Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Table of Contents
2 Chapter 12: Cross-sectoral perspectives ........................................................................ 12-1
3 Executive summary .......................................................................................................... 12-4
4 12.1 Introduction ........................................................................................................... 12-7
5 12.1.1 Chapter overview ........................................................................................... 12-7
6 12.1.2 Chapter content .............................................................................................. 12-7
7 12.1.3 Chapter layout ................................................................................................ 12-8
8 12.2 Aggregation of sectoral costs and potentials ....................................................... 12-12
9 12.2.1 Introduction .................................................................................................. 12-12
10 12.2.2 Costs and potentials of options for 2030...................................................... 12-15
11 12.2.3 Aggregation of sectoral results and comparison with earlier analyses and
12 Integrated Assessment Models ................................................................................... 12-25
13 12.2.4 Sectoral findings on emission pathways until 2050 ..................................... 12-31
14 12.3 Carbon dioxide removal (CDR) .......................................................................... 12-35
15 Cross-Chapter Box 8 Carbon Dioxide Removal: Key characteristics and multiple roles in
16 mitigation strategies .................................................................................................... 12-35
17 12.3.1 CDR methods not assessed elsewhere in this report: DACCS, enhanced
18 weathering and ocean-based approaches .................................................................... 12-42
19 12.3.2 Consideration of methods assessed in sectoral chapters; A/R, biochar, BECCS,
20 soil carbon sequestration ............................................................................................ 12-55
21 12.3.3 CDR governance and policies ......................................................................... 12-62
22 12.4 Food systems ....................................................................................................... 12-65
23 12.4.1 Introduction .................................................................................................. 12-65
24 12.4.2 GHG emissions from food systems ............................................................. 12-68
25 12.4.3 Mitigation opportunities............................................................................... 12-74
26 12.4.4 Enabling food system transformation .......................................................... 12-85
27 12.4.5 Food Systems Governance ........................................................................... 12-93
28 12.5 Land-related impacts, risks and opportunities associated with mitigation options . 12-
29 96
30 12.5.1 Introduction .................................................................................................. 12-96
31 12.5.2 Land occupation associated with different mitigation options .................... 12-96
32 12.5.3 Consequences of land occupation: biophysical and socioeconomic risks,
33 impacts and opportunities ........................................................................................... 12-99
34 12.5.4 Governance of land-related impacts of mitigation options ........................ 12-106
35 Cross-Working Group Box 3: Mitigation and Adaptation via the Bioeconomy ...... 12-112
36 12.6 Other cross-sectoral implications of mitigation ................................................ 12-117
37 12.6.1 Cross-sectoral perspectives on mitigation action ....................................... 12-117
38 12.6.2 Sectoral policy interactions (synergies and trade-offs) .............................. 12-124
Do Not Cite, Quote or Distribute 12-2 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 12.6.3 International trade spill-over effects and competitiveness......................... 12-127
2 12.6.4 Implications of finance for cross-sectoral mitigation synergies and trade-offs 12-
3 130
4 12.7 Knowledge Gaps ............................................................................................... 12-131
5 Frequently Asked Questions (FAQs) ........................................................................... 12-132
6 References .................................................................................................................... 12-134
7
8
9
Do Not Cite, Quote or Distribute 12-3 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Executive summary
2 The total emission mitigation potential achievable by the year 2030, calculated based on sectoral
3 assessments, is sufficient to reduce global greenhouse gas emissions to half of the current (2019)
4 level or less (robust evidence, high agreement). This potential (32 to 44 GtCO2-eq) requires
5 implementation of a wide range of mitigation options. Options with mitigation costs lower than 20 USD
6 tCO2-1 make up more than half of this potential and are available for all sectors {12.2, Table 12.3}
7 Carbon Dioxide Removal (CDR) is a necessary element to achieve net zero CO2 and GHG
8 emissions both globally and nationally, counterbalancing residual emissions from hard-to-
9 transition sectors. It is a key element in scenarios likely to limit warming to 2°C or lower by 2100
10 (robust evidence, high agreement). Implementation strategies need to reflect that CDR methods differ
11 in terms of removal process, timescale of carbon storage, technological maturity, mitigation potential,
12 cost, co-benefits, adverse side-effects, and governance requirements. All Illustrative Mitigation
13 Pathways (IMPs) use land-based biological CDR (primarily Afforestation/Reforestation, A/R) and/or
14 bioenergy with carbon capture and storage (BECCS) and some include direct air carbon capture and
15 storage (DACCS). As a median value (5–95% range) across the scenarios likely limiting warming to
16 2°C or lower, cumulative volumes of BECCS, net CO2 removal on managed land (including A/R), and
17 DACCS reach 328 (168–763) GtCO2, 252 (20–418) GtCO2, and 29 (0–339) GtCO2 for the 2020-2100
18 period, with annual volumes at 2.75 (0.52–9.45) GtCO2 yr-1 for BECCS and 2.98 (0.23–6.38) GtCO2
19 yr-1 for the net CO2 removal on managed land (including A/R), and 0.02 (0–1.74) GtCO2 yr-1 for
20 DACCS, in 2050. {12.3, Cross-Chapter Box 8 in this chapter}
21 Despite limited current deployment, moderate to large future mitigation potentials are estimated
22 for Direct Air Carbon Capture and Sequestration (DACCS), enhanced weathering (EW) and
23 ocean-based CDR methods (including ocean alkalinity enhancement and ocean fertilisation)
24 (medium evidence, medium agreement). The potential for DACCS (5–40 GtCO2 yr-1) is limited mainly
25 by requirements for low-carbon energy and by cost (100-300 (full range: 84–386) USD tCO2-1). DACCS
26 is currently at a medium technology readiness level. EW has the potential to remove 2–4 (full range:
27 <1 to ~100) GtCO2 yr-1, at costs ranging from 50 to 200 (full range: 24–578) USD tCO2-1. Ocean-based
28 methods have a combined potential to remove 1–100 GtCO2 yr-1 at costs of 40–500 USD tCO2-1, but
29 their feasibility is uncertain due to possible side-effects on the marine environment. EW and ocean-
30 based methods are currently at a low technology readiness level. {12.3}
31 Realising the full mitigation potential from the food system requires change at all stages from
32 producer to consumer and waste management, which can be facilitated through integrated policy
33 packages (robust evidence, high agreement). Some 23-42% of global GHG emissions are associated
34 with food systems, while there is still wide-spread food insecurity and malnutrition. Absolute GHG
35 emissions from food systems increased from 14 to 17 GtCO2-eq yr-1 in the period 1990-2018. Both
36 supply and demand side measures are important to reduce the GHG intensity of food systems. Integrated
37 food policy packages based on a combination of market-based, administrative, informative, and
38 behavioural policies can reduce cost compared to uncoordinated interventions, address multiple
39 sustainability goals, and increase acceptance across stakeholders and civil society (limited evidence,
40 medium agreement). {7.2, 7.4, 12.4}
41 Diets high in plant protein and low in meat and dairy are associated with lower GHG emissions
42 (robust evidence, high agreement). Ruminant meat shows the highest GHG intensity. Beef from dairy
43 systems has lower emissions intensity than beef from beef herds (8-23 and 17-94 kgCO2-eq (100g
44 protein)-1, respectively) when a share of emissions is allocated to dairy products. The wide variation in
45 emissions reflects differences in production systems, which range from intensive feedlots with stock
Do Not Cite, Quote or Distribute 12-4 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 raised largely on grains through to rangeland and transhumance production systems. Where appropriate,
2 a shift to diets with a higher share of plant protein, moderate intake of animal-source foods and reduced
3 intake of added sugars, salt and saturated fats could lead to substantial decreases in GHG emissions.
4 Benefits would also include reduced land occupation and nutrient losses to the surrounding
5 environment, while at the same time providing health benefits and reducing mortality from diet-related
6 non-communicable diseases. {7.4.5, 12.4}
7 Emerging food technologies such as cellular fermentation, cultured meat, plant-based
8 alternatives to animal-based food products, and controlled environment agriculture, can bring
9 substantial reduction in direct GHG emissions from food production (limited evidence, high
10 agreement). These technologies have lower land, water, and nutrient footprints, and address concerns
11 over animal welfare. Access to low-carbon energy is needed to realize the full mitigation potential, as
12 some emerging technologies are relatively more energy intensive. This also holds for deployment of
13 cold chain and packaging technologies, which can help reduce food loss and waste, but increase energy
14 and materials use in the food system. (limited evidence, high agreement). {11.4.1.3, 12.4}
15 Scenarios that likely to limit warming to 2°C or lower by 2100 commonly involve extensive
16 mitigation in the AFOLU sector that at the same time provides biomass for mitigation in other
17 sectors. Bioenergy is the most land intensive renewable energy option, but the total land
18 occupation of other renewable energy options can become significant in high deployment
19 scenarios (robust evidence, high agreement). Growing demands for food, feed, biomaterials, and non-
20 fossil fuels increase the competition for land and biomass while climate change creates additional
21 stresses on land, exacerbating existing risks to livelihoods, biodiversity, human and ecosystem health,
22 infrastructure, and food systems. Appropriate integration of bioenergy and other biobased systems, and
23 of other mitigation options, with existing land and biomass uses can improve resource use efficiency,
24 mitigate pressures on natural ecosystems and support adaptation through measures to combat land
25 degradation, enhance food security, and improve resilience through maintenance of the productivity of
26 the land resource base (medium evidence, high agreement). {3.2.5, 3.4.6, 12.5}
27 Bio-based products as part of a circular bioeconomy have potential to support adaptation and
28 mitigation. Key to maximizing benefits and managing trade-offs are sectoral integration,
29 transparent governance, and stakeholder involvement (high confidence). A sustainable bioeconomy
30 relying on biomass resources will need to be supported by technology innovation and international
31 cooperation and governance of global trade to disincentivize environmental and social externalities
32 (medium confidence). {12.5, Cross-Working Group Box 3}
33 Coordinated, cross-sectoral approaches to climate change mitigation should be adopted to target
34 synergies and minimize trade-offs between sectors and with respect to sustainable development
35 (robust evidence, high agreement). This requires integrated planning using multiple-objective-multiple-
36 impact policy frameworks. Strong inter-dependencies and cross-sectoral linkages create both
37 opportunities for synergies and the need to address trade-offs related to mitigation options and
38 technologies. This can only be done if coordinated sectoral approaches to climate change mitigations
39 policies that mainstream these interactions are adopted. Integrated planning and cross-sectoral
40 alignment of climate change policies are particularly evident in developing countries’ NDCs pledged
41 under the Paris Agreement, where key priority sectors such as agriculture and energy are closely aligned
42 between the proposed mitigation and adaptation actions in the context of sustainable development and
43 the SDGs. {12.6.2}
44 Carbon leakage is a critical cross-sectoral and cross-country consequence of differentiated
45 climate policy (robust evidence, medium agreement). Carbon leakage occurs when mitigation measures
Do Not Cite, Quote or Distribute 12-5 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 implemented in one country/sector lead to increased emissions in other countries/sectors. Global
2 commodity value chains and associated international transport are important mechanisms of carbon
3 leakage. Reducing emissions from the value chain and transportation can offer opportunities to mitigate
4 three elements of cross-sectoral spill-overs and related leakage: 1) domestic cross-sectoral spill-overs
5 within the same country; 2) international spill-overs within a single sector resulting from substitution
6 of domestic production of carbon-intensive goods with their imports from abroad; and 3) international
7 cross-sectoral spill-overs among sectors in different countries. {12.6.3}
8 Cross-sectoral considerations in mitigation finance are critical for the effectiveness of mitigation
9 action as well as for balancing the often conflicting social, developmental, and environmental
10 policy goals at the sectoral level (medium evidence, medium agreement). True resource mobilisation
11 plans that properly address mitigation costs and benefits at sectoral level cannot be developed in
12 isolation of their cross-sectoral implications. There is an urgent need for multilateral financing
13 institutions to align their frameworks and delivery mechanisms including the use of blended financing
14 to facilitate cross-sectoral solutions as opposed to causing competition for resources among sectors.
15 {12.6.4}
16 Understanding the co-benefits and trade-offs associated with mitigation is key to supporting
17 societies to prioritize among the various sectoral policy options (medium evidence, medium
18 agreement). For example, CDR options can have positive impacts on ecosystem services and the SDGs,
19 but also potential adverse side-effects; transforming food systems has potential co-benefits for several
20 SDGs, but also trade-offs; and land-based mitigation measures may have multiple co-benefits but may
21 also be associated with trade-offs among environmental, social, and economic objectives. Therefore,
22 the possible implementation of the different sectoral mitigation options would depend on how societies
23 prioritise mitigation versus other products and services including food, material wellbeing, nature
24 conservation and biodiversity protection, as well as on other considerations such as society’s future
25 dependence on CDR and on carbon-based energy and materials. {12.3, 12.4, 12.5, 12.6.1}
26 Governance of CDR, food systems and land-based mitigation can support effective and equitable
27 policy implementation (medium evidence, high agreement). Effectively responding to climate change
28 while advancing sustainable development will require coordinated efforts among a diverse set of state-
29 and non-state-actors on global, multi-national, national, and sub-national levels. Governance
30 arrangements in public policy domains that cut through traditional sectors are confronted with specific
31 challenges, such as establishing reliable systems for monitoring, reporting and verification (MRV) that
32 allow evaluation of mitigation outcomes and co-benefits. Effectively integrating CDR into mitigation
33 portfolios can build on already existing rules, procedures and instruments for emissions abatement.
34 Additionally, to accelerate research, development, and demonstration, and to incentivise CDR
35 deployment, a political commitment to formal integration into existing climate policy frameworks is
36 required, including reliable MRV of carbon flows. Food systems governance may be pioneered through
37 local food policy initiatives complemented by national and international initiatives, but governance on
38 the national level tends to be fragmented, and thus have limited capacity to address structural issues like
39 inequities in access. The governance of land-based mitigation, including land-based CDR, can draw on
40 lessons from previous experience with regulating biofuels and forest carbon; however, integrating these
41 insights requires governance that goes beyond project-level approaches and emphasizes integrated land
42 use planning and management within the frame of the SDGs. {7.4 Box 7.2, 7.6, 12.3.3, 12.4, 12.5}
43
44
Do Not Cite, Quote or Distribute 12-6 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 12.1 Introduction
2 12.1.1 Chapter overview
3 The scope of this chapter was motivated by the need for a succinct bottom-up cross-sectoral view of
4 greenhouse gas (GHG) emissions mitigation coupled with the desire to provide systemic perspectives
5 of critical mitigation potentials and options that go beyond individual sectors and cover cross-sectoral
6 topics such as food systems, land systems, and carbon dioxide removal (CDR) methods. Driven by this
7 motivation, Chapter 12 provides a focused thematic assessment of CDR methods and food systems,
8 followed by consideration of land-related impacts of mitigation options (land-based CDR and other
9 mitigation options that occupy land) and other cross-sectoral impacts of mitigation, with emphasis on
10 synergies and trade-offs between mitigation options, and between mitigation and other environmental
11 and socio-economic objectives. The systems focus is unique to AR6 and is of critical policy relevance
12 as it informs coordinated approaches to planning interventions that deliver multiple benefits and
13 minimise trade-offs, and coordinated policy approaches to support such planning, to tap relatively
14 under-explored areas for the strengthening and acceleration of mitigation efforts in the short to medium
15 term, and for dealing with residual emissions in hard-to-transition sectors in the medium to long term.
16 Table 12.1 presents an overview of the cross-sectoral perspectives addressed in Chapter 12, mapping
17 the chapter main themes to the sectoral and global chapters in this report. These mappings reflect the
18 cross-sectoral aspects of mitigation options in the context of sustainable development, sectoral policy
19 interactions, governance, implications in terms of international trade, spill-over effects, and
20 competitiveness, and cross-sectoral financing options for mitigation. While some cross-sector
21 technologies are covered in more detail in sectoral chapters, this chapter covers important cross-sectoral
22 linkages and provides synthesis concerning costs and potentials of mitigation options, and co-benefits
23 and trade-offs that can be associated with deployment of mitigation options. Additionally, Chapter 12
24 covers CDR methods and specific considerations related to land use and food systems, complementing
25 Chapter 7. The literature assessed in the chapter includes both peer-reviewed and grey literature post
26 IPCC AR5 including IPCC SR1.5, IPCC SRCCL and IPCC SROCC. Knowledge gaps are identified
27 and reflected where encountered, as well as in a separate section. Finally, a strong link is maintained
28 with sectoral chapters and the relevant global chapters of this report to ensure consistency.
29
30 12.1.2 Chapter content
31 Chapters 5 to 11 assess outcomes from mitigation measures that are applicable in individual sectors,
32 and potential co-benefits and adverse side effects of these individual measures. Chapter 12 brings
33 together the cross-sectoral aspects of these assessments including synergies and trade-offs as well as
34 the implications of measures that have application in more than one sector and measures whose
35 implementation in one sector impacts implementation in other sectors.
36 Taking stock of the sectoral mitigation assessments, Chapter 12 provides a summary synthesis of
37 sectoral mitigation costs and potentials in the short and long term along with comparison to the top-
38 down IAM assessment literature of Chapter 3 and the national/regional assessment literature of Chapter
39 4.
40 In the context of cross-sectoral synergies and trade-offs, the chapter identifies a number of mitigation
41 measures that have application in more than one sector. Examples include measures involving product
42 and material circularity, which contribute to mitigation of GHG emissions in a number of ways, such
Do Not Cite, Quote or Distribute 12-7 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 as treatment of organic waste to reduce methane emissions, avoid emissions through generation of
2 renewable energy, and reduce emissions through substitution of synthetic fertilisers. Low carbon energy
3 technologies such as solar and wind may be used for grid electricity supply, as embedded generation in
4 the buildings sector (e.g., rooftop solar) and for energy supply in the agriculture sector. Nuclear and
5 bio-based thermal electric generation can provide multiple synergies including base load to augment
6 solar and wind, district heating, and seawater desalination. Grid-integrated hydrogen systems can buffer
7 variability of solar and wind power and is being explored as a mitigation option in the transport and
8 industry sectors. Carbon Capture and Storage (CCS) has potential application in a number of industrial
9 processes (cement, iron and steel, petroleum refining and pulp and paper) and the fossil fuel electricity
10 sector. When coupled with energy recovery from biomass (BECCS), CCS can help to provide CO2
11 removal from the atmosphere. On the demand side, electric vehicles are also considered an option for
12 balancing variable power, energy efficiency options find application across the sectors, as does reducing
13 demand for goods and services, and improving material use efficiency. Focused inquiry into these areas
14 of cross-sectoral perspectives is provided for CDR, food systems, and land-based mitigation options.
15 A range of examples of where mitigation measures result in cross-sectoral interactions and integration
16 is identified. The mitigation potential of electric vehicles, including plug-in hybrids, is linked to the
17 extent of decarbonisation of the electricity grid, as well as to the liquid fuel supply emissions profile.
18 Making buildings energy positive, where excess energy is used to charge vehicles, can increase the
19 potential of electric and hybrid vehicles. Advanced process control and process optimisation in industry
20 can reduce energy demand and material inputs, which in turn can reduce emissions linked to resource
21 extraction and manufacturing. Trees and green roofs planted to counter urban heat islands reduce the
22 demand for energy for air conditioning and simultaneously sequester carbon. Material and product
23 circularity contributes to mitigation, such as treatment of organic waste to reduce methane emissions,
24 generate renewable energy, and to substitute for synthetic fertilisers.
25 The chapter also discusses cross-sectoral mitigation potential related to diffusion of General-Purpose
26 Technologies (GPT), such as electrification, digitalisation, and hydrogen. Examples include the use of
27 hydrogen as an energy carrier, which, when coupled with low carbon energy, has potential for driving
28 mitigation in energy, industry, transport, and buildings (Box 12.5), and digitalisation has the potential
29 for reducing GHG emissions through energy savings across multiple sectors.
30 The efficient realisation of the above examples of cross-sectoral mitigation would require careful design
31 of government interventions across planning, policy, finance, governance, and capacity building fronts.
32 In this respect, Chapter 12 assesses literature on cross-sectoral integrated policies, cross-sectoral
33 financing solutions, cross-sectoral spill-overs and competitiveness effects, and on cross-sectoral
34 governance for climate change mitigation.
35 Finally, in the context of cross-sectoral synergies and trade-offs, the chapter assesses the non-climate
36 mitigation co-benefits and adverse effects in relation to SDGs, building on the fast-growing literature
37 on the non-climate impacts of mitigation.
38
39 12.1.3 Chapter layout
40 The chapter is mapped into seven sections. Cost and potentials of mitigation technologies are discussed
41 in Section 12.2, where a comparative assessment and a summary of sectoral mitigation cost and
42 potentials is provided in coordination with the sectoral Chapters 5 to 11, along with a comparison to
43 aggregate cost and potentials based on IAM outputs presented in Chapter 3.
Do Not Cite, Quote or Distribute 12-8 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Section 12.3 provides a synthesis of the state and potential contribution of CDR methods for addressing
2 climate change. CDR options associated with the AFOLU and Energy sectors are dealt with in Chapters
3 6 and 7 and synthesised in Section 12.3. Other methods, not dealt with elsewhere, are covered in more
4 detail. A comparative assessment is provided for the different CDR options in terms of costs, potentials,
5 governance, impacts and risks, and synergies and trade-offs.
6 Section 12.4 assesses the literature on food systems and GHG emissions. The term ‘food system’ refers
7 to a composite of elements (environment, people, inputs, processes, infrastructures, institutions, etc.)
8 and activities that relate to the production, processing, distribution, preparation and consumption of
9 food, and the outputs of these activities, including socio-economic and environmental outcomes.
10 Climate change mitigation opportunities and related implications for sustainable development and
11 adaptation are assessed, including those arising from food production, landscape impacts, supply chain
12 and distribution, and diet shifts.
13 Section 12.5 provides a cross-sectoral perspective on land occupation and related impacts, risks and
14 opportunities associated with land-based mitigation options as well as mitigation options that are not
15 designated land-based, yet occupy land. It builds on SRCCL and Chapter 7 in this report, which covers
16 mitigation in agriculture, forestry and other land use (AFOLU), including biomass production for
17 mitigation in other sectors. In addition to an assessment of biophysical and socioeconomic risks, impacts
18 and opportunities, this section includes a cross-chapter box (WGII and WGIII) on Mitigation and
19 Adaptation via the Bioeconomy, and a box on Land Degradation Neutrality as a framework to manage
20 trade-offs in land-based mitigation.
21 Section 12.6 provides a cross-sectoral perspective on mitigation, co-benefits, and trade-offs, including
22 those related to sustainable development and adaptation. The synthesised sectoral mitigation synergies
23 and trade-offs are mapped into options/technologies, policies, international trade, and finance domains.
24 Cross-sectoral mitigation technologies fall into three categories in which the implementation of the
25 technology: (i) occurs in parallel in more than one sector; (ii) could involve interaction between sectors,
26 and/or (iii) could create resource competition among sectors. Policies that have direct sectoral effects
27 include specific policies for reducing GHG emissions and non-climate policies that yield GHG
28 emissions reductions as co-benefits. Policies may also have indirect cross-sectoral effects, including
29 synergies and trade-offs that may, in addition, spill over to other countries.
30 Section 12.7 provides an overview of knowledge gaps, which could be used to inform further research.
Do Not Cite, Quote or Distribute 12-9 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Table 12.1 An overview of cross-sector perspectives addressed in Chapter 12
Sectoral chapters Global chapters
Chapter Chapter 5 Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter Chapter Chapter Chapter Chapter Chapter Chapter
12 Themes 10 11 13 14 15 16 17
Costs & Change in Renewable Land-use Urban Standards Hybridisati Technolog Enabling Finance Synergies
Potentials demand s CCU Change planning on Electric y Biomass of of and trade-
Electrificat vehicles mitigation mitigatio offs to
CCS Cities ion CCU n SDGs
Fuel
Nuclear Demograph economy CCS
ics
Decouplin
g
CDR BECCS Land- C storage Internatio
based CDR in nal
buildings Governan
ce
Food Food Energy Agricultur Urban food Food Food Food Governan Food
Systems demand demand of al systems; transport processing system ce system
some production controlled & transforma and SDGs
Wellbeing emerging environmen packaging tion
mitigation Demand t agriculture
options side
measures
Do Not Cite, Quote or Distribute 12-10 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Mitigation Land use/ A/R, Land use Land use Land use Governan Co-
& land occupation Biomass and and and ce benefits
use : production, biomass biomass biomass and
bioenergy, supply supply supply adverse
hydro, Bioenergy, side
solar, effects
wind, Biochar
nuclear
Cross- Policy Governan Blended General SDGs Co-
sectoral interaction ce financing Purpose benefits
perspectiv s Technologi
es Leakage es Trade-offs
Electrification, Hydrogen, Digitalisation , Circularity, Synergies, Trade-offs, Spill-overs Policy Electrificat
packages ion Adaptatio
Hydrogen n
Case
studies
Value
chain &
carbon
leakage
1
Do Not Cite, Quote or Distribute 12-11 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 12.2 Aggregation of sectoral costs and potentials
2 The aim of this section is to provide a consolidated overview of the net emissions reduction potentials
3 and costs for mitigation options available in the various sectors dealt with in the sectoral chapters 6, 7,
4 9, 10 and 11 of this assessment report. This overview provides policy-makers with an understanding of
5 which options are more or less important in terms of mitigating emissions in the short term (here
6 interpreted as 2030), and which ones are more or less costly. The intention is not to provide a high level
7 of accuracy for each technology cost or potential, but rather to indicate relative importance on a global
8 scale and whether costs are low, intermediate or high. The section starts with an introduction (Section
9 12.2.1), providing definitions and the background. Next, ranges of net emission reduction potentials
10 and the associated costs for the year 2030 are presented (Section 12.2.2) and compared to earlier
11 estimates and with the outputs of integrated assessment models (IAMs) (Section 12.2.3). Finally, an
12 outlook to the year 2050 is provided (Section 12.2.4).
13 12.2.1 Introduction
14 The term ‘mitigation potential’ is used here to report the quantity of net greenhouse gas emissions
15 reductions that can be achieved by a given mitigation option relative to specified reference scenario.
16 The net greenhouse gas emission reduction is the sum of reduced emissions and enhanced sinks. Several
17 types of potential can be distinguished. The technical potential is the mitigation potential constrained
18 by theoretical limits in addition to the availability of technology and practices. Quantification of
19 technical potentials primarily takes into account technical considerations, but social, economic and/or
20 environmental considerations are sometimes also considered, if these represent strong barriers for the
21 deployment of an option. The economic potential, being the potential reported in this section, is the
22 proportion of the technical potential for which the social benefits exceed the social costs, taking into
23 account a social discount rate and the value of externalities (see glossary). In this section, only
24 externalities related to greenhouse gas emissions are taken into account. They are represented by using
25 different cost cut-off levels of options in terms of USD per tonne of avoided CO2-eq emissions. Other
26 potentials, such as market potentials, could also be considered, but they are not included in this section.
27 The analysis presented here is based, as far as possible, on information contained in Chapters 6, 7, 9,
28 10 and 11, where costs and potentials, referred to here as ‘sectoral mitigation potentials’ have been
29 discussed for each individual sector. In the past, these were designated as bottom-up potentials, in
30 contrast to the top-down potentials that are obtained from integrated energy-economic models and
31 IAMs. However, IAMs increasingly include ‘bottom-up’ elements, which makes the distinction less
32 clear. Still, sectoral studies often have more technical and economic detail than IAMs. They may also
33 provide more up-to-date information on technology options and associated costs. However, aggregation
34 of results from sectoral studies is more complex, and although interactions and overlap are corrected
35 for as far as possible in this analysis, it is recognised that such systemic effects are much more rigorously
36 taken into account in IAMs. A comparison is made between the sectoral results and the outcomes of the
37 IAMs in Section 12.2.3.
38 Costs of mitigation options will change over time. For many technologies, costs will reduce as a result
39 of technological learning. An attempt has been made to take into account the average, implementation-
40 weighted costs until 2030. However, the underlying literature did not always allow such costs to be
41 presented. For the year 2030, the results are presented similarly to AR4, with a breakdown of the
42 potential in “cost bins”. For the year 2050, a more qualitative approach is provided. The origins of the
43 cost data in this section mostly are based on studies carried out in the period 2015-2020. Given the wide
44 range of the cost bins that are used in this section it is not meaningful (and often not possible) to convert
Do Not Cite, Quote or Distribute 12-12 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 to USD-values for one specific year. This may lead to some extra uncertainty, but this is expected to be
2 relatively small.
3 As indicated previously, net emission reduction potentials are presented based on comparison with a
4 reference scenario. Unfortunately, not all costs and potentials found in the literature are determined
5 against the same reference scenarios. In this assessment reference scenarios are based on what were
6 assumed current-policy scenarios in the period 2015-2019. Typical reference scenarios are the SSP2
7 scenarios (Fricko et al. 2017) and the Current Policies scenario from the World Energy Outlook 2019
8 (IEA 2019). They can both be considered scenarios with middle-of-the-road expectations on population
9 growth and economic development, but there are still some differences between the two (Table 12.2).
10 The net emission reduction potentials reported here were generally based on analyses carried out before
11 2020, so the impact of the COVID-19 pandemic was not taken into account. For comparison, the Stated
12 Policies scenario of the World Energy Outlook 2020 (IEA 2020a) is also shown, one of the scenarios
13 in which the impact of COVID-19 was considered. Variations of up to 10% between the different
14 reference scenarios exist with respect to macro-variables such as total primary energy use and total
15 GHG emissions. The potential estimates presented below should be interpreted against this background.
16 The total emissions under the reference scenarios in 2030 are expected to be in the range of 54 to 68
17 GtCO2-eq yr-1 with a median of 60 GtCO2-eq yr-1 (Chapter 4, Table 4.1).
18 For the energy sector the potentials are determined using the World Energy Outlook 2019 Current
19 Policies Scenario as a reference (IEA 2019). However, for the economic assessment more recent LCOEs
20 for different electricity generating technologies were used (IEA 2020a). For the agriculture, forestry
21 and other land use (AFOLU) sector, the potentials were derived from a variety of studies. It may be
22 expected that the best estimates, as averages, match with the reference in a middle-of-the-road scenario.
23 For the buildings sector the Current Policies scenario of World Energy Outlook 2019 (IEA 2019) was
24 used as a reference. For the transport sector, the references of the underlying sources were used. For the
25 industry sector, the scenarios used have emissions that are slightly higher than in the Current Policies
26 scenario from the World Energy Outlook 2019 (IEA 2019).
Do Not Cite, Quote or Distribute 12-13 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Table 12.2: Key characteristics of the scenarios that are used as a reference for determining costs and
2 potentials. The values are for the year 2030.
SSP2 All reference WEO- WEO- AR6
reference scenarios 2019 2020 Chapter 4
(MESSAGE median (Current (Stated (Chapter
- (25/75 per- Policies) Policies) 4, Table
GLOBIOM) centiles in 4.1)
parenthesis) (IEA (IEA
(Fricko et al. 2019) 2020a)
2017) (AR6
scenarios
database )
Real GDP (PPP) (1012 USD) 158 159 3.6% p.a.↑ 2.9% p.a.↑
(USD2010) (154–171) (2018 to (2019 to
2030) 2030)
Population (billion) 8.30 8.30 8.60
(8.20–8.34)
Total primary energy use (EJ) 627 670 710 660
(635–718)
Total final energy use (EJ) 499 480 502 472
(457–508)
Energy-related CO2 emissions 33.0 37.9 37.4 33.2* 37
(Gt) (34.7–41.4) (35–45)
CO2 emissions energy and industry 37.9 42.3 36.0
(Gt)
(39.0–45.8)
Total CO2 (emissions Gt) 40.6 45.7 43
(41.8–49.4) (38–51)
Total greenhouse gas emissions 52.7 59.7 60
(GtCO2-eq)
(55.0–65.8) (54-68)
3 *The difference between WEO-2020 and WEO-2019 is partly explained by the fact that WEO-2019 had two
4 different reference scenarios: Current Policies and Stated Policies. WEO-2020 has only one reference: the Stated
5 Policies Scenario (STEPS), which “is based on today’s policy settings”. The Stated Policy scenario in WEO-
6 2019 had energy-related emissions of 34.9 GtCO2.
7
Do Not Cite, Quote or Distribute 12-14 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 12.2.2 Costs and potentials of options for 2030
2 In this section, we present an overview of mitigation options per sector. An overview of net emission
3 reduction potentials for different mitigation options is presented in Table 12.3.
4 Firstly, a brief overview of the process of data collection is presented, with a more detailed overview
5 being found in Supplementary Material SM 12.A.2. For the energy sector, the starting point for the
6 determination of the emission reduction potentials was the Emissions Gap Report (UNEP 2017), but
7 new literature was also assessed, and a few studies that provide updated estimates of the mitigation
8 potentials were included. It was found that higher mitigation potentials than in the UNEP report are
9 now reported for solar and wind energy, but at the same time electricity production by solar and wind
10 energy in the reference scenario has increased, compared to earlier versions of the World Energy
11 Outlook. The net effect is a modest increase of the average value of the potential, and a wider uncertainty
12 range. Costs of electricity generating technologies are discussed in Chapter 6, (Section 6.4.7) with a
13 summary of LCOEs from the literature being presented in Section 6.4.7. Mitigation costs of electricity
14 production technology depend on local conditions and on the baseline technology being displaced, and
15 it is difficult to determine the distribution over the cost ranges used in this assessment. However, it is
16 possible to indicate a broad cost range for these technologies. These cost ranges are presented in Table
17 12.3. For onshore wind and utility scale solar energy, there is strong evidence that despite regional
18 difference in resource potential and cost, a large part of the mitigation potential can be found in the
19 negative cost category or at cost parity with fossil fuel based options. This is also the case for nuclear
20 energy in some regions. Other technologies show mostly positive mitigation potentials, the highest
21 mitigation costs are for CCS, bioelectricity with CCS, for details see Supplementary Material SM
22 12.A.2.
23 For the AFOLU sector, assessments of global net emission reduction studies were provided by Chapter
24 7 (Table 7.3). The number of studies depends on the type of mitigation action, but ranges from 5 to 9.
25 Each of these studies relies on a much larger number of underlying data sources. From these studies,
26 emission reduction ranges and best estimates were derived. The studies presented refer to different years
27 in the period 2020 to 2050, and the mitigation potential presented for AFOLU primarily refers to the
28 average over the period 2020 to 2050. However, because most of the activities involve storage of carbon
29 in stocks that accumulate carbon, or conversely decay over time (e.g., forests, mangroves, peatland
30 soils, agricultural soils, wood products), the 2020 to 2050 average provides a good approximation of
31 the amount of permanent atmospheric CO2 mitigation that could be available at a given price in 2030.
32 The exception is BECCS which is in an early upscaling phase, so the potential estimated by Chapter 7
33 as an average for the 2020 to 2050 period is not included in Table 12.3. Note that for the energy sector
34 a mitigation potential for BECCS is provided in Table 12.3.
35 The emission reduction potentials for the buildings sector were based on the analysis by Chapter 9
36 authors of a large number of sectoral studies for individual countries or regions. In total, the chapter
37 analysed the results of 67 studies that assess the potential of technological energy efficiency and onsite
38 renewable energy production and use, and the results of 11 studies that assess the potential of sufficiency
39 measures helping avoid demand for energy and materials. The sufficiency measures were included in
40 models by reorganization of human activities, efficient design, planning, and use of building space,
41 higher density of building and settlement inhabitancy, redefining and downsizing goods and equipment,
42 limiting their use to health, living, and working standards, and their sharing. Most of these studies
43 targeted 2050 for the decarbonisation of buildings; the potentials in 2030 reported here rely on the
44 estimates for 2030 provided by these studies or on the interpolated estimates targeting these 2050
45 figures. Based on these individual country studies, regional aggregate emission reduction percentages
46 were found. The potential estimates were assembled in the order sufficiency, efficiency, renewable
Do Not Cite, Quote or Distribute 12-15 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 options, correcting the amount of the potential at each step for the interaction with preceding measures.
2 Note that the option ‘Enhanced use of wood products’ was analysed by Chapter 7, but is listed under
3 the buildings sector in Table 12.3, as such enhanced use of wood takes place predominantly in the
4 construction sector.
5 For the transport sector, Chapter 10 provided data on the emission reduction potential for shipping. For
6 the other transportation modes, additional sources were used to achieve a complete overview of
7 emission reduction potentials (for further details, see Supplementary Material 12.A.2). A limited
8 number of estimates for global emission reduction potential is available: the total number of sources is
9 about 10, and some estimates rely on just one source. The data have been coordinated with Chapter 10
10 authors.
11 For the industrial sector, global emission reduction potentials per technology class per sector were
12 derived by Chapter 11 authors, using primarily sectoral or technology-oriented literature. The analysis
13 is based on about 75 studies, including sectoral assessments (Sections 11.4.1, 11.4.2, Figure 11.13).
14 For methane emission reduction from oil and gas operations, coal mining, waste treatment and
15 wastewater, an analysis was done, based on three major data sources in this area (US EPA 2019;
16 Harmsen et al. 2019; Höglund-Isaksson et al. 2020), and for oil and gas operations complemented by
17 (IEA 2021a). A similar analysis for reductions of emissions of fluorinated gases was carried out based
18 on analysis by the same institutes (Purohit and Höglund-Isaksson 2017; US EPA 2019; Harmsen et al.
19 2019). Data for CDR options not discussed previously (such as DACCS and enhanced weathering)
20 were taken from Section 12.3. For more details about data sources and data processing, see
21 Supplementary Material 12.A, Section SM 12.A.2.
22 In Table 12.4 mitigation potentials for all gases are presented in GtCO2-eq. For most sectors the
23 mitigation potentials (notably for methane emissions reductions from coal, oil and gas, waste and
24 wastewater) have been converted to CO2-eq using global warming potentials (GWP) values as presented
25 in the 6th Assessment Report (Cross-Chapter Box 2 in Chapter 2). However, the underlying literature
26 did not always accommodate this, in which cases older GWP values apply. Given the uncertainty ranges
27 in the mitigation potentials in Table 12.3, the impact on the results of using different GWP values is
28 considered to be very small.
Do Not Cite, Quote or Distribute 12-16 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Table 12.3: Detailed overview of global net GHG emission reduction potentials (GtCO2-eq) in the various cost categories for the year 2030. Note that
potentials within and across sectors cannot be summed, as the adoption of some options may affect the mitigation potentials of other options. Only
monetary costs and benefits of options are taken into account. Negative costs occur when the benefits are higher than the costs. For wind energy, for
example, this is the case if production costs are lower than those of the fossil alternatives. Ranges are indicated for each option separately, or
indicated for the sector as a whole (see column “Notes”); they reflect full ranges. Cost ranges are not cumulative, e.g., to obtain the full potential
below 50 USD tCO2-eq-1, the potentials in the cost bins <0, 0–20 and 20–50 USD tCO2-eq-1 need to be summed together.
Emission reduction options (including carbon Cost categories (USD tCO2-eq-1) Notes
sequestration options)
<0 0–20 20–50 50–100 100–200
Energy sector Cost ranges are derived as ranges of LCOEs for
different electricity generating technologies and the
potentials are updated from UNEP (2017).
Wind energy 2.1–5.6 Costs for system integration of intermittent renewables
are not included, but these are expected to have limited
(majority in <0 range) impact until 2030 and will depend on market design
and cross-sectoral integration
Solar energy 2.0–7.0 Ibid.
(majority in <0 range)
Nuclear energy 0.88 ±50%
Bioelectricity 0.86 ±50% Biomass use for indoor heating and industrial heat is
not included here. Currently, about 90% of renewable
industrial heat consumption is biobased, mainly in
industries that can use their own biomass waste and
residues (IEA, 2020)
Hydropower 0.32 ±50% Mitigation costs show large variation and may end up
beyond these ranges.
Do Not Cite, Quote or Distribute 12-17 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Geothermal energy 0.74 ±50% Mitigation costs show large variation and may end up
beyond these ranges.
Carbon capture and storage (CCS) 0.54 ±50%
Bioelectricity with CCS 0.30 ±50%
CH4 emission reduction from coal mining 0.04 0.41 0.03 0.02
(0.01– (0.15– (0.02– (0.01–
0.06) 0.64) 0.05) 0.03)
CH4 emission reduction from oil and gas 0.31 0.61 0.07 0.06 0.10
operations
(0.12– (0.23– (0.03– (0.00– (0–0.29)
0.56) 1.30) 0.20) 0.29)
Land-based mitigation options (including Potentials for AFOLU are averages for the period
agriculture and forestry) 2020–2050, and represent a proxy for mitigation in
2030.
Technical potentials listed below include the potentials
already listed in the previous columns.
Note that in Table 7.3 the same potentials are listed,
but they are cumulative over the cost bins.
Carbon sequestration in agriculture (soil carbon 0.50 0.73 2.21 Technical potential: 9.5 (range 1.1–25.3)
sequestration, agroforestry and biochar
application) (0.38– (0.5–1.0) (0.6–3.9)
0.60)
CH4 and N2O emission reduction in agriculture 0.35 - 0.28 Technical potential: 1.7 (range 0.5–3.2)
(reduced enteric fermentation, improved
Do Not Cite, Quote or Distribute 12-18 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
manure management, nutrient management, (0.11– (0.19– GWPs used from AR4 and AR5
rice cultivation) 0.84) 0.46)
Protection of natural ecosystems (avoid 2.28 0.12 1.63 0.22 Technical potential 6.2 (range 2.8–14.4)
deforestation, loss and degradation of peatlands,
coastal wetlands and grasslands) (1.7–2.9) (0.06– (1.3–4.2) (0.09–
0.18) 045
Restoration (afforestation, reforestation, 0.15 0.57 1.46 0.66 Technical potential 5.0 (range 1.1–12.3)
peatland restoration, coastal wetland
restoration) (0.2–1.5) (0.6–2.3) (0.4–1.1)
Improved forest management, fire management 0.38 - 0.78 Technical potential 1.8 (range 1.1–2.8)
(0.32– (0.32–
0.44) 1.44)
Reduce food loss and food waste Feasible potential 0.5 (0.1–0.9)
Technical potential 0.7 (0.1–1.6)
Estimates reflect direct mitigation from diverted
agricultural production only, not including land-use
effects
Shift to sustainable healthy diets Feasible potential 1.7 (1.0–2.7)
Technical potential 3.5 (2.1–5.5)
Estimates reflect direct mitigation from diverted
agricultural production only, not including land-use
effects
Do Not Cite, Quote or Distribute 12-19 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Buildings The numbers were corrected for the potential overlap
between options in the order “sufficiency, efficiency,
renewable measures” and they could be therefore
added up. In 2050, much larger and cheaper potential
is available (see 9.6 in Chapter 9); the potential in
2030 is lower and more expensive mostly due to
various feasibility constrains.
