innovations among policymakers and the wider public. Managing low
22   carbon transitions is therefore not only a techno-managerial challenge (based on targets, policies, and
23   expert knowledge), but also a broader political project that involves the building of support coalitions
24   that include businesses and civil society (moderate evidence, high agreement).
25   Low-carbon transitions involve cultural changes extending beyond purely technical developments to
26   include changes in consumer practices, business models, and organizational arrangements. The
27   development and adoption of low-carbon innovations will therefore require sustained and effective
28   policies to create appropriate incentives and support. The implementation of such policies entails
     Do Not Cite, Quote or Distribute                 6-113                                 Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    political struggles because actors have different understandings and interests, giving rise to
2    disagreements and conflicts.
 3   Such innovation also involves pervasive uncertainty around technical potential, cost, consumer demand,
 4   and social acceptance. Such uncertainty carries governance challenges. Policy approaches facing deep
 5   uncertainty must protect against and/or prepare for unforeseeable developments, whether it is through
 6   resistance (planning for the worst possible case or future situation), resilience (making sure you can
 7   recover quickly), or adaptation (changes to policy under changing conditions). Such uncertainty can be
 8   hedged in part by learning by firms, consumers, and policymakers. Social interactions and network
 9   building (e.g., supply and distribution chains, intermediary actors) and the articulation of positive
10   visions, such as in long-term, low-emission development strategies, all play a crucial role. This
11   uncertainty extends to the impacts of low carbon innovations on energy demand and other variables,
12   where unanticipated and unintended outcomes are the norm. For instance, rapid investments in public
13   transport networks could restrict car ownership from becoming common in developing countries (Du
14   and Lin 2017).
15   Physical Energy System Lock-In
16   Current investments in fossil infrastructure have committed 500-700 Gt-CO2 of emissions, creating
17   significant risks for limiting warming to 1.5℃ (Callaghan 2020) (high confidence). These current
18   investments combined with emissions from proposed fossil infrastructure exceed the emissions required
19   to limit warming to 1.5℃ (medium confidence). Existing coal and gas fired electricity generation
20   accounts for 200-300 Gt-CO2 of committed emissions. Emissions from coal generation are larger than
21   for gas plants (Tong et al. 2019; Smith et al. 2019). The lifetime of coal-fired power plants is 25-50
22   years, creating long lasting risks to climate goals (Erickson and Tempest 2015). Gas-fired power plants
23   are generally younger than coal-fired power plants. Industry sector lock-in amounts for more than 100
24   Gt-CO2, while buildings and transport sector together contribute another 50-100 Gt-CO2 (Erickson and
25   Tempest 2015).
26   Lock-in is also relevant to fossil resources. Both coal and gas exploration continue, and new permits
27   are being issued, which may cause economic (Erickson et al. 2018) as well as non-economic issues
28   (Boettcher et al. 2019).
29   The nature of lock in varies across the energy system. For example, lock-in in urban and transport
30   sectors is different from the electricity sector. Broadly, urban environments involve infrastructural,
31   institutional, and behavioural lock-in (Ürge-Vorsatz et al. 2018). Addressing lock-in in these sectors
32   requires action by multiple stakeholders and is unlikely with just technological evolution (Table 6.11).
33   Committed carbon emissions are unevenly distributed. The disproportionate high share of committed
34   emissions in emerging economies is the result of rapid growth in recent years, which has led to a
35   comparably young fossil infrastructure with substantial remaining life (Shearer et al. 2017). Mature
36   industrialized countries tend to have older infrastructures, part of which will be up for retirement in the
37   near future (Tong et al. 2019). Coal-fired power plants currently planned or under construction are
38   associated with 150-300 Gt-CO2, of which ~75% and ~10% are located in Asia and the OECD
39   respectively (Pfeiffer et al. 2018; Edenhofer et al. 2018). If implemented, these new fleets will further
40   shorten all coal plants’ lifetimes by another 10 years for meeting climate goals (Cui et al. 2019)

     Do Not Cite, Quote or Distribute                  6-114                                  Total pages: 217
     Final Government Distribution                       Chapter 6                         IPCC WGIII AR6

2      Figure 6.34 Annual emissions from existing, proposed, and future energy system infrastructure (Tong
3                                                  et al. 2019).

4    Despite the imperative to reduce use of fossil fuels and the multiple health and other benefits from
5    closing coal-based infrastructure (Liu et al. 2019a; Portugal-Pereira et al. 2018; Rauner et al. 2020;
6    Karlsson et al. 2020; Cui et al. 2021), coal power plants have continued to be commissioned globally
7    (Jewell et al. 2019; Jakob et al. 2020), most notably in Asian countries. Gas power plants also continue
8    to be built. In many regions, new fossil electricity generation exceeds needed capacity (Shearer et al.
9    2017).
10   Existing policies and the NDCs are insufficient to prevent an increase in fossil infrastructure and
11   associated carbon lock in (high confidence) (Bertram et al. 2015; Johnson et al. 2015). Current
12   investment decisions are critical because there is limited room within the carbon budget required to
13   limit warming to well below 2°C (Rosenbloom 2019; Kalkuhl et al. 2019). Delays in mitigation will
14   increase carbon lock-in and could result in large-scale stranded assets if stringency is subsequently
15   increased to limit warming (Box 6.11). Near-term implementation of stringent GHG mitigation policies
16   are likely to be most effective in reducing carbon lock-in (Haelg et al. 2018). Near-term mitigation
17   policies will also need to consider different energy transition strategies as a result of different resources
18   and carbon budgets between countries (Bos and Gupta 2018; Lucas 2016).
19   Near-term policy choices are particularly consequential for fast-growing economies. For example,
20   Malik et al. (2020) found that 133 to 227 GW of coal capacity would be stranded after 2030 if India
21   were to delay ambitious mitigation through 2030 and then pursue an ambitious, post-2030 climate
22   strategy. (Cui et al. 2021)identified 18% of old, small, inefficient coal plants for rapid near-term
23   retirement in China to help achieve air quality, health, water, and other societal goals and a feasible coal
24   phaseout under climate goals. Comparable magnitudes of stranded assets may also be created in Latin
25   America when adding all announced, authorized, and procured power plants up to 2060 (González-
26   Mahecha et al. 2019). Options to reduce carbon lock in include reducing fossil fuels subsidies (Box
27   6.3), building CCS-ready facilities, or ensuring that facilities are appropriately designed for fuel
28   switching (Budinis et al. 2018). Substantial lock-in may necessitate considerable deployment of CDR
29   to compensate for high cumulative emissions.
30   Past and present energy sector investments have created technological, institutional, and behavioural
31   path dependencies aligned towards coal, oil, and natural gas (high confidence). In several emerging
32   economies, large projects are planned that address poverty reduction and economic development. Coal
33   infrastructure may be the default choice for these investments without policies to invest in low-carbon
34   infrastructure instead (Steckel et al. 2020; Joshua and Alola 2020). Path dependencies frequently have