Sufficiency to avoid demand for energy 0.56
services (e.g., efficient building use and
increased inhabitancy and density) (0.28–
0.84)
Efficient lighting, appliances and equipment, 0.73
including ICT, water heating and cooking
technologies (0.54–
0.91)
New buildings with very high energy
performance (change in construction methods,
management and operation of buildings, 0.35 0.83
efficient heating, ventilation and air
conditioning) (0.26–0.53) (0.62–
1.24)
Onsite renewable production and use (often
backed-up with demand-side flexibility and
digitalization measures, typically installed in 0.20 0.27
very new high energy performance buildings)
(0.15–0.30) (0.20–
0.40)
Do Not Cite, Quote or Distribute 12-20 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Improvement of existing building stock Additionally, there is 0.50 (range 0.37–0.62) GtCO2-eq
(thermal efficiency of building envelopes, of potential above a price of 200 USD tCO2-eq-1
management and operation of buildings, and 0.27
efficient heating, ventilation and air
conditioning leading to “deep” energy savings) (0.20–0.34)
Enhanced use of wood products Technical potential 1.0 (range 0.04–3.7)
Economic potential 0.38 (range 0.3–0.5) (varying
carbon prices). Potential is mainly in the construction
sector.
Transport Options for the transportation sector have an
uncertainty of ±50%.
Light duty vehicles – fuel efficiency 0.6
Light duty vehicles – electric vehicles 0.5–0.7 Depending on the carbon intensity of the electricity
supplied to the vehicles.
Light duty vehicles – shift to public transport 0.5
Light duty vehicles – shift to bikes and e-bikes 0.2
Heavy duty vehicles – fuel efficiency 0.4
Heavy duty vehicles – electric vehicles 0.2
Heavy duty vehicles – shift to rail No data available.
Shipping – efficiency, optimisation, biofuels 0.5
(0.4–0.7)
Do Not Cite, Quote or Distribute 12-21 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Aviation – energy efficiency 0.12– Limited evidence
0.32
Biofuels 0.6–0.8
Industry The numbers for the industry sector typically have an
uncertainty of ±25%.
The numbers are corrected for overlap between the
options, except for the 0.15 GtCO2 potential in the
highest cost bin. For the rest they can be aggregated to
provide full potentials.
Energy efficiency 1.14 This only applies to more efficient use of fuels. More
efficient use of electricity is not included.
Material efficiency 0.93
Circularity (enhanced recycling) 0.48
Fuel switching 1.28 0.67 0.15
Feedstock decarbonisation, process change 0.38
Carbon capture, utilization and storage (CCU 0.15
and CCS)
Cementitious material substitution 0.28
Reduction of non-CO2 emissions 0.2
Do Not Cite, Quote or Distribute 12-22 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Cross-sectorial
Emission reduction of fluorinated gases 0.26 0.68 0.18 0.09 0.03 GWPs not updated
(0.01– (0.55– (0.01– (0–0.20) (0–0.05)
0.50) 0.90) 0.42)
Reduction of CH4 emissions from solid waste 0.33 0.11 0.06 0.04 0.08
(0.24– (0.03– (0.03– (0.01– (0.02–
0.43) 0.15) 0.08) 0.10) 0.12)
Reduction of CH4 emissions from wastewater 0.02 0.03 0.04 0.03 0.07
(0–0.05) (0.01– (0.01– (0.02– (0.01–
0.05) 0.07) 0.04) 0.16)
Direct air carbon capture and storage very There is potential in these categories, but given the
small current technology readiness levels, for 2030 the
potential is limited. Also, it is not certain whether the
Enhanced weathering very costs will already drop below 200 USD tCO2-1 before
small 2030. In the longer term, much larger potentials are
projected, see Section 12.3.1.
Do Not Cite, Quote or Distribute 12-23 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 For all options, uncertainty ranges of the mitigation potentials are given in Table 12.3. As far as possible,
2 the ranges represent the variation in assessments found in the literature. This is the case for wind and
3 solar energy, for the AFOLU options, for the methane mitigation options (coal, oil and gas, waste and
4 wastewater) and for fluorinated gas mitigation. For the latter options, some variability exists for each
5 cost bin, but aggregated over cost ranges the variation is much smaller, typically ±50%. For the
6 buildings sector and the industrial sector options, the uncertainty in the mitigation potential is estimated
7 by the lead authors of Chapters. For options for which only limited sources were available an uncertainty
8 range of ±50% was used. Overall, the uncertainty range per option is typically in the range of ±20% to
9 ±60%.
10 Despite these uncertainties, clearly a number of options with high potentials can be identified, including
11 solar energy, wind energy, reducing conversion of forests and other natural ecosystems, and restoration
12 of forests and other natural ecosystems. As mid-range values, they each represent 4 to 7% of total
13 reference emissions for 2030. Soil carbon sequestration in agriculture and fuel switching in industry
14 can also be considered as options with high potential, although it should be noted that these options
15 consist of a number of discernible sub-options, see Table 12.3. It can be observed that for each sector,
16 a variety of options is available. Many of the smaller options each make up 1 to 2% of the reference
17 emissions for 2030. Within this group of smaller options there are some categories that, summed
18 together, stand out as substantial: the energy efficiency options and the methane mitigations options.
19 Costs are highly variable across the options. All sectors have several options for which at least part of
20 the potential has mitigation costs below 20 USD tCO2-1. The only exception is the industrial sector, in
21 which only energy efficiency is available below this cost level. At the same time, a substantial part of
22 the emission reduction potential comes at higher cost, much being in the 20 to 100 USD tCO2-1 cost
23 ranges. All sectors have substantial additional potential in these cost ranges; only for transportation is
24 this limited. Aggregation of the potentials per cost bin shows that the potential in these cost bins is
25 marginally smaller than in the two cheapest cost bins. For some options, potential was identified in the
26 100 to 200 tCO2-1 cost bin. The mitigation potentials identified in this cost range make up only a small
27 part of the total mitigation potential. It could be that there is limited potential in this range; however, a
28 more plausible explanation, supported by several authors of sectoral chapters, is that this cost range is
29 relatively unexplored.
30 In this assessment, the emphasis is on the specific mitigation costs of the various options, and these are
31 often considered as an indicator to prioritise options. However, in such a prioritisation, other elements
32 will also play a role, like the development of technology for the longer term (Section 12.2.4) and the
33 need to optimise investments over longer time periods, see for example Vogt-Schilb et al. (2018) who
34 argue that sometimes it makes sense to start with implementing the most expensive option.
35 In this section, an overview of emission mitigation options for the year 2030 was presented. The
36 overview of the mitigation potential is based on a variety of approaches, relying on a large number of
37 sources, and the number of sources varied strongly from sector to sector. The main conclusions from
38 this section are: i) there is a variety of options per sector, ii) per sector the options combined show
39 significant mitigation potential, iii) there are a few major options and a lot of smaller ones, and iv) more
40 than half of the potential comes at costs below 20 USD tCO2-1 (between sectors: medium to robust
41 evidence, high agreement).
42
Do Not Cite, Quote or Distribute 12-24 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 12.2.3 Aggregation of sectoral results and comparison with earlier analyses and
2 Integrated Assessment Models
3 In this section, the mitigation potentials are aggregated per sector, and then to the global economy.
4 These potentials, which are based on sectoral analysis, are then compared to the results from earlier
5 assessments and the results from Integrated Assessment Models (IAMs). Given the incompleteness of
6 data on the mitigation potential at mitigation costs larger than 100 USD tCO2-1, the focus will be on
7 options with mitigation costs below 100 USD tCO2-1.
8 As suggested previously, the overview presented in Table 12.3 should be interpreted with care, as the
9 implementation of one option may affect the mitigation potential of another option. Most sectoral
10 chapters have supplied mitigation potentials that were already adjusted for overlap and mutual
11 influences (industry, buildings, AFOLU). For the energy sector, interactions between the options will
12 occur, but parallel implementation of all the options seems to be possible; if all options at costs levels
13 below 100 USD tCO2-1 would be implemented, this would lead to an additional power generation with
14 no direct CO2 emissions of 41% of the total projected generation in 2030. This seems to be possible,
15 but as higher penetrations are relatively unexplored, we apply a smaller uncertainty range at the high
16 end. For the calculation of the aggregate potentials in the energy sector, error propagation rules were
17 applied. For the transport sector, there will be interaction between the technical measures on the one
18 hand and the modal shift measures on the other hand. Given the small mitigation contribution of the
19 modal shift options, these interactions will be negligible. The resulting aggregate mitigation potentials
20 and their uncertainty ranges per (sub)sector are given in Table 12.4 (columns indicated as ‘AR6’). This
21 overview confirms the large potentials per sector, even when taking the uncertainty ranges into account.
22 Calculating aggregated mitigation potentials for the global economy requires that interactions between
23 sectors also need to be taken into account (Section 12.6). First of all, there may be overlap between the
24 electricity supply sector and the electricity demand sectors: if the electricity sector is extensively
25 decarbonized, the avoided emissions due to electricity efficiency measures and local electricity
26 production will be significantly reduced. Therefore, this demand-side mitigation potential is only taken
27 into account for 25% (reflecting the degree of further decarbonisation of the power sector) in the cross-
28 sectoral aggregation. For the other demand sectors, this problem does not arise. The industry sector did
29 not provide estimates for electricity efficiency improvement and in the transport sector the utilization
30 of electricity to date is very low. Electrification options may occur in all sectors, but this enhances the
31 mitigation potential in combination with a decreased carbon intensity of the power sector. For other
32 energy sector options, methane emission reduction from coal, oil and natural gas operations, the
33 situation is more complex. The total emission reduction potential for fossil fuels in the other sectors is
34 high. Should this potential be realised, this would lead to a reduction of the potential reported here.
35 However, reducing fossil fuel use also leads to a reduction in the upstream CH4 emissions, so in the
36 case of reducing fossil fuel use, these upstream emissions will also be avoided, so no overestimate of
37 the aggregate emission reduction potential occurs.
38 The total potential, given these corrections for overlap, leads to a mid-range value for the total
39 mitigation potential at costs below 100 USD tCO2-eq-1 of 38 GtCO2-eq. Given the fact that it is not to
40 be expected that mitigation potentials of the various sectors are mutually correlated, i.e. it is not to be
41 expected that mitigation potentials are all on the high side or all on the low side), the ranges are
42 aggregated using error propagation rules, which leads to a range for the mitigation potential of 32 to 44
43 GtCO2-eq.
Do Not Cite, Quote or Distribute 12-25 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 Table 12.4 Overview of aggregate sectoral net GHG emission reduction potentials (GtCO2-eq)
2 for the year 2030 at costs below 100 USD tCO2-eq-1. Comparisons with earlier assessments are
3 also provided. Note that sectors are not entirely comparable across the three different estimates.
Sector Mitigation potentials at costs less than 100 USD tCO 2-eq-1
AR6 AR6 AR4 UNEP- UNEP- 2017
2017
best range (Barker et al. range
estimate 2007) best
estimate (UNEP 2017)
(UNEP
2017)
Electricity sector 11.0 7.9–12.5 10.3 9.5–11.0
6.2–9.3
Other energy sector 1.6 1.1–2.1 2.2 1.7–2.6
(methane)
Agriculture 4.1 1.7–6.7 2.3–6.4 4.8 3.6–6.0
Forestry and other land- 7.3 3.9–13.1 1.3–4.2 5.3 4.1–6.5
use related options
AFOLU demand-side 2.2 1.1–3.6 1.3–3.4
options (estimates
reflect direct mitigation
from diverted
agricultural
production only, not
including land-use
effects)
Buildings (potentials up Dir 0.7 0.5–1.0 Dir 2.3–2.9 Dir 1.9 Dir 1.6–2.1
to 200 USD tCO2-eq-1
in parentheses) (1.1) (0.7–1.5)
Ind 1.3 0.9–1.8 Ind 3.0–3.8 Ind 4.0
(2.1) (1.5–3.1)
Tot 2.0
(3.2) 1.4–2.9 Tot 5.4–6.7 Tot 5.9
(2.3–4.6)
Do Not Cite, Quote or Distribute 12-26 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
Transport 3.8 1.9–5.7 1.6–2.5 4.7 4.1–5.3
Industry Dir 5.4 4.0–6.7 Dir 2.3–4.9 Dir 3.9 Dir 3.0–4.8
Ind 0.83 Ind 1.9
Tot 3.1–5.7 Tot 5.8
Fluorinated gases (all 1.2 0.7–1.5 NE 1.5 1.2–1.8
sectors)
Waste and wastewater 0.7 0.6–0.8 0.4–1.0 0.4 0.3–0.5
Enhanced weathering - - - 1.0 0.7–1.2
Total of all sectors 38 32–44 15.8–31.1 38 35–41
1 Dir = reduction of direct emissions, Ind = reduction of indirect emissions (related to electricity production), Tot
2 = reduction of total emissions, NA = not applicable, NE = not estimated, AR4: Table 11.3, UNEP-2017: Chapter
3 4.
4
5 Mitigation costs and potentials for 2030 have been presented previously, notably in the AR4 Chapter
6 11 on Mitigation from a cross-sectoral perspective (Barker et al. 2007) and the Emissions Gap Report
7 (UNEP 2017). Note that AR5 did not provide emission reduction potentials in this form. The aggregated
8 potentials reported here are higher than those estimated in AR4. Note however, that AR4 suggested the
9 potentials were underestimated by 10 to 15%, but a higher potential still remains in the current
10 assessment. In a sector-by-sector comparison, higher potentials than in AR4 can be observed especially
11 for the energy sector and the forestry sector, and to a more limited extent for the industry sector and the
12 transport sector. For the energy sector, the change can largely be explained by the higher estimates for
13 wind and solar energy and the improved understanding of how to integrate high shares of intermittent
14 renewable energy sources into power systems. For industry and transport, the higher potentials can be
15 partly explained by the inclusion of more options, like recycling and material efficiency (for industry)
16 and electric transportation and modal shifts for transport. For buildings a lower potential can be
17 observed compared to AR4, one reason is that the 2030 reference direct and indirect emissions were
18 estimated as 45% and 11% higher in AR4 than they were in AR6 (signalling a much quicker actual
19 switch to electricity than was thought 15 to 20 years ago, among other reasons). The other reason for a
20 difference is that the scenarios considered in AR4 had 25 to 30 years between their start year until the
21 target year of 2030 and the scenarios reviewed in AR6 has only 10 to 15 years before 2030. The current
22 retrofitting rates of existing buildings and penetration rates of nearly zero energy buildings do not allow
Do Not Cite, Quote or Distribute 12-27 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 decarbonizing the sector over 10-15 years, but they do over a longer time period. A much larger
2 potential than reported here for 2030 can still be realized in the timeframe up to 2050 (Section 9.6.2).
3 Another global analysis was done by McKinsey (2009) which presents a marginal abatement cost curve
4 for 2030, suggesting a total potential of 38 GtCO2-eq (note that the reference for this study is 70 GtCO2-
5 eq, which is at the high end of the reference range used in this assessment).
6 The potentials reported here are comparable with UNEP (2017). Note that material for the energy sector
7 from the UNEP report was partly reused in this analysis. Furthermore, some options for the transport
8 sector (aviation and biofuels) were identical to the estimates in the UNEP report. The remaining
9 mitigation potentials are all based on new – and much more extended – assessment. There are some
10 notable changes. The AR6 mitigation potential for forestry is substantially larger. For buildings the
11 potential is smaller, mainly related to the smaller mitigation potential for electric appliances than in the
12 UNEP report. But overall, the estimates of the total mitigation potential are well aligned, which
13 confirms there is substantial consistency across various emissions estimates.
14 The results of the sectoral mitigation potentials are also compared with mitigation impacts as calculated
15 by IAMs. To this end, cumulative sectoral potentials over cost ranges were determined, based on the
16 information in Table 12.3. For options that are in various cost ranges, we assumed that they are evenly
17 distributed over these cost ranges. The only exception is wind and solar energy, for which it is indicated
18 that the majority of the mitigation potential is in the negative cost range. It was assumed that the fraction
19 in the negative cost range was 60%; the remainder is evenly distributed over the other cost ranges. These
20 cumulative potentials were compared with emission reductions realized in IAMs at certain price levels
21 for CO2. Note that these price levels selected in IAMs are average price levels – not all IAMs use
22 globally uniform carbon prices, so underlying these cost levels, there may be regional differentiation.
23 Data were taken from the AR6 scenarios database. Note that, strictly speaking, not all models in the
24 database are IAMs; in this analysis all models in the database were used, but the term IAMs is used as
25 shorthand in the text that follows. All scenarios likely to limit warming to 2˚C or lower are included for
26 the comparison (i.e., the categories of scenarios C1-C3 in Chapter 3). A comparison per sector is
27 provided in Figure 12.1. It is important to note that two different things are compared in this figure: on
28 the one hand emission reduction potentials and on the other hand realisations of (part of) the potential
29 within the context of a certain scenario. Having said that, a number of lessons can be learned from the
30 comparison of both.
31
Do Not Cite, Quote or Distribute 12-28 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.1 Comparison of sectoral estimates for the emission reduction potential with the emission
3 reductions calculated using IAMs.
4 The latter are given as box plots of global emissions reduction for each sector (blue and green) at different
5 global carbon cost levels (horizontal axis) for 2030, based on all scenarios likely limiting warming to 2°C
6 or lower (see Chapter 3) in the AR6 scenarios database (IPCC 2021). For IAMs, the cost levels correspond
7 to the levels of the carbon price. Hinges in the blue box plots represent the interquartile ranges and
8 whiskers extend to 5th and 95th percentiles while the hinges in the green box plots describe the full range,
9 and the middle point indicates the mean, not the median. In red, the estimates from the sectoral analysis
10 are given. In all cases, only direct emission reductions are presented, except for the orange boxes (for
11 buildings), which include indirect emission reductions. The orange boxes are only given for reasons of
12 completeness, also for buildings the blue boxes should be compared with the red boxes. Orange and red
13 boxes represent the full ranges of estimates. For IAMs, global carbon prices are applied, which are
14 subject to significant uncertainty.
15
16 For the energy supply sector, the emission reductions projected by the IAMs are for the higher
17 cost levels comparable with the potentials found in the sectoral analysis. But at lower cost
18 levels, the emission reductions as projected by IAMs are smaller than for the sectoral analysis.
19 This is likely due to the fact that high costs for solar energy and wind energy are assumed in
20 IAM models (Krey et al. 2019; Shiraki and Sugiyama 2020). This is not surprising, as the
21 scenario database comprises studies dating back to 2015. A more detailed comparison for the power
Do Not Cite, Quote or Distribute 12-29 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 sector is given in Figure 12.2. Both the sectoral analysis and the IAMs find that both solar and wind
2 energy in particular show strong growth potential, although there is a continuing role for other low-
3 carbon technologies, like nuclear energy and hydropower.
4 For the AFOLU sector, the sectoral studies provide net emission reduction potentials comparable with
5 projections from the IAMs at costs levels up to 50 USD tCO2-eq-1. However, beyond that level the
6 mitigation potential found in the sectoral analysis is larger than in the IAMs. For agriculture, it can be
7 explained by the fact that carbon sequestration options, like soil carbon, biochar and agroforestry have
8 little to no representation in IAMs. Similarly, for forestry and other land-use related options, the
9 protection and restoration of other ecosystems than forests (peatland, coastal wetlands and savannas)
10 are not represented in IAMs. Also note that some IAM baselines already have small carbon prices which
11 induce land-based mitigation, while in others, mitigation, particularly from reduced deforestation is part
12 of the storyline even without an implemented carbon price. Both of these effects dampen the mitigation
13 potential available in the USD 100 tCO2-eq-1 carbon price scenario from IAMs. Furthermore, estimates
14 of mitigation through forestry and other land use related options from the AR6 IAM scenario database
15 represent the net emissions from A/R and deforestation, thus are likely to be lower than the sectoral
16 estimates of A/R potential expressed as gross removals.
17 For the buildings and transport sectors, the sectoral mitigation potentials are higher than those projected
18 by the IAMs. The difference in the transport sector is particularly significant. One possible explanation
19 is that options with negative costs are already included in the reference. In addition, some options, like
20 avoiding demand for energy services in the building sector and model shift in transportation are less
21 well represented in IAMs.
22 For the industry sector, the sectoral emission reduction potentials are somewhat higher than those
23 reported on average by IAMs. The difference can well be explained by the fact that most IAMs do not
24 include circularity options like material efficiency and recycling; these options together account for 1.5
25 GtCO2-eq at costs levels from 20 USD tCO2-eq-1 onwards.
26 For mitigation of emissions of methane and fluorinated gases, the comparability between the sectoral
27 results and IAMs is good.
28 Overall, it is concluded that there are differences between the sectoral analysis and the IAM outcomes,
29 but most of the differences can be explained by the exclusion of specific options in most IAMs. This
30 comparability confirms the reliability of the sectoral analysis of emission reduction potential. It also
31 demonstrates the added value of sectoral analyses of mitigation potentials: they can more rapidly adapt
32 to changes in price levels of technologies and adopt new options for emission mitigation.
Do Not Cite, Quote or Distribute 12-30 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.2 Electricity production in 2030 as calculated by Integrated Assessment Models (blue),
3 compared with electricity production potentials found in the sectoral analysis (red).
4 In both cases cost cut-offs at 100 USD tCO2-1 are applied. Hinges in the blue box plots represent the
5 interquartile ranges and whiskers extend to 5 and 95 percentiles while the hinges in the red box plots
6 describe the full range.
7 In this section, the information on individual options reported in Section 12.2.2 to sectoral and
8 economy-wide totals has been aggregated. It is concluded that, based on the sectoral analysis, the global
9 mitigation potential is in the range of 32 to 44 GtCO2-eq. This mitigation potential is substantially
10 higher than that reported in AR4, but it is comparable to the more recent estimate by UNEP (2017).
11 Differences exist with the results of IAMs, but most of these can be well explained. The conclusion that
12 the global potential is in this range can be drawn with high agreement and robust evidence.
13 Given the median projection of the reference emissions of 60 GtCO2-eq in 2030, the range of mitigation
14 potentials presented here is sufficient to bring down global emissions in the year 2030 to a level of 16
15 to 28 GtCO2-eq. Taking into account that there is a range in reference projections for 2030 of 54 to 68
16 GtCO2-eq, the resulting emissions level shows a wider range: 12 to 31 GtCO2-eq. This is about at or
17 below half of the most recent (2019) emission value of 59±6.6 GtCO2-eq (high confidence).
18
19 12.2.4 Sectoral findings on emission pathways until 2050
20 As noted previously, a more qualitative approach is followed and less quantitative information is
21 presented for 2050. The sectoral results are summarised in Table 12.5. In addition to the many
22 technologies that already play a role by 2030 (Table 12.3) additional technologies may be needed for
23 deep decarbonisation, for example for managing power systems with high shares of intermittent
24 renewable sources and for providing new fuels and associated infrastructure for sectors that are hard to
25 decarbonise. New processes also play an important role, notably for industrial processes. In general,
Do Not Cite, Quote or Distribute 12-31 Total pages: 220
Final Government Distribution Chapter 12 IPCC AR6 WGIII
1 stronger sector coupling is needed, particularly increased integration of energy end-use and supply
2 sectors.
Do Not Cite, Quote or Distribute 12-32 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Table 12.5: Mitigation options and their characteristics for 2050
Sector Major options Degree to which net zero-GHG is possible
Energy sector Range of supply side options possible (see 2030 overview). Zero CO2 energy system is possible
Increased share of electricity in final energy use
Potentially important role for hydrogen, ammonia, etc.
Agriculture, forestry and other land Options comparable to those in 2030. Permanence is Some hard-to-abate activities will still have positive
use important. emissions, but for the sector as a whole, net negative
emissions are possible through carbon sequestration in
agriculture and forestry
Buildings Sufficiency, high performance new and existing buildings At least 8.2 GtCO2 or 61% reduction, as compared to the
with efficient HVAC esp. heat pumps, building baseline is possible with options on demand-side. This is a
management and operation, efficient appliances, onsite low estimate, because in some developing regions literature is
renewables backed up with demand flexibility and not sufficient to derive a comprehensive estimate. Nearly net
digitalisation measures zero CO2 emissions are possible if grid electricity will also be
decarbonised. Carbon storage in buildings provides CDR.
Transport Electrification can become a major option for many To a large extent if the electricity sector is fully decarbonized
transport modes. For long-haul trucking, ships and aviation, and the deployment of alternative fuels for long-haul trucking,
in addition biofuels, hydrogen and potentially synthetic aviation and shipping is successful.
fuels can be applied.
Industry Stronger role for material efficiency and recycling. Approx. 85% reduction is possible. Net zero CO2 emissions
are possible with retrofitting and early retirement.
Full decarbonisation through new processes, CCS, CCU and
hydrogen can become dominant
Do Not Cite, Quote or Distribute 12-33 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
Cross-sectoral Direct air carbon capture and storage Contributes CDR to support net zero GHG by
counterbalancing sectoral emissions
Enhanced weathering
Ocean-based methods
1
Do Not Cite, Quote or Distribute 12-34 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 12.3 Carbon dioxide removal (CDR)
2 CDR refers to a cluster of technologies, practices, and approaches that remove and sequester carbon
3 dioxide from the atmosphere and durably store the carbon in geological, terrestrial, ocean reservoirs, or
4 in products. Despite the common feature of removing carbon dioxide, CDR methods can be very
5 different (Smith et al. 2017). There are proposed methods for removal of non-CO2 greenhouse gases
6 such as methane (Jackson et al. 2019, 2021) but scarcity of literature on these methods prevents
7 assessment here.
8 A number of CDR methods (e.g., Afforestation/Reforestation (A/R), Bioenergy with carbon capture
9 and storage (BECCS), soil carbon sequestration (SCS), biochar, wetland/peatland restoration and
10 coastal restoration) are dealt with elsewhere in this report (Chapters 6 and 7). These methods are
11 synthesised in Section 12.3.2. Others, not dealt with elsewhere, i.e., Direct Air Carbon Capture and
12 Storage (DACCS), enhanced weathering of minerals (EW) and ocean-based approaches including
13 ocean fertilisation (OF) and alkalinity (OA) enhancement, are discussed in Sections 12.3.1.1 to 12.3.1.3
14 below (see also IPCC SROCC and WGI, Section 5.6). Some methods such as BECCS and DACCS
15 involve carbon storage in geological formations, which is discussed in Chapter 6. The climate system
16 and the carbon cycle responses to CDR deployment and each method’s physical and biogeochemical
17 characteristics such as storage form and duration are assessed in Chapters 4 and 5 of the WGI report.
18 START CROSS-CHAPTER BOX 8 HERE
19 Cross-Chapter Box 8 Carbon Dioxide Removal: Key characteristics and multiple roles
20 in mitigation strategies
21 Oliver Geden (Germany), Alaa Al Khourdajie (United Kingdom/Syria), Chris Bataille (Canada), Göran
22 Berndes (Sweden), Holly Jean Buck (the United States of America), Katherine Calvin (the United States
23 of America), Annette Cowie (Australia), Kiane de Kleijne (the Netherlands), Jan Minx (Germany),
24 Gert-Jan Nabuurs (the Netherlands), Glen Peters (Australia/Norway), Andy Reisinger (New Zealand),
25 Peter Smith (United Kingdom), Masahiro Sugiyama (Japan)
26 Carbon Dioxide Removal (CDR) is a necessary element of mitigation portfolios to achieve net zero
27 CO2 and GHG emissions both globally and nationally, counterbalancing residual emissions from ‘hard-
28 to-transition’ sectors such as industry, transport and agriculture. CDR is a key element in scenarios
29 likely to limit warming to 2°C or lower, regardless of whether global emissions reach near-zero, net
30 zero or net-negative levels (Sections 3.3, 3.4, 3.5 in Chapter 3 and Section 12.3 in this chapter). While
31 national mitigation portfolios aiming at net zero or net-negative emissions will need to include some
32 level of CDR, the choice of methods and the scale and timing of their deployment will depend on the
33 ambition for gross emission reductions, how sustainability and feasibility constraints are managed, and
34 how political preferences and social acceptability evolve (Section 12.3.3). This box gives an overview
35 of CDR methods, presents a categorisation based on the key characteristics of removal processes and
36 storage timescales, and clarifies the multiple roles of CDR in mitigation strategies. The term negative
37 emissions is used in this report only when referring to the net emissions outcome at a systems level
38 (e.g., net negative emissions at global, national, sectoral or supply chain levels).
39 Categorisation of the main CDR methods
40 CDR refers to anthropogenic activities that remove CO2 from the atmosphere and store it durably in
41 geological, terrestrial, or ocean reservoirs, or in products. It includes anthropogenic enhancement of
42 biological, geochemical or chemical CO2 sinks, but excludes natural CO2 uptake not directly caused by
43 human activities. Increases in land carbon sink strength due to CO2 fertilisation or other indirect effects
Do Not Cite, Quote or Distribute 12-35 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 of human activities are not considered CDR (see Glossary). Carbon Capture and Storage (CCS) and
2 Carbon Capture and Utilisation (CCU) applied to CO2 from fossil fuel use are not CDR methods as they
3 do not remove CO2 from the atmosphere. CCS and CCU can, however, be part of CDR methods if the
4 CO2 has been captured from the atmosphere, either indirectly in the form of biomass or directly from
5 ambient air, and stored durably in geological reservoirs or products (Sections 11.3.6 in Chapter 11 and
6 Section 12.3 in this chapter).
7 There are many different CDR methods and associated implementation options (Cross-Chapter Box 8,
8 Figure 1). Some of these methods (including afforestation and improved forest management, wetland
9 restoration and SCS) have been practiced for decades to millennia, although not necessarily with the
10 intention of removing carbon from the atmosphere. Conversely, methods such as Direct Air Carbon
11 Capture and Storage (DACCS), Bioenergy with Carbon Capture and Storage (BECCS) and Enhanced
12 Weathering are novel, and while experience is growing, their demonstration and deployment are limited
13 in scale. CDR methods have been categorised in different ways in the literature, highlighting different
14 characteristics. In this report, as in AR6 WGI, the categorisation is based on the role of CDR methods
15 in the carbon cycle, i.e., on the removal process (land-based biological; ocean-based biological;
16 geochemical; chemical) and on the timescale of storage (decades to centuries; centuries to millennia;
17 ten thousand years or longer). The timescale of storage is closely linked to the storage medium: carbon
18 stored in ocean reservoirs (through enhanced weathering, ocean alkalinity enhancement or ocean
19 fertilisation) and in geological formations (through BECCS or DACCS) generally has longer storage
20 times and is less vulnerable to reversal through human actions or disturbances such as drought and
21 wildfire than carbon stored in terrestrial reservoirs (vegetation, soil). Furthermore, carbon stored in
22 vegetation or through SCS has shorter storage times and is more vulnerable than carbon stored in
23 buildings as wood products; as biochar in soils, cement and other materials; or in chemical products
24 made from biomass or potentially through direct air capture (WGI, Section 5.6, Figure 5.36; WGIII,
25 Section 11.3.6; Fuss et al. 2018; Minx et al. 2018; NAS 2019). Within the same category (e.g., land-
26 based biological CDR) options often differ with respect to other dynamic or context-specific dimensions
27 such as mitigation potential, cost, potential for co-benefits and adverse side-effects, and technology
28 readiness level (Section 12.3, Table 12.6).
29
Do Not Cite, Quote or Distribute 12-36 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Cross-Chapter Box 8, Figure 1: Carbon Dioxide Removal taxonomy.
3 Methods are categorised based on removal process (grey shades) and storage medium (for which timescales of storage are given, yellow/brown shades). Main
4 implementation options are included for each CDR method. Note that specific land-based implementation options can be associated with several CDR methods,
5 e.g., agroforestry can support soil carbon sequestration and provide biomass for biochar or BECCS.
6 Source: This figure is an extended version of Figure 2 in (Minx et al. 2018).
Do Not Cite, Quote or Distribute 12-37 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Roles of CDR in mitigation strategies
2 Within ambitious mitigation strategies at global or national levels, CDR cannot serve as a substitute for
3 deep emissions reductions but can fulfil multiple complementary roles: (1) further reduce net CO2 or
4 GHG emission levels in the near-term; (2) counterbalance residual emissions from ‘hard-to-transition’
5 sectors , such as CO2 from industrial activities and long-distance transport (e.g., aviation, shipping), or
6 methane and nitrous oxide from agriculture, in order to help reach net zero CO2 or GHG emissions in
7 the mid-term; (3) achieve and sustain net-negative CO2 or GHG emissions in the long-term, by
8 deploying CDR at levels exceeding annual residual gross CO2 or GHG emissions (Sections 2.7.3 in
9 Chapter 2, 3.3 and 3.5 in Chapter 3).
10 In general, these roles of CDR are not mutually exclusive and can exist in parallel. For example,
11 achieving net zero CO2 or GHG emissions globally might involve some countries already reaching net-
12 negative levels at the time of global net zero, allowing other countries more time to achieve this.
13 Equally, achieving net-negative CO2 emissions globally, which could address a potential temperature
14 overshoot by lowering atmospheric CO2 concentrations, does not necessarily involve all countries
15 reaching net-negative levels (Cross-Chapter Box 3 in Chapter 3; Rajamani et al. 2021; Rogelj et al.
16 2021).
17 Cross-Chapter Box 8, Figure 2 shows these multiple roles of CDR in a stylised ambitious mitigation
18 pathway that can be applied to global and national levels. While such mitigation pathways will differ
19 in their shape and exact composition, they include the same basic components: CO2 emissions from
20 fossil sources, CO2 emissions from managed land, non-CO2 emissions, and various forms of CDR.
21 Figure 2 also illustrates the importance of distinguishing between gross CO2 removals from the
22 atmosphere through deployment of CDR methods and the net emissions outcome (i.e., gross emissions
23 minus gross removals).
24 CDR methods currently deployed on managed land, such as afforestation or reforestation and improved
25 forest management, lead to CO2 removals already today, even when net emissions from land use are
26 still positive, e.g., when gross emissions from deforestation and draining peatlands exceed gross
27 removals from afforestation or reforestation and ecosystem conservation (Sections 2.2 in Chapter 2, 7.2
28 in Chapter 7, Cross-Chapter Box 6 in Chapter 7). As there are currently no removal methods for non-
29 CO2 gases that have progressed beyond conceptual discussions (Jackson et al. 2021), achieving net zero
30 GHG implies gross CO2 removals to counterbalance residual emissions of both CO2 and non-CO2 gases,
31 applying GWP100 as the metric for reporting CO2-equivalent emissions, as required for emissions
32 reporting under the Rulebook of the Paris Agreement (Cross-Chapter Box 2 in Chapter 2).
Do Not Cite, Quote or Distribute 12-38 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Cross-Chapter Box 8, Figure 2: Roles of CDR in global or national mitigation strategies. Stylised pathway
3 showing multiple functions of CDR in different phases of ambitious mitigation: (1) further reducing net
4 CO2 or GHG emission levels in near-term; (2) counterbalancing residual emissions to help reach net zero
5 CO2 or GHG emissions in the mid-term; (3) achieve and sustain net-negative CO2 or GHG emissions in
6 the long-term.
7 Net zero CO2 emissions will be achieved earlier than net zero GHG emissions. As volumes of residual
8 non-CO2 emissions are expected to be significant, this time-lag could reach one to several decades,
9 depending on the respective size and composition of residual GHG emissions at the time of net zero.
10 Furthermore, counterbalancing residual non-CO2 emissions by CO2 removals will lead to net-negative
11 CO2 emissions at the time of net zero GHG emissions (Cross-Chapter Box 3 in Chapter 3).
12 END CROSS-CHAPTER BOX 8 HERE
13 While many governments have included A/R and other forestry measures into their NDCs under the
14 Paris Agreement (Moe and Røttereng 2018; Fyson and Jeffery 2019; Mace et al. 2021), and a few
15 countries also mention BECCS, DACCS and enhanced weathering in their mid-century low emission
16 development strategies (Buylova et al. 2021), very few are pursuing the integration of a broad range of
17 CDR methods into national mitigation portfolios so far (Box 12.1 in Section 12.3.3) (Schenuit et al.
18 2021). There are concerns that the prospect of large-scale CDR could, depending on the design of
19 mitigation strategies, obstruct near-term emission reduction efforts (Lenzi et al. 2018; Markusson et al.
20 2018), mask insufficient policy interventions (Geden 2016; Carton 2019), might lead to an overreliance
21 on technologies that are still in their infancy (Anderson and Peters 2016; Larkin et al. 2018; Grant et al.
22 2021), could overburden future generations (Lenzi 2018; Shue 2018; Bednar et al. 2019) might evoke
23 new conflicts over equitable burden-sharing (Pozo et al. 2020; Lee et al. 2021; Mohan et al. 2021),
24 could impact food security, biodiversity or land rights (Buck 2016; Boysen et al. 2017; Dooley and
25 Kartha 2018; Hurlbert et al. 2019; Dooley et al. 2021), or might be perceived negatively by stakeholders
26 and broader public audiences (Royal Society and Royal Academy of Engineering 2018; Colvin et al.
27 2020). Conversely, without considering different timescales of carbon storage (Fuss et al. 2018;
28 Hepburn et al. 2019) and implementation of reliable measurement, reporting and verification of carbon
29 flows (Mace et al. 2021), CDR deployment might not deliver the intended benefit of removing CO2
30 durably from the atmosphere. Furthermore, without appropriate incentive schemes and market designs
Do Not Cite, Quote or Distribute 12-39 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 (Honegger et al. 2021b), CDR implementation options could see under-investment. The many
2 challenges in research, development and demonstration of novel approaches, to advance innovation
3 according to broader societal objectives and to bring down costs, could delay their scaling up and
4 deployment (Nemet et al. 2018). Depending on the scale and deployment scenario, CDR methods could
5 bring about various co-benefits and adverse side effects (see below). All this highlights the need for
6 appropriate CDR governance and policies (Section 12.3.3).
7 The volumes of future global CDR deployment assumed in IAM-based mitigation scenarios are large
8 compared to current volumes of deployment, which presents a challenge since rapid and sustained
9 upscaling from a small base is particularly difficult (de Coninck et al. 2018; Nemet et al. 2018; Hanna
10 et al. 2021). All Illustrative Mitigation Pathways (IMPs) likely to limit warming to 2°C or lower use
11 some form of CDR. Across the full range of similarly ambitious IAM scenarios (scenario categories
12 C1-C3; see Section 3.3.), the annual net CO2 removal (i.e., gross removals, including A/R, minus gross
13 emissions) on managed land reaches 0.86 [0.01–4.11] GtCO2 yr-1 by 2030, 2.98 [0.23–6.38] GtCO2 yr-
1
14 by 2050, and 4.19 [0.1–6.91] GtCO2 yr-1 by 2100 (values are the medians and bracketed values denote
15 the 5-95 percentile range). The annual BECCS deployment is 0.08 [0–1.09] GtCO2 yr-1, 2.75 [0.52–
16 9.45] GtCO2 yr-1, and 8.96 [2.63–16.15] GtCO2 yr-1 for these years, respectively. The annual DACCS
17 deployment reaches 0 [0–0.02] GtCO2 yr-1 by 2030, 0.02 [0–1.74] GtCO2 yr-1 by 2050, and 1.02 [0–
18 12.6] GtCO2 yr-1 by 2100 (Figure 12.3)1. Cumulative volumes of BECCS, net CO2 removal on managed
19 land, and DACCS reach 328 [168–763] GtCO2, 252 [20–418] GtCO2, and 29 [0–339] GtCO2 for the
20 2020-2100 period, respectively. Reaching the higher end of CDR volumes is subject to issues regarding
21 their feasibility (see below), especially if achieved with only a limited number of CDR methods. Recent
22 studies have identified some drivers for large-scale CDR deployment in IAM scenarios, including
23 insufficient representation of variable renewables, a high discount rate that tends to increase initial
24 carbon budget overshoot and therefore inflates usage of CDR to achieve net-negative emissions at later
25 times, omission of CDR methods aside from BECCS and A/R (Köberle 2019; Emmerling et al. 2019;
26 Hilaire et al. 2019), and limited deployment of demand-side options (Grubler et al. 2018; Daioglou et
27 al. 2019; van Vuuren et al. 2018). The levels of CDR in IAMs in modelled pathways would change
28 depending on the allowable overshoot of policy targets such as temperature or radiative forcing and the
29 costs of non-CDR mitigation options (Johansson et al. 2020; van der Wijst et al. 2021).(see also Section
30 3.2.2)
FOOTNOTE1 We use representative options for labels of each variable reported in the AR6 scenarios database.
Do Not Cite, Quote or Distribute 12-40 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2
3 Figure 12.3 Sequestration of three predominant CDR methods: BECCS, net CO 2 removal on managed
4 land (that is, gross removal through A/R minus emissions from deforestation) , and DACCS (upper
5 panels) annual sequestration and (lower panels) cumulative sequestration.
6 The IAM scenarios correspond to those likely limiting warming to 2°C or lower. The black line in each of
7 the upper panels indicates the median of all the scenarios in categories C1-C3. Hinges in the lower panels
8 represent the interquartile ranges while whiskers extend to 5th and 95th percentiles. The IMPs are
9 highlighted with colours, as shown in the key. The number of scenarios is indicated in the header of each
10 panel. The number of scenarios with a non-zero DACCS value is 146.
11 While many CDR methods are gradually being explored, IAM scenarios have focused mostly on
12 BECCS and A/R (Tavoni and Socolow 2013; Fuhrman et al. 2019; Rickels et al. 2019; Calvin et al.
13 2021; Diniz Oliveira et al. 2021) Although some IAM studies have also included other methods such
14 as DACCS (Chen and Tavoni 2013; Marcucci et al. 2017; Realmonte et al. 2019; Akimoto et al. 2021;
15 Fuhrman et al. 2020, 2021a), enhanced weathering (Strefler et al. 2021), SCS and biochar (Holz et al.
16 2018) there is much less literature compared to studies on BECCS (Hilaire et al. 2019). A large scale,
17 coordinated IAM study on BECCS (“EMF-33”) has been conducted (Muratori et al. 2020; Rose et al.
18 2020a) but none exists for other CDR methods. A recent review proposes a combination of various
19 CDR methods (Fuss et al. 2018) but more in-depth literature on such a portfolio approach is limited
Do Not Cite, Quote or Distribute 12-41 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 (Strefler et al. 2021). A multi-criteria analysis has identified pathways with CDR portfolios different
2 from least-cost pathways often dominated by BECCS and A/R (Rueda et al. 2021).
3 At the national and regional level, the role of land-based biological CDR methods has long been
4 analysed, but there is little detailed technoeconomic assessment of the role of other CDR. There is a
5 small but emerging literature providing such assessments for developed countries (Baik et al. 2018;
6 Sanchez et al. 2018; Patrizio et al. 2018; Larsen et al. 2019; Daggash et al. 2018; Kato and Kurosawa
7 2019; Kraxner et al. 2014; Breyer et al. 2019; McQueen et al. 2020; Bistline and Blanford 2021; Jackson
8 et al. 2021; Kato and Kurosawa 2021; García-Freites et al. 2021; Negri et al. 2021) while the literature
9 outside of developed countries is limited (Alatiq et al. 2021; Fuhrman et al. 2021b; Weng et al. 2021).
10 In IAMs, CDR is contributed mainly by the energy sector (through BECCS) and AFOLU sector
11 (through A/R) (See Figure 12.3). IAMs are starting to include other CDR methods, such as DACCS
12 and enhanced weathering (Section 12.3.1), which are yet to be attributed to specific sectors in IAMs.