     Do Not Cite, Quote or Distribute                   6-115                                   Total pages: 217
     Final Government Distribution                       Chapter 6                         IPCC WGIII AR6

1    sustainability implications beyond carbon emissions. (Box 6.2 and Section 6.7.7). There are several
2    SDG co-benefits associated with decarbonization of energy systems (Section 6.7.7; Sörgel et al. 2021).
3    For example, coal mining communities frequently experience significant health and economic burdens
4    from resource extraction.
5    [START BOX 6.13 HERE]

6                                           Box 6.13 Stranded Assets

 7   Limiting likely warming to 2°C or below will result in stranded assets (high confidence). Stranded assets
 8   can be broadly defined as assets which “suffer from unanticipated or premature write-offs, downward
 9   revaluations or [conversion] to liabilities.” Stranded assets may create risks for financial market stability
10   and macro-economic stability (Mercure et al. 2018, Battiston et al. 2017; Sen and von Schickfus 2020),
11   and they will result in a rapid loss of wealth for the owners of affected assets (Vogt-Schilb and
12   Hallegatte 2017; Ploeg and Rezai 2020).
13   There are two types of stranded assets: fossil-fuel resources that cannot be burned, and premature
14   retirement of fossil infrastructure (e.g., power plants). About 30% of oil, 50% of gas, and 80% of coal
15   reserves will remain unburnable if likely warming is limited to 2°C (Meinshausen et al. 2009; Leaton
16   2011; Leaton Ranger 2013; Pye et al. 2020; IRENA 2017b; Bauer et al. 2016; McGlade and Ekins 2015)
17   (high confidence). Significantly more reserves are expected to remain unburned if warming is limited
18   to 1.5°C. Countries with large oil, gas, and coal reserves are most at risk (Caldecott et al. 2017; Ansari
19   and Holz 2020)
20   About 200 GW of fossil fuel electricity generation per year will likely need to be retired prematurely
21   after 2030 to limit likely warming to 2°C, even if countries achieve their NDCs (Iyer et al. 2015;
22   Johnson et al. 2015; Fofrich et al. 2020) (medium confidence). Limiting warming to 1.5°C will require
23   significantly more rapid premature retirement of electricity generation capacity (Binsted et al. 2020).
24   Coal- and gas-fired power plants will likely need to retire about 25 years earlier than in the past to limit
25   likely warming to 2°C, and 30 years earlier to limit warming to 1.5°C (Cui et al. 2019; Fofrich et al.
26   2020). Coal-fired power plants are at significantly greater risk of stranding compared with gas-fired and
27   oil-fired plants (Iyer et al. 2015; Johnson et al. 2015; Fofrich et al. 2020). The risks of stranded power
28   plants are greatest in countries with newer fossil infrastructure.
29   If likely warming is limited to 2°C, the discounted economic impacts of stranded assets, including
30   unburned fossil reserves, could be as high as USD 1-4 trillion from 2015 through 2050 (USD 10-20
31   trillion in undiscounted terms) (Mercure et al. 2018), IRENA, 2017) (medium confidence). About 40%
32   of these impacts correspond to unburned fossil reserves (IRENA 2017b). If warming is limited to 1.5°C,
33   the economic impacts of stranded assets are expected to be significantly higher (Binsted et al. 2020)
34   Stronger near-term mitigation will reduce premature retirements of fossil infrastructure, because more
35   rapid mitigation will decrease new builds of fossil infrastructure that might later be stranded (Johnson
36   et al. 2015; Bertram et al. 2018) (high confidence). For example, if likely warming is limited to 2°C,
37   strengthening the NDC pledges beyond their 2015 levels could decrease stranded electricity sector
38   assets by more than 50% (Iyer et al. 2015). By contrast, if countries fail to meet their NDCs and continue
39   to build fossil infrastructure, mitigation will need to be accelerated beyond 2030, resulting up to double
40   the amount of stranded electricity generation capacity (Iyer et al. 2015). This corresponds to a total
41   undiscounted cost of about USD 2 trillion from electricity infrastructure alone, from the period 2015
42   to 2050 (IRENA 2017). CCS (6.4) could potentially help reduce hundreds of gigawatts stranded power
43   plant capacity along with other fossil-based capital (Clark and Herzog 2014; Fan et al. 2018; Iyer et al.
44   2017).
45   [END BOX 6.13 HERE]