13 Following IPCC guidance for UNFCCC inventories, A/R and SCS are reported in LULUCF, while
14 BECCS would be reported in the sector where the carbon capture occurs, that is, the energy sector in
15 the case of electricity and heat production, and the industry sector for BECCS linked to manufacturing
16 (e.g., steel or hydrogen) (Tanzer et al. 2020; Bui et al. 2021; Tanzer et al. 2021).
17 12.3.1 CDR methods not assessed elsewhere in this report: DACCS, enhanced
18 weathering and ocean-based approaches
19 This section assesses the CDR methods that are not carried out solely within conventional sectors and
20 so are not covered in other parts of the report: direct air carbon capture and storage, enhanced
21 weathering, and ocean-based approaches. It provides an overview of each CDR method, costs,
22 potentials, risks and impacts, co-benefits, and their role in mitigation pathways. Since these processes,
23 approaches and technologies have medium to low technology readiness levels, they are subject to
24 significant uncertainty.
25 12.3.1.1 Direct Air Carbon Capture and Storage (DACCS)
26 Direct air capture (DAC) is a chemical process to capture ambient CO2 from the atmosphere. Captured
27 CO2 can be stored underground (direct air capture carbon and storage, DACCS) or utilised in products
28 (direct air capture carbon and utilisation, DACCU). DACCS shares with conventional CCS the transport
29 and storage components but is distinct in its capture part. Because CO2 is a well-mixed GHG, DACCS
30 can be sited relatively flexibly, though its locational flexibility is constrained by the availability of low-
31 carbon energy and storage sites. Capturing the CO2 involves three basic steps: a) contacting the air, b)
32 capturing on a liquid or solid sorbent or a liquid solvent, c) regeneration of the solvent or the sorbent
33 (with heat, moisture and/or pressure). After capture, the CO2 stream can be stored underground or
34 utilised. The duration of storage is an important consideration; geological reservoirs or mineralisation
35 result in removal for > 1000 years. The duration of the removal through DACCU (Breyer et al. 2019)
36 varies with the lifetime of respective products (Wilcox et al. 2017; Gunnarsson et al. 2018; Bui et al.
37 2018; Creutzig et al. 2019; Royal Society and Royal Academy of Engineering 2018; Fuss et al. 2018),
38 ranging from weeks to months for synthetic fuels to centuries or more for building materials (e.g.,
39 concrete cured using mineral carbonation) (Hepburn et al. 2019). The efficiency and environmental
40 impacts of DACCS and DACCU options depend on the carbon intensity of the energy input (electricity
41 and heat) and other life-cycle assessment (LCA) considerations (Jacobson 2019; Global CO2 Initiative
42 2018). See Chapters 6 and 11 for further details regarding carbon capture and utilisation. Another key
43 consideration is the net carbon CO2 removal of DACCS over its life cycle (Madhu et al. 2021). Deutz
44 and Bardow (2021) and Terlouw et al. (2021) demonstrated that the life-cycle net emissions of DACCS
Do Not Cite, Quote or Distribute 12-42 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 systems can be negative, even for existing supply chains and some current energy mixes. They found
2 that the GHG-intensity of energy sources is a key factor.
3 DAC options can be differentiated by the specific chemical processes used to capture ambient CO2 from
4 the air and recover it from the sorbent (Fasihi et al. 2019). The main categories are a) liquid solvents
5 with high-temperature regeneration, b) solid sorbents with low temperature regeneration and c)
6 regenerating by moisturising of solid sorbents. Other approaches such as electro-swing (Voskian and
7 Hatton 2019) have been proposed but are less developed. Compared to other CDR methods, the primary
8 barrier to upscaling DAC is its high cost and large energy requirement (high confidence) (Nemet et al.
9 2018), which can be reduced through innovation. It has therefore attracted entrepreneurs and private
10 investments (IEA 2020b).
11 Status: There are some demonstration projects by start-up companies and academic researchers, who
12 are developing various types of DAC, including aqueous potassium solvent with calcium carbonation
13 and solid sorbents with heat regeneration (NASEM 2019). These projects are supported mostly by
14 private investments and grants or sometimes serve utilisation niche markets (e.g., CO 2 for beverages,
15 greenhouses, enhanced oil recovery). As of 2021, there are more than ten plants worldwide, with a scale
16 of ktCO2 yr-1 or smaller (IEA 2020b; NASEM 2019; Larsen et al. 2019). Because of the fundamental
17 difference in the CO2 concentration in the capture stage, DACCS does not benefit directly from RD&D
18 of conventional CCS. Public RD&D programs dedicated to DAC have therefore been proposed
19 (NASEM 2019; Larsen et al. 2019). Possible research topics include development of new liquid solvents,
20 novel solid sorbents, and novel equipment or system designs, and the need for third-party evaluation of
21 techno-economic aspects has also been emphasized (NASEM 2019). However, since basic research
22 does not appear to be a primary barrier, both NASEM (2019) and Larsen et al. (2019) argue for a
23 stronger focus on demonstration in the US context. Though the US and UK governments have begun
24 funding DACCS research (IEA 2020b), the scale of R&D activities is limited.
25 Costs: As the process captures dilute CO2 (~0.04%) from the ambient air, it is less efficient and more
26 costly than conventional carbon capture applied to power plants and industrial installations (high
27 confidence) (with a CO2 concentration of ~10%). The cost of a liquid solvent system is dominated by
28 the energy cost (because of the much higher energy demand for CO2 regeneration, which reduces the
29 efficiency) while capital costs account for a significant share of the cost of solid sorbent systems (Fasihi
30 et al. 2019). The range of the DAC cost estimates found in the literature is wide (60–1000 USD tCO2-
1
31 ) (Fuss et al. 2018) partly because different studies assume different use cases, differing phases (first
32 plant vs. nth plant) (Lackner et al. 2012), different configurations, and disparate system boundaries.
33 Estimates of industrial origin are often on the lower side (Ishimoto et al. 2017). Fuss et al. (2018) suggest
34 a cost range of 600–1000 USD tCO2-1 for first-of-a-kind plants, and 100–300 USD tCO2-1 as experience
35 accumulates. An expert elicitation study found a similar cost level for 2050 with a median of around
36 200 USD tCO2-1 (Shayegh et al. 2021) (medium evidence, medium agreement). NASEM (2019)
37 systematically evaluated the costs of different designs and found a range of 84–386 USD2015 tCO2-1 for
38 the designs currently considered by active technology developers. This cost range excludes the site-
39 specific costs of transportation or storage.
40 Potentials: There is no specific study on the potential of DACCS but the literature has assumed that the
41 technical potential of DACCS is virtually unlimited provided that high energy requirements could be
42 met (medium evidence, high agreement) (Lawrence et al. 2018; Marcucci et al. 2017; Fuss et al. 2018)
43 since DACCS encounters less non-cost constraints than any other CDR method. Focusing only on the
44 Maghreb region, Breyer et al. (2020) reported an optimistic potential 150 GtCO2 at less than 61 USD
45 tCO2-1 for 2050. Fuss et al. (2018) suggest a potential of 0.5–5 GtCO2 yr-1 by 2050 because of
46 environmental side effects and limits to underground storage. In addition to the ultimate potentials,
Do Not Cite, Quote or Distribute 12-43 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Realmonte et al. (2019) noted the rate of scale-up as a strong constraint on deployment. Meckling and
2 Biber (2021) discuss a policy roadmap to address the political economy for upscaling. More systematic
3 analysis on potentials is necessary; first and foremost on national and regional levels, including the
4 requirements for low-carbon heat and power, water and material demand, availability of geological
5 storage and the need for land in case of low-density energy sources such as solar or wind power.
6 Risks and impacts: DACCS requires a considerable amount of energy (high confidence), and depending
7 on the type of technology, water, and make-up sorbents, while its land footprint is small compared to
8 other CDR methods (Smith et al. 2016), but depending on the source of energy for DACCS (e.g.,
9 renewables vs. nuclear), it could require a significant land footprint (NASEM 2019; Sekera and
10 Lichtenberger 2020). The theoretical minimum energy requirement for separating CO2 gas from the air
11 is ~0.5 GJ tCO2-1 (Socolow et al. 2011). Fasihi et al. (2019) reviewed the published estimates of energy
12 requirements and found that for the current technologies, the total energy requirement is ~4–10 GJ tCO2-
1
13 , with heat accounting for about 80% and electricity about 20% (McQueen et al. 2021). At a 10 GtCO2
14 yr-1 sequestration scale, this would translate into 40–100 EJ yr-1 of energy consumption (32–80 EJ yr-1
15 for heat and 8–20 EJ yr-1 electricity), which can be contrasted with the current primary energy supply
16 of ~600 EJ yr-1 and electricity generation of ~100 EJ yr-1. For the solid sorbent technology, low-
17 temperature heat could be sourced from heat pumps powered by low-carbon sources such as renewables
18 (Breyer et al. 2020), waste heat (Beuttler et al. 2019), and nuclear energy (Sandalow et al. 2018). Unless
19 sourced from a clean source, this amount of energy could cause environmental damage (Jacobson 2019).
20 Because DACCS is an open system, water lost from evaporation must be replenished. Water loss varies,
21 depending on technology (including adjustable factors such as the concentration of the liquid solvent)
22 as well as environmental conditions (e.g., temperate vs. tropical climates). For a liquid solvent system,
23 it can be 0–50 tH2O tCO2-1 (Fasihi et al. 2019). A water loss rate of ~1–10 tH2O tCO2-1 (Socolow et al.
24 2011) would translate into ~10–100 GtH2O = 10–100 km3 to capture 1ma0 GtCO2 from the atmosphere.
25 Some solid sorbent technologies actually produce water as a by-product, for example 0.8–2 tH2O tCO2-
1
26 for a solid-sorbent technology with heat regeneration (Beuttler et al. 2019; Fasihi et al. 2019). Large-
27 scale deployment of DACCS would also require a significant quantity of materials, and energy to
28 produce them (Chatterjee and Huang 2020). Hydroxide solutions are currently being produced as a by-
29 product of chlorine but replacement (make-up) requirement of such materials at scale exceeds the
30 current market supply (Realmonte et al. 2019). The land requirements for DAC units are not large
31 enough to be of concern (Madhu et al. 2021). Furthermore, these can be placed on unproductive lands,
32 in contrast to biological CDR. Nevertheless, to ensure that CO2-depleted air does not enter the air
33 contactor of an adjacent DAC system, there must be enough space between DAC units, similar to wind
34 power turbines. Considering this, Socolow et al. (2011) estimated a land footprint of 1.5 km2 MtCO2-1.
35 In contrast, large energy requirements can lead to significant footprints if low-density energy sources
36 (e.g., solar PV) are used (Smith et al. 2016). For the issues associated with CO2 utilisation and storage,
37 see Chapter 6.
38 Co-benefits: While Wohland et al. (2018) proposed solid sorbent-based DAC plants as a Power-to-X
39 technology that could use excess renewable power (at the time of low or even negative prices), such
40 operation would add additional costs. Installations would need to be designed for intermittent operations
41 (i.e., at low load factors) which would negatively affect capital and operation costs (Sandalow et al.
42 2018; Daggash et al. 2018) as a high time-resolution model suggests a high utilisation rate (Breyer et
43 al. 2020). Solid sorbent DAC designs can potentially remove more water from the ambient air than
44 needed for regeneration, thereby delivering surplus water that would contribute to Sustainable
45 Development Goal (SDG) 6 (Clean Water and Sanitation) in arid regions (Sandalow et al. 2018; Fasihi
46 et al. 2019).
Do Not Cite, Quote or Distribute 12-44 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Trade-offs and spill over effects: Liquid solvent DACCS systems need substantial amounts of water
2 (Fasihi et al. 2019), although much less than BECCS systems (Smith et al. 2016), which could
3 negatively affect SDG 6 (Clean Water and Sanitation). Although the high energy demand of DACCS
4 could affect SDG 7 (Affordable and Clean Energy) negatively through potential competition or
5 positively through learning effects (Beuttler et al. 2019), its impact has not been thoroughly assessed
6 yet.
7 Role in mitigation pathways: There are a few IAM studies that have explicitly incorporated DACCS.
8 Stringent emissions constraints in these studies lead to high carbon prices, allowing DACCS to play an
9 important role in mitigation. Chen and Tavoni (2013) examined the role of DACCS in an IAM
10 (WITCH) and found that incorporating DACCS in their IAM reduces the overall cost of mitigation and
11 tends to postpone the timing of mitigation. The scale of capture goes up to 37 GtCO2 yr-1 in 2100.
12 (Akimoto et al. 2021) introduced DACCS in the integrated assessment model DNE21+, and also found
13 the long-term marginal cost of abatement is significantly reduced by DACCS. Marcucci et al. (2017)
14 ran MERGE-ETL, an integrated model with endogenous learning, and showed that DACCS allows for
15 a model solution for the 1.5°C target, and that DACCS substitutes for BECCS under stringent targets.
16 In their analysis, DACCS captures up to 38.3 GtCO2 yr-1 in 2100. Realmonte et al. (2019) modelled two
17 types of DACCS (based on liquid and solid sorbents) with two IAMs (TIAM-Grantham and WITCH),
18 and showed that in deep mitigation scenarios, DACCS complements, rather than substitutes, other CDR
19 methods such as BECCS, and that DACCS is effective at containing mitigation costs. At the national
20 scale, Larsen et al. (2019) utilised the Regional Investment and Operations (RIO) Platform coupled with
21 the Energy PATHWAYS model, and explicitly represented DAC in US energy systems scenarios. They
22 found that in a scenario that reaches net zero emissions by 2045, about 0.6 GtCO2 or 1.8 GtCO2 of
23 DACCS would be deployed, depending on the availability of biological carbon sinks and bioenergy.
24 The modelling supporting the European Commission’s initial proposal for net zero GHG emissions by
25 2050 incorporated DAC, whose captured CO2 is used for both synthetic fuel production (DACCU) and
26 storage (DACCS) (Capros et al. 2019). Fuhrman et al. (2021a) evaluated the role of DACCS across 5
27 shared socioeconomic pathways with the GCAM modelling framework and identified a substantial role
28 of DACCS in mitigation and a decreased pressure on land and water resources from BECCS, even under
29 the assumption of limited energy efficiency improvement and conservative cost declines of DACCS
30 technologies. The newest iteration of the World Economic Outlook by IEA (2021b) deploys CDR on
31 a limited scale, and DACCS removes 0.6 GtCO2 in 2050 for its Net Zero CO2 Emissions scenario.
32 Status, costs, potentials, risk and impacts, co-benefits, trade-offs and spillover effects and the role in
33 mitigation pathways of DACCS are summarised in Table 12.6.
34 12.3.1.2 Enhanced weathering
35 Enhanced weathering involves a) the mining of rocks containing minerals that naturally absorb CO 2
36 from the atmosphere over geological timescales (as they become exposed to the atmosphere through
37 geological weathering), b) the comminution of these rocks to increase the surface area, and c) the
38 spreading of these crushed rocks on soils (or in the ocean/coastal environments; Section 12.3.1.3) so
39 that they react with atmospheric CO2 (Schuiling and Krijgsman 2006; Hartmann et al. 2013; Beerling
40 et al. 2018; Goll et al. 2021). Construction waste, and waste materials from mining can also be used as
41 a source material for enhanced weathering. Silicate rocks such as basalt, containing minerals rich in
42 calcium and magnesium and lacking metal ions such as nickel and chromium, are most suitable for
43 enhanced weathering (Beerling et al. 2018); they reduce soil solution acidity during dissolution, and
44 promote the chemical transformation of CO2 to bicarbonate ions. The bicarbonate ions can precipitate
45 in soils and drainage waters as a solid carbonate mineral (Manning 2008), or remain dissolved and
46 increase alkalinity levels in the ocean when the water reaches the sea (Renforth and Henderson 2017).
Do Not Cite, Quote or Distribute 12-45 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 The modelling study by Cipolla et al. (2021) found that rate of weathering is greater in high rainfall
2 environments, and was increased by organic matter amendment.
3 Status: Enhanced weathering has been demonstrated in the laboratory and in small scale field trials
4 (TRL 3–4) but has yet to be demonstrated at scale (Beerling et al. 2018; Amann et al. 2020). The
5 chemical reactions are well understood (Gillman 1980; Gillman et al. 2001; Manning 2008), but the
6 behaviour of the crushed rocks in the field and potential co-benefits and adverse-side effects of
7 enhanced weathering require further research (Beerling et al. 2018). Small scale laboratory experiments
8 have calculated weathering rates that are orders of magnitude slower than the theoretical limit for mass
9 transfer-controlled forsterite (Renforth et al. 2015; Amann et al. 2020) and basalt dissolution (Kelland
10 et al. 2020). Uncertainty surrounding silicate mineral dissolution rates in soils, the fate of the released
11 products, the extent of legacy reserves of mining by-products that might be exploited, location and
12 availability of rock extraction sites, and the impact on ecosystems remain poorly quantified and require
13 further research to better understand feasibility (Renforth 2012; Moosdorf et al. 2014; Beerling et al.
14 2018). Closely monitored, large-scale demonstration projects would allow these aspects to be studied
15 (Smith et al. 2019a; Beerling et al. 2020).
16 Costs: Fuss et al. (2018), in a systematic review of the costs and potentials of CDR methods including
17 enhanced weathering, note that costs are closely related to the source of the rock, the technology used
18 for rock grinding and material transport (Hartmann et al. 2013; Renforth 2012; Strefler et al. 2018). Due
19 to differences in the methods and assumptions between studies, literature ranges are highly uncertain
20 and range from 15–40 USD tCO2-1 to 3460 USD tCO2-1 (Köhler et al. 2010; Taylor et al. 2016). Renforth
21 (2012) reported operational costs in the UK of applying mafic rocks (rocks with high magnesium and
22 iron silicate mineral concentrations) of 70–578 USD tCO2-1, and for ultramafic rocks (rocks rich in
23 magnesium and iron silicate minerals but with very low silica content - the low silica content enhances
24 weathering rates ) of 24–123 USD tCO2-1. Beerling et al. (2020) combined a spatially resolved
25 weathering model with a technoeconomic assessment to suggest costs of between 54–220 USD tCO2-1
26 (with a weighted mean of 118–128 USD tCO2-1). Fuss et al. (2018) suggested an author judgement cost
27 range of 50–200 USD tCO2-1 for a potential of 2–4 GtCO2 yr−1 from 2050, excluding biological storage.
28 Potentials: In a systematic review of the costs and potentials of enhanced weathering, Fuss et al. (2018)
29 report a wide range of potentials (limited evidence, low agreement). The highest reported regional
30 sequestration potential, 88.1 GtCO2 yr−1, is reported for the spreading of pulverised rock over a very
31 large land area in the tropics, a region considered promising given the higher temperatures and greater
32 rainfall (Taylor et al. 2016). Considering cropland areas only, the potential carbon removal was
33 estimated by Strefler et al. (2018) to be 95 GtCO2 yr−1 for dunite and 4.9 GtCO2 yr−1 for basalt. Slightly
34 lower potentials were estimated by Lenton (2014) where the potential of carbon removal by enhanced
35 weathering (including adding carbonate and olivine to both oceans and soils) was estimated to be 3.7
36 GtCO2 yr–1 by 2100, but with mean annual removal an order of magnitude less at 0.2 GtC-eq yr–1
37 (Lenton 2014). The estimates reported in Smith et al. (2016) are based on the potential estimates of
38 Lenton (2014). Beerling et al. (2020) estimate that up to 2 GtCO2 yr–1 could be removed by 2050 by
39 spreading basalt onto 35-59% (weighted mean 53%) of agricultural land of 12 countries. Fuss et al.
40 (2018) provide an author judgement range for potential of 2–4 GtCO2 yr−1 for 2050.
41 Risks and impacts: Mining of rocks for enhanced weathering will have local impacts and carries risks
42 similar to that associated with the mining of mineral construction aggregates, with the possible
43 additional risk of greater dust generation from fine comminution and land application. In addition to
44 direct habitat destruction and increased traffic to access mining sites, there could be adverse impacts on
45 local water quality (Younger and Wolkersdorfer 2004).
Do Not Cite, Quote or Distribute 12-46 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Co-benefits: Enhanced weathering can improve plant growth by pH modification and increased mineral
2 supply (Kantola et al. 2017; Beerling et al. 2018), can enhance SCS in some soils (Beerling et al. 2018)
3 thereby protecting against soil erosion (Wrigth and Upadhyaya 1998), and increasing the cation
4 exchange capacity, resulting in increased nutrient retention and availability (Baldock and Skjemstad
5 2000; Yu et al. 2017; Guntzer et al. 2012; Tubana et al. 2016; Manning 2010; Haque et al. 2019; Smith
6 et al. 2019a; Gillman 1980; Gillman et al. 2001). Through these actions, it can contribute to the UN
7 SDGs 2 Zero Hunger, 15 Life of Land (by reducing land demand for croplands), 13 Climate Action
8 (through CDR), 14 Life Below Water (by ameliorating ocean acidification) and 6 Clean Water and
9 Sanitation (Smith et al. 2019a). To more directly ameliorate ocean acidification while increasing CDR
10 and reducing impacts on land ecosystems, alkaline minerals could instead be directly added to the ocean
11 (Section 12.3.1.3). There are potential benefits in poverty reduction through employment of local
12 workers in mining (Pegg 2006).
13 Trade-offs and spill over effects: Air quality could be adversely affected by the spreading of rock dust
14 (Edwards et al. 2017), though this can partly be ameliorated by water-spraying (Grundnig et al. 2006).
15 As noted above, any significant expansion of the mining industry would require careful assessment to
16 avoid possible detrimental effects on biodiversity (Amundson et al. 2015). The processing of an
17 additional 10 billion tonnes of rock would require up to 3000 TWh, which could represent
18 approximately 0.1-6 % of global electricity in 2100. The emissions associated with this additional
19 energy generation may reduce the net carbon dioxide removal by up to 30% with present day grid
20 average emissions, but this efficiency loss would decrease with low-carbon power (Beerling et al.
21 2020).
22 Role in mitigation pathways: Only one study to date has included enhanced weathering in an integrated
23 assessment model to explore mitigation pathways (Strefler et al. 2021).
24 Status, costs, potentials, risk and impacts, co-benefits, trade-offs and spill over effects and the role in
25 mitigation pathways of enhanced weathering are summarised in Table 12.6.
26 12.3.1.3 Ocean-based methods
27 The ocean, which covers over 70% of the Earth’s surface, contains ~38,000 GtC, some 45 times more
28 than the present atmosphere, and oceanic uptake has already consumed close to 30-40% of
29 anthropogenic C emissions (Gruber et al. 2019, Sabine et al. 2004). The ocean is characterised by
30 diverse biogeochemical cycles involving carbon, and ocean circulation has much longer timescales than
31 the atmosphere, meaning that additional anthropogenic carbon could potentially be stored in the ocean
32 for centuries to millennia for methods that increase deep ocean dissolved carbon concentrations or
33 temporarily bury the carbon; or essentially permanently (over ten thousand years) for methods that store
34 the carbon in mineral forms or as ions by increasing alkalinity (Siegel et al., 2021) (Cross-Chapter Box
35 8 Figure 1). A wide range of methods and implementation options for marine CDR have been proposed
36 (Gattuso et al. 2018; Hoegh-Guldberg et al. 2018; GESAMP 2019). The most studied ocean-based CDR
37 methods are ocean fertilisation, alkalinity enhancement (including electrochemical methods) and
38 intensification of biologically driven carbon fluxes and storage in marine ecosystems, referred to as
39 “blue carbon”. The mitigation potentials, costs, co-benefits and trade-offs of these three options are
40 discussed below. Less well studied are methods including artificial upwelling, terrestrial biomass
41 dumping into oceans, direct CO2 removal from seawater (with CCS), and sinking marine biomass into
42 the deep ocean or harvesting it for bioenergy (with CCS) or biochar (GESAMP 2019). These methods
43 are summarized briefly below. Potential climate response and influence on the carbon budget of ocean-
44 based CDR methods are discussed in Chapter 5 in WGI AR6.
Do Not Cite, Quote or Distribute 12-47 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Ocean fertilisation (OF). One natural mechanism of carbon transfer from the atmosphere to the deep
2 ocean is the ocean biological pump, which is driven by the sinking of organic particles from the upper
3 ocean. These particles derive ultimately from primary production by phytoplankton and most of them
4 are remineralised within the upper ocean with only a small fraction reaching the deep ocean where the
5 carbon can be sequestered on centennial and longer timescales. Increasing nutrient availability would
6 stimulate uptake of CO2 through phytoplankton photosynthesis producing organic matter, some of
7 which would be exported into the deep ocean, sequestering carbon. In areas of the ocean where
8 macronutrients (nitrogen, phosphorus) are available in sufficient quantities (about 25% of the total
9 area), the growth of phytoplankton is limited by the lack of trace elements such as iron. Thus, OF CDR
10 can be based on two implementation options to increase the productivity of phytoplankton (Minx et al.
11 2018): macronutrient enrichment and micronutrient enrichment. A third option highlighted in
12 GESAMP (2019) is based on fertilisation for fish stock enhancement, for instance, as naturally occurs
13 in eastern boundary current systems. Iron fertilisation is the best studied OF option to date, but
14 knowledge so far is still inadequate to predict global ecological and biogeochemical consequences.
15 Status: OF has a natural analogue: periods of glaciation in the geological past are associated with
16 changes in deposition of dust containing iron into the ocean. Increased formation of phytoplankton has
17 also been observed during seasonal deposition of dust from the Arabian Peninsula and ash deposition
18 on the ocean surface after volcanic eruptions (Jaccard et al., 2013; Achterberg et al. 2013; Olgun et al.
19 2013; Martínez-García et al. 2014). OF options may appear technologically feasible, and enhancement
20 of photosynthesis and CO2 uptake from surface waters is confirmed by a number of field experiments
21 conducted in different areas of the ocean, but there is scientific uncertainty about the proportion of
22 newly formed organic carbon that is transferred to deep ocean, and the longevity of storage (Blain et al.
23 2008; Williamson et al. 2012; Trull et al. 2015). The efficiency of OF also depends on the region and
24 experimental conditions, especially in relation to the availability of other nutrients, light and
25 temperature (Aumont and Bopp 2006). In the case of macronutrients, very large quantities are needed
26 and the proposed scaling of this technique has been viewed as unrealistic (Williamson and Bodle 2016).
27 Costs: Ocean fertilisation costs depend on nutrient production and its delivery to the application area
28 (Jones 2014). The costs range from 2 USD tCO2-1 for fertilisation with iron (Boyd 2008) to 457 USD
29 tCO2-1 for nitrate (Harrison 2013). Reported costs for macronutrient application at 20 USD tCO2-1 (Jones
30 2014), contrast with higher estimates by (Harrison 2013) reporting that low costs are due to
31 overestimation of sequestration capacity and underestimation of logistical costs. The median of OF
32 cost estimates, 230 USD tCO2-1 (Gattuso et al., 2021) indicates low cost-effectiveness, albeit
33 uncertainties are large.
34 Potentials: Theoretical calculations indicate that organic carbon export increases 2–20 kg per gram of
35 iron added, but experiments indicate much lower efficiency: a significant part of the CO2 can be emitted
36 back the atmosphere because much of the organic carbon produced is remineralised in the upper ocean.
37 Efficiency also varies with location (Bopp et al. 2013). Between studies, there are substantial
38 differences in the ratio of iron added to carbon fixed photosynthetically, and in the ratio of iron added
39 to carbon eventually sequestered (Trull et al. 2015), which has implications both for the success of this
40 strategy, and its cost. Estimates indicate potentially achievable net sequestration rates of 1–3 GtСО2 yr-
1
41 for iron fertilisation, translating into cumulative CDR of 100–300 GtCO2 by 2100 (Ryaboshapko and
42 Revokatova 2015; Minx et al. 2018), whereas OF with macronutrients has a higher theoretical potential
43 of 5.5 GtCO2 yr-1 (Harrison 2017; Gattuso et al. 2021). Modelling studies show a maximum effect on
44 atmospheric CO2 of 15–45 ppmv in 2100 (Zeebe and Archer 2005; Aumont and Bopp 2006; Keller et
45 al. 2014; Gattuso et al. 2021).
Do Not Cite, Quote or Distribute 12-48 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Risks and impacts: Several of the mesoscale iron enrichment experiments have seen the emergence of
2 potentially toxic species of diatoms (Silver et al. 2010; Trick et al. 2010). There is also (limited)
3 evidence of increased concentrations of other GHGs such as methane and nitrous oxide during the
4 subsurface decomposition of the sinking particles from iron-stimulated blooms (Law 2008). Impacts on
5 marine biology and food web structure are not well known, however OF at large scale could cause
6 changes in nutrient distributions or anoxia in subsurface water (Fuhrman and Capone 1991; DFO 2010).
7 Other potential risks are perturbation to marine ecosystems via reorganisation of community structure,
8 enhanced deep ocean acidification (Oschlies et al. 2010) and effects on human food supply.
9 Co-benefits: Co-benefits of OF include a potential increase in fish biomass through enhanced biological
10 production (Minx et al. 2018) and reduced ocean acidification in the short term in the upper ocean (by
11 CO2 removal), though it could be enhanced in the long term in the ocean interior (by CO2 release)
12 (Oschlies et al., 2010; Gattuso et al. 2018).
13 Trade-offs and spill-over effects: Potential drawbacks include subsurface ocean acidification and
14 deoxygenation (Oschlies et al., 2010; Cao and Caldeira 2010; Williamson et al. 2012); altered regional
15 meridional nutrient supply and fundamental alteration of food webs (GESAMP 2019); and increased
16 production of N2O and CH4 (Jin and Gruber 2003; Lampitt et al. 2008). Ocean fertilisation is considered
17 to have negative consequences for eight SDGs, and a combination of both positive and negative
18 consequences for seven SDGs (Honegger et al. 2020).
19 Ocean Alkalinity enhancement (OAE). CDR through ‘ocean alkalinity enhancement’ or ‘artificial
20 ocean alkalinisation’ (Renforth and Henderson 2017) can be based on: 1) the dissolution of natural
21 alkaline minerals that are added directly to the ocean or coastal environments; 2) the dissolution of such
22 minerals upstream from the ocean (e.g., ‘enhanced weathering’, Section 12.3.1.2); 3) the addition of
23 synthetic alkaline materials directly to the ocean or upstream; and 4) electrochemical processing of
24 seawater. In the case of 2), minerals are dissolved on land and the dissolution products are conveyed to
25 the ocean through runoff and river flow. These processes result in chemical transformation of CO2 and
26 sequestration as bicarbonate and carbonate ions (HCO3-, CO32-) in the ocean. Imbalances between the
27 input and removal fluxes of alkalinity can result in changes in global oceanic alkalinity and therefore
28 the capacity of the ocean to store C. Such alkalinity-induced changes in partitioning of C between
29 atmosphere and ocean are thought to play an important role in controlling climate change on timescales
30 of 1000 years and longer (e.g., Zeebe 2012). The residence time of dissolved inorganic carbon in the
31 deep ocean is around 100,000 years. However, residence time may decrease if alkalinity is reduced by
32 a net increase in carbonate minerals by either increased formation (precipitation) or reduced dissolution
33 of carbonate (Renforth and Henderson 2017). The alkalinity of seawater could potentially also be
34 increased by electrochemical methods, either directly by reactions at the cathode that increase the
35 alkalinity of the surrounding solution that can be discharged into the ocean, or by forcing the
36 precipitation of solid alkaline materials (e.g., hydroxide minerals) that can then be added to the ocean
37 (e.g., Rau et al. 2013; La Plante et al. 2021).
38 Status: OAE has been demonstrated by a small number of laboratory experiments (in addition to
39 enhanced weathering, Section 12.3.1.2). The use of enhanced ocean alkalinity for C storage was first
40 proposed by Kheshgi (1995) who considered the creation of highly reactive lime that would readily
41 dissolve in the surface ocean and sequester CO2. An alternative method proposed the dissolution of
42 carbonate minerals (e.g., CaCO3) in the presence of waste flue gas CO2 and seawater as a means
43 capturing CO2 and converting it to bicarbonate ions (Rau and Caldeira 1999; Rau 2011). House et al.
44 (2007) proposed the creation of alkalinity in the ocean through electrolysis. The fate of the stored carbon
45 is the same for these proposals (i.e., HCO3- and CO32- ions), but the reaction pathway is different.
46 Enhanced weathering of silicate minerals such as olivine could add alkalinity to the ocean, for example,
Do Not Cite, Quote or Distribute 12-49 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 by placing olivine sand in coastal areas (Montserrat et al. 2017; Meysman and Montserrat 2017). Some
2 authors suggest use of maritime transport to discharge calcium hydroxide (slaked lime, SL) (Caserini
3 et al. 2021).
4 Costs: Techno-economic assessments of OAE largely focus on quantifying overall energy and carbon
5 balances. Cost ranges are 40–260 USD tCO2-1 (Fuss et al. 2018). Considering life cycle carbon and
6 energy balances for various OA options, adding lime (or other reactive calcium or magnesium
7 oxide/hydroxides) to the ocean would cost 64–260 USD tCO2-1 (Renforth et al. 2013; Renforth &
8 Kruger 2013; Caserini et al. 2019). Rau (2008) and Rau et al. (2018) estimate that electrochemical
9 processes for increasing ocean alkalinity may have a net cost of 3–160 USD tCO2-1, largely depending
10 on energy cost and co-product (H2) market value. In the case of direct addition of alkaline minerals to
11 the ocean (i.e., without calcination), the cost is estimated to be 20–50 USD tCO2-1 (Harvey 2008; Köhler
12 et al. 2013; Renforth and Henderson 2017).
13 Potentials: For OAE, the ocean theoretically has the capacity to store thousands of GtCO2
14 (cumulatively) without exceeding pre-industrial levels of carbonate saturation (Renforth and Henderson
15 2017) if the impacts were distributed evenly across the surface ocean. The potential of increasing ocean
16 alkalinity may be constrained by the capability to extract, process, and react minerals (Section 12.3.1.2);
17 the demand for co-benefits (see below), or to minimise impacts around points of addition. Important
18 challenges with respect to the detailed quantification of the CO2 sequestration efficiency include
19 nonstoichiometric dissolution, reversed weathering and potential pore water saturation in the case of
20 adding minerals to shallow coastal environments (Meysman and Montserrat 2017). Fuss et al. (2018)
21 suggest storage potentials of 1–100 GtCO2 yr-1. (González and Ilyina 2016) suggested that addition of
22 114 Pmol of alkalinity to the surface ocean could remove 3400 GtCO2 from the atmosphere.
23 Risks and impacts: For OAE, the local impact of increasing alkalinity on ocean chemistry can depend
24 on the speed at which the impacted seawater is diluted/circulated and the exchange of CO 2 from the
25 atmosphere (Bach et al. 2019). Also, more extreme carbonate chemistry perturbations due to non-
26 equilibrated alkalinity could affect local marine biota (Bach et al. 2019), although biological impacts
27 are largely unknown . Air-equilibrated seawater has a much lower potential to perturb seawater
28 carbonate chemistry. However, seawater with slow air-sea gas exchange, in which alkalinity increases,
29 consumes CO2 from the surrounding water without immediate replenishment from the atmosphere,
30 which would increase seawater pH and saturation states and may impact marine biota (Meysman and
31 Montserrat 2017; Montserrat et al. 2017). It may be possible to use this effect to ameliorate ocean
32 acidification. Like enhanced weathering, some proposals may result in the dissolution products of
33 silicate minerals (e.g., Si, Fe, K, Ni) being supplied to ocean ecosystems (Meysman and Montserrat
34 2017; Montserrat et al. 2017). Ecological and biogeochemical consequences of OA largely depend on
35 the minerals used. When natural minerals such as olivine are used, the release of additional Si and Fe
36 could have fertilising effects (Bach et al. 2019). In addition to perturbations to marine ecosystems via
37 reorganisation of community structure, potentially adverse effects of OA that should be studied include
38 the release of toxic trace metals from some deposited minerals (Hartmann et al. 2013).
39 Co-benefits: Intentional addition of alkalinity to the oceans through OAE would decrease the risk to
40 ocean ecosystems caused by the CO2-induced impact of ocean acidification on marine biota and the
41 global carbon cycle (Doney et al. 2009; Köhler et al. 2010; Rau et al. 2012; Williamson and Turley
42 2012; Albright et al. 2016; Bach et al. 2019). OA could be jointly implemented with enhanced
43 weathering (see section 12.3.1.2), spreading the finely crushed rock in the ocean rather than land.
44 Regional alkalinisation could be effective in protecting coral reefs against acidification (Feng et al.
45 2016) (Mongin et al., 2021) and coastal OA could be part of a broader strategy for geochemical
Do Not Cite, Quote or Distribute 12-50 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 management of the coastal zone, safeguarding specific coastal ecosystems from the adverse impact of
2 ocean acidification, such as important shellfisheries (Meysman and Montserrat 2017).
3 Trade-offs and spill-over effects: There is a paucity of research on biological effects of alkalinity
4 addition. The very few studies that have explored the impact of elevated alkalinity on ocean ecosystems
5 have largely been limited to single species experiments (Cripps et al. 2013; Gore et al. 2019) and a
6 constrained field study quantifying the net calcification response of a coral reef flat to alkalinity
7 enhancement (Albright et al. 2016). The addition rate would have to be great enough to overcome
8 mixing of the local seawater with the ambient environment, but not sufficient to detrimentally impact
9 ecosystems. More research is required to assess locations in which this may be feasible, and how such
10 a scheme may operate (Renforth and Henderson 2017). The environmental impact of large-scale release
11 of natural dissolution products into the coastal environment will strongly depend on the scale of olivine
12 application, the characteristics of the coastal water body (e.g., residence time) and the particular biota
13 present (e.g., coral reefs will react differently compared with seagrasses) (Meysman and Montserrat
14 2017). Model simulations (González et al. 2018) suggest that termination of OA implemented on a
15 massive scale under a high CO2 emission scenario (RCP8.5) might pose high risks to biological systems
16 sensitive to rapid environmental changes because it would cause a sharp increase in ocean acidification.
17 For example, OA termination would lead to a decrease in surface pH in warm shallow regions where
18 vulnerable coral reefs are located, and a drop in the carbonate saturation state. However, other studies
19 with lower levels of OA have shown no termination effect (Keller et al., 2014).
20 Blue carbon management. The term “blue carbon” was used originally to refer to biological carbon
21 sequestration in all marine ecosystems, but it is increasingly applied to CDR associated with rooted
22 vegetation in the coastal zone, such as tidal marshes, mangroves and seagrasses. Potential for carbon
23 sequestration in other coastal and non-coastal ecosystems, such as macroalgae (e.g., kelp), is debated
24 (Krause-Jensen et al., 2018; Krause-Jensen and Duarte, 2016). In this report, blue carbon refers to CDR
25 through coastal blue carbon management.
26 Status: In recent years, there has been increasing research on the potential, effectiveness, risks, and
27 possibility of enhancing CO2 sequestration in shallow coastal ecosystems (Duarte, 2017). About 20%
28 of the countries that are signatories to the Paris Agreement refer to blue carbon approaches for climate
29 change mitigation in their NDCs and are moving toward measuring blue carbon in inventories. About
30 40% of those same countries have pledged to manage shallow coastal ecosystems for climate change
31 adaptation (Kuwae and Hori 2019).
32 Costs: There are large differences in cost of CDR applying blue carbon management methods between
33 different ecosystems (and at the local level). Median values are estimated as 240, 30,000, and 7,800
34 USD tCO2-1, respectively for mangroves, salt marsh and seagrass habitats (Gattuso et al. 2021).
35 Currently estimated cost effectiveness (for climate change mitigation) is very low (Siikamäki et al.
36 2012; Bayraktarov et al. 2016; Narayan et al. 2016).
37 Potentials: Globally, the total potential carbon sequestration rate through blue carbon CDR is estimated
38 in the range 0.02–0.08 GtCO2 yr-1 (Wilcox et al. 2017; National Academies of Sciences 2019). Gattuso
39 et al. (2021) estimate the theoretical cumulative potential of coastal blue carbon management by 2100
40 to be 95 GtCO2, taking into account the maximum area that can be occupied by these habitats and
41 historic losses of mangroves, seagrass and salt marsh ecosystems.
42 Risks and impacts: For blue carbon management, potential risks relate to the high sensitivity of coastal
43 ecosystems to external impacts associated with both degradation and attempts to increase carbon
44 sequestration. Under expected future warming, sea-level rise and changes in coastal management, blue
45 carbon ecosystems are at risk, and their stored carbon is at risk of being lost (Bindoff et al. 2019).
Do Not Cite, Quote or Distribute 12-51 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Co-benefits: Blue carbon management provides many non-climatic benefits and can contribute to
2 ecosystem-based adaptation, also reducing emissions associated with habitat degradation and loss
3 (Howard et al. 2017; Hamilton and Friess 2018). Shallow coastal ecosystems have been severely
4 affected by human activity; significant areas have already been deforested or degraded and continue to
5 be denuded. These processes are accompanied by carbon emissions. The conservation and restoration
6 of coastal ecosystems, which will lead to increased carbon sequestration, is also essential for the
7 preservation of basic ecosystem services, and healthy ecosystems tend to be more resilient to the effects
8 of climate change.
9 Trade-offs and spill-over effects: Blue carbon management schemes should consist of a mix of
10 restoration, conservation and areal increase, including complex engineering interventions that enhance
11 natural capital, safeguard their resilience and the ecosystem services they provide, and decrease the
12 sensitivity of such ecosystems to further disturbances.
13 Overview of other ocean-based CDR approaches
14 Artificial Upwelling This concept uses pipes or other methods to pump nutrient-rich deep ocean water
15 to the surface where it has a fertilizing effect (see OF section). To achieve CO 2 removal at a Gt
16 magnitude, modelling studies have shown that artificial upwelling would have to be implemented on a
17 massive scale (over 50% of the ocean to deliver maximum rate of 10GtCO2 yr-1 under RCP 8.5)
18 (Oschlies et al., 2010, Keller et al. 2014). Because the deep water is much colder than surface water, at
19 massive scale this could cool the Earth’s surface by several degrees, but the cooling effect would cease
20 as the deeper ocean warms, and would reverse, leading to rapid warming, if the pumping ceased
21 (Oschlies et al., 2010, Keller et al. 2014).
22 Furthermore, the cooling would also severely alter atmospheric circulation and precipitation patterns
23 (Kwiatkowski et al. 2015). Several upwelling approaches have been developed and tested (Pan et al.,
24 2016) and more R&D is underway.
25 Terrestrial biomass dumping There are proposals to sink terrestrial biomass (crop residues or logs)
26 into the deep ocean as a means of sequestering carbon (Strand and Benford 2009). Sinking biochar has
27 also been proposed (Miller and Orton, 2021). Decomposition would be inhibited by the cold and
28 sometimes hypoxic/anoxic environment on the ocean floor, and absence of bacteria that decompose
29 terrestrial lignocellulosic biomass, so storage timescale is estimated at hundreds to thousands of years
30 (Strand and Benford 2009)(Burdige 2005). Potential side-effects on marine ecosystems, chemistry, or
31 circulation have not been thoroughly assessed. Neither have these concepts been evaluated with respect
32 to the impacts on land from enhanced transfer of nutrients and organic matter to the ocean, nor the
33 relative merits of alternative applications of residues and biochar as an energy source or soil amendment
34 (Chapter 7).