     Do Not Cite, Quote or Distribute                   6-116                                   Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    6.7.4 Fossil fuels in a low-carbon transition
 2   Global fossil fuel use will need to decline substantially by 2050 to limit likely warming to 2°C or below,
 3   and it must decline substantially by 2030 to limit warming to 1.5°C with no or limited overshoot (high
 4   confidence). Failing to reduce global fossil fuel use below today’s levels by 2030 will make it more
 5   challenging to limit likely warming to below 2°C. (high confidence). Fossil fuel use declines by 260-
 6   330 EJ (52-73% from 2020 levels, interquartile range) through 2050 to limit warming to 1.5°C with no
 7   or limited overshoot and 124-231 EJ (23-51% reduction compared to 2020 levels) to limit likely
 8   warming to 2°C; this will require a significant reduction in coal, oil and gas investments. Fossil fuels
 9   account for about 80% of primary energy today. In scenarios limiting warming to 1.5°C with limited or
10   no overshoot, fossil energy provides 59-69% (interquartile range) primary energy in 2030 and 25-40s%
11   primary energy in 2050 (AR6 database). In scenarios limiting likely warming to 2°C or below, fossil
12   energy provides 71-75% (interquartile range) primary energy in 2030 and 41-57% primary energy in
13   2050 (AR6 database). The timeline for reducing production and usage varies across coal, oil, and gas
14   due to their differing carbon intensities and their differing uses.
15   Global coal consumption without CCS needs to be largely eliminated by 2040-2050 to limit warming
16   to 1.5°C, and 2050-2060 to limit likely warming to 2°C (high confidence). New investments in coal-
17   fired electricity without CCS are inconsistent with limiting likely warming to 2°C or below (high
18   confidence) (Spencer et al. 2018; Pfeiffer et al. 2018; Edenhofer et al. 2018; Cui et al. 2019). Coal
19   consumption declines 130 EJ yr-1 to 140 EJ yr-1 in 2050 (79% to 99% compared to 2020 levels,
20   interquartile range) in scenarios limiting warming to 1.5°C and 118 EJ yr-1 to 139 EJ yr-1 (66% to 98%
21   compared to 2020 levels) in scenarios limiting likely warming to 2°C. Coal consumption without CCS
22   falls by 67% to 82% (interquartile range) in 2030 in scenarios limiting warming to 1.5°C with no or
23   limited overshoot. Studies indicate that coal use may decline substantially in the US and Europe over
24   the coming decade based on the increasing competitiveness of low-carbon sources and near-term policy
25   actions (Oei et al. 2020; Grubert and Brandt 2019). In several developing economies, the relative youth
26   of the coal-fired electricity fleet will make a complete phaseout before 2050 difficult (Garg and Shukla
27   2009; Jewell et al. 2016). There are considerable differences in projected coal phaseout timelines in
28   major Asian economies. Some studies suggest that coal may continue to be a part of the Chinese energy
29   mix composing around a third of the total primary energy consumption by 2050 even if emissions are
30   reduced by 50% by 2030 (He et al. 2020). Others indicate that a strategic transition would decrease the
31   risk of stranded assets and enable a near-complete phaseout by 2050 (Wang et al. 2020a; Cui et al.
32   2021). This would entail prioritizing earlier retirements of plants based on technical (efficiency),
33   economic (profitability, local employment) and environmental considerations (e.g., water scarcity for
34   cooling).
35   Natural gas may remain part of energy systems through mid-century, both for electricity generation and
36   use in industry and buildings, and particularly in developed economies, even if likely warming is limited
37   to 2°C or less (medium confidence). The decline in natural gas use in from 2020 to 2050 is 38 EJ yr-1 to
38   78 EJ yr-1 (21% to 61% decline from 2020 levels, interquartile range) in scenarios limiting warming to
39   1.5C with no or limited overshoot and -22 EJ yr-1 to 46 EJ yr-1 (-13% to 36% decline from 2020 levels)
40   in scenarios limiting likely warming to 2°C. Scenarios indicate that gas use in electricity will likely
41   peak around 2035 and 2050 if warming is limited to 1.5°C with limited or no overshoot or likely below
42   2°C, respectively. There is variability in the role gas would play in future scenarios based on national
43   climate commitments and availability of cheap renewable (Vrontisi et al. 2020, (Vishwanathan and
44   Garg 2020; Malik et al. 2020). Note that these differences are not only present in the electricity sector
45   but also in other end-uses.
46   While oil use is anticipated to decline substantially, due to changes in the transport sector, its use will
47   likely continue through the mid-century, even if likely warming is limited to 2°C or less (medium
48   confidence). Oil use declines by 43 EJ yr-1 to 91 EJ yr-1 (19% to 54% from 2020 levels, interquartile
     Do Not Cite, Quote or Distribute                  6-117                                  Total pages: 217
     Final Government Distribution                                                                                                                                                                       Chapter 6                                                                                                                                    IPCC WGIII AR6

1    range) in scenarios that limit warming to 1.5°C with no or limited overshoot and 46 EJ yr-1 to 109 EJ
2    yr-1 (21% 60% from 2020 levels) by 2050 in scenarios that limit likely warming to 2°C. While oil use
3    is anticipated to decline immediately in scenarios limiting warming to 1.5°C, it is likely to continue to
4    be used through 2050. Oil use continues to be a significant source of transport fuels in most scenarios
5    limiting likely warming to 2°C (Welsby et al, 2021). Oil use may reduce to about half of the current
6    levels as a transport fuels (Feijoo et al. 2020) if likely warming is limited to 2°C, because of the
7    availability of other options (biofuels, green hydrogen) and rapid deployment of EVs. In the absence of
8    rapid transport electrification, the decline is slower with some studies projecting peak oil use around
9    2035 (Delgado et al. 2020; Pan et al. 2020).
10   There is a lack of consensus about how CCS might alter fossil fuel transitions for limiting likely
11   warming to 2°C or below. CCS deployment will increase the shares of fossil fuels associated with
12   limiting warming, and it can ease the economic transition to a low-carbon energy system (Muratori et
13   al. 2016; Marcucci et al. 2019).While some studies find a significant role for fossil fuels with CCS by
14   2050 (Koelbl et al. 2014; Eom et al. 2015; Vishwanathan and Garg 2020), others find that retirement of
15   unabated coal far outpaces the deployment of coal with CCS (McJeon et al, 2021; Budinis et al. 2018;
16   Xie et al. 2020) Moreover, several models also project that with availability of CO2 capture technology,
17   BECCS might become significantly more appealing than fossil CCS even before 2050 (Luderer et al.
18   2018b; Muratori et al, 2017).
                                                          Coal without CCS
                                                                                                                                                                 Gas without CCS                                                                                                     Oil                                                                               Total fossil
                                                                                                                                                       140                                                                                                   140                                                                                              120
        Primary energy relative to 2020 (%)

                                                                                                                 Primary energy relative to 2020 (%)

                                                                                                                                                                                                                      Primary energy relative to 2020 (%)

                                               100                                                                                                     120                                                                                                   120

                                                                                                                                                       100                                                                                                   100

                                                                                                                                                        80                                                                                                    80

                                                                                                                                                        60                                                                                                    60

                                                                                                                                                                                                                                                                                                                        Primary energy relative to 2020 (%)
                                                40                                                                                                                                                                                                                                                                                                             80
                                                                                                                                                        40                                                                                                    40

                                                                                                                                                        20                                                                                                    20

                                                    0                                                                                                    0                                                     CH4                                             0                                                                                               60
                                                          2030       2040                                 2050                                                  2030             2040           2050                                                                     2030               2040               2050

                                                        Total fossil without CCS                                                                                     Coal with CCS                                                                                              Gas with CCS
                                              120                                                                                                      90                                                                                                      140
                                                                            Emissions (Mt-CO2/year)

                                                                                                      6000                                             80
     Primary energy relative to 2020 (%)

                                              100                                                                                                                                                                                                              120                                                                                             40
                                                                                                                         Primary energy (EJ)

                                                                                                                                                                                                                                        Primary energy (EJ)

                                                                                                      5000                                                                                                                                                     100
                                               80                                                                                                      60

                                                                                                      4000                                             50                                                                                                       80
                                                                                                                                                       40                                                                                                       60
                                                                                                      3000                                                                                                                                                                                                                                                     20
                                               40                                                                                                      30
                                                                                                      1000                                             10

                                                0                                                                                                       0                                                                                                          0                                                                                            0
                                                         2030        2040                                02050                                                2020        2030          2040      2050                                                                  2020         2030          2040          2050                                               2030       2040   2050
                                                                                                                                                       2020                                     2030                                                                             2040                                   2050

                                                                                                                                                                     1.5°C without overshoot   1.5°C with overshoot                                         Likely below 2°C     Likely below 2°C with delay
20    Figure 6.35 Global fossil fuel pathways for the 1.5°C scenarios (without and with overshoot), and likely
21       2°C scenarios (without and with delayed policy action) during 2020-2050 (Source: AR6 Scenario
22     Database). Results for total consumption are expressed as a percentage relative to 2020 consumption.
23     Results for fossil energy with CCS are expressed in total energy consumption. Boxes indicate 25th and
24   75th percentiles while whiskers indicate 5th and 95th percentiles. Oil use with CCS is not shown here as it
25                                         remains below 5% of total use.

26   6.7.5 Policy and Governance
27   Policy and governance frameworks are essential for shaping near- and medium-term low-emissions
28   energy system transitions (high confidence). While policy interventions are necessary to achieve low-
29   carbon energy system transitions, appropriate governance frameworks are crucial to ensure policy
30   implementation (high confidence). The policy environment in energy transition pathways relate to
     Do Not Cite, Quote or Distribute               6-118                                 Total pages: 217
     Final Government Distribution                     Chapter 6                        IPCC WGIII AR6