35 Marine biomass CDR options Proposals have been made to grow macroalgae (Duarte et al., 2017) for
36 BECCS (N‘Yeurt et al. 2012; Duarte et al. 2013; Chen et al., 2015), to sink cultured macroalgae into
37 the deep sea, or to use marine algae for biochar (Roberts et al., 2015). Naturally growing sargassum
38 has also been considered for these purposes (Bach et al., 2021). Froehlich et al. (2019) found a
39 substantial area of the ocean (ca. 48 million km2) suitable for farming seaweed. N’Yeurt et al. (2012)
40 suggested that converting 9% of the oceans to macroalgal aquaculture could take up 19 GtCO 2 in
41 biomass, generate 12 Gt per annum of biogas, and the CO₂ produced by burning the biogas could be
42 captured and sequestered. Productivity of farmed macroalgae in the open ocean could potentially be
43 enhanced through fertilizing via artificial upwelling (Fan et al., 2020) or through cultivation platforms
44 that dive at night to access nutrient-rich waters below the, often nutrient-limited, surface ocean. If the
Do Not Cite, Quote or Distribute 12-52 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 biomass were sunk, it is unknown how long the carbon would remain in the deep ocean and what the
2 additional impacts would be. Research and development on macroalgae cultivation and use is currently
3 underway in multiple parts of the world, though not necessarily directly focused on CDR.
4 Extraction of CO2 from seawater (with storage) CO2 can be extracted by applying a vacuum, or by
5 purging with a gas low in CO2 (Koweek et al., 2016). CO2 stripping can also be accomplished by
6 acidifying seawater with a mineral acid, or through electrodialysis and electrolysis, to convert
7 bicarbonate ions (HCO3-) to CO2 (Eisaman et al., 2018; Eisaman 2020; Willauer et al., 2017, Digdaya
8 et al., 2020) Sharifian et al., 2021). The removal of CO2 from the ocean surface leads to undersaturation
9 in the water, thus forcing CO2 to move from the atmosphere into the ocean to restore equilibrium.
10 Electrochemical seawater CO2 extraction has been modeled, prototyped, and analyzed from a techno-
11 economic perspective (Eisaman et al., 2012; Willauer et al., 2017; de Lannoy et al., 2018; Eisaman et
12 al., 2018a; Eisaman et al., 2018b).
13 Status, costs, potentials, risk and impacts, co-benefits, trade-offs and spill-over effects and the role in
14 mitigation pathways of ocean-based approaches are summarised in Table 12.6.
15 12.3.1.4 Feasibility assessment
16 Following the framework presented in Section 6.4 and Annex II.11, a multi-dimensional feasibility
17 assessment on the CDR methods covered here is provided in Figure 12.4, taking into account the
18 assessment presented in this section. Both DACCS and EW perform positively on the geophysical and
19 technological dimensions while for ocean-based approaches performance is mixed. There is limited
20 evidence to assess social-cultural, environmental/ecological, and institutional dimensions as the
21 literature is still nascent for DACCS and EW, while these aspects are positive for blue carbon and mixed
22 or negative for ocean fertilization. On the economic dimension, the cost is assessed negatively for all
23 CDR methods.
Do Not Cite, Quote or Distribute 12-53 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.4 Summary of the extent to which different factors would enable or inhibit the deployment of the carbon dioxide methods DACCS, EW, ocean
3 fertilisation and blue carbon management.
4 Blue bars indicate the extent to which the indicator enables the implementation of the CDR method (E) and orange bars indicate the extent to which an indicator is
5 a barrier (B) to the deployment of the method, relative to the maximum possible barriers and enablers assessed. An ‘X’ signifies the indicator is not applicable or
6 does not affect the feasibility of the method, while a forward slash indicates that there is no or limited evidence whether the indicator affects the feasibility of the
7 method. The shading indicates the level of confidence, with darker shading signifying higher levels of confidence. Supplementary Material 12.B provides an
8 overview of the factors affecting the feasibility of CDR methods and how they differ across context (e.g., region), time (e.g., 2030 versus 2050), and scale (e.g., small
9 versus large), and includes a line of sight on which the assessment is based. The assessment methodology is explained in Annex II, Part II, Section 11.
Do Not Cite, Quote or Distribute 12-54 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 12.3.2 Consideration of methods assessed in sectoral chapters; A/R, biochar, BECCS,
3 soil carbon sequestration
4 Status: BECCS, A/R, soil carbon sequestration (SCS) and biochar are land-based biological CDR
5 methods (Smith et al. 2016). BECCS combines biomass use for energy with CCS to capture and store
6 the biogenic carbon geologically (Section 6.4.2.6); A/R and SCS involve fixing atmospheric carbon in
7 biomass and soils, and biochar involves converting biomass to biochar and using it as a soil amendment.
8 These CDR methods can be associated with both co-benefits and adverse side-effects, see Section 7.4,
9 12.5 and (Schleicher et al. 2019; Smith et al. 2019b; Hurlbert et al. 2019; Mbow et al. 2019; Olsson et
10 al. 2019; Smith et al. 2016; Babin et al. 2021; Dooley et al. 2021).
11 Among CDR methods, BECCS and A/R are most commonly selected by IAMs to meet the requirements
12 of likely limiting warming to 2°C or lower. This is partially because of the long lead time required to
13 refine IAMs to include additional methods and update technoeconomic parameters. Currently, few
14 IAMs represent SCS or biochar (Frank et al. 2017). Given the removal potential of SCS and biochar
15 and some potential co-benefits, more efforts should be made to include these methods within IAMs, so
16 that their mitigation potential can be compared to other CDR methods, along with possible co-benefits
17 and adverse side effects (Smith et al. 2016; Rogelj et al. 2018) (Section 12.5).
18 Potential: The technical potential for BECCS by 2050 is estimated at 0.5–11.3 GtCO2-eq yr-1 (Chapter
19 7, Table 7.3). These potentials do not include avoided emissions resulting from the use of heat,
20 electricity and/or fuels provided by the BECCS system, which depends on substitution patterns,
21 conversion efficiencies, and supply chain emissions for the BECCS and substituted energy systems (see
22 Box 7.7 in Chapter 7). The mitigation effect of BECCS also depends on how deployment affects land
23 carbon stocks and sink strength (see section 7.4.4).
24 As detailed in Chapter 7, the technical potential for gross removals realised through A/R in 2050 is 0.5–
25 10.1 GtCO2-eq yr-1 , and for improved forest management the potential is 1–2.1 GtCO2-eq yr-1 (including
26 both CDR and emissions reduction). Technical potential for SCS in 2050 is estimated to be 0.6–9.4
27 GtCO2-eq yr-1, for agroforestry it is 0.3–9.4 GtCO2-eq yr-1, and for biochar it is 0.2–6.6 GtCO2-eq yr-1.
28 Peatland and coastal wetland restoration have a technical potential of 0.5–2.1 GtCO2-eq yr-1 in 2050,
29 with an estimated 80% of the potential being CDR. Note that these potentials reflect only biophysical
30 and technological conditions and become reduced when factoring in economic, environmental, socio-
31 cultural and institutional constraints (Table 12.6).
32 Costs: Costs across technologies vary substantially (Smith et al. 2016) and were estimated to be 15–
33 400 USD tCO2-1 for BECSS, 0–240 USD tCO2-1 for A/R, -45–100 USD tCO2-1 for SCS and10–345 USD
34 tCO2-1 for biochar. Fuss et al. (2018), estimated abatement cost ranges for BECCS, A/R, SCS and
35 biochar to be 100–200, 5–50, 0–100, and 30–120 tCO2-eq−1 respectively, corresponding to 2100
36 potentials. Ranges for economic potential (<100 USD tCO2-1) reported in Chapter 7 are 0.5–3.0 GtCO2
37 yr-1 (A/R); 0.6–1.9 GtCO2 yr-1 (improved forest management); 0.7–2.5 GtCO2 yr-1 (SCS); 0.4–1.1
38 GtCO2 yr-1 (agroforestry); 0.3–1.8 GtCO2 yr-1 (biochar); 0.2–0.8 GtCO2 yr-1 (peatland and coastal
39 wetland restoration).
40 Risks, impacts, and co-benefits: a brief summary of risks, impacts and co-benefits is provided here and
41 more detail is provided in chapter 7 and Section 12.5. A/R and biomass production for BECCS and
42 biochar potentially compete for land, water and other resources, implying possible adverse outcomes
43 for ecosystem health, biodiversity, livelihoods and food security (medium evidence, high agreement)
44 Smith et al. 2016; Heck et al. 2018; Hurlbert et al. 2019; Mbow et al. 2019) (Chapter 7). SCS requires
Do Not Cite, Quote or Distribute 12-55 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 addition of nitrogen and phosphorus to maintain stoichiometry of soil organic matter, leading to a
2 potential risk of eutrophication (Fuss et al. 2018). Apart from possible negative effects associated with
3 biomass supply, adverse side-effects from biochar are relatively low if the biomass is uncontaminated
4 (Tisserant and Cherubini 2019).
5 Possible climate risks relate to direct and/or indirect land carbon losses (A/R, BECCS, biochar),
6 increased N2O emissions (BECCS, SCS), saturation and non-permanence of carbon storage (A/R, SCS)
7 (Jia et al. 2019; Smith et al. 2019b) (Chapter 7), and potential CO2 leakage from deep geological
8 reservoirs (BECCS) (Chapter 6). Land cover change associated with A/R and biomass supply for
9 BECCS and biochar may cause albedo changes that reduce mitigation effectiveness (Jia et al. 2019;
10 Fuss et al. 2018). Potentially unfavourable albedo change resulting from biochar use can be minimised
11 by incorporating biochar into the soil (Fuss et al. 2018)(Chapter 7)
12 Concerning co-benefits, A/R and biomass production for BECCS or biochar could improve soil carbon,
13 nutrient and water cycling (robust evidence, high agreement), and contribute to market opportunities,
14 employment and local livelihoods, economic diversification, energy security, and technology
15 development and transfer (medium evidence, high agreement) (Chapter 7)(Fuss et al. 2018). It may
16 contribute to reduction of other air pollutants, health benefits, and reduced dependency on imported
17 fossil fuels. A/R can improve biodiversity if native and diverse species are used, and (Fuss et al. 2018).
18 For biochar, additional co-benefits include increased crop yields and reduced drought impacts, reduced
19 CH4 and N2O emissions from soils (Section 7.4.5.2) (Joseph et al., 2021). SCS can improve soil quality
20 and resilience and improve agricultural productivity and food security (Frank et al. 2017; Smith et al.
21 2019c).
22 Role in Mitigation Pathways: Biomass use for BECCS in 2050 is 61 EJ yr-1 (13–208 EJ yr-1, 5-95
23 percentile range) in scenarios limiting warming to 1.5˚C with no or limited overshoot (C1, excluding
24 traditional energy). This corresponds to 5.3 GtCO2 yr-1 (1.1-18 GtCO2 yr-1) CDR, if assuming 28 kg C
25 GJ-1 biomass carbon content and 85% capture rate in BECCS systems. In scenarios likely to limit
26 warming to 2˚C (C3), biomass use for BECCS in 2050 is 28 EJ yr-1 (0–96 EJ yr-1, 5-95 percentile
27 range), corresponding to 2.4 GtCO2 yr-1 (0-8.3 GtCO2 yr-1) CDR. Cumulative net CO2 removals on
28 managed land (CDR through A/R minus land C losses due to deforestation) in the period 2020-2100 is
29 262 GtCO2 (17–397 GtCO2) and 209 GtCO2 (20–415 GtCO2) in C1 and C3 scenarios, respectively (5-
30 95 percentile range).
31 Uncertainties remain in two main areas: the availability of land and biomass, which is affected by many
32 factors (see Chapter 7) (Anandarajah et al. 2018), and the role of other mitigation measures including
33 CDR methods other than A/R and BECCS. Strong near-term climate change mitigation to limit
34 overshoot, and deployment of other CDR methods than A/R and BECCS, may significantly reduce the
35 contribution of these CDR methods in scenarios limiting warming to 1.5˚C or 2˚C (Köberle 2019;
36 Hasegawa et al. 2021).
37 Trade-offs and spill-overs: Some land-based biological CDR methods, such as BECCS and A/R,
38 demand land. Combining mitigation strategies has the potential to increase overall carbon sequestration
39 rates (Humpenöder et al. 2014). However, these CDR methods may also compete for resources (Frank
40 et al. 2017). Land-based mitigation approaches currently propose the use of forests (i) as a source of
41 woody biomass for bioenergy and various biomaterials, and (ii) for carbon sequestration in vegetation,
42 soils, and forest products. Forests are therefore required to provide both provisioning (biomass
43 feedstock) and regulating (carbon sequestration) ecosystem services. This multifaceted strategy has the
44 potential to result in trade-offs (Makkonen et al. 2015). Some land-based mitigation options could
45 conflict with biodiversity goals, e.g., A/R using monoculture plantations can reduce species richness
Do Not Cite, Quote or Distribute 12-56 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 when introduced into (semi-)natural grasslands (Smith et al. 2019a; Dooley et al. 2021). When trade-
2 offs exist between biodiversity protection and mitigationobjectives, biodiversity is typically given a
3 lower priority, especially if the mitigation option is considered risk-free and economically feasible
4 (Pörtner et al. 2021). Approaches that promote synergies, such as sustainable forest management
5 (SFM), reducing deforestation rates, cultivation of perennial crops for bioenergy in sustainable farming
6 practices, and mixed-species forests in A/R, can mitigate biodiversity impacts and even improve
7 ecosystem capacity to support biodiversity while mitigating climate change (Pörtner et al. 2021)
8 (Section 12.5). Systematic land-use planning could help to deliver land-based mitigation options that
9 also limit trade-offs with biodiversity (Longva et al. 2017) (Cross-Working Group Box 3: Mitigation
10 and Adaptation via the Bioeconomy).
11 Status, costs, potentials, risk and impacts, co-benefits, trade-offs and spill-over effects and the role in
12 mitigation pathways of A/R, biochar, SCS, peatland and coastal wetland restoration, agroforestry and
13 forest management are summarised in Table 12.6. See also 12.5.
Do Not Cite, Quote or Distribute 12-57 Total pages: 220
Second Order Draft Chapter 12 IPCC AR6 WGIII
1 Table 12.6 Summary of status, costs, potentials, risk and impacts, co-benefits, trade-offs and spill over effects and the role in mitigation pathways for CDR
2 methods. TRL = Technology Readiness Level. Author judgement ranges (assessed by authors in the literature) are shown, with full literature ranges shown in
3 brackets
CDR option Status Cost (USD Mitigation Risk & Impacts Co-benefits Trade-offs and spill Role in Section
(TRL) tCO ) 2 Potential
-1
over effects modelled
(GtCO yr 2
-
mitigation
) 1
pathways
DACCS 6 100–300 5–40 Increased energy and water Water produced (solid Potentially increased In a few IAMs; {12.3.1.1}
(84–386) use. sorbent DAC designs emissions from water DACCS
only). supply and energy complements
generation. other CDR
methods.
Enhanced weathering 3–4 50–200 2–4 (<1– Mining impacts; air quality Enhanced plant growth, Potentially increased In a few IAMs; {12.3.1.2}
(24–578) 95) impacts of rock dust when reduced erosion, emissions from water EW
spreading on soil. enhanced soil carbon, supply and energy complements
reduced pH, soil water generation. other CDR
retention. methods.
Ocean alkalinity 1–2 40–260 1–100 Increased seawater pH and Limiting ocean Potentially increased No data. {12.3.1.3}
enhancement saturation states and may acidification. emissions of CO2 and
impact marine biota. Possible dust from mining,
release of nutritive or toxic transport and
elements and compounds. deployment
Mining impacts. operations.
Ocean fertilisation 1–2 50–500 1–3 Nutrient redistribution, Increased productivity Subsurface ocean No data. {12.3.1.3}
restructuring of the and fisheries, reduced acidification,
ecosystem, enhanced oxygen upper ocean acidification. deoxygenation;
consumption and acidification altered meridional
in deeper waters, potential for supply of macro-
decadal-to-millennial-scale nutrients as they are
return to the atmosphere of utilized in the iron-
nearly all the extra carbon fertilized region and
become unavailable
for transport to, and
Do Not Cite, Quote or Distribute 12-58 Total pages: 220
Second Order Draft Chapter 12 IPCC AR6 WGIII
removed, risks of unintended utilization in other
side effects. regions, fundamental
alteration of food
webs, biodiversity
Blue carbon management in 2–3 Insufficient <1 If degraded or lost, coastal Provide many non- If degraded or lost, Not {12.3.1.3},
coastal wetlands data, blue carbon ecosystems are climatic benefits and can coastal blue carbon incorporated in Chapter 7,
estimates likely to release most of their contribute to ecosystem- ecosystems are likely IAMs, but in Section
range from carbon back to the based adaptation, coastal to release most of some bottom-up 7.4
~ 100 to ~ atmosphere; potential for protection, increased their carbon back to studies: small
10000 sediment contaminants, biodiversity, reduced the atmosphere. The contribution.
toxicity, bioaccumulation and upper ocean acidification; full delivery of the
biomagnification in could potentially benefit benefits at their
organisms; issues related to human nutrition or maximum global
altering degradability of produce fertiliser for capacity will require
coastal plants; use of subtidal terrestrial agriculture, years to decades to be
areas for tidal wetland carbon anti-methanogenic feed achieved
removal; effect of shoreline additive, or as an
modifications on sediment industrial or materials
redeposition and natural feedstock.
marsh accretion; abusive use
of coastal blue carbon as
means to reclaim land for
purposes that degrade
capacity for carbon removal.
BECCS 5–6 15–400 0.5–11 Competition for land and Reduction of air Competition for land Substantial Chapter 7,
water resources, to grow pollutants; fuel security, with biodiversity contribution in Section
biomass feedstock. optimal use of residues, conservation and food IAMs and 7.4
Biodiversity and carbon stock additional income, health production bottom -up
loss if from unsustainable benefits and if sectoral studies
biomass harvest. implemented well can
enhance biodiversity, soil
health and land carbon
Do Not Cite, Quote or Distribute 12-59 Total pages: 220
Second Order Draft Chapter 12 IPCC AR6 WGIII
Afforestation/Reforestation 8–9 0–240 0.5–10 Reversal of carbon removal Enhanced employment Inappropriate Substantial Chapter 7,
through wildfire, disease, and local livelihoods, deployment at large contribution in Section
pests may occur. improved biodiversity, scale can lead to IAMs and also 7.4
Reduced catchment water improved renewable competition for land in bottom-up
yield and lower groundwater wood products provision, with biodiversity sectoral studies.
level if species and biome are soil carbon and nutrient conservation and food
inappropriate. cycling. Possibly less production.
pressure on primary
forest.
Biochar 6–7 10–345 0.3–6.6 Particulate and GHG Increased crop yields and Environmental In development Chapter 7,
emissions from production; reduced non- impacts associated - not yet in Section
biodiversity and carbon stock CO emissions from soil; particulate matter;
2 global 7.4
loss from unsustainable and resilience to drought. competition for mitigation
biomass harvest. biomass resource. pathways
simulated by
IAMs.
Soil Carbon 8–9 45–100 0.6–9.3 Risk of increased nitrous Improved soil quality, Attempts to increase In development Chapter 7,
Sequestration in croplands oxide emissions due to higher resilience and agricultural carbon sequestration - not yet in Section
and grasslands levels of organic nitrogen in productivity. potential at the global 7.4
the soil; risk of reversal of expense of mitigation
carbon sequestration. production. Net pathways
addition per hectare is simulated by
very small; hard to IAMs; in
monitor. bottom-up
studies: with
medium
contribution.
Peatland and coastal 8–9 Insufficient 0.5–2.1 Reversal of carbon removal in Enhanced employment Competition for land Not in IAMs but Chapter 7,
wetland restoration data drought or future disturbance. and local livelihoods, for food production some bottom-up Section
Risk of increased methane increased productivity of on some peatlands studies with 7.4
emissions. fisheries, improved used for food medium
biodiversity, soil carbon production. contribution.
and nutrient cycling.
Do Not Cite, Quote or Distribute 12-60 Total pages: 220
Second Order Draft Chapter 12 IPCC AR6 WGIII
Agroforestry 8–9 Insufficient 0.3–9.4 Risk that some land area lost Enhanced employment Some trade-off with No data from Chapter 7,
data from food production; and local livelihoods, agricultural crop IAMs, but in Section
requires high skills. variety of products production, but bottom-up 7.4
improved soil quality, enhanced sectoral studies.
more resilient systems. biodiversity, and with medium
resilience of system. contribution.
Improved Forest 8–9 Insufficient 0.1–2.1 If improved management is In case of sustainable If it involves No data from Chapter 7,
management data understood as merely forest management, it increased fertiliser use IAMs, but in Section
intensification involving leads to enhanced and introduced bottom-up 7.4
increased fertiliser use and employment and local species it could sectoral studies
introduced species, then it livelihoods, enhanced reduce biodiversity with medium
could reduce biodiversity and biodiversity, improved and increase contribution.
increase eutrophication. productivity. eutrophication and
upstream GHG
emissions.
1
2
Do Not Cite, Quote or Distribute 12-61 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 12.3.3 CDR governance and policies
2 As shown in Cross-Chapter Box 8 in this Chapter, CDR fulfils multiple functions in different
3 phases of ambitious mitigation: (1) further reducing net CO2 or GHG emission levels in the
4 near-term; (2) counterbalancing residual emissions (from hard-to-transition sectors like
5 transport, industry, or agriculture) to help reach net zero CO2 or GHG emissions in the mid-
6 term; (3) achieving and sustaining net-negative CO2 or GHG emissions in the long-term. While
7 inclusion of emissions and removals on managed land (LULUCF) is mandatory for developed
8 countries under UNFCCC inventory rules (Grassi et al. 2021), not all Annex I countries have
9 included land-based biological removals when setting domestic mitigation targets in the past,
10 but updated NDCs for 2030 indicate a shift, most notably in the European Union (Gheuens and
11 Oberthür 2021; Schenuit et al. 2021). The early literature on CDR governance and policy has
12 been mainly conceptual rather than empirical, focusing on high-level principles (see the
13 concerns listed in introduction of Section 12.3) and the representation of CDR in global
14 mitigation scenarios (see Section 3.2.2). However, with the widespread adoption of net zero
15 targets and the recognition that CDR is a necessary element of mitigation portfolios to achieve
16 net zero CO2 or GHG emissions, countries with national net zero emissions targets have begun
17 to integrate CDR into modelled national mitigation pathways, increase research, development
18 & demonstration (RD&D) efforts on CDR methods, and consider CDR-specific incentives and
19 policies (Honegger et al. 2021b; Schenuit et al. 2021), (Box 12.1). Nevertheless, this increasing
20 consideration of CDR has not yet extended to net-negative targets and policies to achieve these.
21 While the use of CDR at levels that would lead to net negative CO2 or GHG emissions in the
22 long-term has been assumed in most global mitigation scenarios that limit warming to 1.5°C,
23 net-negative emissions trajectories and BECCS as the main CDR method modelled to achieve
24 these have not been mirrored by corresponding UNFCCC decisions so far (Fridahl 2017;
25 Mohan et al. 2021). Likewise, only a few national long-term mitigation plans or legal acts
26 already entail a vision for net-negative GHG emissions (Buylova et al. 2021), for example
27 Finland, Sweden, Germany and Fiji).
28 For countries with emissions targets aiming for net zero or lower, the core governance question
29 is not whether CDR should be mobilised or not, but which CDR methods governments want to
30 see deployed by whom, by when, at which volumes and in which ways (Bellamy and Geden
31 2019; Minx et al. 2018). The choice of CDR methods and the scale and timing of their
32 deployment will depend on the respective ambitions for gross emission reductions, how
33 sustainability and feasibility constraints are managed, and how political preferences and social
34 acceptability evolve (Bellamy 2018; Forster et al. 2020; Fuss et al. 2020; Waller et al. 2020;
35 Clery et al. 2021; Iyer et al. 2021; Rogelj et al. 2021). As examples of emerging CDR
36 policymaking at (sub-)national levels show, policymakers are beginning to incorporate CDR
37 methods beyond those currently dominating global mitigation scenarios, i.e. BECCS and
38 afforestation/reforestation (Box 12.1) (Bellamy and Geden 2019; Buylova et al. 2021; Schenuit
39 et al. 2021; Uden et al. 2021). CDR policymaking is faced with the need to consider method-
40 specific timescales of CO2 storage, as well as challenges in MRV and accounting, potential co-
41 benefits, adverse side effects, interactions with adaptation and trade-offs with SDGs (Table
42 12.6) (Dooley and Kartha 2018; McLaren et al. 2019; Buck et al. 2020; Honegger et al. 2020;
43 Brander et al. 2021; Dooley et al. 2021; Mace et al. 2021). Therefore, CDR governance and
44 policymaking is expected to focus on responsibly incentivising RD&D and targeted
Do Not Cite, Quote or Distribute 12-62 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 deployment, building on both technical and governance experience with already widely
2 practiced CDR methods like afforestation/reforestation (Lomax et al. 2015; Field and Mach
3 2017; Bellamy 2018; Carton et al. 2020; VonHedemann et al. 2020), as well as learning from
4 two decades of slow-moving CCS deployment (Buck 2021; Martin-Roberts et al. 2021; Wang
5 et al. 2021). For some less well understood methods and implementation options, such as ocean
6 alkalinisation or enhanced weathering, investment in RD&D can help in understanding the
7 risks, rewards, and uncertainties of deployment (Nemet et al. 2018; Fajardy et al. 2019; Burns
8 and Corbett 2020; Goll et al. 2021).
9
10 START BOX 12.1 HERE
11 Box 12.1 Case Study: Emerging CDR policy, research and development in the United
12 Kingdom
13 Climate change mitigation policies in the UK have been motivated since 2008 by a domestic,
14 legally-binding framework. This framework includes a 2050 target for net greenhouse gas
15 emissions, interim targets and an independent advisory body called the Climate Change
16 Committee (Muinzer 2019). It has led successive UK governments to publish mitigation plans
17 to 2050, causing policy to be more forward-looking (Averchenkova et al. 2021).
18 The UK’s targets include emissions and removals from LULUCF. In 2008 the target for 2050
19 was an economy-wide net emissions reduction of at least 80% below 1990 levels. Even the
20 first government plans to achieve this target proposed deployment of removal methods,
21 specifically afforestation and wood in construction, increased soil carbon and BECCS (HM
22 Government 2011).
23 Adoption of the Paris Agreement in 2015 caused the government to change the legislated 2050
24 target to a reduction of at least 100% (i.e. net zero). Since then, removal of CO2 and other
25 greenhouse gases has received greater prominence as a distinct topic. The most recent national
26 plan (published October 2021) proposes deployment not only of the methods mentioned above,
27 but also DACCS, biochar and enhanced weathering. The government has committed to amend
28 accounting of UK targets to include a wider range of removal methods beyond LULUCF, and
29 set a target of 5 MtCO2 yr-1 from methods such as BECCS, DACCS and enhanced weathering
30 by 2030. It is consulting on markets and incentives for deployment, and exploring new
31 requirements for MRV (HM Government 2021).
32 In parallel to these policy developments, the UK funds research into technical, environmental
33 and social aspects of removal (Lezaun et al. 2021). Research on some elements (e.g., forestry,
34 CCS, soils, bioenergy) have been funded for well over a decade, but the first programme
35 dedicated to greenhouse gas removal ran during 2017-2021. This has been followed by two
36 new programmes with greater focus on demonstration, totalling £100m over four years (HM
37 Government 2021). A wide variety of methods is supported in these programmes, covering
38 approaches such as CO2 capture from seawater and capture of methane from cattle, in addition
39 to those included already in national mitigation scenarios.
Do Not Cite, Quote or Distribute 12-63 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Deployment of removal methods has lagged expectations, as national targets for tree planting
2 are not being met and infrastructure for CO2 transport and storage is not yet in place (Climate
3 Change Committee 2021). While public awareness around carbon removal is low, studies
4 indicate support in general, provided it is perceived as enhancing rather than impeding action
5 to reduce emissions (Cox et al. 2020a).
6 END BOX 12.1 HERE
7 Since the enhancement of carbon sinks is a form of climate change mitigation (Honegger et al. 2021a),
8 CDR governance challenges will in many respects be similar to those around emissions reduction
9 measures, as will policy instruments like RD&D funding, carbon pricing, tax or investment credits,
10 certification schemes, and public procurement (see Sections 13.4, 13.6, 14.4, 14.5). Effectively
11 integrating CDR into mitigation portfolios can build on already existing rules, procedures and
12 instruments for emissions abatement (Torvanger 2019; Fridahl et al. 2020; Zakkour et al. 2020;
13 Honegger et al. 2021b; Mace et al. 2021; Rickels et al. 2021). Additionally, to accelerate RD&D and to
14 incentivise CDR deployment a political commitment to formal integration into existing climate policy
15 frameworks is required (robust evidence, high agreement) (Lomax et al. 2015; Geden et al. 2018;
16 Honegger and Reiner 2018; VonHedemann et al. 2020; Schenuit et al. 2021) To avoid that CDR is
17 misperceived as a substitute for deep emissions reductions, the prioritisation of emissions cuts can be
18 signalled and achieved with differentiated target setting for reductions and removals (Geden et al. 2019;
19 McLaren et al. 2019). Similarly, sub-targets are conceivable for different types of CDR, to prioritise
20 preferred methods according to characteristics such as removal processes or timescales of storage
21 (Smith 2021).
22 IPCC guidance on quantifying removals is available for land-based biological CDR methods (IPCC
23 2006, 2019), but has yet to be developed for other CDR methods (Royal Society and Royal Academy
24 of Engineering 2018). Challenges with development of estimation algorithms, data collection, and
25 attribution between sectors and countries will need to be overcome (Luisetti et al. 2020; Wedding et
26 al. 2021). Trusted methodologies for MRV, required to enable private sector participation will need to
27 address the permanence, leakage, and saturation challenges with land and ocean-based biological
28 methods (Mace et al. 2021). Protocols that also capture social and ecological co-benefits, could
29 encourage the adoption of biological CDR methods such as SCS, biochar, A/R and blue carbon
30 management (robust evidence, high agreement) (VonHedemann et al. 2020; Macreadie et al. 2021).
31 Private capital and companies, impact investors, and philanthropy will play a role in technical
32 demonstrations and bringing down costs, as well as creating demand for carbon removal products on
33 voluntary markets, which companies may purchase to fulfil corporate social responsibility-driven
34 targets (Friedmann 2019; Fuss et al. 2020; Joppa et al. 2021). Niche markets can provide entry points
35 for limited deployment of novel CDR methods (Cox and Edwards 2019), but targeting currently existing
36 revenue streams by using CO2 captured from the atmosphere in Enhanced Oil Recovery and other
37 utilisation routes (Mackler et al. 2021; Meckling and Biber 2021) is contested, and highlights the
38 importance of choosing appropriate system boundaries when assessing supply chains (Tanzer and
39 Ramírez 2019; Brander et al. 2021). While the private sector will play a distinct role in scaling CDR,
40 governments will need to commit to developing infrastructure for the transport and storage of CO 2,
41 including financing, permitting, and regulating liabilities (Sanchez et al. 2018; Mace et al. 2021;
42 Mackler et al. 2021).
43 International governance considerations include global technology transfer around CDR
44 implementation options (Batres et al. 2021); land use change that could affect food production and land
45 condition, and cause conflict around land tenure and access (Dooley and Kartha 2018; Hurlbert et al.
Do Not Cite, Quote or Distribute 12-64 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 2019; Milne et al. 2019); and efforts to create sustainable and just supply chains for CDR (Fajardy and
2 Mac Dowell 2020; Tan et al. 2021), such as resources used for BECCS, enhanced weathering, or ocean
3 alkalinisation. International governance would be particularly important for methods posing
4 transboundary risks, especially for ocean-based methods. Specific regulations have so far only been
5 developed in the context of the London Protocol, an international treaty that explicitly regulates ocean
6 fertilisation and allows parties to govern other marine CDR methods like ocean alkalinity enhancement
7 (GESAMP 2019; Burns and Corbett 2020; Boettcher et al. 2021)(see Section 14.4.5).
8 Engagement of civil society organisations and publics will be important for shaping CDR policy and
9 deployment (medium evidence, high agreement). Public awareness of CDR and its role in national net
10 zero emissions strategies is generally very low (Cox et al. 2020a), and perceptions differ across
11 countries and between methods (Bertram and Merk 2020; Spence et al. 2021; Sweet et al. 2021; Wenger
12 et al. 2021). When awareness increases, social processes will shape political attitudes on CDR (Shrum
13 et al. 2020), as will efforts to frame particular CDR methods as ‘natural’ or ‘technological’ (Osaka et
14 al. 2021), and the policy instruments chosen to support CDR (Bellamy et al. 2019). Lack of confidence
15 in CDR implementation options from both publics and investors, and lack of trust in project developers
16 (Cox et al. 2020b) have hampered support for CCS (Thomas et al. 2018) and is expected to affect
17 deployment of CDR methods with geological storage (Gough and Mander 2019). On local and regional
18 scales, CDR projects will need to consider air and water quality, impacts to human health, energy needs,
19 land use and ecological integrity, and local community engagement and procedural justice. Bottom-up
20 and community driven strategies are important for deploying equitable carbon removal projects
21 (Hansson et al. 2021; Batres et al. 2021).
22
23 12.4 Food systems
24 12.4.1 Introduction
25 This section complements Chapter 7 by reviewing recent estimates of food system emissions
26 and assessing options beyond the agriculture, forestry and land use sectors to mitigate food
27 systems GHG emissions. A food system approach enables identification of cross-sectoral
28 mitigation opportunities including both technological and behavioural options. Further, a
29 system approach permits evaluation of policies that do not necessarily directly target primary
30 producers or consumers, but other food system actors with possibly higher mitigation
31 efficiency. A food system approach was introduced in the IPCC Special Report on Climate
32 Change and Land (SRCCL) (Mbow et al. 2019). Besides major knowledge gaps in the
33 quantification of food system GHG emissions (Section 12.4.2), the SRCCL authors identified
34 as major knowledge gaps the understanding of the dynamics of dietary change (including
35 behavioural patterns, the adoption of plant-based dietary patterns, and interaction with human
36 health and nutrition of sustainable healthy diets and associated feedbacks); and instruments and
37 mechanisms to accelerate transitions towards sustainable and healthy food systems.
38 Sufficient food and adequate nutrition are fundamental human needs (HLPE 2020; Ingram
39 2020). Food needs to be grown and processed, transported and distributed, and finally prepared
40 and consumed. Food systems range from traditional, involving only few people and short
41 supply chains, to modern food systems, comprising complex webs involving large numbers of
42 stakeholders and processes that grow and transform food commodities into food products and
43 distribute them globally (HLPE 2017; Gómez and Ricketts 2013). A ‘food system’ includes all
Do Not Cite, Quote or Distribute 12-65 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 food chain activities (production, processing, distribution, preparation, consumption of food)
2 and the management of food loss and wastes. It also includes institutions and infrastructures
3 influencing any of these activities, as well as people and systems impacted (HLPE 2017; FAO
4 2018a). Food choices are determined by the food environment consisting of the “physical,
5 economic, political and socio-cultural context in which consumers engage with the food system
6 to acquire, prepare and consume food” (HLPE 2017). Food system outcomes encompass food
7 and nutrition, productivity, profit and livelihood of food producers and other actors in food
8 value chains, but also social outcomes and the impact on the environment (Zurek et al. 2018).
9 ‘Sustainable healthy diets’ have been defined by FAO and WHO (FAO and WHO 2019) as
10 “dietary patterns that promote all dimensions of individuals’ health and wellbeing; have low
11 environmental pressure and impact; are accessible, affordable, safe and equitable; and are
12 culturally acceptable.”
13 The SRCCL estimated overall global anthropogenic emissions from food systems to range
14 between 10.8 and 19.1 GtCO2-eq yr-1, equivalent to 21-37% of total anthropogenic emissions
15 (Rosenzweig et al. 2020a; Mbow et al. 2019). The authors identified major knowledge gaps for
16 the GHG emission inventories of food systems, particularly in providing disaggregated
17 emissions from the food industry and transportation. The food system approach taken in the
18 SRCCL (Mbow et al. 2019) evaluates the synergies and trade-offs of food system response
19 options and its implications for food security, climate change adaptation and mitigation. This
20 integrated framework allows the identification of fundamental attributes of responses to
21 maximise co-benefits, while avoiding maladaptation measures and adverse side effects. A food
22 system approach supports the design of interconnected climate policy responses to tackle
23 climate change, incorporating perspectives of producers and consumers. The SRCCL (Mbow
24 et al. 2019) found that the technical mitigation potential by 2050 of demand-side responses at
25 0.7–8.0 GtCO2-eq yr-1 is comparable to supply-side options at 2.3–9.6 GtCO2-eq yr-1. This
26 shows that mitigation actions need to go beyond food producers and suppliers to incorporate
27 dietary changes and consumers’ behavioural patterns and reveals that producers and consumers
28 need to work together to reduce GHG emissions.
29 Though total production of calories is sufficient for the world population (Wood et al. 2018;
30 Benton et al. 2019), availability and access to food is unequally distributed, and there is a lack
31 of nutrient-dense foods, fruit and vegetables (Berners-Lee et al. 2018; Kc et al. 2018). In 2019,
32 close to 750 million people were food insecure. An estimated 2 billion people lacked adequate
33 access to safe and nutritious food in both quality and quantity (FAO et al. 2020). Two billion
34 adults are overweight or obese through inadequate nutrition, with an upward trend globally
35 (FAO et al. 2019). Low intake of fruit and vegetables is further aggravated by high intake rates
36 of refined grains, sugar and sodium together leading to a high risk of non-communicable
37 diseases such as cardiovascular disease and type 2 diabetes (Springmann et al. 2016; Clark et
38 al. 2018, 2019; GBD 2017 Diet Collaborators et al. 2019; Willett et al. 2019) (robust evidence,
39 high agreement). At least 340 million children under 5 years of age experience lack of vitamins
40 or other essential bio-available nutrients, including almost 200 million suffering from stunting,
41 wasting or overweight (UNICEF 2019).
42 Bodirsky et al. (Bodirsky et al. 2020) find that global prevalence of overweight will increase
43 to 39–52% of world population in 2050 (from 29% in 2010; range across the Socioeconomic
Do Not Cite, Quote or Distribute 12-66 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Pathways studied), and 13–20% obese people (9% in 2010). The prevalence of underweight
2 people was predicted to approximately halve, with absolute numbers stagnating at 0.4–0.7
3 billion. Although many studies represent future pathways of diets and food systems, there are
4 few holistic and consistent narratives and quantification of the future pathways of diets and
5 food systems (Mora et al. 2020; Mitter et al. 2020). Alternative pathways for improved diets
6 and food systems have been developed, emphasising climate, environmental and health co-
7 benefits (Bajželj et al. 2014; Hedenus et al. 2014; Damerau et al. 2016; Weindl et al. 2017a,b;
8 Springmann et al. 2018a; Bodirsky et al. 2020; Prudhomme et al. 2020; Hamilton et al. 2021),
9 reduced food waste and closing yield gaps (Pradhan et al. 2014; Bajželj et al. 2014), nitrogen
10 management (Bodirsky et al. 2014), urban and peri-urban agriculture (Kriewald et al. 2019)
11 and different sustainability targets (Henry et al. 2018b). The FAO has examined three
12 alternative food system scenarios: “business as usual”, “towards sustainability”, and “stratified
13 societies” (FAO 2018b). Others have identified research priorities or changes in legislation
14 needed to support adoption of improved food systems (Mylona et al. 2018).
15 Malnutrition aggravates susceptibility of children to various infectious diseases (Farhadi and
16 Ovchinnikov 2018; França et al. 2009), and infectious diseases can also decrease nutrient
17 uptake, thereby promoting malnutrition (Farhadi and Ovchinnikov 2018). Contamination of
18 food with bacteria, viruses, parasites and microbial toxins can cause foodborne illnesses
19 (Abebe et al. 2020; Ricci et al. 2017; Gallo et al. 2020), foodborne substances such as food
20 additives and specific proteins can cause adverse reactions, and contamination with toxic
21 chemical substances used in agriculture and food processing, can lead to poisoning or chronic
22 diseases (Gallo et al. 2020). Further, health risks from food systems may originate from the use
23 of antibiotics in livestock production and the occurrence of anti-microbial resistance in
24 pathogens (ECDC et al. 2015; Bennani et al. 2020), or zoonotic diseases such as COVID-19
25 (Vågsholm et al. 2020; Gan et al. 2020; Patterson et al. 2020).
26 Modern food systems are highly consolidated, through vertical and horizontal integration
27 (Swinnen and Maertens 2007; Folke et al. 2019). This consolidation has led to uneven
28 distribution of power across the food value chain, with influence concentrated among a few
29 actors in the post-farm gate food supply chain (e.g., large food processors and retailers), and
30 has contributed to a loss of indigenous agriculture and food systems, for example on Pacific
31 Islands (Vogliano et al. 2020). While agricultural producers contribute a higher proportion of
32 GHG emissions compared with other actors in the supply chain, they have relatively little
33 power to change the system (Leip et al. 2021; Clapp 2019; Group of Chief Scientific Advisors
34 2020).
35 In 2016, the agriculture, fisheries, and forestry sectors employed 29% of working people;
36 employment within these sectors was 4% in developed countries, down from 9% in 1995, and
37 57% in least developed countries, down from 71% in 1995 (World Bank 2021). Employment
38 in other (non-agriculture) food system sectors, such as the food processing industry and service
39 sectors, differs between food systems. The share of total non-farm food system employment
40 ranges from 10% in traditional food systems (e.g., Sub-Saharan Africa), to over 50% in food
41 systems in transition (e.g., Brazil), to high shares (80%) in modern food systems (e.g., U.S.)
42 (Townsend et al. 2017). The share of the food expenditures that farmers receive is decreasing;
43 at the global level, this share has been estimated at 27% in 2015 (Yi et al. 2021).