1    climate policy goals, the characteristics of the policy regimes and measures to reach the policy goals
2    including implementation limits and obstacles, and the timing of the climate instrument (Kriegler et al.
3    2014b)
 4   The literature discusses a broad set of policy approaches. Environmental economics focuses mainly on
 5   market-based approaches as the least-cost policy to achieve emission reductions (Kube et al. 2018).
 6   Many countries, however, have implemented policy mixes with a diverse set of complementary policies
 7   to achieve energy and climate policy targets. One example is the German Energiewende, which includes
 8   substantial support for renewables, an action plan for energy efficiency, and phase-out processes for
 9   nuclear- and coal-based power generation next to carbon pricing (Löschel et al. 2019). The halving of
10   CO2 emissions in UK power generation reflects multiple policies, particularly within the UK’s Climate
11   Change Act 2008 (Grubb and Newbery 2018). More generally, the implementation of the NDCs under
12   the Paris Agreement are all characterized by diverse climate policy mixes.
13   These policy mixes (or policy packages) are shaped by different factors, including policy goals and
14   objectives (including political, social and technological influences), multiple market, governance or
15   behavioural failures or previous policy choices of earlier policy eras (Rogge 2017). When pursuing
16   multiple policy goals or targeting some type of imperfection, well designed policy mixes can in
17   principle reduce mitigation costs (Corradini et al. 2018) or address distributional concerns, especially
18   vulnerable populations. For example, the interaction between carbon pricing and the support for clean
19   energy technologies in the EU clean low-carbon strategy for 2050 can reduce mitigation costs and allow
20   for the early adoption of more stringent climate targets (Vandyck et al. 2016). Policy efforts to promote
21   adoption of low-carbon technologies are more successful if they focus not only on economic incentives
22   but include behavioural interventions that target relevant cognitive and motivational factors (Khanna et
23   al. 2021; Mundaca et al. 2019; see also Section 6.7.6). Overlapping nudges might not necessarily lead
24   to lower effectiveness (Brandon et al. 2019).
25   Well-designed policy mixes can support the pursuit of multiple policy goals, target effectively different
26   types of imperfections and framework conditions and take into account the technological, economical,
27   and societal situation (high confidence). Accounting for the different development stages of new
28   technologies will enhance low-emissions transitions (Graaf and Sovacool 2020). For prototype
29   technologies and technologies in the demonstration phase, research subsidies and demonstration
30   projects are most important. For technologies experiencing early adoption, infrastructure development
31   and strengthening of markets are increasingly important, while retiring or repurposing of existing assets
32   is important for mature technologies (IEA 2020h) Effective policy mixes will address different market
33   frictions and deal with various uncertainties, for example, those pertaining to technological, climate,
34   and socio-economic developments (Aldy 2020), but also with respect to outcomes of individual policies
35   (e.g. Borenstein et al. 2019). Therefore, policy mixes may balance the trade-off between stability and
36   the flexibility to change individual policies (Gawel and Lehmann 2019) and the policy mix over time
37   (Rayner et al. 2017). Some policy instruments may become feasible over time, for example, as
38   technological advancements reduce the transaction costs of comprehensive market-based approaches
39   (Andoni et al. 2019; Di Silvestre et al. 2020), or as weakened barriers to stringency enable policy
40   sequencing (Pahle et al. 2018) . Energy system policy mixes often include sector-specific regulation.
41   Compared to economy-wide approaches, sectoral policies may be able to directly target specific sectors
42   or mitigation options. However, uncoordinated implementation or limited coordination across sectors
43   may lead to efficiency losses (e.g. Rosendahl et al. 2017). These losses also depend on other policies,
44   such as pre-existing taxes (Goulder et al. 2016; Marten et al. 2018) or research and development policies
45   (Acemoglu et al. 2016). Moreover, unilateral policies – those taken by individual countries in the
46   absence of coordination with other countries – could raise carbon leakage risks, while balancing
47   potential issues of (industrial) competitiveness (Martin et al. 2014; Rosendahl et al. 2017). Energy
48   leakage may become more important during low-carbon energy systems. Numerous studies have

     Do Not Cite, Quote or Distribute                 6-119                                 Total pages: 217
     Final Government Distribution                      Chapter 6                          IPCC WGIII AR6

1    identified pathways for carbon leakage in electricity markets with incomplete emission markets (Caron
2    et al. 2015; Thurber et al. 2015; Murray and Maniloff 2015; Fell and Maniloff 2017; Duan et al. 2017;
3    Qian et al. 2018). Well-designed policy mixes will need to target the whole life-cycle or value chains,
4    for example, through policies on limiting fossil fuel extraction (Asheim et al. 2019), or they will need
5    to include measures to limit carbon leakage (e.g. Cosbey et al. 2019).
 6   Interactions between policy measures including their scope, stringency, and timing, influence the costs
 7   of reducing emissions (Corradini et al. 2018). In particular, some policy instruments may lead to lock-
 8   in effects (Section 6.7.3.), compete with other regulations (Graaf and Sovacool 2020), or trigger
 9   negative policy interactions (Perino 2015; Jarke-Neuert and Perino 2020). Existing policy mixes often
10   reflect different political economy constraints and sometimes not well coordinated goals. The resulting
11   policy mixes are often economically inefficient. However, comprehensive evaluation of policy mixes
12   requires a broader set of criteria that reflect different considerations, such as broader goals (e.g., SDGs)
13   and the feasibility of policies (high confidence).
14   Policy mixes might rather emerge piece-by-piece over time out of individual policy interventions rather
15   than be designed as a whole from the outset (Howlett 2014; Rogge 2017) and may reflect differences
16   across jurisdictions and sectors (Howlett 2014). For example, taking into account country-specific
17   objectives, failures, and limitations, carbon prices may be only one part of a broader policy mix and
18   thereby may not be uniform across countries (Bataille 2020). This lack of consistency makes it more
19   difficult to assess economic outcomes since costs of complementary policies are often less visible and
20   are often targeted at high-cost mitigation options (Borenstein et al. 2019).
21   Effective assessment of policy mixes requires comprehensive, validated international data,
22   methodologies, and indicators. Existing policy mixes are difficult to evaluate because they target
23   multiple objectives, and the evaluation must consider various criteria (Chapter 13, 6.7.7), such as
24   environmental and economic effectiveness, distributional effects, transformative potential, institutional
25   requirements, and feasibility. Economic outcomes depend on policy goals and implementation. Existing
26   studies on policy mixes suggest the benefits of a comprehensive approach (Rosenow et al. 2017), while
27   also highlighting that an “excessive” number of instruments may reduce overall effectiveness
28   (Costantini et al. 2017). Combining environmental regulation and innovation policies may be of
29   particular importance to tackle both emissions and innovation market failures (Fabrizi et al. 2018). The
30   consistency and credibility of policy mixes is positively associated with green innovation (Rogge and
31   Schleich 2018).
32   Potential future policies are difficult to evaluate due to methodological challenges (high confidence).
33   Recent model-based analyses of future policy mixes based on “current policy scenarios” try to
34   implement existing policies besides explicit or implicit carbon prices (den Elzen et al. 2016; Roelfsema
35   et al. 2020; van Soest et al. 2017; Rogelj et al. 2016). Many assessments of future low-carbon energy
36   transitions are still based on cost-optimal evaluation frameworks and include only limited analysis of
37   interactions between policy measures. Hence they are often not describing real-world energy transitions
38   properly, but rather differences in implied carbon prices, constraints in technology deployment, and
39   timing of policies (Trutnevyte 2016).
41   6.7.6 Behaviour and Societal Integration
42   Members of societies, including individuals, civil society, and businesses, will all need to engage with
43   and be affected by low-carbon energy system transitions (high confidence). This raises questions about
44   the extent to which different strategies and policy would effectively promote mitigation behaviours and
45   the factors that increase the social acceptability of mitigation options, policies, and system changes.