Do Not Cite, Quote or Distribute 12-67 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 12.4.2 GHG emissions from food systems
3 12.4.2.1 Sectoral contribution of GHG emissions from food systems
4 New calculations using EDGAR.v6 (Crippa et al. 2021a) and FAOSTAT (FAO 2021) databases provide
5 territorial-based food system GHG emissions by country globally for the time period 1990 to 2018
6 (Crippa et al. 2021b). The data are calculated based on a combination of country-specific data and
7 aggregated information as described by Crippa et al. (2021b) and Tubiello et al. (2021). The data show
8 that, in 2018, 17 GtCO2-eq yr-1 (95% confidence range 13–23 GtCO2-eq yr-1, calculated according to
9 Solazzo et al. (2020) were associated with the production, processing, distribution, consumption of food
10 and management of food system residues. This corresponded to 31% (range 23-42%) of total
11 anthropogenic GHG emissions of 54 GtCO2-eq yr-1. Based on the IPCC sectoral classification (Table
12 12.7 and Figure 12.5), the largest contribution of food systems GHG emissions in 2018 was from
13 agriculture, i.e. livestock and crop production systems (6.3 GtCO2-eq yr-1, range 2.6–11.9) and land use,
14 land use change and forestry (LULUCF) (4.0 GtCO2-eq yr-1, range 2.1–5.9) (Figure 12.5). Emissions
15 from energy use were 3.9 GtCO2-eq yr-1 (3.6–4.4), waste management 1.7 GtCO2-eq yr-1 (0.9–2.6), and
16 industrial processes and product use 0.9 GtCO2-eq yr-1 (0.6–1.1). The share of GHG emissions from
17 food systems generated outside the AFOLU (agriculture and LULUCF) sectors has increased over
18 recent decades, from 28% in 1990 to 39% in 2018.
19 Energy. Emissions from energy use occur throughout the food supply chain. In 2018, the main
20 contributions came from energy industries supplying electricity and heat (970 MtCO 2-eq yr-1),
21 manufacturing and construction (920 MtCO2-eq yr-1, of which 29% was attributable to the food,
22 beverage, and tobacco industry), and transport (760 MtCO2-eq yr-1). These emissions were almost
23 entirely as CO2. Energy emissions from forestry and fisheries amounted to 480 MtCO2-eq yr-1, with
24 91% of emissions as CO2. Emissions from residential and commercial fuel combustion contributed 250
25 MtCO2-eq yr-1 (79% of emissions as CO2, and with emissions of 1.7 MtCH4 yr-1) and 130 MtCO2-eq yr-
1
26 (with 98% of emissions as CO2), respectively.
27 Refrigeration uses an estimated 43% of energy in the retail sector (Behfar et al. 2018) and significantly
28 increases fuel consumption during distribution. Besides being energy intensive, supermarket
29 refrigeration also contributes to GHG emissions through leakage of refrigerants (F-gases), although
30 their contribution to food system GHG emissions is estimated to be minor (Crippa et al. 2021b). The
31 cold chain accounts for approximately 1% of global GHG emissions, but as the number of refrigerators
32 per capita in developing countries is reported to be one order of magnitude lower than the number in
33 developed countries (19 m3 versus 200 m3 refrigerated storage capacity per 1000 inhabitants), the
34 importance of refrigeration to total GHG emissions is expected to increase (James and James 2010).
35 Although refrigeration gives rise to GHG emissions, both household refrigeration and effective cold
36 chains could contribute to a substantial reduction in losses of perishable food and thus in emissions
37 associated with food provision (University of Birmingham 2018; James and James 2010). A trade-off
38 exists between reducing food waste and increased refrigeration emissions, with the benefits depending
39 on type of produce, location and technologies used (Wu et al. 2019; Sustainable Cooling for All 2018).
40 Transport has overall a minor importance for food system GHG emissions, with a share of 5% to 6%
41 (Poore and Nemecek 2018a; Crippa et al. 2021b). The largest contributor to food system transport GHG
42 emissions was road transport (92%), followed by marine shipping (4%), rail (3%), and aviation (1%).
43 Only looking at energy needs, air or road transport consumes one order of magnitude higher energy
44 (road: 70–80 MJ t-1 km-1; aviation: 100–200 MJ t-1 km-1) than marine shipping (10–20 MJ t-1 km-1) or
Do Not Cite, Quote or Distribute 12-68 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 rail (8–10 MJ t-1 km-1) (FAO 2011). For specific food products with high water content, relatively low
2 agricultural emissions and high average transport distances, the share of transport in total GHG
3 emissions can be over 40% (e.g., bananas, with total global average GHG emissions of 0.7 kgCO2-eq
4 kg-1) (Poore and Nemecek 2018a), but transport is a minor source of GHG emissions for most food
5 products (Poore and Nemecek 2018a).
6 Industry. Direct industrial emissions associated with food systems are generated by the refrigerants
7 industry (580 MtCO2-eq yr-1 as F-gases) and the fertiliser industry for ammonia production (280
8 MtCO2-eq yr-1 as CO2) and nitric acid (60 MtCO2-eq yr-1 as N2O). The industry sector data account for
9 CO2 stored in urea (-50 MtCO2-eq yr-1). Packaging contributed about 6% of total food system emissions
10 (0.98 GtCO2-eq yr-1, 91% as CO2, with CH4 emissions of 2.8 Mt CH4 yr-1). Major emissions sources are
11 pulp and paper (60 MtCO2-eq yr-1) and aluminium (30 MtCO2-eq yr-1), with ferrous metals, glass, and
12 plastics making a smaller contribution. High shares of emissions from packaging are found for
13 beverages and some fruit and vegetables (Poore and Nemecek 2018a).
14 Waste. Management of waste generated in the food system (including food waste, wastewater,
15 packaging waste etc.) leads to biogenic GHG emissions, and contributed 1.7 GtCO2-eq yr-1 to food
16 systems’ GHG emissions in 2018. Of these emissions, 55% were from domestic and commercial
17 wastewater (30 MtCH4 yr-1 and 310 ktN2O yr-1), 36% from solid waste management (20 MtCH4 yr-1
18 and 310 ktN2O yr-1), and 8% from industrial wastewater (4 MtCH4 yr-1 and 80 ktN2O yr-1). Emissions
19 from waste incineration and other waste management systems contributed 1%.
20 Table 12.7: GHG emissions from food systems by sector according to IPCC classification in Mt gas yr -1
21 and food systems’ share of total anthropogenic GHG emissions in 1990 and 2015.
CO2 CH4 N2O F- GHG CO2 CH4 N2O F- GHG
Sector gases gases
Emissions (Mt gas yr-1) Share of total sectoral emissions (%)
1990
1 Energy 2212 10 0 - 2583 10.5 10.2 26.7 - 10.7
2 Industrial Processes 190 0 0 0 263 14.5 0 38 4.8 16.2
3 Solv + Product Use 0 - - - 0 0.2 - - - 0.2
4 Agriculture 102 142 5 - 5370 100 100 99.2 - 99.8
5 LULUCF 4946 - 0 - 5080 181 - 194 - 182
6 Waste 3 40 0 - 1155 29 72.4 99.1 - 73.2
Total 7453 192 6 0 14452 29.3 65.2 84.5 4.8 40.3
Total [MtCO2-eq yr- 7453 5243 1755 0 14452 29.3 63.9 84.5 0.3 40.3
1
]
2018
1 Energy 3449 13 0 - 3927 10.1 9.5 24.1 - 10.2
2 Industrial Processes 242 0 0 0 881 7.9 0 28.6 58 20.1
Do Not Cite, Quote or Distribute 12-69 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
3 Solv + Product Use 7 - - - 7 4.1 - - - 3.6
4 Agriculture 140 161 7 - 6326 100 100 99.1 - 99.7
5 LULUCF 3823 - 1 - 3982 190 - 229 - 191
6 Waste 5 58 0 - 1699 30.6 71.8 99.1 - 72.9
Total 7666 231 8 0 16821 19.3 61.6 83.7 58 31.1
Total [MtCO2-eq yr- 7666 6317 2256 581 16821 19.3 60.2 83.7 53.6 31.1
1
]
1 Notes: Agricultural emissions include the emissions from the whole sector; biomass production for non-food use
2 currently not differentiated. Non-food system AFOLU emissions are a carbon sink, therefore the share of AFOLU
3 food system emissions is > 100%. Source: EDGARv5 (Crippa et al. 2019, 2021b), and FAOSTAT (FAO 2021).
4 Solv+Produc Use = Solvent and Other Product Use; LULUCF: Land Use, Land-Use Change & Forestry.
5
6
7 Figure 12.5: Food system GHG emissions from the agriculture, LULUCF, waste, and energy & industry
8 sectors. Source: Crippa et al. (2021b).
9
10 12.4.2.2 GHG intensities of food commodities
11 There is high variability in the GHG emissions of different food products and production systems
12 (Figure 12.6). GHG emissions intensities – measured using attributional Life Cycle Assessment,
Do Not Cite, Quote or Distribute 12-70 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 considering the full supply-chain, expressed as CO2-eq per kg of product or per kg of protein – are
2 generally highest for ruminant meat, cheese, and certain crustacean species (e.g., farmed shrimp and
3 prawns, trawled lobster) (Nijdam et al. 2012; Clark and Tilman 2017; Clune et al. 2017; Hilborn et al.
4 2018; Poore and Nemecek 2018) (robust evidence, high agreement). Generally, beef from dairy systems
5 has a lower footprint (8–23 kgCO2-eq (100g protein)-1 than beef from beef herds (17–94 kgCO2-eq
6 (100g protein)-1 (Figure 12.6, re-calculated from Poore and Nemecek (2018) using AR6 GWPs based
7 on a 100 year horizon) (medium evidence, high agreement). The wide variation in beef emissions
8 reflects differences in production systems, which range from intensive feedlots with stock raised largely
9 on grains through to rangeland and transhumance production systems. Dairy systems are generally more
10 intensive production systems, with higher digestibility feed than beef systems. Further, emissions from
11 dairy systems are shared between milk and meat, which brings GHG footprints of beef from dairy herds
12 beef closer to those of meat from monogastric animals with emissions intensities of pork (4.4–13
13 kgCO2-eq per 100g protein) and poultry meat (2.3–11 kgCO2-eq per 100g protein) (Poore and Nemecek
14 2018a).
15 Emission intensities for farmed fish ranged from 2.4–11 kgCO2-eq per 100g protein (Poore and
16 Nemecek 2018a). For Norwegian seafood, large differences have been found ranging from 1.1 kgCO2-
17 eq per kg edible product for herring to more than 8 kgCO2-eq per kg edible product for salmon shipped
18 by road and ferry from Oslo to Paris (Winther et al. 2020). For capture fish, large differences in
19 emissions have been found, ranging from 0.2–7.9 kgCO2-eq per kg landed fish (Parker et al. 2018),
20 although an environmental comparison of capture fish to farmed foods should include other indicators
21 such as overfishing. Plant-based foods generally have lower GHG emissions (-2.2–4.5 kgCO2-eq per
22 100g protein) than farmed animal based foods (Clune et al. 2017; Hilborn et al. 2018; Clark and Tilman
23 2017; Nijdam et al. 2012; Poore and Nemecek 2018a) (robust evidence, high agreement). Several plant-
24 based foods are associated with emissions from land use change, for example, palm oil, soy and coffee
25 (Poore and Nemecek 2018a), although emissions intensities are context-specific (Meijaard et al. 2020)
26 and for plant-based proteins, GHG footprints per serving remain lower than those of animal source
27 proteins (Kim et al. 2019).
28 In traditional production systems, especially in developing countries, livestock serve multiple functions,
29 providing draught power, fertiliser, investment and social status, besides constituting an important
30 source of nutrients (Weiler et al. 2014). In landscapes dominated by forests or cropland, semi-natural
31 pastures grazed by ruminants provide heterogeneity that supports biodiversity (Röös et al. 2016).
32 Grazing on marginal land and the use of crop residues and food waste can provide human-edible food
33 with lower demands for cropland (Röös et al. 2016; van Zanten et al. 2018; Van Hal et al. 2019). Animal
34 protein requires more land than vegetable protein, so switching consumption from animal to vegetable
35 proteins could reduce the pressure on land resources and potentially enable additional mitigation
36 through expansion of natural ecosystems, storing carbon while supporting biodiversity, or reforestation
37 to sequester carbon and enhance wood supply capacity for the production of biobased products
38 substituting fossil fuels, plastics, cement, etc. (Searchinger et al. 2018; Schmidinger and Stehfest 2012;
39 Hayek et al. 2021). At the same time, alternatives to animal-based meat and other livestock products
40 are being developed (Figure 12.6). Their increasing visibility in supermarkets and catering services, as
41 well as falling production prices, could make meat substitutes competitive in one to two decades
42 (Gerhardt et al. 2019). However, uncertainty around their uptake creates uncertainty around their effect
43 on future GHG emissions.
Do Not Cite, Quote or Distribute 12-71 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.6: Ranges of GHG intensities [kgCO2-eq per 100 g of protein, 10th-90th percentile] in protein-
3 rich foods, quantified via a meta-analysis of attributional Life Cycle Assessment studies using economic
4 allocation
5 Aggregation of CO2, CH4, and N2O emissions in Poore and Nemecek, (2018) updated to use IPCC-AR6
6 100-year GWP. Data for capture fish, crustaceans, and cephalopods from Parker et al. (2018), with
7 post-farm data from (Poore and Nemecek 2018a), where the ranges represent differences across species
8 groups. CH4 emissions include emissions from manure management, enteric fermentation, and flooded
9 rice only.
10 *Grains are not generally classed as protein-rich, but they provide ~41% of global protein intake. Here
11 grains are a weighted average of wheat, maize, oats, and rice by global protein intake (FAO Food
12 Balance Sheets).
13 $$
Conversion of annual to perennial crops can lead to carbon sequestration in woody biomass and soil,
14 shown as negative emissions intensity.
15 Source: Poore and Nemecek, 2018; Parker et al., 2018
16
17 12.4.2.3 Territorial national per capita GHG emissions from food systems
18 Food systems are connected to other societal systems, such as the energy system, financial system, and
19 transport system (Leip et al. 2021). Also, food systems are dynamic and continuously changing and
20 adapting to existing and anticipated future conditions. Food production systems are very diverse and
21 vary by farm size, intensity level, farm specialisation, technological level, production methods (e.g.,
22 organic, conventional, etc.), with environmental and social consequences (Herrero et al. 2017; Fanzo
23 2017; Václavík et al. 2013; Herrero et al. 2021).
Do Not Cite, Quote or Distribute 12-72 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Various frameworks have been proposed to assess sustainability of food systems, including metrics and
2 indicators on environmental, health, economic and equity issues, pointing to the importance of
3 recognizing the multi-dimensionality of food system outcomes (Béné et al. 2020; Chaudhary et al. 2018;
4 Gustafson et al. 2016; Eme et al. 2019; Hallström et al. 2018; Hebinck et al. 2021; Zurek et al. 2018).
5 Data platforms are being developed, but so far comprehensive data for evidence-based food system
6 policy are lacking (Fanzo et al. 2020).
7 To visualise several food systems dimensions in a GHG context, Figure 12.7 shows GHG emissions
8 per capita and year for regional country aggregates (Crippa et al. 2021a,b), indicated by the size of the
9 bubbles. The GHG emissions presented here are based on territorial accounting similar to the UNFCCC
10 GHG inventories: emissions are assigned to the country where they occur, not where food is consumed
11 (Section 12.4.2.1 and Crippa et al., 2021a, b). The colours of the bubbles indicate the relative
12 contribution of one of the following risk factors to deaths, according to the classification used in the
13 Global Burden of Disease Study: Child and maternal malnutrition (red, deficiencies of iron, zinc or
14 Vitamin A, or low birth weight or child growth failure), Dietary risks (yellow, for example diets low in
15 vegetables, legumes, whole grains or diets high in red and processed meat and sugar-sweetened
16 beverages) or High body-mass index (blue). The combined contribution of these three risk factors to
17 total deaths varies strongly and is between 28% and 88% of total deaths. Figure 12.7 shows that dietary
18 risk factors are prevalent throughout all regions. Though not a complete measure of the health impact
19 of food, these were selected as a proxy for nutritional adequacy and balance of diets, avoidance of food
20 insecurity, over- or mal-nutrition and associated non communicable diseases (GBD 2017 Diet
21 Collaborators 2018; GBD 2017 Diet Collaborators et al. 2019).
22 The data are plotted in a matrix with share of GHG emissions from energy use (Crippa et al. 2021b) on
23 the y-axis and the wholesale cost of food (Springmann et al. 2021) on the x-axis. The share of GHG
24 emissions from energy use is taken as a proxy for the structure of food supply in a region (Section
25 12.4.1), and the cost for food as a proxy for the structure of the demand side and the access to (healthy)
26 food (Chen et al. 2016; Hirvonen et al. 2019; Finaret and Masters 2019; HLPE 2020; Springmann et al.
27 2021), though acknowledging the limitations of such a simplification.
28 While total food system emissions in 2018 range between 0.9 and 8.5 tCO2-eq cap-1 yr-1 between
29 regions, the share of energy emissions relative to energy and land-based (agriculture and food system
30 land use change) emissions ranges between 3% and 78%. Regional expenditures for food range from
31 3.0–8.8 USD cap-1 day-1 (Figure 12.7), though there is high variability within countries and the costs of
32 nutrient-adequate diets often exceeds those of diets delivering adequate energy (Bai et al. 2020;
33 Hirvonen et al. 2019; FAO et al. 2020). Thus, low-income households in industrialised countries can
34 also be affected by food insecurity (Penne and Goedemé 2020).
35
36
Do Not Cite, Quote or Distribute 12-73 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.7: Regional differences in health outcome, territorial per capita GHG emissions from national
3 food systems, and share of food system GHG emission from energy use.
4 GHG emissions are calculated according to the IPCC Tier 1 approach and are assigned to the country
5 where they occur, not necessarily where the food is consumed. Health outcome is expressed as relative
6 contribution of each of the following risk factors to their combined risk for deaths: Child and maternal
7 malnutrition (red), Dietary risks (yellow) or High body-mass index (blue).
8 Source: cost for food (whole sale price) per capita (Springmann et al. 2021); Territorial food system GHG
9 emissions: EDGAR v.6 (Crippa et al. 2021a), recalculated according to Crippa et al. (2021) using AR6-GWPs;
10 Deaths attributed to dietary factors: (IHME 2018; GBD 2017 Diet Collaborators et al. 2019).
11
12 12.4.3 Mitigation opportunities
13 GHG emissions from food systems can be reduced by targeting direct or indirect GHG emissions in the
14 supply chain including enhanced carbon sequestration, by introducing sustainable production methods
15 such as agro-ecological approaches which can reduce system-level GHG emissions of conventional
16 food production and also enhance resilience (HLPE 2019), substituting food products with high GHG
17 intensities with others of lower GHG intensities, by reducing food over-consumption or by reducing
18 food loss and waste. The substitution of food products with others that are more sustainable and/or
19 healthier is often called ‘dietary shift’.
20 Clark et al. (2020) showed that even if fossil fuel emissions were eliminated immediately, food system
21 emissions alone would jeopardize the achievement of the 1.5ºC target and threaten the 2ºC target. They
22 concluded that both demand-side and supply-side strategies are needed, including a shift to a diet with
23 lower GHG intensity and rich in plant-based ‘conventional’ foods (e.g., pulses, nuts), or new food
24 products that could support dietary shift. Such dietary shift needs to overcome socio-cultural,
25 knowledge, and economic barriers to significantly achieve GHG mitigation (Section 12.4.5).
26
27 Food losses occur at the farm, post-harvest and food processing/wholesale stages of a food supply chain,
28 while in the final retail and consumption stages the term food waste is used (HLPE 2014). Typically,
29 food losses are linked to technical issues such as lack of infrastructure and storage while food waste is
Do Not Cite, Quote or Distribute 12-74 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 often caused by socio-economic and behavioural factors. Mitigation opportunities through reducing
2 food waste and loss exist in all food supply chain stages and are described in the sub-sections below.
3 Food system mitigation opportunities are divided into five categories as given in Table 12.8:
4 • Food production from agriculture, aquaculture, and fisheries (Chapter 7.4 and Section 12.4.3.1)
5 • Controlled environment agriculture (Section 12.4.3.2)
6 • Emerging food production technologies (Section 12.4.3.3)
7 • Food processing industries (Section 12.4.3.4)
8 • Storage and distribution (Section 12.4.3.5)
9 Food system mitigation opportunities can be either incremental or transformative (Kugelberg et al.
10 2021). Incremental options are based on mature technologies, for which processes and causalities are
11 understood, and their implementation is generally accepted by society. They do not require a substantial
12 change in the way food is produced, processed, or consumed and might lead to a (slight) shift in
13 production systems or preferences. Transformative mitigation opportunities have wider food system
14 implications and usually coincide with a significant change in food choices. They are based on
15 technologies that are not yet mature and are expected to require further innovation (Klerkx and Rose
16 2020), and/or mature technologies that might already be part of some food systems but are not yet
17 widely accepted and have transformative potential if applied at large scale, e.g. consumption of insects
18 (Raheem et al. 2019a). Many emerging technologies might be seen as a further step in agronomic
19 development where land-intensive production methods relying on the availability of naturally available
20 nutrients and water are successively replaced with crop variants and cultivation practices reducing these
21 dependencies at the cost of larger energy input (Winiwarter et al. 2014). Others suggest a shift to agro-
22 ecological approaches combining new scientific insights with local knowledge and cultural values
23 (HLPE 2019). Food system transformation can lead to regime shifts or (fast) disruptions (Pereira et al.
24 2020) if driven by events that are out of control of private or public measures and have a ‘crisis’
25 character (e.g., BSE, Skuce et al. 2013).
26 Table 12.8 summarises the main characteristics of food system mitigation opportunities, their effect on
27 GHG emissions, and associated co-benefits and adverse effects.
Do Not Cite, Quote or Distribute 12-75 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Table 12.8: Food system mitigation opportunities
2 $ Direct and indirect GHG effects: D – Direct emissions except emissions from energy use, E – Energy demand, M – Material demand, FL – food losses, FW – food waste; direction of effect on
3 GHG mitigation: (+) increased mitigation, (0) neutral, (-) decreased mitigation.
4 & Co-benefits/Adverse effects: H - health aspects, A - Animal welfare, R - resource use, L - Land demand, E – Ecosystem services; (+) co-benefits, (-) adverse effects.
Food system mitigation options (I: Direct and indirect effect on GHG mitigation Co-benefits / Adverse effects & Source
incremental; T: transformative) (+/0/-) $
Food from (I) Dietary shift, in D+ ↓ GHG footprint A+ Animal welfare 1-5
agricultural, particular increased
aquaculture share of plant-based L+ Land sparing
and fisheries protein sources
H+ Good nutritional properties, potentially ↓ risk from zoonotic
diseases, pesticides and antibiotics
(I/T) Digital agriculture D+ ↑ logistics L+ Land sparing 6-7
R+ ↑ resource use efficiencies
(T) Gene technology D+ ↑ productivity or efficiency H+ ↑ nutritional quality 7-11
E0 ↓ use of agrochemicals; ↑ probability of off-target impacts
(I) Sustainable intensific. D+ ↓ GHG footprint L+ Land sparing 7, 12
land use optimisation
E0 Mixed effects R- Might ↑ pollution/biodiversity loss
(I) Agroecology D+ ↓ GHG/area, positive micro-climatic effects E+ Focus on co-benefits/ecosystem services 13-17
E+ ↓ energy, possibly ↓ transport
FL+ Circular approaches R+ Circular, ↑ nutrient and water use efficiencies
Controlled (T) Soilless agriculture D+ ↑ productivity, weather independent R+ Controlled loops ↑ nutrient and water use efficiency 18-24
environment
agriculture FL+ Harvest on demand L+ Land sparing
E- Currently ↑ energy demand, but ↓ transport, H+ Crop breeding can be optimised for taste and/or nutritional quality
building spaces can be used for renewable energy
Do Not Cite, Quote or Distribute 12-76 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
Food system mitigation options (I: Direct and indirect effect on GHG mitigation Co-benefits / Adverse effects & Source
incremental; T: transformative) (+/0/-) $
Emerging (T) Insects D0 Good feed conversion efficiency H0 Good nutritional qualities but attention to allergies and food safety 25-28
Food issues required
Production FW+ Can be fed on food waste
technologies
(I/T) Algae and bivalves D+ ↓ GHG footprints A+ Animal welfare 29-32
L+ Land sparing
H+ Good nutritional qualities; risk of heavy metal and pathogen
contamination
R+ Biofiltration of nutrient-polluted waters
(I/T) Plant-based D+ No emissions from animals, ↓ inputs for feed A+ Animal welfare 31-33
alternatives to animal-
based food products L+ Land sparing
H+ Potentially ↓ risk from zoonotic diseases, pesticides and
antibiotics; but ↑ processing demand
(T) Cellular agriculture D+ No emissions from animals, high protein A+ Animal welfare 3, 24,
(including cultured conversion efficiency
meat, microbial R+ ↓ emissions of reactive nitrogen or other pollutants 34-42
protein) E- ↑ energy need
H0 Potentially ↓ risk from zoonotic diseases, pesticides and
FLW+ ↓ food loss & waste antibiotics; ↑ research on safety aspects needed
Food (I) Valorisation of by- M+ Substitution of bio-based materials 43-44
processing and products, FLW logistics
packaging and management FL+ ↓ of food losses
(I) Food conservation FW+ ↓ of food waste 45-46
E0 ↑ energy demand but also energy savings
possible (e.g., refrigeration, transport)
Do Not Cite, Quote or Distribute 12-77 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
Food system mitigation options (I: Direct and indirect effect on GHG mitigation Co-benefits / Adverse effects & Source
incremental; T: transformative) (+/0/-) $
(I) Smart packaging and FW+ ↓ of food waste H+ Possibly ↑ freshness/reduced food safety risks 46-49
other technologies
M0 ↑ material demand and ↑ material-efficiency
E0 ↑ energy demand; energy savings possible
(I) Energy efficiency E+ ↓ energy 50
Storage and (I) Improved logistics D+ ↓ transport emissions 46-47
distribution
FL+ ↓ losses in transport 51-53
FW- Easier access to food could ↑ food waste
(I) Specific measures to FW+↓ of food waste 54-56
reduce food waste in
retail and food catering E+ ↓ downstream energy demand
M+ ↓ downstream material demand
(I) Alternative D+ ↓ emissions from transport
fuels/transport modes
(I) Energy efficiency E+ ↓ energy in refrigeration, lightening, 57-58
climatisation
(I) Replacing refrigerants D+ ↓ emissions from the cold chain 50,
59-60
1 [1] (McDermott and Wyatt 2017); [2] (Foyer et al. 2016); [3] (Semba et al. 2021); [4] (Weindl et al. 2020); [5] (Hertzler et al. 2020); [6] (Finger et al. 2019); [7] (Herrero et al.
2 2020); [8] Steinwand and Ronald (2020); [9] Zhang et al. (2020a); [10] (Ansari et al. 2020); [11] (Eckerstorfer et al. 2021); [12] (Folberth et al. 2020); [13] (HLPE 2019); [14]
3 (Wezel et al. 2009); [15] (Van Zanten et al. 2018); [16] (van Zanten et al. 2019); [17] (van Hal et al. 2019); [18] (Beacham et al. 2019); [19] (Benke and Tomkins 2017); [20]
4 (Gómez and Gennaro Izzo 2018); [21] (Maucieri et al. 2018); [22] (Rufí-Salís et al. 2020); [23] (Shamshiri et al. 2018); [24] (Graamans et al. 2018); [25] (Fasolin et al. 2019);
5 [26] (Garofalo et al. 2019); [27] (Parodi et al. 2018); [28] (Varelas 2019); [29] (Gentry et al. 2020); [30] (Peñalver et al. 2020); [31] (Torres-Tiji et al. 2020); [32] (Willer and
6 Aldridge 2020); [33] (Fresán et al. 2019); [34] (Mejia et al. 2019); [35] (Tuomisto 2019); [36] (Thorrez and Vandenburgh 2019); [37] (Tuomisto and Teixeira de Mattos 2011);
Do Not Cite, Quote or Distribute 12-78 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 [38] (Mattick et al. 2015); [39] (Mattick 2018); [40] (Souza Filho et al. 2019); [41] Chriki and Hocquette (Chriki and Hocquette 2020); [42] Hadi and Brightwell (Hadi and
2 Brightwell 2021); [43] (Göbel et al. 2015); [44] (Caldeira et al. 2020); [45] (Silva and Sanjuán 2019); [46] (FAO 2019a); [47] (Molina-Besch et al. 2019); [48] (Poyatos-
3 Racionero et al. 2018); [49] (Müller and Schmid 2019); [50] (Niles et al. 2018); [51] (Lindh et al. 2016); [52] (Wohner et al. 2019); [53] (Bajželj et al. 2020); [54]. (Buisman
4 et al. 2019); [55] (Albizzati et al. 2019); [56] (Liu et al. 2016); [57] (Chaomuang et al. 2017); [58] (Lemma et al. 2014); [59] (McLinden et al. 2017); [60] (Gullo et al. 2017).
Do Not Cite, Quote or Distribute 12-79 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 12.4.3.1 Food from agriculture, aquaculture, and fisheries
2 Agricultural food production systems range from smallholder subsistence farms to large animal
3 production factories, in open spaces, greenhouses, rural areas or urban settings.
4 Dietary shift. Studies demonstrate that a shift to diets rich in plant-based foods, particularly pulses, nuts,
5 fruits & vegetables, such as vegetarian, pescatarian or vegan diets, could lead to substantial reduction
6 of greenhouse gas emissions as compared to current dietary patterns in most industrialized countries,
7 while also providing health benefits and reducing mortality from diet-related non-communicable
8 diseases (Ernstoff et al. 2020; Semba et al. 2020; Theurl et al. 2020; Costa Leite et al. 2020; Chen et al.
9 2019; Jarmul et al. 2020; Willett et al. 2019; Bodirsky et al. 2020; Hamilton et al. 2021; Springmann et
10 al. 2018a).
11 Pulses such as beans, chickpeas, or lentils, have a protein composition complementary to cereals,
12 providing together all essential amino acids (McDermott and Wyatt 2017; Foyer et al. 2016). Bio-
13 availability of proteins in foods is influenced by several factors, including amino-acid composition,
14 presence of anti-nutritional factors, and preparation method (Hertzler et al. 2020; Weindl et al. 2020;
15 Semba et al. 2021). Soy beans, in particular, have a well-balanced amino acid profile with high bio-
16 availability (Leinonen et al. 2019). Pulses are part of most traditional diets (Semba et al. 2021) and
17 supply up to 10-35% of protein in low-income countries, but consumption decreases with increasing
18 income and they are globally only a minor share of the diet (McDermott and Wyatt 2017). Pulses play
19 a key role in crop rotations, fixing nitrogen and breaking disease cycles, but yields of pulses are
20 relatively low and have seen small yield increases relative to those of cereals (Barbieri et al. 2021;
21 McDermott and Wyatt 2017; Foyer et al. 2016; Semba et al. 2021).
22 Technological innovations have made food production more efficient since the onset of agriculture
23 (Winiwarter et al. 2014; Herrero et al. 2020). Emerging technologies include digital agriculture (using
24 advanced sensors, big data), gene technology (crop bio-fortification, genome editing, crop innovations),
25 sustainable intensification (automation of processes, improved inputs, precision agriculture) (Herrero
26 et al. 2020), or multi-trophic aquaculture approaches (Sanz-Lazaro and Sanchez-Jerez 2020; Knowler
27 et al. 2020), though literature on aquaculture and fisheries in the context of GHG mitigation is limited.
28 Such technologies may contribute to a reduction of GHG emission at the food system level enhanced
29 provision of food, better consideration of ecosystem services, or contribute to nutrition sensitive
30 agriculture, for example, by increasing the nutritional quality of staple crops, increasing the palatability
31 of leguminous crops such as lupines, or the agronomic efficiency or resilience of crops with good
32 nutritional characteristics.
33 For details on agricultural mitigation opportunities refer to Chapter 7.4.
34 12.4.3.2 Controlled-environment agriculture
35 Controlled-environment agriculture is mainly based on hydroponic or aquaponic cultivation systems
36 that do not require soil. Aquaponics combine hydroponics with a re-circulating aquaculture
37 compartment for integrated production of plants and fish (Junge et al. 2017; Maucieri et al. 2018), while
38 aeroponics is a further development of hydroponics that replaces water as a growing medium with a
39 mist of nutrient solution (Al-Kodmany 2018). Aquaponics could potentially produce proteins in urban
40 farms, but the technology is not yet mature and its economic and environmental performance is unclear
41 (O’Sullivan et al. 2019; Love et al. 2015).
Do Not Cite, Quote or Distribute 12-80 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Controlled-environmental agriculture is often undertaken in urban environments to take advantage of
2 short supply chains (O’Sullivan et al. 2019), and might use abandoned buildings or be integrated in
3 supermarkets, producing for example herbs ‘on demand’.
4 Optimising growing conditions, hydroponic systems achieve higher yields than un-conditioned
5 agriculture (O’Sullivan et al. 2019); and yields can be further enhanced in CO2-enriched atmospheres
6 (Armanda et al. 2019; Shamshiri et al. 2018). By using existing spaces or modular systems that can be
7 vertically stacked, this technology minimises land demand, however it is energy intensive and requires
8 large financial investments. So far, only a few crops are commercially produced in vertical farms,
9 including lettuce and other leafy greens, herbs and some vegetables due to their short growth period
10 and high value (Benke and Tomkins 2017; Beacham et al. 2019; O’Sullivan et al. 2019; Armanda et al.
11 2019). Through breeding, other crops could reach commercial feasibility, or crops with improved taste
12 or nutritional characteristics can be grown (O’Sullivan et al. 2019).
13 In controlled-environment agriculture, photosynthesis is fuelled by artificial light through LEDs or a
14 combination of natural light with LEDs. Control of the wave band and light cycle of the LEDs and
15 micro-climate can be used to optimise photosynthetic activity, yield and crop quality (Gómez and
16 Gennaro Izzo 2018; Shamshiri et al. 2018).
17 Co-benefits of controlled-environment agriculture include minimising water and nutrient losses as well
18 as agro-chemical use (Farfan et al. 2019; Shamshiri et al. 2018; O’Sullivan et al. 2019; Armanda et al.
19 2019; Al-Kodmany 2018; Rufí-Salís et al. 2020) (robust evidence, high agreement). Water is recycled
20 in a closed system and additionally some plants generate fresh water by evaporation from grey or black
21 water, and high nutrient use efficiencies are possible. Food production from controlled-environment
22 agriculture is independent of weather conditions and able to satisfy some consumer demand for locally-
23 produced fresh and diverse produce throughout the year (O’Sullivan et al. 2019; Al-Kodmany 2018;
24 Benke and Tomkins 2017).
25 Controlled-environment agriculture is a very energy intensive technology (mainly for cooling) and its
26 GHG intensity depends therefore crucially on the source of the energy. Options for reducing GHG
27 intensity include reducing energy use through improved lighting and cooling efficiency or by employing
28 low-carbon energy sources, potentially integrated into the building structure (Benke and Tomkins
29 2017).
30 Comprehensive studies assessing the GHG balance of controlled-environment agriculture are lacking.
31 The overall GHG emissions from controlled-environment agriculture is therefore uncertain and depends
32 on the balance of reduced GHG emissions from production and distribution and reduced land
33 requirements, versus increased external energy needs.
34 12.4.3.3 Emerging foods and production technologies
35 A diverse range of novel food products and production systems are emerging, that are proposed to
36 reduce GHG emissions from food production, mainly by replacing conventional animal-source food
37 with alternative protein sources. Assessments of the potential of dietary changes are given in Chapter
38 5.3 and Chapter 7.4. Here, we assess the GHG intensities of emerging food production technologies.
39 This includes products such as insects, algae, mussels and products from bio-refineries, some of which
40 have been consumed in certain societies and/or in smaller quantities (Pikaar et al. 2018; Jönsson et al.
41 2019; Govorushko 2019; Raheem et al. 2019a; Souza Filho et al. 2019). The novel aspect considered
42 here is the scale at which they are proposed to replace conventional food with the aim to reduce both
43 negative health and environmental impact. To fully realize the health benefits, dietary shifts should also
Do Not Cite, Quote or Distribute 12-81 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 encompass a reduction in consumption of added sugars, salt, and saturated fats, and potentially harmful
2 additives (Curtain and Grafenauer 2019; Fardet and Rock 2019; Petersen et al. 2021).
3 Meat analogues have attracted substantial venture capital, and production costs have dropped
4 considerably in the last decade, with some reaching market maturity (Mouat and Prince 2018; Santo
5 et al. 2020), but there is uncertainty whether they will ‘disrupt’ the food market or remain niche
6 products. According to Kumar et al. (2017), the demand for plant-based meat analogues is expected to
7 increase as their production is relatively cheap and they satisfy consumer demands with regard to health
8 and environmental concerns as well as ethical and religious requirements. Consumer acceptance is still
9 low for some options, especially insects (Aiking and de Boer 2019) and cultured meat (Siegrist and
10 Hartmann 2020; Chriki and Hocquette 2020).
11 Insects. Farmed edible insects have a higher feed conversion ratio than other animals farmed for food,
12 and have short reproduction periods with high biomass production rates (Halloran et al. 2016). Insects
13 have good nutritional qualities (Parodi et al. 2018). They are suited as a protein source for both humans
14 and livestock, with high protein contents and favourable fatty acid composition (Raheem et al. 2019b;
15 Fasolin et al. 2019). If used as feed, they can grow on food waste and manure; if used as food, food
16 safety concerns/regulations can restrict the use of manure (Raheem et al. 2019b) or food waste (Varelas
17 2019) as growing substrates, and the dangers of pathogenic or toxigenic microorganisms and incidences
18 of anti-microbial resistance need to be managed (Garofalo et al. 2019).
19 Algae and bivalves have a high protein content and a favourable nutrient profile and can play a role in
20 providing sustainable food. Bivalves are high in omega-3 fatty acids and vitamin B12 and therefore
21 well-suited as replacement of conventional meats, and have a lower GHG footprint (Willer and Aldridge
22 2020; Parodi et al. 2018). Micro- and macro algae are rich in omega-3 and omega-6 fatty acids, anti-
23 oxidants and vitamins (Peñalver et al. 2020; Parodi et al. 2018; Torres-Tiji et al. 2020). Kim et al. (2019)
24 show that diets with modest amounts of low-food chain animals such as forage fish, bivalves, or insects
25 have similar GHG intensities to vegan diets. Algae and bi-valves can be used to filter nutrients from
26 waters, though care is required to avoid accumulation of hazardous substances (Willer and Aldridge
27 2020; Gentry et al. 2020).
28 Plant-based meat, milk and egg analogues. Demand for plant-based proteins is increasing and
29 incentivising the development of protein crop varieties with improved agronomic performance and/or
30 nutritional quality (Santo et al. 2020). There is also an emerging market for meat replacements based
31 on plant proteins, such as pulses, cereals, soya, algae and other ingredients mainly used to imitate the
32 taste, texture and nutritional profiles of animal-source food (Boukid 2021; Kumar et al. 2017).
33 Currently, the majority of plant-based meat analogues is based on soy (Semba et al. 2021). While other
34 products still serve a ‘niche’ market, their share is growing rapidly and some studies project a sizeable
35 share within a decade (Kumar et al. 2017; Jönsson et al. 2019). In particular, plant-based milk
36 alternatives have seen large increases in the market share (Jönsson et al. 2019). A LCA of 56 plant-
37 based meat analogues showed mean GHG intensities (farm to factory) of 0.21–0.23 kgCO2-eq per 100
38 g of product or 20 g of protein for all assessed protein sources (Fresán et al. 2019). Higher footprints
39 were found in the meta-review by Santo et al. (2020). Including preparation, Meija et al. (2019) found
40 higher emissions for burgers and sausages as compared to minced products.
41 Cellular agriculture. The use of fungi, algae and bacteria is an old process (beer, bread, yoghurt) and
42 serves, among others, for the preservation of products. The concept of cellular agriculture (Mattick
43 2018) covers bio-technological processes that use micro-organisms to produce acellular (fermentation
44 based cellular agriculture) or cellular products. Yeasts, fungi or bacteria can synthesise acellular
45 products such as haem, milk and egg proteins, or protein-rich animal feed, other food ingredients, and
Do Not Cite, Quote or Distribute 12-82 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 pharmaceutical and material products (Rischer et al. 2020; Mendly-Zambo et al. 2021). Cellular
2 products include cell tissues such as muscle cells to grow cultured meat, fish or other cells (Post 2012;
3 Rischer et al. 2020) and products where the micro-organisms will be eaten themselves (Pikaar et al.
4 2018; Sillman et al. 2019; Schade et al. 2020). Single cell proteins, combined with photovoltaic
5 electricity generation and direct air capture of carbon dioxide are proposed as highly land- and energy-
6 efficient alternatives to plant-based protein (Leger et al. 2021). Some microbial proteins are produced
7 in a ‘bioreactor’ and use Haber-Bosch nitrogen and vegetable sugars or atmospheric CO2 as source of
8 N and C (Simsa et al. 2019; Pikaar et al. 2018). Cultured meat is currently in the research stage and
9 some challenges remain, such as the need for animal-based ingredients to ensure fast/effective growth
10 of muscle cells, tissue engineering to create different meat products, the production at scale and at
11 competitive costs, and regulatory barriers (Rubio et al. 2019; Stephens et al. 2018; Post et al. 2020; Post
12 2012; Tuomisto 2019). Only a few studies to date have quantified the GHG emissions of microbial
13 proteins or cultured meat, suggesting GHG emissions at the level of poultry meat (Tuomisto and
14 Teixeira de Mattos 2011; Mattick et al. 2015; Souza Filho et al. 2019; Tuomisto 2019).
15 A review of LCA studies on different plant-based, animal source and nine ‘future food’ protein sources
16 (Parodi et al. 2018) concluded that insects, macro-algae, mussels, myco-proteins and cultured meat
17 show similar GHG intensities per unit of protein (mean values ranging 0.3–3.1 kgCO2-eq per 100 g of
18 protein), comparable to milk, eggs, and tuna (mean values ranging 1.2–5.4 kgCO2-eq per 100 g of
19 protein); while chlorella and spirulina consume more energy per unit of protein and were associated
20 with higher GHG emissions (mean values ranging 11–13 kgCO2-eq per 100 g of protein). As the main
21 source of GHG emissions from insects and cellular agriculture foods is energy consumption, their GHG
22 intensity improves with increased use of low-carbon energy (Smetana et al. 2015; Pikaar et al. 2018;
23 Parodi et al. 2018).
24 Future foods offer other benefits such as lower land requirements, controlled systems with reduced
25 losses of water and nutrients, increased resilience, and possibly reduced hazards from pesticide and
26 antibiotics use and zoonotic diseases, although more research is needed including allergenic and other
27 safety aspects, and possibly reduced protein bioavailability (Tzachor et al. 2021; Alexander et al. 2017;
28 Stephens et al. 2018; Parodi et al. 2018; Santo et al. 2020; Fasolin et al. 2019; Chriki and Hocquette
29 2020; Hadi and Brightwell 2021) (medium evidence, high agreement). Research is needed also on the
30 effect of processing (Wickramasinghe et al. 2021), though a randomized crossover trial comparing
31 appetizing plant foods with meat alternatives found several beneficial and no adverse effects from the
32 consumption of the plant-based meats (Crimarco et al. 2020).
33 12.4.3.4 Food processing and packaging
34 Food processing includes preparation and preservation of fresh commodities (fruit and vegetables, meat,
35 seafood and dairy products), grain milling, production of baked goods, and manufacture of pre-prepared
36 foods and meals. Food processors range from small local operations to large multi-national food
37 producers, producing food for local to global markets. The importance of food processing and
38 preservation is particularly evident in developing countries lacking cold chains for the preservation and
39 distribution of fresh perishable products such as fresh fish (Adeyeye 2017; Adeyeye and Oyewole
40 2016).