     Do Not Cite, Quote or Distribute                  6-120                                   Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1 Strategies to encourage climate mitigation actions
2    Climate policy will be particularly effective if it targets key factor inhibiting, enabling, and motivating
3    mitigation behaviours. As barriers differ across mitigation options, regions, and groups, tailored
4    approaches are more effective (Grubb et al. 2017). When people face important barriers to change (e.g.,
5    high costs, legal barriers), policy would be needed make low carbon actions more attractive, or to make
6    high carbon actions less attractive. As people generally face multiple barriers for change, combinations
7    of policies would be more effective (Rosenow et al. 2017).
 8   Financial incentives can motivate mitigation actions (Santos 2008; Thøgersen 2009; Eliasson 2014;
 9   Maki et al. 2016; Bolderdijk et al. 2011), particularly when actions are costly (Mundaca 2007). In many
10   countries, more residential solar PV were installed after the introduction of favourable financial
11   schemes such as feed-in-tariffs, federal income tax credits, and net metering (Wolske and Stern 2018).
12   Similarly, a subsidy promoted the installation of solar water heaters in Asia (Chang et al. 2009). Yet,
13   financial incentives may underperform expectations when other factors are overlooked. For example,
14   people may not respond to financial incentives when they do not trust the organization sponsoring the
15   program or when it takes too much effort to receive the incentive (Mundaca 2007; Stern et al. 2016a).
16   Financial incentives are more effective if combined with strategies addressing non-financial barriers.
17   Communicating financial consequences of behaviour seems less effective than emphasizing social
18   rewards (Handgraaf et al. 2013) or benefits of actions for people (e.g., public health, comfort) and the
19   environment (Asensio and Delmas 2015, 2016; Schwartz et al. 2015; Ossokina 2020; Bolderdijk et al.
20   2013). Financial appeals may have limited effects because they reduce people’s focus on environmental
21   consequences, weaken intrinsic motivation to engage in mitigation actions, provide a license to pollute
22   (Agrawal et al. 2015; Bolderdijk and Steg 2015; Schwartz et al. 2015), and because pursuing small
23   financial gains is perceived not worth the effort (Dogan et al. 2014; Bolderdijk et al. 2013).
24   Providing information on the causes and consequences of climate change or on effective mitigation
25   actions increases people’s knowledge and awareness, but generally does not promote mitigation actions
26   by individuals (Abrahamse et al. 2005) or organizations (Anderson and Newell 2004). Fear-inducing
27   representations of climate change may inhibit action when they make people feel helpless (O´Neill and
28   Nicholson-Cole 2009). Energy-related advice and feedback can promote energy savings, load shifting
29   in electricity use and sustainable travel, particularly when framed in terms of losses rather than gains
30   (Gonzales et al. 1988; Wolak 2011; Bradley et al. 2016; Bager and Mundaca 2017). Also, credible and
31   targeted information at the point of decision can promote action (Stern et al. 2016a). Information is
32   more effective when delivered by a trusted source, such as peers (Palm 2017), advocacy groups (Schelly
33   2014), and community organizations (Noll et al. 2014), and when tailored to actors’ personal situation
34   and core values (Abrahamse et al. 2007; Boomsma and Steg 2014; van den Broek et al. 2017; Daamen
35   et al. 2001; Wolsko et al. 2016; Bolderdijk et al. 2013). This explains why home energy audits promoted
36   energy savings (Delmas et al. 2013; Alberini and Towe 2015), and investments in resource efficiency
37   and renewable energy generation (Kastner and Stern 2015).
38   Energy use feedback can promote energy saving behaviour within households (Grønhøj and Thøgersen
39   2011; Fischer 2008; Karlin et al. 2015; Delmas et al. 2013; Zangheri et al. 2019) and at work (Young
40   et al. 2015), particularly when provided in real-time or immediately after the action so that people learn
41   the impact of different actions (Faruqui et al. 2009; Delmas et al. 2013; Abrahamse et al. 2005;
42   Tiefenbeck et al. 2016; Stern et al. 2016a; Yu et al. 2015). Energy labels (Banerjee and Solomon 2003;
43   Stadelmann 2017), visualization techniques (Pahl et al. 2016), and ambient persuasive technology
44   (Midden and Ham 2012) can encourage energy savings as they immediately make sense and hardly
45   require users’ conscious attention. Feedback can make people aware of their previous mitigation
46   behaviours, which can strengthen their environmental self-identity, and motivate them to engage in
47   other mitigation actions as well as to act in line with their self-image (Van der Werff et al. 2014).