41 Mitigation in food processing largely focuses on reducing food waste and fossil energy usage during
42 the processing itself, as well as in the transport, packaging and storage of food products for distribution
43 and sale (Silva and Sanjuán 2019). Reducing food waste provides emissions savings by reducing
44 wastage of primary inputs required for food production. Another mitigation route, contributing to the
45 circular economy (Sections 12.6.1.2 and Cross-Working Group Box 3 in this Chapter), is by valorisation
Do Not Cite, Quote or Distribute 12-83 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 of food processing by-products through recovery of nutrients and/or energy. No global analyses of the
2 emissions savings potential from the processing step in the value chain could be found.
3 Reduced food waste during food processing can be achieved by seeking alternative processing routes
4 (Atuonwu et al. 2018), improved communication along the food value chain (Göbel et al. 2015),
5 optimisation of food processing facilities, reducing contamination, and limiting damages and spillage
6 (HLPE 2014). Optimisation of food packaging also plays an important role in reducing food waste, in
7 that it can extend product shelf life; protect against damage during transport and handling; prevent
8 spoilage; facilitate easy opening and emptying; and communicate storage and preparation information
9 to consumers (Molina-Besch et al. 2019).
10 Developments in smart packaging are increasingly contributing to reducing food waste along the food
11 value chain. Strategies for reducing the environmental impact of packaging include using less, and more
12 sustainable, materials and a shift to re-usable packaging (Coelho et al. 2020). Active packaging
13 increases shelf life through regulating the environment inside the packaging, including levels of oxygen,
14 moisture and chemicals released as the food ages (Emanuel and Sandhu 2019). Intelligent packaging
15 communicates information on the freshness of the food through indicator labels (Poyatos-Racionero et
16 al. 2018), and data carriers can store information on conditions such as temperature along the entire
17 food chain (Müller and Schmid 2019).
18 LCA can be used to evaluate the benefits and trade-offs associated with different processing or
19 packaging types (Silva and Sanjuán 2019). Some options, such as aluminium, steel and glass, require
20 high energy investment in manufacture when produced from primary materials, with significant savings
21 in energy through recycling being possible (Camaratta et al. 2020). However, these materials are inert
22 in landfill. Other packaging options, such as paper and biodegradable packaging, may require a lower
23 energy investment during manufacture, but may require larger land area and can release methane when
24 consigned to anaerobic landfill where there is no methane recovery. Nevertheless, packaging accounts
25 for only 1-12% (typically around 5%) of the GHG emissions in the life cycle of a food system (Wohner
26 et al. 2019; Crippa et al. 2021b), suggesting that its benefits can often outweigh the emissions associated
27 with the packaging itself.
28 The second component of mitigation in food processing relates to reduction in fossil energy use.
29 Opportunities include energy efficiency in processes (also discussed in Chapter 11.3), the use of heat
30 and electricity from low-carbon energy sources in processing (Chapter 6), through off-grid thermal
31 processing (sun drying, food smoking) and improving logistics efficiencies. Energy intensive processes
32 with energy saving potential include milling and refining (oil seeds, corn, sugar), drying, and food safety
33 practices such as sterilisation and pasteurisation (Niles et al. 2018). Packaging also plays a role: reduced
34 transport energy can be achieved through reducing the mass of goods transported and improving
35 packing densities in transport vehicles (Lindh et al. 2016; Molina-Besch et al. 2019; Wohner et al.
36 2019). Choice of packaging also influences refrigeration energy requirements during transport and
37 storage.
38 12.4.3.5 Storage and distribution
39 Transport mitigation options along the supply chain include improved logistics, the use of alternative
40 fuels and transport modes, and reduced transport distances. Logistics and alternative fuels and transport
41 modes are discussed in Chapter 10. Transport emissions might increase with increasing demand for a
42 diversity of foods as developing countries become more affluent. New technologies that enable food on
43 demand or online food shopping systems might further increase emissions from food transport;
44 however, the consequences are uncertain and might also entail a shift from individual traffic to bulk
45 transport. The impact on food waste is also uncertain as more targeted delivery options could reduce
Do Not Cite, Quote or Distribute 12-84 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 food waste, but easier access to a wider range of food could also foster over-supply and increase food
2 waste. Mitigation opportunities in food transport are inherently linked to decarbonisation of the
3 transport sector (Chapter 10).
4 Retail and the food service industry are the main factors shaping the external food environment or ‘food
5 entry points’; they are the “physical spaces where food is obtained; the built environment that allows
6 consumers to access these spaces” (HLPE 2017). These industries have significant influence on
7 consumers’ choices and can play a role in reducing GHG emissions from food systems. Opportunities
8 are available for optimisation of inventories in response to consumer demands through advanced IT
9 systems (Niles et al. 2018), and for discounting foods close to sell-by dates, which can both serve to
10 reduce food spoilage and wastage (Buisman et al. 2019).
11 As one of the highest contributors to energy demand at this stage in the food value chain, refrigeration
12 has received a strong focus in mitigation. Efficient refrigeration options include advanced refrigeration
13 temperature control systems, and installation of more efficient refrigerators, air curtains and closed
14 display fridges (Chaomuang et al. 2017). Also related to reducing emissions from cooling and
15 refrigeration is the replacement of hydrofluorocarbons which have very high GWPs with lower GWP
16 alternatives (Niles et al. 2018). The use of propane, isobutane, ammonia, hydrofluoroolefins and CO2
17 (refrigerant R744) are among those that are being explored, with varying success (McLinden et al.
18 2017). In recent years, due to restrictions on high GWP-refrigerants, a considerable growth in the market
19 availability of appliances and systems with non-fluorinated refrigerants has been seen (Eckert et al.
20 2021)
21 Energy efficiency alternatives generic to buildings more broadly are also relevant here, including
22 efficient lighting, HVAC systems and building management, with ventilation being a particularly high
23 energy user in retail, that warrants attention (Kolokotroni et al. 2015).
24 In developing countries particularly, better infrastructure for transportation and expansion of processing
25 and manufacturing industries can significantly reduce food losses, particularly of highly perishable food
26 (Niles et al. 2018; FAO 2019a).
27 12.4.4 Enabling food system transformation
28 Food system mitigation potentials in AFOLU are assessed in Chapter 7.4, and food system mitigation
29 potentials linked to demand side measures are assessed in Chapter 5. Studies suggest that using supply
30 and demand-side policies are implemented in combination makes ambitious mitigation targets easier to
31 achieve (Latka et al. 2021a; Temme et al. 2020; Global Panel on Agriculture and Food Systems for
32 Nutrition 2020; Clark et al. 2020) (high agreement; limited evidence).
33 The trends in the global and national food systems towards a globalisation of food supply chains and
34 increasing dominance of supermarkets and large corporate food processors (Dries et al. 2004; Neven
35 and Reardon 2004; Baker and Friel 2016; Andam et al. 2018; Popkin and Reardon 2018; Reardon et al.
36 2019; Pereira et al. 2020) has led to environmental, food insecurity and malnutrition problems. Studies
37 therefore call for a transformation of current global and national food systems to solve these problems
38 (Schösler and Boer 2018; McBey et al. 2019; Kugelberg et al. 2021). This has not yet been successful,
39 including due to insufficient coordination between relevant food system policies (Weber et al. 2020)
40 (medium evidence, high agreement).
41 Different elements of food systems are currently governed by separate policy areas that in most
42 countries scarcely interact or cooperate (iPES Food 2019; Termeer et al. 2018). This
43 compartmentalisation makes the identification of synergetic and antagonistic effects difficult and faces
Do Not Cite, Quote or Distribute 12-85 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 the possibility of failure due to unintended and unanticipated negative impacts on other policy areas
2 and consequently lack of agreement and social acceptance (Mylona et al. 2018; Brouwer et al. 2020;
3 Mausch et al. 2020; Hebinck et al. 2021) (Section 12.4.5). This could be overcome through cooperation
4 across several policy areas (Sections 12.6.2; 13.7), in particular agriculture, nutrition, health, trade,
5 climate, environment policies, and an inclusive and transparent governance structure (Bhunnoo 2019;
6 Diercks et al. 2019; Herrero et al. 2021; iPES Food 2019; Termeer et al. 2018; Mausch et al. 2020;
7 Kugelberg et al. 2021), making use of potential spill-over effects (Kanter et al. 2020; OECD 2021).
8 Transformation of food systems may come from technological, social or institutional innovations that
9 start as niches but can potentially lead to rapid changes, including changes in social conventions
10 (Centola et al. 2018; Benton et al. 2019).
11 Where calories and ruminant animal-source food are consumed in excess of health guidelines, reduction
12 of excess meat (and dairy) consumption is amongst the most effective measures to mitigate GHG
13 emissions, with a high potential for environment, health, food security, biodiversity, and animal welfare
14 co-benefits (Stylianou et al. 2021; Chai et al. 2019; Semba et al. 2020; Willett et al. 2019; Chen et al.
15 2019; Hamilton et al. 2021; Hedenus et al. 2014; Kim et al. 2019; Theurl et al. 2020; Springmann et al.
16 2018a) (robust evidence, high agreement). Dietary changes are relevant for several SDGs, apart from
17 SDG 13 (climate action), including SDG 2 (zero hunger), SDG 3 (good health and well-being), SDG 6
18 (clean water and sanitation), SDG 12 (responsible consumption and production), SDG 14 (life below
19 water) and SDG 15 (life on land) (Bruce M et al. 2018; Vanham et al. 2019; Mbow et al. 2019; Herrero
20 et al. 2021) (Section 12.6.1). However, behavioural change towards diets of lower environmental impact
21 and higher nutritional qualities faces barriers both from agricultural producers and consumers
22 (Apostolidis and McLeay 2016; Aiking and de Boer 2018; de Boer et al. 2018; Milford et al. 2019), and
23 requires policy packages that combine informative instruments with behavioural, administrative and/or
24 market-based instruments, and are attentive to the needs of, and engage, all food system stakeholders
25 including civil society networks, and change the food environment (see Section 12.4.1) (Stoll-
26 Kleemann and Schmidt 2017; Kraak et al. 2017; El Bilali 2019; Cornelsen et al. 2015; iPES Food 2019;
27 Milford et al. 2019; Temme et al. 2020) (robust evidence, high agreement).
28 Table 12.9 summarizes the implications of a range of policy instruments discussed in more detail in the
29 following sub-sections and highlights the benefits of integrated policy packages. Furthermore, Table
30 12.9 assesses transformative potential, environmental effectiveness, feasibility, distributional effect,
31 cost, and cost-benefits and trade-offs of individual policy instruments, as well as their potential role as
32 part of coherent policy packages. Table 12.9 shows that information and behavioural policy instruments
33 can have significant but small effects in changing diets (robust evidence, medium agreement), but are
34 mutually enforcing and might be essential to lower barriers and increase acceptance of market-based
35 and administrative instruments (medium evidence, high agreement).
36 The policy instruments are assessed in relation to shifting food consumption and production towards
37 increased sustainability and health. This includes lowering GHG emissions, although not in all cases is
38 this the primary focus of the instrument, and in some cases lowering GHG emission may not even be
39 explicitly mentioned.
Do Not Cite, Quote or Distribute 12-86 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Table 12.9: Assessment of food system policies targeting (post-farm gate) food chain actors and consumers
Transformative
Feasibility
effective.
potential
Environ.
Distribution Co-benefits$ and adverse side- Implications for coordination, coherence and consistency in
Level
al effects Cost effect policy package &
Cost Reduces cost of uncoordinated interventions; increases
Integrated food can be + balanced, addresses multiple
NL efficien acceptance across stakeholders and civil society (robust evidence,
policy packages controlled sustainability goals
t high agreement)
High enforcing effect on other food policies; higher acceptance if
Taxes on food
GN regressive low#1 - unintended substitution effects compensation or hypothecated taxes (medium evidence, high
products
agreement)
-unintended substitution effects Supportive, enabling effect on other food policies,
GHG taxes on agricultural/fishery policies; requires changes in power
GN regressive low#2
food distribution and trade agreements
+high spillover effect
(medium evidence, medium agreement)
+ counters leakage effects
impacts comple
Requires changes in existing trade agreements (medium evidence,
Trade policies G global x
+/- effects on market structure high agreement)
distribution effects
and jobs
Investment into + high spillover effect
mediu Can fill targeted gaps for coordinated policy packages (e.g.
research & GN none + converging with digital
m monitoring methods) (robust evidence, high agreement)
innovation society
Food and Can be supportive; might be supportive to realise innovation;
marketing N low voluntary standards might be less effective (medium evidence,
regulations medium agreement)
Do Not Cite, Quote or Distribute 12-87 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
Organisational
+ can address multiple Enabling effect on other food policies; reaches large share of
level procurement NL low
sustainability goals population (medium evidence, high agreement)
policies
Little attention so far on environmental aspects; can serve as
Sustainable food-
GN + can address multiple benchmark for other policies (labels, food formulation standards,
based dietary none low
L sustainability goals etc.)
guidelines
(medium evidence, medium agreement).
+ empowers citizens Effective mainly as part of a policy package; incorporation of
education
Food labels/ GN other objectives (e.g. animal welfare, fair trade...); higher effect if
level low + increases awareness
information L mandatory
relevant
+ multiple objectives (medium evidence, medium agreement).
+ possibly counteracting
High enabling effect on other food policies, (medium evidence,
Nudges NL none low information deficits in
high agreement)
population subgroups
1
2 Colour code: Effect of measures: negative , none/unclear p , slightly positive e , positive ; Level: G: global/multinational, N: national, L: local; #1 Minimum level
3 to be effective 20% price increase; #2 Minimum level to be effective 50-80USD per tCO2eq. $ In addition, all interventions are assumed to address health and climate
4 change mitigation. & Requires coordination between policy areas, participation of stakeholders, transparent methods and indicators to manage trade-offs and prioritisation
5 between possibly conflicting objectives; and suitable indicators for monitoring and evaluation against objectives.
6
Do Not Cite, Quote or Distribute 12-88 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 12.4.4.1 Market based instruments
2 Taxes and subsidies: Food-based taxes have largely been implemented to reduce non-communicable
3 diseases and sugar intake, particularly those targeting sugar-sweetened beverages (WHO 2019). Many
4 health-related organisations recommend the introduction of such taxes to improve the nutritional quality
5 of marketed products and consumers' diets (Park and Yu 2019; Wright et al. 2017; WHO 2019), even
6 though the impacts of food taxes are complex due to cross-price and substitution effects and supplier
7 reactions (Blakely et al. 2020; Gren et al. 2019; Cornelsen et al. 2015) and their regressive effect (WHO
8 2019). Subsidies and taxes are found to be effective in changing dietary behaviour at levels above 20%
9 price increase (Niebylski et al. 2015; Mozaffarian et al. 2018; Nakhimovsky et al. 2016; Hagenaars et
10 al. 2017; Cornelsen et al. 2015), even though longer term effects are scarcely studied (Cornelsen et al.
11 2015) and effects of sugar tax with tax rates lower than 20% have been observed for low-income groups
12 (Temme et al. 2020).
13 Modelling results show only small consumption shifts with moderate meat price increases; and high
14 price increases are required to reach mitigation targets, even though model predictions become highly
15 uncertain due to lack of observational data (Zech and Schneider 2019; Fellmann et al. 2018; Bonnet et
16 al. 2018; Mazzocchi 2017; Latka et al. 2021b). Taxes applied at the consumer level are found to be
17 more effective than levying the taxes at the production side (Springmann et al. 2017).
18 Unilateral taxes on food with high GHG intensities have been shown to induce increases in net export
19 flows, which could reduce global prices and increase global demand. Indirect effects on GHG mitigation
20 therefore could be reduced by up to 70–90% of national results (Fellmann et al. 2018; Zech and
21 Schneider 2019) (limited evidence, high agreement). The global mitigation potential for GHG taxation
22 of food products at 52 USD kgCO2-eq-1 has been estimated at 1 GtCO2-eq yr-1 (Springmann et al. 2017).
23 Studies have shown that taxes can improve the nutritional quality of diets and reduce GHG emissions
24 from the food system, particularly if accompanied by other policies that increase acceptance and
25 elasticity, and reduce regressive and distributional problems (Niebylski et al. 2015; Hagenaars et al.
26 2017; Mazzocchi 2017; Springmann et al. 2017; Wright et al. 2017; Henderson et al. 2018; Säll 2018;
27 FAO et al. 2020; Penne and Goedemé 2020) (robust evidence, high agreement).
28 Trade: Since the middle of the last century, global trade of agricultural products has contributed to
29 boosting productivity and reducing commodity prices, while also incentivising national subsidies for
30 farmers to remain competitive in the global market (Benton et al. 2019). Trade liberalisation has been
31 coined as an essential element of sustainable food systems, and as one element required to achieve
32 sustainable development, that can shift pressure to regions where the resources are less scarce (Traverso
33 and Schiavo 2020; Wood et al. 2018). However, Clapp (2017) argues that the main economic benefit
34 of trade liberalisation flows to large transnational firms. Benton and Bailey (2019) argue that low food
35 prices in the second half of last century contributed to both yield and food waste increases, and to a
36 focus on staple crops to the disadvantage of nutrient dense foods. However, global trade can also
37 contribute to economic benefits such as jobs and income, reduce food insecurity and facilitate access to
38 nutrients (Wood et al. 2018; Hoff et al. 2019; Traverso and Schiavo 2020; Geyik et al. 2021) and has
39 contributed to increased food supply diversity (Kummu et al. 2020). The relevance of trade for food
40 security, and adaptation and mitigation of agricultural production, has also been discussed in Mbow et
41 al. (2019).
42 Trade policies can be used to protect national food system measures, by requiring front-of-package
43 labels, or to impose border taxes on unhealthy products (Thow and Nisbett 2019). For example, in the
44 frame of the Pacific Obesity Prevention in Communities (OPIC), the Fijian government implemented
45 three measures (out of seven proposed) that eliminated import duties on fruits and vegetables, and
Do Not Cite, Quote or Distribute 12-89 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 imposed 15% import duties on unhealthy oils (Latu et al. 2018). Trade agreements, however, have the
2 potential to undermine national efforts to improve public health (Unar-Munguía et al. 2019). GHG
3 mitigation efforts in food supply chains can be counteracted by GHG leakage, with a general increase
4 of environmental and social impact in developing countries exporting food products, and a decrease in
5 the developed countries importing food products (Wiedmann and Lenzen 2018; Sandström et al. 2018;
6 Fellmann et al. 2018). The demand for agricultural commodities has also been associated with tropical
7 deforestation, though a robust estimate on the extent of embodied deforestation in food commodities is
8 not available (Pendrill et al. 2019).
9 Investment into research & innovation: El Bilali (2019) assessed research gaps in the food system
10 transition literature and found a need to develop comparative studies that enable the assessment of
11 spatial variability and scalability of food system transitions. The author found also that the role of
12 private industry and corporate business is scarcely researched, although they could play a major role in
13 food system transitions.
14 The InterAcademy Partnership assessed how research can contribute to providing the required evidence
15 and opportunities for food system transitions, with a focus on climate change impacts and mitigation
16 (IAP 2018). The project builds on four regional assessments of opportunities and challenges on food
17 and nutrition security in Africa (NASAC 2018), the Americas (IANAS 2018), Asia (AASSA 2018),
18 and Europe (EASAC 2017). The Partnership concludes with a set of research questions around food
19 systems, that need to be better understood: (i) how are sustainable food systems constituted in different
20 contexts and at different scales, (ii) how can transition towards sustainable food systems be achieved,
21 and (iii) how can success and failure be measured along sustainability dimensions including climate
22 mitigation?
23
24 12.4.4.2 Regulatory and administrative instruments
25 Marketing regulations: Currently, 16 countries regulate marketing of unhealthy food to children, mainly
26 on television and in schools (Taillie et al. 2019), and many other efforts are ongoing across the globe
27 (European Commission 2019). The aim to counter the increase in obesity in children and target products
28 high in saturated fats, trans-fatty acids, free sugars and/or salt (WHO 2010) was endorsed by 192
29 countries (Kovic et al. 2018). Nutrition and health claims for products are used by industry to increase
30 sales, for example in the sport sector or for breakfast cereals. They can be informative, but can also be
31 misleading if misused for promoting unhealthy food (Ghosh and Sen 2019; Sussman et al. 2019;
32 Whalen et al. 2018).
33 Strong statutory marketing regulations can significantly reduce the exposure of children to, and sales
34 of, unhealthy food compared with voluntary restrictions (Kovic et al. 2018; Temme et al. 2020). Data
35 on effectiveness of marketing regulations with a broader food sustainability scope are not available. On
36 the other hand, regulations that mobilise private investment into emerging food production technologies
37 can be instrumental in curbing the cost and making them competitive (Bianchi et al. 2018a).
38 Voluntary sustainability standards: Voluntary sustainability standards are developed either by a public
39 entity or by private organisations to respond to consumers’ demands for social and environmental
40 standards (Fiorini et al. 2019). For example, the Dutch “Green Protein Alliance”, an alliance of
41 government, industry, NGOs and academia, formulated a goal to shift the ratio of protein consumption
42 from 60% animal source proteins currently to 40% by 2050 (Aiking and de Boer 2020), and Cool Food
43 Pledge signatories (organisations that serve food, such as restaurants, hospitals and universities)
44 committed to a 25% reduction in GHG emissions by 2030, compared with 2015 (Cool Food 2020). For
Do Not Cite, Quote or Distribute 12-90 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 firms, obtaining certification under such schemes can be costly, and costs are generally borne by the
2 producers and/or supply chain stakeholders (Fiorini et al. 2019). The effectiveness of private voluntary
3 sustainability standards is uncertain. Cazzolla Gatti et al. (2019) have investigated the effectiveness of
4 the Roundtable on Sustainable Palm Oil on halting forest loss and habitat degradation in Southeast Asia
5 and concluded that production of certified palm oil continued to lead to deforestation.
6 Organisational procurement: Green public procurement is a policy that aims to create additional
7 demand for sustainable products (Bergmann Madsen 2018; Mazzocchi and Marino 2019) or decrease
8 demand for less sustainable products (e.g., the introduction of “Meatless Monday” by the Norwegian
9 Armed Forces (Milford and Kildal 2019; Cheng et al. 2018; Wilts et al. 2019; Gava et al. 2018)). To
10 improve dietary choices, organisations can increase the price of unsustainable options while decreasing
11 the price of sustainable ones, or employ information or choice architecture measures (Goggins and Rau
12 2016; Goggins 2018). Procurement guidelines exist at global, national, organisational or local levels
13 (Neto and Gama Caldas 2018; Noonan et al. 2013). Procurement rules in schools or public canteens
14 increase the accessibility of healthy food and can improve dietary behaviour and decrease purchases of
15 unhealthy food (Cheng et al. 2018; Temme et al. 2020).
16 Food regulations: Novel foods based on insects, microbial proteins or cellular agriculture must go
17 through authorisation processes to ensure compliance with food safety standards before they can be
18 sold to consumers. Several countries have ‘novel food’ regulations governing the approval of foods for
19 human consumption. For example, the European Commission, in its update of the Novel Food
20 Regulation in 2015, expanded its definition of novel food to include food from cell cultures, or that
21 produced from animals by non-traditional breeding techniques (EU 2015).
22 For animal product analogues, regulatory pathways and procedures (Stephens et al. 2018) and
23 terminology issues (defining equivalence questions) (Carrenõ and Dolle 2018; Pisanello and Ferraris
24 2018) need clarification, as does their relation to religious rules (Chriki and Hocquette 2020).
25 Examples of legislation targeting food waste include the French ban on wasting food approaching best-
26 before dates, requiring its donation to charity organisations (Global Alliance for the Future of Food
27 2020). In Japan, the Food Waste Recycling Law set targets for food waste recycling for industries in
28 the food sector for 2020, ranging between 50% for restaurants and 95% for food manufacturers (Liu et
29 al. 2016).
30 12.4.4.3 Informative instruments.
31 Sustainable Food-Based Dietary Guidelines: National food based dietary guidelines (FBDGs) provide
32 science-based recommendations on food group consumption quantities. They are available for 94,
33 mostly upper- and middle-income countries globally (Wijesinha-Bettoni et al. 2021), adapted to
34 national cultural and socio-economic context, and can be used as a benchmark for food formulation
35 standards or public and private food procurement, or to inform citizens (Bechthold et al. 2018; Temme
36 et al. 2020). Most FBDGs are based on health considerations and only a few mention environmental
37 sustainability aspects (Bechthold et al. 2018; Ritchie et al. 2018; Ahmed et al. 2019; Springmann et al.
38 2020). Implementation of FBDGs so far focuses largely in the education and health sectors, with few
39 countries also using their potential for guiding food system policies in other sectors (Wijesinha-Bettoni
40 et al. 2021).
41 Despite the fact that 1.5 billion people follow a vegetarian diet from choice or necessity and the position
42 statements of various nutrition societies point out that vegetarian diets are adequate if well planned, few
43 FBDGs give recommendations for vegetarian diets (Costa Leite et al. 2020). An increase in
44 consumption of plant-based food is a recurring recommendation in FBDGs, though an explicit reduction
Do Not Cite, Quote or Distribute 12-91 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 or limit of animal source proteins is not often included, with the exception of red or processed meat
2 (Temme et al. 2020). To account for changing dietary trends, however, FBDGs need to incorporate
3 sustainability aspects (Herforth et al. 2019). A healthy diet respecting planetary boundaries has been
4 proposed by Willett et al. (2019), though some authors have questioned the validity of the nutritional
5 (Zagmutt et al. 2019) or environmental implications, such as water use (Vanham et al. 2020). In October
6 2019, 14 global cities pledged to adhere to this ‘planetary health diet’ (C40 Cities 2019).
7 Education on food/nutrition and environment: Some consumers are reluctant to adopt sustainable
8 healthy dietary patterns because of a lack of awareness of the environmental and health consequences
9 of what they eat, but also out of suspicion towards alternatives that are perceived as not ‘natural’ and
10 that seem to be difficult to integrate into their daily dietary habits (Hartmann and Siegrist 2017;
11 Stephens et al. 2018; McBey et al. 2019; Siegrist and Hartmann 2020) or simply lack of knowledge on
12 how to prepare or eat unfamiliar foods (Aiking and de Boer 2020; El Bilali 2019; Temme et al. 2020).
13 Misconceptions may contribute, for example, the belief that packaging or ‘food miles’ dominate the
14 climate impact of food (Macdiarmid et al. 2016). However, spill-over effects can induce sustainable
15 behaviour from ‘entry points’ such as concerns about food waste (El Bilali 2019). Early-life experiences
16 are crucial determinants for adopting healthy and sustainable lifestyles (Bascopé et al. 2019; McBey et
17 al. 2019), so improved understanding of sustainability aspects in the education of public health
18 practitioners and in university education is proposed (Wegener et al. 2018). Investment in education,
19 particularly of women (Vermeulen et al. 2020), might lower the barrier for stronger policies to be
20 accepted and effective (McBey et al. 2019; Temme et al. 2020) (medium evidence, high agreement).
21 Food labels: Instruments to improve transparency and information on food sustainability aspects are
22 based on the assumption of the ’rational’ consumer. Information gives the necessary freedom of choice,
23 but also the responsibility to make the ‘right choice’ (Kersh 2015; Bucher et al. 2016). Studies find a
24 lack of consumer awareness about the link between own food choices and environmental effect
25 (Grebitus et al. 2016; Leach et al. 2016; de Boer et al. 2018; Hartmann and Siegrist 2017) and so
26 effective messaging is required to raise awareness and acceptance of potentially stricter food system
27 policies.
28 Back-of-package labels usually provide detailed nutritional information (Temple 2019). Front-of-
29 package labels simplify and interpret the information: for example, the traffic light system or the Nutri-
30 Score label used in France (Kanter et al. 2018b) and the health star rating used in Australia and New
31 Zealand (Shahid et al. 2020) provide an aggregate rating based on product attributes such as energy,
32 sugar, saturated fat and fibre content; other labels warn against frequent consumption (e.g., in the 1990s
33 Finland introduced a mandatory warning for products high in salt; the keyhole label was introduced in
34 Sweden in 1989 (Storcksdieck genannt Bonsmann et al. 2020); and ‘high in’ (energy/ saturated fat/
35 sugar) labels were introduced in Chile in 2016 to reduce obesity (Corvalán et al. 2019)). Front-of-
36 package labels serve also as an incentive to industry to produce healthier or more sustainable products,
37 or can serve as a marketing strategy (Van Loo et al. 2014; Kanter et al. 2018b; Apostolidis and McLeay
38 2016). Carbon footprint labels can be difficult for consumers to understand (Hyland et al. 2017), and
39 simple, interpretative summary indicators used in front-of-package labels (e.g., traffic lights) are more
40 effective than more complex ones (Tørris and Mobekk 2019; Ikonen et al. 2019; Bauer and Reisch
41 2019; Temple 2019) (robust evidence, high agreement). Reviews find mixed results but overall a
42 positive effect of food labels in improving direct purchasing decisions (Sarink et al. 2016; Anastasiou
43 et al. 2019; Shangguan et al. 2019; Hieke and Harris 2016; Temple 2019), and in raising level of
44 awareness, thus possibly increasing success of other policy instruments (Al-Khudairy et al. 2019;
45 Samant and Seo 2016; Miller et al. 2019; Temple 2019; Apostolidis and McLeay 2016) (medium
46 evidence, high agreement).
Do Not Cite, Quote or Distribute 12-92 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 12.4.4.4 Behavioural instruments.
3 Choice architecture: Information is more effective if accompanied by reinforcement through structural
4 changes or by changing the food environment, such as through product placement in supermarkets, to
5 overcome the intention-behaviour gap (Bucher et al. 2016; Broers et al. 2017; Tørris and Mobekk 2019).
6 Behavioural change strategies have also been shown to improve efficiencies of school food programs
7 (Marcano-Olivier et al. 2020).
8 Environmental considerations rank behind financial, health, or sensory factors for determining citizens’
9 food choices (Leach et al. 2016; Hartmann and Siegrist 2017; Rose 2018; Neff et al. 2018; Gustafson
10 et al. 2019). There is evidence that choice architecture (“nudging”) can be effective in influencing
11 purchase decisions, but regulators do not normally explore this option (Broers et al. 2017). Examples
12 of green nudging include making the sustainable option the default option, enhancing visibility,
13 accessibility of, or exposure to, sustainable products and reducing visibility and accessibility of un-
14 sustainable products, or increasing the salience of healthy sustainable choices through social norms or
15 food labels (Bucher et al. 2016; Wilson et al. 2016; Broers et al. 2017; Al-Khudairy et al. 2019; Bauer
16 and Reisch 2019; Ferrari et al. 2019; Weinrich and Elshiewy 2019; Cialdini and Jacobson 2021).
17 Available evidence suggests that choice architecture measures are relatively inexpensive and easy to
18 implement (Ferrari et al. 2019; Tørris and Mobekk 2019), they are a preferred solution if a restriction
19 of choices is to be avoided (Wilson et al. 2016; Kraak et al. 2017; Vecchio and Cavallo 2019), and can
20 be effective (Arno and Thomas 2016; Bianchi et al. 2018b; Cadario and Chandon 2018; Bucher et al.
21 2016) if embedded in policy packages (Wilson et al. 2016; Tørris and Mobekk 2019) (medium evidence,
22 high agreement).
23 Choice architecture measures are also facilitated by growing market shares of animal-free protein
24 sources taken up by discount chains and fast food companies, that enhance visibility of new products
25 and ease integration into daily life for consumers, particularly if sustainable products are similar to the
26 products they substitute (Slade 2018). This effect can be further increased by media and role models
27 (Elgaaied-Gambier et al. 2018).
28 12.4.5 Food Systems Governance
29 To support the policies outlined in Section 12.4.4, food system governance depends on the cooperation
30 of actors across traditional sectors in several policy areas, in particular agriculture, nutrition, health,
31 trade, climate, and environment (Termeer et al. 2018; Bhunnoo 2019; Diercks et al. 2019; iPES Food
32 2019; Rosenzweig et al. 2020b). Top-down integration, mandatory mainstreaming, or boundary-
33 spanning structures like public-private partnerships may be introduced to promote coordination
34 (Termeer et al. 2018). “Flow-centric” rather than territory-centric governance combined with private
35 governance mechanisms has enabled codes of conduct and certification schemes (Eakin et al. 2017),
36 for example the Roundtable for Sustainable Palm Oil (RSPO), as well as commodity chain transparency
37 initiatives and platforms like Trase (Pirard et al. 2020; Meijaard et al. 2020). Trade agreements are an
38 emerging arena of governance in which improving GHG performance may be an objective, and trade
39 agreements can involve sustainability assessments.
40 Research on food system governance is mostly non-empirical or case study based, which means that
41 there is limited understanding of which governance arrangements work in specific social and ecological
42 contexts to produce particular food system outcomes (Delaney et al. 2018). Research has identified a
43 number of desirable attributes in food systems governance, including adaptive governance (Termeer et
44 al. 2018), a systems perspective (Whitfield et al. 2018), governance that considers food system
Do Not Cite, Quote or Distribute 12-93 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 resilience (Moragues-Faus et al. 2017; Ericksen 2008; Meyer 2020), transparency, participation of civil
2 society (Duncan 2015; Candel 2014), and cross-scale governance (Moragues-Faus et al. 2017).
3 Food systems governance has multiple targets and objectives, not least contributing the achievement of
4 the SDGs. GHG emissions from food systems can be impacted by both interventions targeted at
5 different parts of the food system and interventions in other systems, such as reducing deforestation or
6 promoting reforestation (Lee et al. 2019). For example, policies targeting health can contribute to diet
7 shifts away from red meat, while also influencing GHG emissions (Springmann et al. 2018b; Semba et
8 al. 2020); national and local food self-sufficiency policies may also have GHG impacts (Loon et al.
9 2019; Kriewald et al. 2019). Cross-sectoral governance could enhance synergies between reduced GHG
10 emissions from food systems and other goals; however, integrative paradigms for cross-sectoral
11 governance between food and other sectors have faced implementation challenges (Delaney et al. 2018).
12 For example, in the late 2000s, the water-energy-food nexus emerged as a framework for cross-sectoral
13 governance, but has not been well-integrated into policy (Urbinatti et al. 2020), perhaps because of
14 perceptions that it is an academic concept, or that it takes a technical-administrative view of governance;
15 simply adopting the paradigm is not sufficient to develop effective nexus governance (Cairns and
16 Krzywoszynska 2016; Weitz et al. 2017; Pahl-Wostl et al. 2018). Other policy paradigms and
17 theoretical frameworks that aim to integrate food systems governance include system transition,
18 agroecology, multifunctionality in agriculture (Andrée et al. 2018), climate-smart agriculture (Taylor
19 2018) and the circular economy (Box 12.2). Cross-sectoral coordination on food systems and climate
20 governance could be aided by internal recognition and ownership by agencies, dedicated budgets for
21 cross-sectoral projects, and consistency in budgets (Pardoe et al. 2018); see also Box 12.1 and Box 12.2.
22 Food systems governance is still fragmented at national levels, which means that there may be a
23 proliferation of efforts that cannot be scaled and are ineffective (Candel 2014). National policies can be
24 complemented or possibly pioneered by initiatives at the local level (de Boer et al. 2018; Rose 2018).
25 The city-region has been proposed as a useful focus for food system governance (Vermeulen et al.
26 2020); for example, the Milan Urban Food Policy Pact involves 180 global cities committed to
27 integrative food system strategies (Candel 2019; Moragues-Faus 2021). Local food policy groups and
28 councils that assemble stakeholders from government, civil society, and the private sector have formed
29 trans-local networks of place-based local food policy groups, with over two hundred food policy
30 councils worldwide (Andrée et al. 2018). However, the fluidity and lack of clear agendas and
31 membership structures may hinder their ability to confront fundamental structural issues like
32 unsustainable diets or inequities in food access (Santo and Moragues-Faus 2019).
33 Early characterisations of food systems governance featured a binary distinction between global and
34 local scales, but this has been replaced by a relational approach where the local governance is seen a
35 process that relies on the interconnections between scales (Lever et al. 2019). Cross-scalar governance
36 is not simply an aggregation of local groups, but involves the telecoupling of distant systems; for
37 example, transnational NGO networks have been able to link coffee retailers in the global North with
38 producers in the global South via international NGOs concerned about deforestation and social justice
39 (Eakin et al. 2017). Global governance institutions like the Committee on World Food Security can
40 promote policy coherence globally and reinforce accountability at all levels (McKeon 2015), as can
41 norm-setting efforts like the ‘Voluntary Guidelines for the Responsible Governance of Tenure of Land,
42 Fisheries and Forests’ (FAO 2012). Global multi-stakeholder processes like the UN Food Systems
43 Summit can foster the development of principles for guiding further actions based on sound scientific
44 evidence. The European Commission’s Farm to Fork strategy aims to promote policy coherence in food
45 policy at EU and national levels, and could be the exemplar of a genuinely integrated food policy
46 (Schebesta and Candel 2020).
Do Not Cite, Quote or Distribute 12-94 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 START BOX 12.2 HERE
2 Box 12.2: Case Study: The Finnish Food2030 Strategy
3
4 Until 2016, the strategic goals of Finnish food policy were split between different programs and
5 Ministries, resulting in fragmented national oversight of the Finnish food system. To enable policy
6 coordination, a national food strategy was adopted in 2017 called Food2030 (Government of Finland
7 2017). Food2030 embodies a holistic food system approach and addresses multiple outcomes of the
8 food system, including the competitiveness of the food supply chain and the development of local,
9 organic and climate-friendly food production, as well as responsible and sustainable consumption.
10
11 The specific policy mix covers a range of policy instruments to enable changes in agro-food supply,
12 processing and societal norms (Kugelberg et al. 2021). The government provides targeted funding and
13 knowledge support to drive technological innovations on climate solutions to reduce emissions from
14 food and in the agriculture, forestry and land-use sectors. In addition, the Finnish government applies
15 administrative means, such as legislation, advice, guidance on public procurement and support schemes
16 to diversify and increase organic food production to 20% of arable land, which in turn improve the
17 opportunities of small-scale food production and steer public bodies to purchase local and organic food.
18 The Finnish government applies educational and informative instruments to enable a shift to healthy
19 and sustainable dietary behaviours. The policy objective is to reduce consumption of meat and replace
20 it with other sources of protein, aligned with nutrition recommendations and avoiding food waste. The
21 Ministry of Agriculture and Forestry, in collaboration with the Finnish Farmer’s unions (MTK) and the
22 Union of Swedish-speaking Farmers and Forest Owners in Finland (SLC), ran a two-year multi-media
23 campaign in 2018 with key messages on sustainability, traceability and safety of the locally-produced
24 food (Ministry of Agriculture and Forestry 2021). A “Food Facts website project” (Luke 2021), funded
25 by the Ministry of Agriculture and Forestry in collaboration with the Natural Resources Institute Finland
26 and the Finnish Food Safety Authority, helps to raise knowledge about food, which could shape
27 responsible individual food behaviour, for example choosing local and sustainable foods and reducing
28 food waste.
29
30 A critical enabler for developing a shared food system strategy across sectors and political party
31 boundaries was the implementation of a one-year inclusive, deliberative and consensual stakeholder
32 engagement process. A wide range of stakeholders could exert real influence during the vision-building
33 process, resulting in strong agreement on key policy objectives, and subsequently an important leverage
34 point to policy change (Kugelberg et al. 2021). Moreover, cross-sectoral coordination of Food2030 and
35 the government’s wider climate action programs are enabled by a number of institutional mechanisms
36 and collaborative structures, for example the Advisory board for the food chain, formally established
37 during the agenda-setting stage of Food2030, inter-ministerial committees to guide and assess policy
38 implementation, and Our common dining table, a multi-stakeholder partnership that assembles 18 food
39 system actors to engage in reflexive discussions about the Finnish food system.
40
41 Critical barriers to strategy and policy formulation include a lack of attention to integrated impact
42 assessments (Kugelberg et al. 2021), which blurs a transparent overview of potential trade-offs and
43 hidden conflicts. There were few policy evaluations from independent organisations to inform
44 policymaking, reducing the opportunities for more progressive policy approaches. Monitoring and food
45 policy evaluation is very close to the ministry in charge, which hampers critical thinking about policy
46 measures (Hildén et al. 2014). In addition, there is a lack of standardised indicators covering the whole
47 food system, which hinders comprehensive oversight of government’s progress towards a sustainable
48 food system (Kanter et al. 2018a). Some of the problems related to monitoring, reporting and
49 verification (MRV) are typical for countries in the EU. To improve MRV will probably require
Do Not Cite, Quote or Distribute 12-95 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 structural changes, such as efforts to build up institutional capacity and application of new technology,
2 development of standardised indicators covering the whole food system, regulations on transparency
3 and verification, and mechanisms to enable reflexive discussions between business, farmers, public,
4 NGOs and the government (Meadowcroft and Steurer 2018; Kanter et al. 2020).
5
6 END BOX 12.2 HERE
7
8 12.5 Land-related impacts, risks and opportunities associated with
9 mitigation options
10 12.5.1 Introduction
11 This Section provides a cross-sectoral perspective on land occupation and related impacts, risks and
12 opportunities associated with land-based mitigation options as well as mitigation options that are not
13 designated land-based, yet occupy land. It builds on Chapter 7, that covers mitigation in agriculture,
14 forestry and other land use (AFOLU), including future availability of biomass resources for mitigation
15 in other sectors. It complements Section 12.4, which covers mitigation inherent in the food system, as
16 well as Chapters 6, 9, 10 and 11 that cover mitigation in the energy, transport, building and industry
17 sectors, and Chapters 3 and 4 that cover land and biomass use, primarily in energy applications, in
18 mitigation and development pathways in the near- to mid-term (Chapter 4) and in pathways compatible
19 with long-term goals (Chapter 3).
20 The deployment of climate change mitigation options often affects land and water conditions, and
21 ecosystem capacity to support biodiversity and a range of ecosystem services (IPCC 2019; IPBES 2019)
22 (robust evidence, high agreement). It can increase or decrease terrestrial carbon stocks and sink strength,
23 hence impacting the mitigation effect positively or negatively. As for any other land uses, impacts, risks
24 and opportunities associated with mitigation options that occupy land depend on deployment strategy
25 and on contextual factors that vary geographically and over time (Doelman et al. 2018; Hurlbert et al.
26 2019; Smith et al. 2019a; Wu et al. 2020) (robust evidence, high agreement).