     Do Not Cite, Quote or Distribute                  6-121                                  Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    Social influence approaches that communicate what other people do or think can encourage mitigation
2    actions (Clayton et al. 2015), as can social models of desired actions (Osbaldiston and Schott 2012;
3    Abrahamse and Steg 2013; Wolske et al. 2020; Sussman and Gifford 2013). Feedback on one’s own
4    energy use relative to others can be effective (Nolan et al. 2008; Allcott 2011; Schultz et al. 2015),
5    although not always, and effect sizes are small (Abrahamse and Steg 2013) compared to other types of
6    feedback (Karlin et al. 2015).
 7   Interventions that capitalize on people’s motivation to be consistent can promote mitigation actions
 8   (Steg 2016). Examples are commitment strategies where people pledge to act (Abrahamse and Steg
 9   2013; Lokhorst et al. 2013), implementation intentions where they additionally explicate how and when
10   they will perform the relevant action and how they would cope with possible barriers (Rees et al. 2018;
11   Bamberg 2000, 2002), and hypocrisy-related strategies that make people aware of inconsistencies
12   between their attitudes and behaviour (Osbaldiston and Schott 2012).
13   Bottom-up approaches can promote mitigation action (Abrahamse and Steg 2013). Indeed, community
14   energy initiatives can encourage members’ low carbon behaviour (Middlemiss 2011; Seyfang and
15   Haxeltine 2012; Abrahamse and Steg 2013; Sloot et al. 2018). Organizations can promote mitigation
16   behaviour among their employees and customers by communicating their mission and strategies to
17   mitigate climate change (van der Werff et al. 2021; Ruepert et al. 2017).
18   Default options, where a preset choice is implemented if users do not select another option, can promote
19   mitigation actions such as energy savings, green electricity uptake, and meat-free meals options (Liebe
20   et al. 2021; Pichert and Katsikopoulos 2008; Ölander and Thøgersen 2014; Kunreuther and Weber
21   2014; Bessette et al. 2014; Ebeling and Lotz 2015; Liebe et al. 2018; Campbell-Arvai et al. 2014).
22 Acceptability of policy, mitigation options and system changes
23   Public acceptability reflects the extent to which the public evaluates climate policy, mitigation options,
24   and system changes (un)favourably, which can shape, enable, or prevent low-carbon energy system
25   transitions. Public acceptability of policy and mitigation options is higher when people expect these
26   have more positive and less negative consequences for self, others, and the environment (Demski et al.
27   2015; Drews and Van den Bergh 2016; Perlaviciute and Steg 2014). Public opposition may result when
28   a culturally valued landscape is affected by renewable energy development (Warren et al. 2005; Devine-
29   Wright and Howes 2010), particularly place-based identities are threatened (Devine-Wright 2009, 2013;
30   Boudet 2019). Acceptability can increase after a policy or change has been implemented and the
31   consequences appear to be more positive than expected (Carattini et al. 2018; Schuitema et al. 2010;
32   Eliasson 2014; Weber 2015); effective policy trials can thus build public support.
33   Next, climate policy and low carbon options are evaluated as more fair and acceptable when costs and
34   benefits are distributed equally, and when nature, the environment and future generations are protected
35   (Schuitema et al. 2011; Drews and Van den Bergh 2016). Compensating affected groups for losses due
36   to policy or systems changes enhanced public acceptability in some cases (Perlaviciute and Steg 2014),
37   but people may disagree on which compensation would be worthwhile (Aitken 2010b; Cass et al. 2010),
38   on the distribution of compensation (Devine-Wright and Sherry-Brennan 2019; Leer Jørgensen et al.
39   2020), or feel they are being bribed (Perlaviciute and Steg 2014; Cass et al. 2010). Pricing policies are
40   more acceptable when revenues are earmarked for environmental purposes (Steg et al. 2006; Sælen and
41   Kallbekken 2011) or redistributed towards those affected (Schuitema and Steg 2008).
42   Climate policy and mitigation options, such as renewable energy projects, are also perceived as more
43   fair and acceptable when the public (Dietz 2013; Bidwell 2014; Bernauer et al. 2016b) or public society
44   organizations (Terwel et al. 2010; Bernauer et al. 2016b) could participate in the decision making
45   (Devine-Wright 2005; Terwel et al. 2012; Perlaviciute and Squintani 2020; Arvai 2003; Walker and
46   Baxter 2017). People are more motivated to participate in decision making on local projects than on
47   national or general policy goals (Perlaviciute and Squintani 2020). Public acceptability is also higher
     Do Not Cite, Quote or Distribute                 6-122                                  Total pages: 217
     Final Government Distribution                      Chapter 6                         IPCC WGIII AR6

1    when people can influence major rather than only minor decisions, particularly when trust in responsible
2    parties is low (Liu et al. 2019a). Public participation can enhance the quality and legitimacy of decisions
3    by including local knowledge and views that may otherwise be missed (Bidwell 2016; Dietz 2013).
4    Public support is higher when people trust responsible parties (Perlaviciute and Steg 2014; Jiang et al.
5    2018; Drews and Van den Bergh 2016; Michaels and Parag 2016; Liu et al. 2019a). Public support for
6    unilateral climate policy is rather strong and robust (Bernauer et al. 2016a), even in the absence of
7    reciprocal commitments by other states (Bernauer and Gampfer 2015).
 8   Public acceptability of climate policy and low carbon options differs across individuals. Climate policy
 9   and low carbon options are more acceptable when people strongly value protecting other people and
10   the environment, and support egalitarian worldviews, left-wing or green political ideologies, while
11   acceptability is lower when people strongly endorse self-centered values, and support individualistic
12   worldviews (Dietz et al. 2007; Perlaviciute and Steg 2014; Drews and Van den Bergh 2016). Similarly,
13   public decision makers support climate policy more when they endorse environmental values (Nilsson
14   et al. 2016). Climate and energy policy is more acceptable when people are more concerned about
15   climate change (Hornsey et al. 2016), when they believe their actions would help mitigating climate
16   change, and feel responsible to mitigate climate change (Steg 2005; Jakovcevic and Steg 2013; Ünal et
17   al. 2019; Eriksson et al. 2006; Drews and Van den Bergh 2016; Kim and Shin 2017).
19   6.7.7 The Costs and Benefits of Low-Carbon Energy System Transitions in the Context
20         of Sustainable Development
21   The attractiveness of energy sector mitigation ultimately depends on the way that it provides benefits
22   and reduces the costs for the many different priorities that societies value(Wei et al. 2018, 2020; Yang
23   et al. 2018a). While costs and benefits of climate mitigation are often considered in the context of pure
24   economic outcomes – for example, GDP effects or changes in value of consumption – costs and benefits
25   should be viewed with a broader lens that accounts for the many ways that the energy system interacts
26   with societal priorities (Karlsson et al. 2020). Climate mitigation is not separate from countries’ broader
27   growth and development strategies, but rather as a key element of those strategies.
28   Cost reductions in key technologies, particularly in electricity and light-duty transport, have increased
29   the economic attractiveness of near-term low-carbon energy system transitions (high confidence). The
30   near-term, economic outcomes of low-carbon energy system transitions in some sectors and regions
31   may be on par with or superior to those of an emissions-intensive future (high confidence). Even in
32   cases when system costs are higher for low-carbon transitions, these transitions may still be
33   economically favourable when accounting for health impacts and other co-benefits (Gielen et al. 2019).
34   Past assessments have quantified the aggregate economic costs for climate change mitigation using
35   different metrics, for example carbon prices, GDP losses, investments in energy infrastructure, and
36   energy system costs. Assessments of mitigation costs from integrated assessment and energy system
37   models vary widely. For example, scenarios include carbon prices in 2030 of less than USD 20/t-CO2,
38   but also more than USD 400/t-CO2 depending on the region, sector boundary, and methodology (e.g.
39   (Bauer et al. 2016; Oshiro et al. 2017; Vaillancourt et al. 2017; Chen et al. 2019; Brouwer et al. 2016).
40   Those arise both from different methodologies (Guivarch and Rogelj 2017) and assumptions about
41   uncertainties in key factors that drive costs (Meyer et al. 2021)
42   Recent developments, however, raise the prospect that economic outcomes could be substantially
43   superior to prior estimates, particularly if key technologies continue to improve rapidly. In some regions
44   and circumstances, particularly in the electricity sector, near-term mitigation may well lead to superior
45   economic outcomes than continuing to invest in and utilize emissions-intensive infrastructure (e.g.
46   (Brown et al. 2017; Kumar et al. 2020). Given the importance of electricity decarbonization in near-
47   term mitigation strategies (see Section 6.7.1), decreasing costs of solar PV, wind power, and batteries
     Do Not Cite, Quote or Distribute                    6-123                                Total pages: 217
     Final Government Distribution                     Chapter 6                         IPCC WGIII AR6

1    to support their integration, have an outsized influence on near-term economic outcomes from
2    mitigation. At the same time, economic outcomes may vary across regions depending, among other
3    things, on the characteristics of the current energy systems, energy resources, and needs for integrating
4    VRE technologies.
 5   The long-term economic characteristics of low-emissions energy system transitions are not well
 6   understood and depend on policy design and implementation along with future costs and availability of
 7   technologies in key sectors (e.g., process heat, long-distance transport), and the ease of electrification
 8   in end-use sectors (high confidence). The long-term aggregate economic outcomes from a low-
 9   emissions future are not likely to be substantially worse than in an emissions-intensive future and may
10   prove superior (see, e.g., Bogdanov et al. 2021, Child et al. 2019, Farmer et al. 2020) (medium
11   confidence). For the whole economy, the interquartile range of estimated mitigation costs is between
12   USD2015 140 and USD2015 340/t-CO2 in 2050 in scenarios limiting likely warming to 2°C and between
13   USD2015 430 and USD2015 990/tCO2 in scenarios limiting likely warming to 1.5°C (Chapter 3). For
14   energy sectors in various regions and globally, different scenarios show a wide range of implied carbon
15   prices in 2050 to limit warming to 1.5°C, from below USD 50/t-CO2 to more than USD 900/t-CO2
16   (Brouwer et al. 2016; Rogelj et al. 2018a). Mitigation costs for scenarios limiting likely warming to 2°C
17   were 3-11% in consumption losses in AR5, but the median in newer studies is about 3% in GDP losses
18   (Su et al. 2018; Gambhir et al. 2019).
19   Estimates of long run mitigation costs are highly uncertain and depend on various factors. Both faster
20   technological developments and international cooperation are consistently found to improve economic
21   outcomes (Paroussos et al. 2019). Long-term mitigation is likely to be more challenging than near-term
22   mitigation because low-cost opportunities get utilized first and efforts after that would require
23   mitigation in more challenging sectors (Section 6.6). Advances in low-carbon energy resources and
24   carriers such as next-generation biofuels, hydrogen produced from electrolysis, synthetic fuels, and
25   carbon-neutral ammonia would substantially improve the economics of net zero energy systems (high
26   confidence). Current estimates of cumulative mitigation costs are comparably high for developing
27   countries, amounting to up to 2-3% of GDP, indicating difficulties for mitigation without adequate
28   support from developed countries (Fujimori et al. 2020; Dorband et al. 2019). In scenarios involving
29   large amounts of stranded assets, the overall costs of low-carbon transitions also include the additional
30   costs of early retirements (Box 6.11).
31   Focusing only on aggregate economic outcomes neglects distributional impacts, impacts on broader
32   SDGs, and other outcomes of broad societal importance. Strategies to increase energy efficiency and
33   energy conservation are, in most instances, mutually reinforcing with strategies to support sustainable
34   development. Improving efficiency and energy conservation will promote sustainable consumption and
35   production of energy and associated materials (SDG-12) (high confidence). Contrastingly, successful
36   implementation of demand-side options requires sustainable partnerships (SDG-17) between different
37   actors in energy systems, for example governments, utilities, distributors, and consumers. Many authors
38   have argued that energy efficiency has a large untapped potential in both supply and demand (Lovins
39   2018; Méjean et al. 2019). For example, improved fossil power plant efficiency has been estimated to
40   lower the costs of CCS from USD 80-100/t-CO2 for a subcritical plant to