27 The SR1.5 found that large areas may be utilised for A/R and energy crops in modelled pathways
28 limiting warming to 1.5 C (Rogelj et al. 2018). The SRCCL investigated the implications of land-based
29 mitigation measures for land degradation, food security and climate change adaptation. It focussed on
30 identification of synergies and trade-offs associated with individual land-based mitigation measures
31 (Smith et al. 2019b). In this section we expand beyond the scope of the SRCCL assessment to include
32 also mitigation measures that occupy land while not considered land-based measures, we discuss ways
33 to minimise potential adverse effects, and we consider the potential for synergies through integrating
34 mitigation measures with other land uses, by applying a systems perspective that seeks to meet multiple
35 objectives from multi-functional landscapes. Mitigation measures with zero land occupation, e.g.,
36 offshore wind and kelp farming, are not considered,
37 12.5.2 Land occupation associated with different mitigation options
38 As reported in Chapter 3, in scenarios limiting warming to 1.5ºC with no or limited overshoot, median
39 area dedicated for energy crops in 2050 is 1.99 (0.56 to 4.82) Mkm2 and median forest area increased
40 3.22 (-0.67 to 8.90) Mkm2 in the period 2019-2050 (5-95 percentile range, scenario category C1). For
41 comparison, the total global areas of forests, cropland and pasture (year 2015) are in the SRCCL
42 estimated at about 40 Mkm2, 15.6 Mkm2, and 27.3 Mkm2, respectively (additionally, 21 Mkm2 of
43 savannahs and shrublands are also used for grazing) (IPCC 2019). The SRCCL concluded that
Do Not Cite, Quote or Distribute 12-96 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 conversion of land for A/R and bioenergy crops at the scale commonly found in pathways limiting
2 warming to 1.5°C or 2°C is associated with multiple feasibility and sustainability constraints, including
3 land carbon losses (high confidence). Pathways in which warming exceeds 1.5°C require less land-
4 based mitigation, but the impacts of higher temperatures on regional climate and land, including land
5 degradation, desertification, and food insecurity, become more severe (Smith et al. 2019b).
6 Depending on emission-reduction target, the portfolio of mitigation options chosen, and the policies
7 developed to support their implementation, different land-use pathways can arise with large differences
8 in resulting agricultural and forest area. Some response options can be more effective when applied
9 together (Smith et al. 2019c); for example, dietary change, efficiency increases, and reduced wastage
10 can reduce emissions as well as the pressure on land resources, potentially enabling additional land-
11 based mitigation such as A/R and cultivation of biomass crops for biochar, bioenergy and other bio-
12 based products. The SRCCL (Smith et al. 2019c) report that dietary change combined with reduction
13 in food loss and waste can reduce the land requirement for food production by up to 5.8 Mkm2 (0.8–2.4
14 Mkm2 for dietary change; about 2 Mkm2 for reduced post-harvest losses, and 1.4 Mkm2 for reduced
15 food waste (see also Sections 7.4 and 12.4 and Parodi et al. 2018; Springmann et al. 2018; Clark et al.
16 2020; Rosenzweig et al. 2020b). Stronger mitigation action in the near term targeting non-CO2
17 emissions reduction and deployment of other CDR options (DACCS, enhanced weathering, ocean-
18 based approaches, see 12.3) can reduce the land requirement for land-based mitigation (Obersteiner et
19 al. 2018; van Vuuren et al. 2018).
20 Global Integrated Assessment Models (IAMs) provide insights about the roles of land-based mitigation
21 in pathways limiting warming to 1.5°C or 2°C; interaction between land-based and other mitigation
22 options such as wind and solar power; influence of land-based mitigation on food markets, land use and
23 land carbon; and the role of BECCS vis- à-vis other CDR options (See Chapter 3). However, IAMs do
24 not capture more subtle changes in land management and in the associated industrial/energy systems
25 due to relatively coarse temporal and spatial resolution, and limited representation of land quality and
26 feedstocks/management practices, interactions between biomass production and conversion systems,
27 and local context, e.g., governance of land use (Daioglou et al. 2019; Rose et al. 2020; Welfle et al.
28 2020; Calvin et al. 2021). A/R have generally been modelled as forests managed for carbon
29 sequestration alone, rather than forestry providing both carbon sequestration and biomass supply
30 (Calvin et al. 2021). Because IAMs do not include options to integrate new biomass production with
31 existing agricultural and forestry systems (Paré et al. 2016; Mansuy et al. 2018; Cossel et al. 2019;
32 Braghiroli and Passarini 2020; Moreira et al. 2020; Djomo et al. 2020; Strapasson et al. 2020; Rinke
33 Dias de Souza et al. 2021), they may over-estimate the total additional land area required for biomass
34 production. On the other hand, some integrated biomass production systems may prove less attractive
35 to landholders than growing biomass crops in large blocks, from logistic, economic, or other points of
36 view (Ssegane et al. 2016; Busch 2017; Ferrarini et al. 2017).
37 Land occupation associated with mitigation options other than A/R and bioenergy is rarely quantified
38 in global scenarios. Stressing large uncertainties (e.g., type of biomass used and share of solar PV
39 integrated in buildings), (Luderer et al. 2019) modelled land occupation and land transformation
40 associated with a range of alternative power system decarbonisation pathways in the context of a global
41 2°C climate stabilisation effort. On a per-MWh basis, bioelectricity with CCS was most land-intensive,
42 followed by hydropower, coal with CCS, and concentrated solar power (CSP), which in turn were
43 around five times as land-intensive as wind and solar photovoltaics (PV). A review of studies of power
44 densities (electricity generation per unit land area) confirmed the relatively larger land occupation
45 associated with biopower, although hydropower overlaps with biopower (van Zalk and Behrens 2018).
46 This study also quantifies the low land occupation of nuclear energy, similar to fossil energy sources.
Do Not Cite, Quote or Distribute 12-97 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 The land occupation of PV depends on the share of ground-mounted vs. buildings-integrated PV, the
2 latter assumed to reach 75% share by 2050 in (Luderer et al. 2019). van de Ven et al. (2021) assumed a
3 3% share of urbanized land in 2050 available for rooftop PV, referring to (Capellán-Pérez et al. 2017;
4 Dupont et al. 2020) reporting 2-3% availability of urbanized surface area, when considering factors
5 such as roof slopes and shadows between buildings, and threshold relating to energy return on
6 investment. Referring to (De Castro et al. 2013; MacKay 2013; Ong et al. 2013; Smil; Capellán-Pérez
7 et al. 2017) state that land occupation of solar technologies is underestimated in studies assuming ideal
8 conditions, with real occupation being five to ten times higher.
9 Production of hydrogen and synthetic hydrocarbon fuels via electrolysis and hydrocarbon synthesis is
10 subject to conversion losses that vary depending on technology, system integration and source of
11 carbon (Wulf et al. 2020; Ince et al. 2021)(cross-ref 6.4.4.1 and 6.4.5.1). Indicative electricity-to-
12 hydrocarbon fuel efficiency loss is estimated at about 60% (Ueckerdt et al. 2021). The advantage of
13 smaller land occupation for solar, wind, hydro and nuclear, compared with biomass-based options, is
14 therefore smaller for hydrocarbon fuels than for electricity. Furthermore, biofuels are often co-
15 produced with other bio-based products, which further reduces their land occupation, although
16 comparisons are complicated by inconsistent approaches to allocating land occupation between co-
17 products (Ahlgren et al. 2015; Czyrnek-Delêtre et al. 2017).
18
19 Note that comparisons on a per-MWh basis do not reflect the GHG emissions associated with the power
20 options, or that the different options serve different functions in power systems. Reservoir hydropower
21 and biomass-based dispatchable power can complement other balancing options (e.g., battery storage,
22 grid extensions and demand-side management (Göransson and Johnsson 2018; Chapter 6) to provide
23 power stability and quality needed in power systems with large amounts of variable electricity
24 generation from wind and solar power plants. Furthermore, the requirements of transport in grids,
25 pipelines etc. differ. For example, electricity from buildings-integrated PV can be used in the same
26 location as it is generated.
27 The character of land occupation, and, consequently, the associated impacts (see 12.5.3), vary
28 considerably among mitigation options and also for the same option depending on geographic location,
29 scale, system design and deployment strategy (Olsson et al. 2019; Ioannidis and Koutsoyiannis 2020;
30 van de Ven et al. 2021). Land occupation associated with different mitigation options can be large
31 uniform areas (e.g., large solar farms, reservoir hydropower dams, or tree plantations), or more
32 distributed occupation, such as wind turbines, solar PV, and patches of biomass cultivation integrated
33 with other land uses in heterogeneous landscapes (Cacho et al. 2018; Jager and Kreig 2018; Correa et
34 al. 2019; Englund et al. 2020a). Studies with broader scope, covering total land use requirement induced
35 by plant infrastructure, provide a more complete picture of land footprints. For example, Wu et al.
36 (2021) quantified a land footprint by the infrastructure of a pilot solar plant being three times the onsite
37 land area. Sonter et al. (2020b) found significant overlap of mining areas (82% targeting materials
38 needed for renewable energy production) and biodiversity conservation sites and priorities, suggesting
39 that strategic planning is critical to address mining threats to biodiversity (See section 12.5.4) along
40 with recycling and exploration of alternative technologies that use that use abundant minerals (See
41 Chapter 11, Box Critical Minerals and The Future of Electro-Mobility and Renewables)
42 There are also situations where expanding mitigation is more or less decoupled from additional land
43 use. The use of organic consumer waste, harvest residues and processing side-streams in the agriculture
44 and forestry sectors can support significant volumes of bio-based products with relatively lower land-
45 use change risks than dedicated biomass production systems (Hanssen et al. 2019; Spinelli et al. 2019;
46 Mouratiadou et al. 2020). Such uses can provide waste management solutions while increasing the
47 mitigation achieved from the land that is already used for agricultural and forest production. Bioenergy
Do Not Cite, Quote or Distribute 12-98 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 accounts for about 90% of renewable heat used in industrial applications, mainly in industries that can
2 use their own biomass waste and residues, such as the pulp and paper industry, food industry, and
3 ethanol production plants (see Chapters 6 and 11) (IEA 2020c). Heat and electricity produced on-site
4 from side-streams but not needed for the industrial processes can be sold to other users, e.g., district
5 heating systems. Surplus waste and residues can also be used to produce solid and liquid biofuels, or be
6 used as feedstock in other industries such as the petrochemical industry (IRENA 2018; Lock and Whittle
7 2018; Thunman et al. 2018; IRENA 2019; Haus et al. 2020; Chapters 6 and 11). Electrification and
8 improved process efficiencies can reduce GHG emissions and increase the share of harvested biomass
9 that is used for production of bio-based products (Johnsson et al. 2019; Madeddu et al. 2020; Lipiäinen
10 and Vakkilainen 2021; Rahnama Mobarakeh et al. 2021; Silva et al. 2021; Chapter 11). Besides
11 integrating solar thermal panels and solar PV into buildings and other infrastructure, floating solar PV
12 panels in, e.g., hydropower dams (Ranjbaran et al. 2019; Cagle et al. 2020; Haas et al. 2020; Lee et al.
13 2020; Gonzalez Sanchez et al. 2021), and over canals (Lee et al. 2020; McKuin et al. 2021) could
14 decouple renewable energy generation from land use while simultaneously reducing evaporation losses
15 and potentially mitigating aquatic weed growth and climate change impacts on water body temperature
16 and stratification (Cagle et al. 2020; Exley et al. 2021; Gadzanku et al. 2021; Solomin et al. 2021).
17
18 12.5.3 Consequences of land occupation: biophysical and socioeconomic risks, impacts
19 and opportunities
20 Land occupation associated with mitigation options can present challenges related to impacts and trade-
21 offs, but can also provide opportunities and in different ways support the achievement of additional
22 societal objectives, including adaptation to climate change. This section focuses on mitigation options
23 that have significant risks, impacts and/or co-benefits with respect to land resources, food security and
24 the environment. Bioenergy (with or without CCS), biochar and bio-based products require biomass
25 feedstocks that can be obtained from purpose-grown crops, residues from conventional agriculture and
26 forestry systems, or from biomass wastes, each with different implications for the land. Here we
27 consider separately (i) “biomass-based systems”, including dedicated biomass crops (e.g., perennial
28 grasses, short rotation woody crops) and biomass produced as a co-product of conventional agricultural
29 production (e.g. maize stover), and (ii) “afforestation/reforestation”, including forests established for
30 ecological restoration, plantations grown for forest products and agroforestry, where biomass may also
31 be a co-product. We then discuss impacts and opportunities common to both systems, before
32 considering impacts and opportunities associated with non-land-based mitigation options that
33 nevertheless occupy land.
34
35 Biomass-based systems
36 Mitigation options that are based on the use of biomass, that is, bioenergy/BECCS, biochar, wood
37 buildings, and other bio-based products, can have different positive and negative effects depending on
38 the character of the mitigation option, the land use, the biomass conversion process, how the bio-based
39 products are used and what other product they substitute (Leskinen et al. 2018; Howard et al. 2021;
40 Myllyviita et al. 2021). The impacts of the same mitigation option can therefore vary significantly and
41 the outcome in addition depends on previous land/biomass use (Cowie et al. 2021). As biomass-based
42 systems commonly produce multiple food, material and energy products, it is difficult to disentangle
43 impacts associated with individual bio-based products (Ahlgren et al. 2015; Djomo et al. 2017;
44 Obydenkova et al. 2021). As for other mitigation options, governance has a critical influence on
45 outcome, but larger scale and higher expansion rate generally translates into higher risk for negative
46 outcomes such as competition for scarce land, freshwater and phosphorous resources, displacement of
Do Not Cite, Quote or Distribute 12-99 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 natural ecosystems, and diminishing capacity of agro-ecosystems to support biodiversity and essential
2 ecosystem services, especially if produced without sustainable land management and in inappropriate
3 contexts (Popp et al. 2017; Dooley and Kartha 2018; Hasegawa et al. 2018; Heck et al. 2018;
4 Humpenöder et al. 2018; Fujimori et al. 2019; Hurlbert et al. 2019; IPBES 2019; Smith et al. 2019b;
5 Drews et al. 2020; Hasegawa et al. 2020; Schulze et al. 2020; Stenzel et al. 2021) (medium evidence,
6 high agreement).
7 Removal of crop and forestry residues can cause land degradation through soil erosion and decline in
8 nutrients and soil organic matter (Cherubin et al. 2018) (robust evidence, high agreement). These risks
9 can be reduced by retaining a proportion of the residues to protect the soil surface from erosion and
10 moisture loss and maintain or increase soil organic matter (See Section 7.4.3.6); incorporating a
11 perennial groundcover into annual cropping systems (Moore et al. 2019); and by replacing nutrients
12 removed, such as by applying ash from bioenergy combustion plants (Kludze et al. 2013; Harris et al.
13 2015; Warren Raffa et al. 2015; de Jong et al. 2017) while safeguarding against contamination risks
14 (Pettersson et al. 2020) (medium evidence, high agreement). Besides topography, soil, and climate
15 conditions, sustainable residue removal rates also depend on the fate of extracted biomass. For example,
16 to maintain the same level of soil organic carbon, the harvest of straw, if used for combustion (which
17 would return no carbon to fields), was estimated to be only 26% of the rate that could be extracted if
18 used for anaerobic digestion involving return of recalcitrant carbon to fields (Hansen et al. 2020).
19 Similarly, biomass pyrolysis produces biochar which can be returned to soils to counteract C losses
20 associated with biomass extraction (Joseph et al. 2021; Lehmann et al. 2021).
21 Expansion of biomass crops, especially monocultures of exotic species, can pose risks to natural
22 ecosystems and biodiversity through introduction of invasive species and land use change, also
23 impacting the mitigation value (robust evidence, high agreement) ((Liu et al. 2014; El Akkari et al.
24 2018). Cultivation of conventional oil, sugar, and starch crops tends to have larger negative impact than
25 lignocellulosic crops (Núñez-Regueiro et al. 2020). Social and environmental outcomes can be
26 enhanced through integration of suitable plants (such as perennial grasses and short rotation woody
27 crops) into agricultural landscapes (within crop rotations or through strategic localization, e.g., as
28 contour belts, along fencelines and riparian buffers). Such integrated systems can provide shelter for
29 livestock, retention of nutrients and sediment, erosion control, pollination, pest and disease control, and
30 flood regulation (robust evidence, high agreement) (See Figure 12.8 below; Box 12.3 and Cross-
31 Working Group Box 3) (Berndes et al. 2008; Christen and Dalgaard 2013; Asbjornsen et al. 2014;
32 Holland et al. 2015; Ssegane et al. 2015; Dauber and Miyake 2016; Milner et al. 2016; Ssegane and
33 Negri 2016; Styles et al. 2016; Crews et al. 2018; Zalesny et al. 2019; Englund et al. 2020b, 2021).
34 (Zheng et al. 2016; Osorio et al. 2019). (Ferrarini et al. 2017; Henry et al. 2018a). Many of the land use
35 practices described above align with agroecology principles [cross ref WGII CCB Nature-Based
36 Solutions, WGII 5.14 box 5.11] and can simultaneously contribute to climate change mitigation, climate
37 change adaptation and reduced risk of land degradation (IPCC 2019) (robust evidence, high agreement).
38
Do Not Cite, Quote or Distribute 12-100 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2 Figure 12.8 Overview of opportunities related to selected land-based climate change mitigation options
3 Afforestation/Reforestation (A/R)
4 When A/R activities comprise the establishment of natural forests, the risk to land is primarily
5 associated with potential displacement of previous land use to new locations, which could indirectly
6 cause land use change including deforestation (see Sections 7.4.2 and 7.6.2.4). A/R (including
7 agroforestry) aimed at providing timber, fibre, biomass, non-timber resources and other ecosystem
8 services can provide renewable resources to society and long-term livelihoods for communities. Forest
9 management and harvesting regimes around the world will adjust in different ways as society seeks to
10 meet climate goals. The outcome depends on forest type, climate, forest ownership and the character
11 and product portfolio of the associated forest industry (Lauri et al. 2019; Favero et al. 2020). How forest
12 carbon stocks, biodiversity, hydrology, etc. are affected by changes in forest management and
13 harvesting in turn depends on both management practices and the characteristics of the forest
14 ecosystems (Eales et al. 2018; Griscom et al. 2018; Kondo et al. 2018; Nieminen et al. 2018; Thom et
15 al. 2018; Runting et al. 2019; Tharammal et al. 2019) (robust evidence, medium agreement). As
16 described above, the GHG savings achieved from producing and using bio-based products will in
17 addition depend on the character of existing societal systems, including technical infrastructure and
18 markets, as this determines the product substitution patterns.
19 Environmental and socio-economic co-benefits are enhanced when ecological restoration principles are
20 applied (Gann et al. 2019) along with effective planning at landscape level and strong governance
21 (Morgan et al., 2020). For example, restoration of natural vegetation and establishing plantations on
22 degraded land enable organic matter to accumulate in the soil and have potential to deliver significant
23 co-benefits for biodiversity, land resource condition and livelihoods (See Box 12.3 and Cross-Working
24 Group Box 3). Tree planting and agroforestry on cleared land can deliver biodiversity benefits (Seddon
25 et al. 2009; Kavanagh and Stanton 2012; Law et al. 2014), with biodiversity outcomes influenced by
26 block size, configuration and species mix (Cunningham et al. 2015; Paul et al. 2016)(robust evidence,
27 high agreement).
28 Risks and opportunities common to biomass production and A/R mitigation options
29 Biomass-based systems and A/R can contribute to addressing land degradation through land
30 rehabilitation or restoration (Box 12.3). Land-based mitigation options that produce biomass for
31 bioenergy/BECCS or biochar through land rehabilitation rather than land restoration imply a trade-off
32 between production / carbon sequestration and biodiversity outcomes (Hua et al. 2016; Cowie et al.
Do Not Cite, Quote or Distribute 12-101 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 2018). Restoration, seeking to establish native vegetation with the aim to maximise ecosystem integrity,
2 landscape connectivity, and conservation of on-ground C stock, will have higher biodiversity benefits
3 than rehabilitation measures (Lin et al. 2013). However, sequestration rate declines as forests mature,
4 and the sequestered C is vulnerable to loss through disturbance such as wildfire, so there is a higher risk
5 of reversal of the mitigation benefit compared with use of biomass for substitution of fossil fuels and
6 GHG-intensive building materials (Russell and Kumar 2017; Dugan et al. 2018; Anderegg et al. 2020).
7 Trade-offs between different ecosystem services, and between societal objectives including climate
8 change mitigation and adaptation, can be managed through integrated landscape approaches that aim to
9 create a mosaic of land uses, including conservation, agriculture, forestry and settlements (Freeman et
10 al. 2015; Nielsen 2016; Reed et al. 2016; Sayer et al. 2017) where each is sited with consideration of
11 land potential and socioeconomic objectives and context (Cowie et al. 2018) (limited evidence, high
12 agreement).
13 Impacts of biomass production and A/R on the hydrological cycle and water availability and quality,
14 depend on scale, location, previous land use/cover and type of biomass production system. For example,
15 extraction of logging residues in forests managed for timber production has little effect on hydrological
16 flows, while land use change to establish dedicated biomass production can have a significant effect
17 (Teter et al. 2018; Drews et al. 2020). Deployment of A/R can affect temperature, albedo and
18 precipitation locally and regionally, and can mitigate or enhance the effects of climate change in the
19 affected areas (Stenzel et al. 2021b) and Section 7.2.4). A/R activities can increase evapotranspiration
20 impacting groundwater and downstream water availability, but can also result in increased infiltration
21 to groundwater and improved water quality (Farley et al. 2005; Zhang et al. 2016, 2017; Lu et al. 2018)
22 and can be beneficial where historical clearing has caused soil salinisation and stream salinity
23 (Farrington and Salama 1996; Marcar 2016). There is limited evidence that very large-scale land use or
24 vegetation cover changes can alter regional climate and precipitation patterns, e.g., downwind
25 precipitation depends on upwind evapotranspiration from forests and other vegetation ( Keys et al.
26 2016; Ellison et al. 2017; van der Ent and Tuinenburg 2017).
27 Another example of beneficial effects includes perennial grasses and woody crops planted to intercept
28 runoff and subsurface lateral flow, reducing nitrate entering groundwater and surface waterbodies (e.g
29 Woodbury et al. 2018; Femeena et al. 2018; Griffiths et al. 2019). In India, (Garg et al. 2011) found
30 desirable effects as a result of planting Jatropha on wastelands previously used for grazing (which could
31 continue in the Jatropha plantations): soil evaporation was reduced, as a larger share of the rainfall was
32 channelled to plant transpiration and groundwater recharge, and less runoff resulted in reduced soil
33 erosion and improved downstream water conditions. Thus, adverse effects can be reduced and synergies
34 achieved when plantings are sited carefully, with consideration of potential hydrological impacts (Davis
35 et al. 2013).
36 Several biomass conversion technologies can generate co-benefits for land and water. Anaerobic
37 digestion of organic wastes (e.g., food waste, manure) produces a nutrient-rich digestate and biogas that
38 can be utilised for heating and cooking or upgraded for use in electricity generation, industrial processes,
39 or as transportation fuel (See Chapter 6) (Parsaee et al. 2019; Hamelin et al. 2021). The digestate is a
40 rich source of nitrogen, phosphorus and other plant nutrients, and its application to farmland returns
41 exported nutrients as well as carbon (Cowie 2020b). Studies have identified potential risks, including
42 Mn toxicity, Cu and Zn contamination, and ammonia emission, compared with application of
43 undigested animal manure (Nkoa 2014). Although the anaerobic digestion process reduces pathogen
44 risk compared with undigested manure feedstocks, it does not destroy all pathogens (Nag et al. 2019).
45 Leakage of methane is a significant risk that needs to be managed, to ensure mitigation potential is
46 achieved (Bruun et al. 2014). Anaerobic digestion of wastewater, such as sugarcane vinasse, reduces
47 methane emissions and pollution loading as well as producing biogas (Parsaee et al. 2019).
Do Not Cite, Quote or Distribute 12-102 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Biorefineries can convert biomass to food, feed and biomaterials along with bioenergy (Aristizábal‐
2 Marulanda and Cardona Alzate 2019; Schmidt et al. 2019). Biorefinery plants are
3 commonly characterised by high process integration to achieve high resource use efficiency, minimise
4 waste production and energy requirements, and maintain flexibility towards changing markets for raw
5 materials and products (Schmidt et al. 2019). Emerging technologies can convert biomass that is
6 indigestible for monogastric animals or humans (e.g., algae, grass, clover or alfalfa) into food and feed
7 products. For example, Lactic acid bacteria can facilitate the use of green plant biomass such as grasses
8 and clover to produce a protein concentrate suitable for animal feed and other products for material or
9 energy use (Lübeck and Lübeck 2019). Selection of crops suitable for co-production of protein feed
10 along with biofuels and other bio-based products can significantly reduce the land conversion pressure
11 by reducing the need to cultivate other crops (e.g., soybean) for animal feeding (Bentsen and Møller
12 2017; Solati et al. 2018). Thus, such solutions, using alternatives to high-input, high-emission grain-
13 based feed, can enable sustainable intensification of agricultural systems with reduced environmental
14 impacts (Jørgensen and Lærke 2016). The use of seaweed and algae as biorefinery feedstock can
15 facilitate recirculation of nutrients from waters to agricultural land, thus reducing eutrophication while
16 substituting purpose-grown feed (Thomas et al. 2021).
17 Pyrolysis can convert organic wastes, including agricultural and forestry residues, food waste, manure,
18 poultry litter and sewage sludge, into combustible gas and biochar, which can be used as a soil
19 amendment (Joseph et al. 2021; Schmidt et al. 2021; Chapter 7). Pyrolysis facilitates nutrient recovery
20 from biomass residues, enabling return to farmland as biochar, noting, however, that a large fraction of
21 nitrogen is lost during pyrolysis (Joseph et al. 2021). Conversion to biochar aids the logistics of
22 transport and land application of materials such as sewage sludge, by reducing mass and volume,
23 improving flow properties, stability and uniformity, and decreasing odour. Pyrolysis is well-suited for
24 materials that may be contaminated with pathogens, microplastics, per- and polyfluoroalkyl substances,
25 such as abattoir and sewage wastes, removing these risks, and reduces availability of heavy metals in
26 feedstock (Joseph et al. 2021). Applying biochar to soil sequesters biochar-carbon for hundreds to
27 thousands of years and can further increase soil carbon by reducing mineralisation of soil organic matter
28 and newly added plant carbon (Singh et al. 2012; Wang et al. 2016a; Weng et al. 2017; Lehmann et al.
29 2021). Biochars can improve a range of soil properties, but effects vary depending on biochar properties,
30 which are determined by feedstock and production conditions (Singh et al. 2012; Wang et al. 2016a),
31 and on the soil properties where biochar is applied (e.g. Razzaghi et al. 2020). Biochars can increase
32 nutrient availability, reduce leaching losses (Singh et al. 2010; Haider et al. 2017) and enhance crop
33 yields particularly in infertile acidic soils (Jeffery et al. 2017), thus supporting food security under
34 changing climate. Biochars can enhance infiltration and soil water-holding capacity, reducing runoff
35 and leaching, increasing water retention in the landscape and improving drought tolerance and resilience
36 to climate change (Quin et al. 2014; Omondi et al. 2016). (See Chapter 7 for review of biochar’s
37 potential contribution to climate change mitigation).
38 Both A/R and dedicated biomass production could have adverse impacts on food security and cause
39 indirect land use change if deployed in locations used for food production (IPCC 2019). But the degree
40 of impact associated with a certain mitigation option also depends on how deployment takes place and
41 also the rate and total scale of deployment. The highest increases in food insecurity due to deployment
42 of land-based mitigation are expected to occur in Sub-Saharan Africa and Asia (Hasegawa et al. 2018).
43 The land area that could be used for bioenergy or other land-based mitigation options with low to
44 moderate risks to food security depends on patterns of socioeconomic development, reaching limits
45 between 1 and 4 million km2 (IPCC 2019; Hurlbert et al. 2019; Smith et al. 2019c).
46 The use of less-productive, degraded/marginal lands has received attention as an option for biomass
47 production and other land-based mitigation that can improve the productive and adaptive capacity of
Do Not Cite, Quote or Distribute 12-103 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 the lands (Liu et al. 2017; Qin et al. 2018; Dias et al. 2021; Kreig et al. 2021) (Section 7.4.4 and Cross-
2 Working Group Box 3). The potential is however uncertain as biomass growth rates may be low, a
3 variety of assessment approaches have been used, and the identification of degraded/marginal land as
4 “available” has been contested, as much low productivity land is used informally by impoverished
5 communities, particularly for grazing, or may be economically infeasible or environmentally
6 undesirable for development of energy crops (Baka 2013, 2014; Haberl et al. 2013; Fritz et al. 2013)
7 (medium evidence, low agreement).
8 As many of the SDGs are closely linked to land use, the identification and promotion of mitigation
9 options that rely on land uses described above can support a growing use of bio-based products while
10 advancing several SDGs, e.g., SDG2 “Zero hunger”, SDG6 “Clean water and sanitation”, SDG7
11 “Affordable and Clean Energy” and SDG15 “Life on Land” (Fritsche et al. 2017; IRP 2019; Blair et al.
12 2021). Policies supporting the target of Land Degradation Neutrality (LDN; SDG 15.3) encourage
13 planning of measures to counteract loss of productive land due to unsustainable agricultural practices
14 and land conversion, through sustainable land management, and strategic restoration and rehabilitation
15 of degraded land (Cowie et al. 2018). LDN can thus be an incentive for land-based mitigation measures
16 that build carbon in vegetation and soil, and can provide impetus for land use planning to achieve
17 multifunctional landscapes that integrate land-based mitigation with other land uses (see Box 12.3). The
18 application of sustainable land management practices that build soil carbon will enhance productivity
19 and resilience of crop and forestry systems, thereby enhancing biomass production (Henry et al. 2018a).
20 Non-bio-based mitigation options can enhance land-based mitigation: enhanced weathering, that is,
21 adding ground silicate rock to soil to take up atmospheric CO2 through chemical weathering (Section
22 12.3), could supply nutrients and alleviate soil acidity, thereby boosting productivity of biomass crops
23 and A/R, particularly when combined with biochar application (Haque et al. 2019; De Oliveira Garcia
24 et al. 2020; Buss et al. 2021) Land rehabilitation and enhanced landscape diversity through production
25 of biomass crops could simultaneously contribute to climate change mitigation, climate change
26 adaptation, addressing land degradation, increasing biodiversity and improving food security in the
27 longer term (Mackey et al. 2020; Chapter 7).
28 Wind power
29 The land requirement and impacts (including visual and noise impacts) of on-shore wind turbines
30 depend on the size and type of installation, and location (Ioannidis and Koutsoyiannis 2020). Wind
31 power and agriculture can coexist in beneficial ways and wind power production on agriculture land is
32 well established ( Fritsche et al. 2017; Miller and Keith 2018a). Spatial planning and local stakeholder
33 engagement can reduce opposition due to visual landscape impacts and noise (Frolova et al. 2019;
34 Hevia-Koch and Ladenburg 2019). Repowering, i.e., replacing with higher capacity wind turbines, can
35 mitigate additional land requirement associated with deployment towards higher share of wind in power
36 systems (Pryor et al. 2020).
37 Mortality and disturbance risks to birds, bats and insects are major ecological concerns associated with
38 wind farms (Thaxter et al. 2017; Cook et al. 2018; Heuck et al. 2019; Coppes et al. 2020; Choi et al.
39 2020; Fernández-Bellon 2020; Marques et al. 2020; Voigt 2021). Careful siting is critical (May et al.
40 2021), while painting blades to increase the visibility can also reduce mortality due to collision (May et
41 al. 2020). Theoretical studies have suggested that wind turbines could lead to warmer night temperatures
42 due to atmospheric mixing (Keith et al. 2004), later confirmed through observation (Zhou et al. 2013),
43 although Vautard et al. (2014) found limited impact at scales consistent with climate policies. More
44 recent studies report mixed results; indications that the warming effect could be substantial with
45 widespread deployment Miller and Keith 2018b and conversely limited impacts on regional climate at
46 20% of US electricity from wind. (Pryor et al. 2020).
Do Not Cite, Quote or Distribute 12-104 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Solar power
2 As for wind power, land impacts of solar power depend on the location, size and type of installation
3 (Ioannidis and Koutsoyiannis 2020). Establishment of large-scale solar farms could have positive or
4 negative environmental effects at the site of deployment, depending on the location. Solar PV and CSP
5 power installations can lock away land areas, displacing other uses (Mohan 2017). Solar PV can be
6 deployed in ways that enhance agriculture: for example, Hassanpour Adeh et al. (2018) found that
7 biomass production and water use efficiency of pasture increased under elevated solar panels. PV
8 systems under development may achieve significant power generation without diminishing agricultural
9 output (Miskin et al. 2019). Global mapping of solar panel efficiency showed that croplands, grasslands
10 and wetlands are located in regions with the greatest solar PV potential (Adeh et al. 2019). Dual-use
11 agrivoltaic systems are being developed that overcome previously recognised negative impact on crop
12 growth, mainly due to shadows (Armstrong et al. 2016; Marrou et al. 2013b,a), thus facilitating
13 synergistic co-location of solar photovoltaic power and cropping (Miskin et al. 2019; Adeh et al. 2019).
14 Assessment of the potential for optimising deployment of solar PV and energy crops on abandoned
15 cropland areas produced an estimate of the technical potential for optimal combination at 125 EJ per
16 year (Leirpoll et al. 2021).
17 Deserts can be well-suited for solar PV and CSP farms, especially at low latitudes where global
18 horizontal irradiance is high, as there is lower competition for land and land carbon loss is minimal,
19 although remote locations may pose challenges for power distribution (Xu et al. 2016). Solar arrays can
20 reduce the albedo, particularly in desert landscapes, which can lead to local temperature increases and
21 regional impacts on wind patterns (Millstein and Menon 2011). Modelling studies suggest that large-
22 scale wind and solar farms, for example in the Sahara (Li et al. 2018), could increase rainfall through
23 reduced albedo and increased surface roughness, stimulating vegetation growth and further increasing
24 regional rainfall (Li et al. 2018) (limited evidence). Besides impacts at the site of deployment, wind
25 and solar power affect land through mining of critical minerals required by these technologies (Viebahn
26 et al. 2015; McLellan et al. 2016; Carrara et al. 2020).
27 Nuclear power
28 Nuclear power has land impacts and risks associated with mining operations (Falck 2015; Winde et al.
29 2017; Srivastava et al. 2020) and disposal of spent fuel (IAEA 2006a; Ewing et al. 2016; Bruno et al.
30 2020), but the land occupation is small compared to many other mitigation options. Substantial volumes
31 of water are required for cooling (Liao et al. 2016), as for all thermal power plants, but most of this
32 water is returned to rivers and other water bodies after use (Sesma Martín and Rubio-Varas 2017).
33 Negative impacts on aquatic systems can occur due to chemical and thermal pollution loading (Fricko
34 et al. 2016; Raptis et al. 2016; Bonansea et al. 2020). The major risk to land from nuclear power is that
35 a nuclear accident leads to radioactive contamination. An extreme example, the 1986 Chernobyl
36 accident in Ukraine, resulted in radioactive contamination across Europe. Most of the fallout
37 concentrated near Belarus, Ukraine and Russia, where some 125,000 km2 of land (more than a third of
38 which was in agricultural use) was contaminated. About 350,000 people were relocated away from
39 these areas (Sovacool 2008; IAEA 2006b). About 116,000 people were permanently evacuated from
40 the 4,200 km Chernobyl exclusion zone (IAEA 2006a). New reactor designs with passive and enhanced
41 safety systems reduce the risk of such accidents significantly (Section 6.4.2.4). An example of
42 alternatives to land reclamation for productive purposes, a national biosphere reserve has been
43 established around Chernobyl to conserve, enhance and manage carbon stocks and biodiversity
44 (Deryabina et al. 2015; Ewing et al. 2016), although invertebrate and plant populations area affected
45 (Mousseau and Møller 2014, 2020).
Do Not Cite, Quote or Distribute 12-105 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Hydropower
2 Reservoir hydropower projects submerge areas as dams are established for water storage. Hydropower
3 can be associated with significant and highly varying land occupation and carbon footprint (Poff and
4 Schmidt 2016; Scherer and Pfister 2016a; Ocko and Hamburg 2019; dos Santos et al. 2017). The
5 flooding of land causes CH4 emissions due to the anaerobic decomposition of submerged vegetation
6 and there is also a loss of C sequestration due to mortality of submerged vegetation. The size of GHG
7 emissions depends on the amount of vegetation submerged. The carbon in accumulated sediments in
8 reservoirs may be released to the atmosphere as CO2 and CH4 upon decommissioning of dams, and
9 while uncertain, estimates indicate that these emissions can make up a significant part of the cumulative
10 GHG emissions of hydroelectric power plants (Almeida et al. 2019; Moran et al. 2018; Ocko and
11 Hamburg 2019). Positive radiative forcing due to lower albedo of hydropower reservoirs compared to
12 surrounding landscapes can reduce mitigation contribution significantly (Wohlfahrt et al. 2021).
13 Hydropower can have high water usage due to evaporation from dams (Scherer and Pfister 2016b).
14 Hydropower projects may impact aquatic ecology and biodiversity, necessitate the relocation of local
15 communities living within or near the reservoir or construction sites and affect downstream
16 communities (in positive or negative ways) (Barbarossa et al. 2020; Moran et al. 2018). Displacement
17 as well as resettlement schemes can have both socio-economic and environmental consequences
18 including those associated with establishment of new agricultural land (Nguyen et al. 2017; Ahsan and
19 Ahmad 2016). Dam construction may also stimulate migration into the affected region, which can lead
20 to deforestation and other negative impacts (Chen et al. 2015). Impacts can be mitigated through basin-
21 scale dam planning that considers GHG emissions along with social and ecological effects (Almeida et
22 al. 2019). Land occupation is minimal for run-of-river hydropower installations, but without storage
23 they have no resilience to drought and installations inhibit dispersal and migration of organisms (Lange
24 et al. 2018). Reservoir hydropower schemes can regulate water flows and reduce flood damage to
25 agricultural production (Amjath-Babu et al. 2019). On the other hand, severe flooding due to failure of
26 hydropower dams has caused fatalities, damage to infrastructure and loss of productive land (Lu et al.
27 2018; Zhang et al. 2016, 2017; Farley et al. 2005)(Farrington and Salama 1996; Marcar 2016; Kalinina
28 et al. 2018).
29
30 12.5.4 Governance of land-related impacts of mitigation options
31 The land sector (Chapter 7) contributes to mitigation via emissions reduction and enhancement of land
32 carbon sinks, and by providing biomass for mitigation in other sectors. Key challenges for governance
33 of land-based mitigation include social and environmental safeguards (Larson et al. 2018; Sills et al.
34 2017; Duchelle et al. 2017); insufficient financing (Turnhout et al. 2017); capturing co-benefits;
35 ensuring additionality, addressing non-permanence of carbon sequestration; monitoring, reporting, and
36 verification (MRV) of emissions reduction and carbon dioxide removals; and avoiding leakage or spill-
37 over effects. Governance approaches to addressing these challenges are discussed in section 7.6, and
38 include MRV systems and integrity criteria for project-level emissions trading; payments for ecosystem
39 services; land use planning and land zoning; certification schemes, standards and codes of practice.
40
41 With respect to renewable energy options that occupy land, the focus of governance has been directed
42 to technological adoption and, public acceptance (Sequeira and Santos 2018), rather than land use.
43 Recent work has found that spatial processes shape the emerging energy transition, creating zones of
44 friction between global investors, national and local governments, and civil society (McEwan 2017;
Do Not Cite, Quote or Distribute 12-106 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Jepson and Caldas 2017). For example, hydropower and ground-based solar parks in India have
2 involved enclosure of lands designated as degraded, displacing pastoral use by vulnerable communities,
3 constituting forms of spatial injustice (Yenneti et al. 2016). Hydropower leads to dam-induced
4 displacement, and though this can be addressed through compensation mechanisms, governance is
5 complicated by a lack of transparency in resettlement data (Kirchherr et al. 2016, 2019). Renewable
6 energy production is resulting in new land conflict frontiers where degraded land is framed as having
7 mitigation value such as for palm oil production and wind power in Mexico (Backhouse and Lehmann
8 2020); land use conflict as well as impacts on wildlife from large-scale solar installations have also
9 emerged in the southwestern United States (Mulvaney 2017). The renewable energy transition also
10 involves the extraction of critical minerals used in renewable energy technologies, such as lithium and
11 cobalt. Governance challenges include the lack of transparent greenhouse gas accounting for mining
12 activities (Lee et al. 2020a), and threats to biodiversity from land disturbance, which require strategic
13 planning to address (Sonter et al. 2020a). Strategic spatial planning is needed more generally to address
14 trade-offs between using land for renewable energy and food: for example, agriculture can be co-located
15 with solar photovoltaics (Barron-Gafford et al. 2019) or wind power (Miller and Keith 2018a).
16 Integrative spatial planning can integrate renewable energy with not just agriculture, but mobility and
17 housing (Hurlbert et al. 2019). Integrated planning is needed to avoid scalar pitfalls, and local and
18 regional contextualised governance solutions need to be sited within a planetary frame of reference
19 (Biermann et al. 2016). Greater planning and coordination are also needed to ensure co-benefits from
20 land-based mitigation (see Box 12.3) as well as from CDR and efforts to reduce food systems emissions.
21 In emerging domains for governance such as land-based mitigation, global institutions, private sector
22 networks and civil society organisations are also playing key roles in terms of norm-setting. The shared
23 languages and theoretical frameworks, or cognitive linkages (Pattberg et al. 2018) that arise with
24 polycentric governance can not only be helpful in creating expectations and establishing benchmarks
25 for (in)appropriate practices where enforceable ‘hard law’ is missing (Karlsson-Vinkhuyzen et al. 2018;
26 Gajevic Sayegh 2020), but can also form the basis of voluntary guidelines or niche markets (See also
27 the case study in Box 12.3) However, the ability to apply participatory processes for developing
28 voluntary guidelines and other participatory norm-setting endeavours varies from place to place. Social
29 and cultural norms shape the ability of women, youth, and different ethnic groups to participate in
30 governance fora, such as those around agroecological transformation (Anderson et al. 2019).
31 Furthermore, establishing new norms alone does not solve structural challenges such as lack of access
32 to food, confront power imbalances, or provide mechanisms to deal with uncooperative actors
33 (Morrison et al. 2019).
34 START BOX 12.3 HERE
35 Box 12.3 Land Degradation Neutrality as a framework to manage trade-offs in land-based
36 mitigation
37 The UNCCD introduced the concept of Land Degradation Neutrality (LDN), defined as “a state
38 whereby the amount and quality of land resources necessary to support ecosystem functions and
39 services and enhance food security remain stable or increase within specified temporal and spatial scales
40 and ecosystems” (UNCCD 2015), and it has been adopted as a target of Goal 15 of the SDGs, Life on
41 Land. At December 2020, 124 (mostly developing) countries have committed to pursue voluntary LDN
42 targets.
43
44 The goal of LDN is to maintain or enhance land-based natural capital, and its associated ecosystem
45 services such as provision of food and regulation of water and climate, while enhancing the resilience
46 of the communities that depend on the land. LDN encourages a dual-pronged approach promoting
47 sustainable land management (SLM) to avoid or reduce land degradation, combined with strategic effort
Do Not Cite, Quote or Distribute 12-107 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 in land restoration and rehabilitation to reverse degradation on degraded lands and thereby deliver the
2 target of “no net loss” of productive land (Orr et al. 2017).
3
4 In the context of LDN, land restoration refers to actions undertaken with the aim of reinstating
5 ecosystem functionality, whereas land rehabilitation refers to actions undertaken with a goal of
6 provision of goods and services (Cowie et al. 2018). Restoration interventions can include destocking
7 to encourage regeneration of native vegetation; shelter belts of local species established from seed or
8 seedlings, strategically located to provide wildlife corridors and link habitat; and rewetting drained
9 peatland. “Farmer-managed natural regeneration” is a low-cost restoration approach in which
10 regeneration of tree stumps and roots is encouraged, stabilising soil and enhancing soil nutrients and
11 organic matter levels (Chomba et al. 2020; Lohbeck et al. 2020). Rehabilitation actions include
12 establishment of energy crops, or afforestation with fast-growing exotic trees to sequester carbon or
13 produce timber. Application of biochar can facilitate rehabilitation by enhancing nutrient retention and
14 water holding capacity, and stimulating microbial activity (Cowie 2020a).
15
16 SLM, rehabilitation and restoration activities undertaken towards national LDN targets have potential
17 to deliver substantial CDR through carbon sequestration in vegetation and soil. In addition, biomass
18 production, for bioenergy or biochar, could be an economically viable land use option for reversing
19 degradation, through rehabilitation. Alternatively, a focus on ecological restoration (Gann et al. 2019)
20 as the strategy for reversing degradation will deliver greater biodiversity benefits.
21
22 Achieving neutrality requires estimating the likely impacts of land-use and land management decisions,
23 to determine the area of land, of each land type, that is likely to be degraded (Orr et al. 2017). This
24 information is used to plan interventions to reverse degradation on an equal area of the same land type.
25 Therefore, pursuit of LDN requires concerted and coordinated efforts to integrate LDN objectives into
26 land-use planning and land management, underpinned by sound understanding of the human-
27 environment system and effective governance mechanisms.
28
29 Countries are advised to apply a landscape-scale approach for planning LDN interventions, in which
30 land uses are matched to land potential, and resilience of current and proposed land uses is considered,
31 to ensure that improvement in land condition is likely to be maintained (Cowie 2020a). A participatory
32 approach that enables effective representation of all stakeholders is encouraged, to facilitate equitable
33 outcomes from planning decisions, recognising that decisions on LDN interventions are likely to
34 involve trade-offs between various environmental and socio-economic objectives (Schulze et al. 2021).
35 Planning and implementation of LDN programmes provides a framework in which locally-adapted
36 land-based mitigation options can be integrated with use of land for production, conservation and
37 settlements, in multifunctional landscapes where trade-offs are recognised and managed, and
38 synergistic opportunities are sought. LDN is thus a vehicle to focus collaboration in pursuit of the
39 multiple land-based objectives of the multilateral environmental agreements and the SDGs.
40
41
42
Do Not Cite, Quote or Distribute 12-108 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2
3 Box 12.3, Figure 1 Schematic illustrating the elements of the Land Degradation Neutrality conceptual
4 framework.
5 Source: (Cowie et al. 2018)
6
7 END BOX 12.3 HERE
8
9 Table 12.10 collates risks, impacts and opportunities associated with different mitigation
10 options that occupy land.
11
Do Not Cite, Quote or Distribute 12-109 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Table 12.10 Summary of impacts, risks and co-benefits associated with land occupation by mitigation
2 options considered in section 12.5.
Mitigation option Impacts and risks Opportunities for co-benefits
Non-bio-based options that may displace food production
Solar farms Land use competition; Loss of soil Target areas unsuitable for agriculture
carbon; heat island effect (scale such as deserts
dependent) {12.5.3}
{12.5.3, 12.5.4}
Hydropower (dams) Land use competition, displacement Water storage (including for irrigation)
of natural ecosystems, CO2 and CH4 and regulation of water flows; Pumped
emissions storage can store excess energy from
{12.5.3, 12.5.4} other renewable generation sources.
{12.5.3}
Non-bio-based options that can (to a varying degree) be integrated with food production
Wind turbines May affect local/regional weather and Design and siting informed by t visual
climate (scale dependent) landscape impacts, relevant habitats, and
flight trajectories of migratory birds.
Impact on wildlife and visual impacts {12.5.3}
{12.5.3}
Solar panels Land use competition Integration with buildings and other
{12.5.3} infrastructure. integration with food
production is being explored
{12.5.2}
Enhanced weathering Disturbance at sites of extraction; Increase crop yields and biomass
Ineffective in low rainfall regions production through nutrient supply and
{12.3.1.2} increasing pH of acid soils; synergies
with biochar
{12.5.3}
Bio-based options that may displace existing food production
A/R Land use competition, potentially Strategic siting to minimise adverse
leading to indirect land use change; impacts on hydrology, land use,
reduced water availability; loss of biodiversity
biodiversity {12.5.3}
{12.5.3}
Biomass crops Land use competition, potentially Strategic siting to minimise adverse
leading to indirect land use change; impacts / enhance beneficial effects on
reduced water availability; reduced land use, landscape variability,
soil fertility; loss of biodiversity biodiversity, soil organic matter,
{12.5.3} hydrology and water quality
{12.5.3}
Bio-based options that can (to a varying degree) be combined with food production
Do Not Cite, Quote or Distribute 12-110 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
Agroforestry Competition with adjacent crops and Shelter for stock and crops,
pastures reduces yields diversification, biomass production,
{7.4.3.3} increases soil organic matter and soil
fertility. Increased biodiversity and
perennial vegetation enhance beneficial
organisms; can reduce need for pesticides
{7.4.3.3, 12.5.3}
Soil carbon management Increase in nitrous oxide emissions if Increasing soil organic matter improves
in croplands and fertiliser used to enhance crop soil health, increases crop and pasture
grasslands production; Reduced cereal yields, and resilience to drought, can
production through increased crop reduce fertiliser requirement, nutrient
legumes and pasture phases could leaching and need for land use change.
lead to indirect land use change {7.4.3.1}
{7.4.3.1, 7.4.3.6}
Biochar addition to soil Land use competition if biochar is Facilitate beneficial use of organic
produced from purpose-grown residues, to return nutrients to farmland.
biomass. Loss of forest carbon stock Increase land productivity to increase C
and impacts on biodiversity if sequestration in vegetation and soil.
biomass is harvested unsustainably. Increase nutrient-use efficiency, and
{12.5.3} reduce requirement for chemical
fertiliser.
{7.4.3.2, 12.5.3}
Harvest residue extraction Decline in soil organic matter and soil Retain portion of stubble; return nutrients
and use for bioenergy, fertility e.g. as ash
biochar and other bio- {12.5.3} Utilising forest residues for bioenergy
products reduces fuel load and wildfire risk
{7.4.3.2, 12.5.3}
Manure management (i.e., Risk of fugitive emissions Biogas as renewable energy source.
for biogas) Apply digestate as soil amendment
Can contain pathogens {12.5.3}
{7.4.3.7, 12.5.3}
Options that don’t occupy land used for food production
Management of organic Can contain contaminants (heavy Processing using anaerobic digestion or
waste (food waste, bio- metals, persistent organic pollutants, pyrolysis produces renewable gas and
solids, organic component pathogens) soil amendment, enabling return of
of MSW) {12.5.3} nutrients to farmland. (note that some
feedstock nitrogen is lost in pyrolysis)
{12.5.3}
A/R and biomass High labour and material inputs can Application of biochar can re-establish
production on degraded be needed to restore productivity on nutrient cycling; bioenergy crops can add
non-forested land (e.g., degraded land. Abandoned land can organic matter, restoring soil fertility,
abandoned agricultural support informal grazing and have and can remove heavy metals, enabling
land) significant biodiversity value. food production.
Reduced water availability. {7.4.3.2, 12.5.3}
{12.5.3}
1
Do Not Cite, Quote or Distribute 12-111 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 START CROSS-WORKING GROUP BOX 3 HERE
2 Cross-Working Group Box in Working Group II, Chapter 5
3 Cross-Working Group Box 3: Mitigation and Adaptation via the Bioeconomy
4 Henry Neufeldt (Denmark/Germany), Göran Berndes (Sweden), Almut Arneth (Germany), Rachel
5 Bezner Kerr (USA/Canada), Luisa F Cabeza (Spain), Donovan Campbell (Jamaica), Jofre Carnicer Cols
6 (Spain), Annette Cowie (Australia), Vassilis Daioglou (Greece), Joanna House (UK), Adrian Leip
7 (Italy/Germany), Francisco Meza (Chile), Michael Morecroft (UK), Gert-Jan Nabuurs (Netherlands),
8 Camille Parmesan (UK/USA), Julio C Postigo (USA/Peru), Marta G. Rivera-Ferre (Spain), Raphael
9 Slade (UK), Maria Cristina Tirado von der Pahlen (USA/Spain), Pramod K. Singh (India), Peter Smith
10 (UK)
11 Summary statement
12 The growing demand for biomass offers both opportunities and challenges to mitigate and adapt to
13 climate change and natural resource constraints (high confidence). Increased technology innovation,
14 stakeholder integration and transparent governance structures and procedures at local to global scales
15 are key to successful bioeconomy deployment maximizing benefits and managing trade-offs (high
16 confidence).
17 Limited global land and biomass resources accompanied by growing demands for food, feed, fibre, and
18 fuels, together with prospects for a paradigm shift towards phasing out fossil fuels, set the frame for
19 potentially fierce competition for land 2 and biomass to meet burgeoning demands even as climate
20 change increasingly limits natural resource potentials (high confidence).
21 Sustainable agriculture and forestry, technology innovation in bio-based production within a circular
22 economy and international cooperation and governance of global trade in products to reflect and
23 disincentivize their environmental and social externalities, can provide mitigation and adaptation via
24 bioeconomy development that responds to the needs and perspectives of multiple stakeholders to
25 achieve outcomes that maximize synergies while limiting trade-offs (high confidence).
26 Background
27 There is high confidence that climate change, population growth and changes in per capita consumption
28 will increase pressures on managed as well as natural and semi-natural ecosystems, exacerbating
29 existing risks to livelihoods, biodiversity, human and ecosystem health, infrastructure, and food systems
30 (Conijn et al. 2018; IPCC 2018, 2019; Lade et al. 2020). At the same time, many global mitigation
31 scenarios presented in IPCC assessment reports rely on large GHG emissions reduction in the AFOLU
32 sector and concurrent deployment of reforestation/afforestation and biomass use in a multitude of
33 applications (Rogelj et al. 2018; Hanssen et al. 2020; AR6 WG1 Ch.4 and Ch.5; AR6 WG3 Ch.3 and
34 Ch.7).
35 Given the finite availability of natural resources, there are invariably trade-offs that complicate land-
36 based mitigation unless land productivity can be enhanced without undermining ecosystem services
37 (e.g., Obersteiner et al. 2016; Campbell et al. 2017; Conijn et al. 2018; Caron et al. 2018; WRI 2018;
38 Heck et al. 2018; Smith et al. 2019). Management intensities can often be adapted to local conditions
FOOTNOTE2 For lack of space the focus is on land only although the bioeconomy also includes sea-related
bioresources.
Do Not Cite, Quote or Distribute 12-112 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 with consideration of other functions and ecosystem services, but at a global scale the challenge remains
2 to avoid further deforestation and degradation of intact ecosystems, in particular biodiversity-rich
3 systems (AR6 WGII Cross-Chapter Box on NATURAL), while meeting the growing demands. Further,
4 increased land-use competition can affect food prices and impact food security and livelihoods (To and
5 Grafton 2015; Chakravorty et al. 2017), with possible knock-on effects related to civil unrest (Abbott
6 et al. 2017; D’Odorico et al. 2018).
7 Developing new bio-based solutions while mitigating overall biomass demand growth
8 Many existing bio-based products have significant mitigation potential. Increased use of wood in
9 buildings can reduce GHG emissions from cement and steel production while providing carbon storage
10 (Churkina et al. 2020). Substitution of fossil fuels with biomass in manufacture of cement and steel can
11 reduce GHG emissions where these materials are difficult to replace. Dispatchable power based on
12 biomass can provide power stability and quality as the contribution from solar and wind power increases
13 (See AR6 WG3 Ch.6), and biofuels can contribute to reducing fossil fuel emissions in the transport and
14 industry sectors (See AR6 WG3 Ch.10 and Ch.11). The use of bio-based plastics, chemicals and
15 packaging could be increased, and biorefineries can achieve high resource-use efficiency in converting
16 biomass into food, feed, fuels and other bio-based products (Aristizábal‐Marulanda and Cardona Alzate
17 2019; Schmidt et al. 2019). There is also scope for substituting existing bio-based products with more
18 benign products. For example, cellulose-based textiles can replace cotton, which requires large amounts
19 of water, chemical fertilizers and pesticides to ensure high yields.
20 While increasing and diversified use of biomass can reduce the need for fossil fuels and other GHG-
21 intensive products, unfavourable GHG balances may limit the mitigation value. Growth in biomass use
22 may in the longer term also be constrained by the need to protect biodiversity and ecosystems’ capacity
23 to support essential ecosystem services. Biomass use may also be constrained by water scarcity and
24 other resource scarcities, and/or challenges related to public perception and acceptance due to impacts
25 caused by biomass production and use. Energy conservation and efficiency measures and deployment
26 of technologies and systems that do not rely on carbon, e.g., carbon-free electricity supporting, inter
27 alia, electrification of transport as well as industry processes and residential heating (IPCC 2018; UNEP
28 2019), can constrain the growth in biomass demand when countries seek to phase out fossil fuels and
29 other GHG-intensive products while providing an acceptable standard of living. Nevertheless, demand
30 for bio-based products may become high where full decoupling from carbon is difficult to achieve (e.g.,
31 aviation, bio-based plastics and chemicals) or where carbon storage is an associated benefit (e.g., wood
32 buildings, BECCS, biochar for soil amendments), leading to challenging trade-offs (e.g., food security,
33 biodiversity) that need to be managed in environmentally sustainable and socially just ways.
34 Changes on the demand side as well as improvements in resource-use efficiencies within the global
35 food and other bio-based systems can also reduce pressures on the remaining land resources. For
36 example, dietary changes toward more plant-based food (where appropriate) and reduced food waste
37 can provide climate change mitigation along with health benefits (Willett et al. 2019; WG3 Ch 7.4 and
38 Ch 12.4) and other co-benefits with regard to food security, adaptation and land use (Smith et al. 2019a;
39 Mbow et al. 2019; WG2 Ch.5). Advancements in the provision of novel food and feed sources (e.g.,
40 cultured meat, insects, grass-based protein feed and cellular agriculture) can also limit the pressures on
41 finite natural resources (Parodi et al. 2018; Zabaniotou 2018; WG3 Ch 12.4).
42
43 Circular bioeconomy
44
45 Circular economy approaches (AR6 WG3 Ch 12.6) are commonly depicted by two cycles, where the
46 biological cycle focuses on regeneration in the biosphere and the technical cycle focuses on reuse,
Do Not Cite, Quote or Distribute 12-113 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 refurbishment and recycling to maintain value and maximize material recovery (Mayer et al. 2019a).
2 Biogenic carbon flows and resources are part of the biological carbon cycle, but carbon-based products
3 can be included in, and affect, both the biological and the technical carbon cycles (Velenturf et al. 2019;
4 Winans et al. 2017; Kirchherr et al. 2017). The integration of circular economy and bioeconomy
5 principles has been discussed in relation to organic waste management (Teigiserova et al. 2020), societal
6 transition and policy development (Bugge et al. 2019; European Commission 2018) as well as COVID-
7 19 recovery strategies (Palahí et al. 2020). To maintain the natural resource base, circular bioeconomy
8 emphasizes sustainable land use and the return of biomass and nutrients to the biosphere when it leaves
9 the technical cycle.
10
11 Biomass scarcity is an argument for adopting circular economy principles for the management of
12 biomass as for non-renewable resources. This includes waste avoidance, product reuse and material
13 recycling, which keep down resource use while maintaining product and material value. However, reuse
14 and recycling is not always feasible, e.g., when biofuels are used for transport and bio-based
15 biodegradable chemicals are used to reduce ecological impacts where losses to the environment are
16 unavoidable. A balanced approach to management of biomass resources could take departure in the
17 carbon cycle from a value-preservation perspective and the possible routes that can be taken for biomass
18 and carbon, considering a carbon budget defined by the Paris Agreement, principles for sustainable land
19 use and natural ecosystem protection.
20
21 Land use opportunities and challenges in the bioeconomy
22 Analyses of synergies and trade-offs between adaptation and mitigation in the agriculture and forestry
23 sectors show that outcomes depend on context, design and implementation, so actions have to be
24 tailored to the specific conditions to minimize adverse effects (Kongsager 2018). This is supported in
25 literature analyzing the nexus between land, water, energy and food in the context of climate change
26 which consistently concludes that addressing these different domains together rather than in isolation
27 would enhance synergies and reduce trade-offs (Obersteiner et al. 2016; D’Odorico et al. 2018; Soto
28 Golcher and Visseren-Hamakers 2018; Momblanch et al. 2019; Froese et al. 2019).
29 Nature-based solutions addressing climate change can provide opportunities for sustainable livelihoods
30 as well as multiple ecosystem services, such as flood risk management through floodplain restoration,
31 saltmarshes, mangroves or peat renaturation (UNEP 2021; AR6 WGII Cross-Chapter Box on
32 NATURAL). Climate-smart agriculture can increase productivity while enhancing resilience and
33 reducing GHG emissions inherent to production (Lipper et al. 2014; Singh and Chudasama 2021). (Bell
34 et al. 2018; FAO 2019b; Singh and Chudasama 2021) Similarly, climate-smart forestry considers the
35 whole value chain and integrates climate objectives into forest sector management through multiple
36 measures (from strict reserves to more intensively managed forests) providing mitigation and adaptation
37 benefits (Nabuurs et al. 2018; Verkerk et al. 2020) (AR6 WG3 Ch 7.3).
38 Agroecological approaches can be integrated into a wide range of land management practices to support
39 a sustainable bioeconomy and address equity considerations (HLPE 2019). Relevant land-use practices,
40 such as agroforestry, intercropping, organic amendments, cover crops and rotational grazing, can
41 provide mitigation and support adaption to climate change via food security, livelihoods, biodiversity
42 and health co-benefits (Bezner Kerr et al. 2019; Bharucha et al. 2020; Clark et al. 2019b; D’Annolfo et
43 al. 2017; Garibaldi et al. 2016; Ponisio et al. 2015; Renard and Tilman 2019; HLPE 2019; Sinclair et
44 al. 2019; Córdova et al. 2019; Mbow et al. 2019; Bezner Kerr et al. 2021; and AR6 WGII Chapter 2
45 Cross-Chapter Box NATURAL). Strategic integration of appropriate biomass production systems into
46 agricultural landscapes can provide biomass for bioenergy and other bio-based products while providing
47 co-benefits such as enhanced landscape diversity, habitat quality, retention of nutrients and sediment,
Do Not Cite, Quote or Distribute 12-114 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 erosion control, climate regulation, flood regulation, pollination and biological pest and disease control
2 (Christen and Dalgaard 2013; Asbjornsen et al. 2014; Englund et al. 2020; Cacho et al. 2018; Dauber
3 and Miyake 2016; Holland et al. 2015; Milner et al. 2016; Ssegane et al. 2015; Ssegane and Negri 2016;
4 Styles et al. 2016; Zalesny et al. 2019; Zumpf et al. 2017; HLPE 2019; Cubins et al. 2019; Alam and
5 Dwivedi 2019; Olsson et al. 2019) (AR6 WGIII Chapter 12 Box 12.3 on UNCCD-LDN). Such
6 approaches can help limit environmental impacts from intensive agriculture while maintaining or
7 increasing land productivity and biomass output.
8 Cross-Working Group Box 3, Figure 1 Left: High-input intensive agriculture, aiming for high yields of a
9 few crop species, with large fields and no semi-natural habitats. Right: Agroecological agriculture,
10 supplying a range of ecosystem services, relying on biodiversity and crop and animal diversity instead of
11 external inputs, and integrating plant and animal production, with smaller fields and presence of semi-
12 natural habitats.
13 Credit: Jacques Baudry (left); Valérie Viaud (right), published in (van der Werf et al. 2020)
14 Transitions from conventional to new biomass production and conversion systems include challenges
15 related to cross-sector integration and limited experience with new crops and land use practices,
16 including needs for specialized equipment (Thornton and Herrero 2015; HLPE 2019; AR6 WG2
17 Section 5.10). Introduction of agroecological approaches and integrated biomass/food crop production
18 can result in lower food crop yields per hectare, particularly during transition phases, potentially causing
19 indirect land use change, but can also support higher and more stable yields, reduce costs, and increase
20 profitability under climate change (Muller et al. 2017; Seufert and Ramakutty 2017; Barbieri et al. 2019;
21 HLPE 2019; Sinclair et al. 2019; Smith et al. 2019a, 2020). Crop diversification, organic amendments,
22 and biological pest control (HLPE 2019) can reduce input costs and risks of occupational pesticide
23 exposure and food and water contamination (González-Alzaga et al. 2014; EFSA 2017; Mie et al. 2017),
24 reduce farmers’ vulnerability to climate change (e.g., droughts and spread of pests and diseases affecting
25 plant and animal health (Delcour et al. 2015; FAO 2020) and enhance provisioning and sustaining
26 ecosystem services, such as pollination (D’Annolfo et al. 2017; Sinclair et al. 2019).
27 Barriers toward wider implementation include absence of policies that compensate land owners for
28 providing enhanced ecosystem services and other environmental benefits, which can help overcome
29 short term losses during the transition from conventional practices before longer term benefits can
30 accrue. Other barriers include limited access to markets, knowledge gaps, financial, technological or
31 labour constraints, lack of extension support and insecure land tenure (Jacobi et al. 2017; Kongsager
32 2017; Hernández-Morcillo et al. 2018; Iiyama et al. 2018; HLPE 2019). Regional-level agroecology
33 transitions may be facilitated by co-learning platforms, farmer networks, private sector, civil society
34 groups, regional and local administration and other incentive structures (e.g. price premiums, access to
35 credit, regulation) (Coe et al. 2014; Pérez-Marin et al. 2017; Mier y Terán Giménez Cacho et al. 2018;
Do Not Cite, Quote or Distribute 12-115 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 HLPE 2019; Valencia et al. 2019; SAEPEA 2020). With the right incentives, improvements can be
2 made with regard to profitability, making alternatives more attractive to land owners.
3 Governing the solution space
4 Literature analysing the synergies and trade-offs between competing demands for land suggest that
5 solutions are highly contextualized in terms of their environmental, socioeconomic and governance-
6 related characteristics, making it difficult to devise generic solutions (Haasnoot et al. 2020). Aspects of
7 spatial and temporal scale can further enhance the complexity, for instance where transboundary effects
8 across jurisdictions or upstream-downstream characteristics need to be considered, or where climate
9 change trajectories might alter relevant biogeophysical dynamics (Postigo and Young 2021).
10 Nonetheless, there is broad agreement that taking the needs and perspectives of multiple stakeholders
11 into account in a transparent process during negotiations improves the chances of achieving outcomes
12 that maximize synergies while limiting trade-offs (Ariti et al. 2018; Metternicht 2018; Favretto et al.
13 2020; Kopáček 2021; Muscat et al. 2021). Yet differences in agency and power between stakeholders
14 or anticipated changes in access to or control of resources can undermine negotiation results even if
15 there is a common understanding of the overarching benefits of more integrated environmental
16 agreements and the need for greater coordination and cooperation to avoid longer-term losses to all
17 (Aarts and Leeuwis 2010; Weitz et al. 2017). There is also the risk that strong local participatory
18 processes can become disconnected from broader national plans, and thus fail to support the
19 achievement of national targets. Thus, connection between levels is needed to ensure that ambition for
20 transformative change is not derailed at local level (Aarts and Leeuwis 2010; Postigo and Young 2021).
21 Decisions on land uses between biomass production for food, feed, fibre or fuel, as well as nature
22 conservation or restoration and other uses (e.g., mining, urban infrastructure), depend on differences in
23 perspectives and values. Because the availability of land for diverse biomass uses is invariably limited,
24 setting priorities for land-use allocations therefore first depends on making the perspectives underlying
25 what is considered as ‘high-value’ explicit (Fischer et al. 2007; Garnett et al. 2015; De Boer and Van
26 Ittersum 2018; Muscat et al. 2020). Decisions can then be made transparently based on societal norms,
27 needs and the available resource base. Prioritization of land-use for the common good therefore requires
28 societal consensus-building embedded in the socioeconomic and cultural fabric of regions, societies
29 and communities. Integration of local decision-making with national planning ensures local actions
30 complement national development objectives.
31 International trade in the global economy today provides important opportunities to connect producers
32 and consumers, effectively buffering price volatilities and potentially offering producers countries
33 access to global markets, which can be seen as an effective adaptation measure (Baldos and Hertel 2015;
34 Costinot et al. 2016; Hertel and Baldos 2016; Gouel and Laborde 2021; AR6 WG2 Ch 5.11). But there
35 is also clear evidence that international trade and the global economy can enhance price volatility, lead
36 to food price spikes and affect food security due to climate and other shocks, as seen recently due to the
37 COVID-19 pandemic (Cottrell et al. 2019; WFP-FSIN 2020; Verschuur et al. 2021; AR6 WG2 Ch
38 5.12). The continued strong demand for food and other bio-based products, mainly from high- and
39 middle-income countries, therefore requires better cooperation between nations and global governance
40 of trade to more accurately reflect and disincentivize their environmental and social externalities. Trade
41 in agricultural and extractive products driving land-use change in tropical forest and savanna biomes is
42 of major concern because of the biodiversity impacts and GHG emissions incurred in their provision
43 (Hosonuma et al. 2012; Forest Trends 2014; Smith et al. 2014; Henders et al. 2015; Curtis et al. 2018;
44 Pendrill et al. 2019; Seymour and Harris 2019; Kissinger et al. 2021; AR6 WG2 CCP Tropical Forests).
Do Not Cite, Quote or Distribute 12-116 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 In summary, there is significant scope for optimizing use of land resources to produce more biomass
2 while reducing adverse effects (high confidence). Context-specific prioritisation, technology innovation
3 in bio-based production, integrative policies, coordinated institutions and improved governance
4 mechanisms to enhance synergies and minimize trade-offs can mitigate the pressure on managed as
5 well as natural and semi-natural ecosystems (medium confidence). Yet, energy conservation and
6 efficiency measures, and deployment of technologies and systems that do not rely on carbon-based
7 energy and materials, are essential for mitigating biomass demand growth as countries pursue ambitious
8 climate goals (high confidence).
9 END CROSS-WORKING BOX 3 HERE
10
11 12.6 Other cross-sectoral implications of mitigation
12 This section presents further cross-sectoral considerations related to GHG mitigation. Firstly,
13 various cross-sectoral perspectives on mitigation actions are presented. Then, sectoral policy
14 interactions are presented. Finally, implications in terms of international trade spill-over effects
15 and competitiveness, and finance flows and related spill-over effects at the sectoral level are
16 addressed.
17 12.6.1 Cross-sectoral perspectives on mitigation action
18 Chapters 5 to 11 present mitigation measures applicable in individual sectors, and potential co-benefits
19 and adverse side effects3 of these individual measures. This section builds on the sectoral analysis of
20 mitigation action from a cross-sectoral perspective. Firstly, Section 12.6.1.1 brings together some of
21 the observations presented in the sectoral chapters to show how different mitigation actions in different
22 sectors can contribute to the same co-benefits and result in the same adverse side effects, thereby
23 demonstrating the potential synergistic effects. The links between these co-benefits and adverse side
24 effects and the SDGs is also demonstrated. In Section 12.6.1.2, the focus turns from sector-specific
25 mitigation measures to mitigation measures which have cross-sectoral implications, including measures
26 that have application in more than one sector and measures where implementation in one sector impacts
27 on implementation in another. Finally, Section 12.6.1.3 notes the cross-sectoral relevance of a selection
28 of General-Purpose Technologies, a topic that is covered further in Chapter 16.
29 12.6.1.1 A cross-sectoral perspective on co-benefits and adverse side effects of mitigation measures,
30 and links the SDGs
31 A body of literature has been developed which addresses the co-benefits of climate mitigation action,
32 (Karlsson et al. 2020). Adverse side effects of mitigation are also well documented. Co-benefits and
33 adverse side-effects in individual sectors and associated with individual mitigation measures are
34 discussed in the individual sector chapters (Sections 5.2, 6.7.7, 7.4, 7.6, 8.2, 8.4, 9.8, 10.1.1, 11.5.3), as
35 well as in previous IPCC General and Special Assessment reports. The term co-impacts has been
36 proposed to capture both the co-benefits and adverse side-effects of mitigation. An alternative framing
37 is one of multiple objectives, where climate mitigation is placed alongside other objectives when
FOOTNOTE3 Here, the term co-benefits is used to refer to the additional benefits to society and the environment
that are realised in parallel with emissions reductions, while an understanding of adverse side effects highlights
where policy and decision makers are required to make trade-offs between mitigation benefits and other impacts.
The choice of language differs to some degree in other chapters.
Do Not Cite, Quote or Distribute 12-117 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 assessing policy decisions (Ürge-Vorsatz et al. 2014; Mayrhofer and Gupta 2016; Cohen et al. 2017;
2 Bhardwaj et al. 2019).
3 The identification and assessment of co-benefits has been argued to serve a number of functions
4 (Section 1.4) including using them as a leverage for securing financial support for implementation,
5 providing justification of actions which provide a balance of both short and long-term benefits and
6 obtaining stakeholder buy-in (robust evidence, low agreement) (Karlsson et al. 2020). Assessment of
7 adverse side-effects has been suggested to be useful in avoiding unforeseen negative impacts of
8 mitigation and providing policy and decision makers with the information required to make informed
9 trade-offs between climate and other benefits of actions (Ürge-Vorsatz et al. 2014; Bhardwaj et al. 2019;
10 Cohen et al. 2019) (high evidence, low agreement).
11 Various approaches to identifying and organising co-impacts in specific contexts and across sectors
12 have been proposed towards providing more comparable and standardised analyses. However,
13 consistent quantification of co-impacts, including cost-benefit analysis, and the utilisation of the
14 resulting information, remains a challenge (Ürge-Vorsatz et al. 2014; Floater et al. 2016; Mayrhofer
15 and Gupta 2016; Cohen et al. 2019; Karlsson et al. 2020). This challenge is further exacerbated when
16 considering that co-impacts of a mitigation measure in one sector can either enhance or reduce the co-
17 impacts associated with mitigation in another, or the achievement of co-benefits in one geographic
18 location can lead to adverse side effects in another. For example, the production of lithium for batteries
19 for energy storage has the potential to contribute to protecting water resources and reducing wastes
20 associated with coal fired power in many parts of the world, but mining of lithium has the potential for
21 creating water and waste challenges if not managed properly (Agusdinata et al. 2018; Kaunda 2020).
22 While earlier literature has suggested that co-impacts assessments can support adoption of climate
23 mitigation action, a more recent body of literature has suggested limitations in such framing (Ryan
24 2015; Bernauer and McGrath 2016; Walker et al. 2018). Presenting general information on co-impacts
25 as a component of a mitigation analysis does not always lead to increased support for climate mitigation
26 action. Rather, the most effective framing is determined by factors relating to local context, type of
27 mitigation action under consideration and target stakeholder group. More work has been identified to
28 be required to bring context into planning co-impacts assessments and communication thereof ( Ryan
29 2015; Bernauer and McGrath 2016; Walker et al. 2018) (low evidence, low agreement).
30 An area where the strong link between the cross-sectoral co-impacts of mitigation action and global
31 government policies is being clearly considered is in the achievement of the SDGs (Chapters 1 and 17,
32 individual sectoral chapters) (Obergassel et al. 2017; Doukas et al. 2018; Markkanen and Anger-Kraavi
33 2019; Smith et al. 2019; van Soest et al. 2019). Figure 12.9 demonstrates these relationships from a
34 cross-sectoral perspective. It shows the links between sectors which give rise to emissions, the
35 mitigation measures that can find application in the sector, co-benefits and adverse side effects of
36 mitigation measures and the SDGs (noting that the figure is not intended to be comprehensive). Such a
37 framing of co-impacts from a cross-sectoral perspective in the context of the SDGs could help to further
38 support climate mitigation action, particularly within the context of the Paris Agreement (Gomez-
39 Echeverri 2018) (medium evidence, medium agreement). Literature sources utilised in the compilation
40 of this diagram are presented in Supplementary Material 12.C.
Do Not Cite, Quote or Distribute 12-118 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1
2
3 Figure 12.9 Co-benefits and adverse side effects of mitigation actions with links to the SDGs.
4 The inner circle represents the sectors in which mitigation occurs. The second circle shows different
5 generic types of mitigation actions (A to G), with the symbols showing which sectors they are applicable
6 to. The third circle indicates different types of climate related co-benefits (green letters) and adverse side
7 effects (red letters) that may be observed as a result of implementing each of the mitigation actions (as
8 indicated by the letters A-G). Here I relates to climate resilience, II-IV economic co-impacts, V-VII
9 environmental, VIII-XII social, and XIII political and institutional. The final circle maps co-benefits and
10 adverse side-effects to the SDGs (Cohen et al. 2021).
11
12 12.6.1.2 Mitigation measures from a cross-sectoral perspective
13 Three aspects of mitigation from a cross-sectoral perspective are considered, following (Barker et al.
14 2007):
15 ▪ mitigation measures used in more than one sector;
16 ▪ implications of mitigation measures for interaction and integration between sectors; and
17 ▪ competition among sectors for scarce resources.
18 A number of mitigation measures find application in more than one sector. Renewable energy
19 technologies such as solar and wind may be used for grid electricity supply, as embedded generation in
20 the buildings sector and for energy supply in the agriculture sector (Chapters 6, 7 and 8) (Shahsavari
21 and Akbari 2018). Hydrogen and fuel cells, coupled with low carbon energy technologies for producing
22 the hydrogen, is being explored in transport, urban heat, industry and for balancing electricity supply
23 (Chapters 6, 8, 11) (Dodds et al. 2015; Staffell et al. 2019). Electric vehicles are considered an option
24 for balancing variable power (Kempton and Tomić 2005; Liu and Zhong 2019). Carbon Capture and
25 Storage (CCS) and Carbon Capture and Utilisation (CCU) has potential application in a number of
Do Not Cite, Quote or Distribute 12-119 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 industrial processes (cement, iron and steel, petroleum refining and pulp and paper) (Chapters 6 and 11)
2 (Leeson et al. 2017; Garcia and Berghout 2019) and the fossil fuel electricity sector (Chapter 6). When
3 coupled with energy recovery from biomass , CCS can provide a carbon sink (BECCS) (Section 12.5).
4 On the demand side, energy efficiency options find application across the sectors (Chapters 6, 8, 9, 10,
5 and 11), as does reducing demand for goods and services (Chapter 5), and improving material efficiency
6 (Section 11.3.2).
7 A range of examples of where mitigation measures result in cross-sectoral interactions and integration
8 is identified. The mitigation potential of electric vehicles, including plug-in hybrid hybrids, is linked to
9 the extent of decarbonisation of the electricity grid, as well as to the liquid fuel supply emissions profile
10 (Lutsey 2015). Making buildings energy positive, where excess energy is used to charge vehicles, can
11 increase the potential of electric and hybrid vehicles (Zhou et al. 2019). Advanced process control and
12 process optimisation in industry can reduce energy demand and material inputs (Section 11.3), which
13 in turn can reduce emissions linked to resource extraction and manufacturing. Reductions in coal-fired
14 power generation through replacement with renewables or nuclear power result in a reduction in coal
15 mining and its associated emissions. Increased recycling results in a reduction in emissions from
16 primary resource extraction. CCU can contribute to the transition to more renewable energy systems
17 via power-to-X technologies, which enables the production of CO2-based fuels/e-fuels and chemicals
18 using carbon dioxide and hydrogen ( Breyer et al. 2015; Anwar et al. 2020). Certain reductions in the
19 AFOLU sector are contingent on energy sector decarbonisation. Trees and green roofs planted to
20 counter urban heat islands reduce the demand for energy for air conditioning and simultaneously
21 sequester GHGs (Kim and Coseo 2018; Kuronuma et al. 2018). Recycling of organic waste avoids
22 methane generation if the waste would have been disposed of in landfill sites, can generate renewable
23 energy if treated through anaerobic digestion and can reduce requirements for synthetic fertiliser
24 production if the nutrient value is recovered (Creutzig et al. 2015). Liquid transport biofuels links to the
25 land, energy and transport sectors (Section 12.5.2.2).
26 Demand-side mitigation measures, discussed in Chapter 5, also have cross-sectoral implications which
27 need to be taken into account when calculating mitigation potentials. Residential electrification has the
28 potential to reduce emissions associated with lighting and heating particularly in developing countries
29 where this is currently met by fossil fuels and using inefficient technologies, but will increase demand
30 for electricity (Chapters 5, 8 and Sections 6.6.2.3, 8.4.3.1). Many industrial processes can also be
31 electrified in the move away from fossil reductants and direct energy carriers (Chapter 11). The impact
32 of electrification on electricity sector emissions will depend on whether electricity generation is based
33 on fossil fuels in the absence of CCS or low carbon energy sources (Chapter 5).
34 At the same time, saving electricity in all sectors reduces the demand for electricity, thereby reducing
35 mitigation potential of renewables and CCS. Demand side flexibility measures and electrification of
36 vehicle fleets are supportive of more intermittent renewable energy supply options (Sections 6.3.7,
37 6.4.3.1 and 10.3.4). Production of maize, wheat, rice and fresh produce requires lower energy inputs on
38 a life cycle basis than poultry, pork and ruminant based meats (Section 12.4) (Clark and Tilman 2017).
39 They also require less land and area per kilocalorie or protein output (Clark and Tilman 2017; Poore
40 and Nemecek 2018), and so replacing meat with these products makes land available for sequestration,
41 biodiversity or other societal needs. However, production of co-products of the meat industry, such as
42 leather and wool, is reduced, resulting in a need for substitutes. Further discussion and examples of
43 cross-sectoral implications of mitigation, with respect to cost and potentials, are presented in Section
44 12.2. One final example on this topic included here is that of Circular Economy (Box 12.4).
45 Finally, in terms of competition among sectors for scarce resources, this issue is often considered in the
46 assessments of mitigation potentials linked to bioenergy and diets (vegetable vs. animal food products),
Do Not Cite, Quote or Distribute 12-120 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 land use and water (Section 12.5, Cross-Working Group Box 3 in this Chapter) (robust evidence, high
2 agreement). It is, however, also relevant elsewhere. Constraints have been identified in the supply of
3 indium, tellurium, silver, lithium, nickel and platinum that are required for implementation of some
4 specific renewable energy technologies (Watari et al. 2018; Moreau et al. 2019). Other studies have
5 shown constraints in supply of cobalt, one of the key elements used in production of lithium-ion
6 batteries, which has been assessed for mitigation potential in energy, transport and buildings sectors
7 (Jaffe 2017; Olivetti et al. 2017) (medium evidence, high agreement), although alternatives to cobalt are
8 being developed (Olivettie et al. 2017; Watari et al. 2018).
9 START BOX 12.4 HERE
10
11 Box 12.4: Circular Economy from a Cross-Sectoral Perspective
12
13 Circular economy approaches consider the entire life cycle of goods and services, and seek to design
14 out waste and pollution, keep products and materials in use, and regenerate natural systems ( The Ellen
15 MacArthur Foundation 2013; CIRAIG 2015). The use of Circular Economy for rethinking how
16 society’s needs for goods and services is delivered in such a way as to minimise resource use and
17 environmental impact and maximise societal benefit has been discussed elsewhere in this assessment
18 report (Chapter 5 and Section 5.3.4). A wide range of potential application areas is identified, from food
19 systems to bio-based products to plastics to metals and minerals to manufactured goods. Circular
20 economy approaches are implicitly cross-sectoral, impacting the energy, industrial, AFOLU, waste and
21 other sectors. They will have climate and non-climate co-benefits and trade-offs. The scientific
22 literature mainly investigates incremental measures claiming but not demonstrating mitigation; highest
23 mitigation potential is found in the industry, energy, and transport sector; mid-range potential in the
24 waste and building sector; and lowest mitigation gains in agriculture (Cantzler et al. 2020). Circular
25 economy thinking has been identified to support increased resilience to the physical effects of climate
26 change and contribute to meeting other UN SDGs, notably SDG12 (responsible consumption and
27 production) (The Ellen MacArthur Foundation 2019).
28
29 Circular economy approaches to deployment of low-carbon infrastructure have been suggested to be
30 important to optimise resource use and mitigate environmental and societal impacts caused by
31 extraction and manufacturing of composite and critical materials as well as infrastructure
32 decommissioning (Jensen and Skelton 2018; Sica et al. 2018; Salim et al. 2019; Watari et al. 2019;
33 Jensen et al. 2020; Mignacca et al. 2020). The circular carbon economy is an approach inspired by the
34 circular economy principles that rely on a combination of technologies, including CCU, CCS and CDR,
35 to enable transition pathways especially relevant in economies dependent on fossil fuel exports (Lee et
36 al. 2017; Alshammari 2020; Morrow and Thompson 2020; Zakkour et al. 2020). The integration of
37 circular economy and bioeconomy principles (See Cross-Working Group Box 3 in this Chapter on
38 mitigation and adaption via the bioeconomy) is conceptualised in relation to policy development
39 (European Commission 2018) as well as COVID-19 recovery strategies (Palahí et al. 2020)
40 emphasising the use of renewable energy sources and sustainable management of ecosystems with
41 transformation of biological resources into food, feed, energy and biomaterials.
42 At this stage, however, there is no single globally agreement of how circular economy principles are
43 best to be implemented, and differential government support for circular economy interventions is
44 observed in different jurisdictions.
45 END BOX 12.4 HERE
46
Do Not Cite, Quote or Distribute 12-121 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 12.6.1.3 Cross-sectoral considerations relating to emerging general purpose technologies
2 General Purpose Technologies (GPTs) include, but are not limited to, additive manufacturing, artificial
3 intelligence, biotechnology, hydrogen, digitalisation, electrification, nanotechnology and robots (de
4 Coninck et al. 2018). Many of the individual sectoral chapters have identified the roles that such
5 technologies can have in supporting mitigation of GHG emissions. Section 16.2.2.3 presents an
6 overview of the individual technologies and specific applications thereof.
7 In this chapter, which focuses on cross-sectoral implications of mitigation, it is highlighted that certain
8 of these GPTs will find application across the sectors, and there will be synergies and trade-offs when
9 utilising these technologies in more than sector. One example here is the use of hydrogen as an energy
10 carrier, which when coupled with low carbon energy, has potential for driving mitigation in energy,
11 industry, transport, and buildings. The increased uptake of hydrogen across the economy requires
12 establishment of hydrogen production, transport and storage infrastructure which could simultaneously
13 support multiple sectors, although there is the potential to utilise existing infrastructure in some parts
14 of the world (Alanne and Cao 2017).
15 Box 12.5 provides for further details on hydrogen in the context of cross-sectoral mitigation specifically,
16 while further details on the role of hydrogen in individual sectors are provided in Chapters 6, 8. 9, 10
17 and 11. In contrast, the benefits of digitalisation, which could potentially give rise to substantial energy
18 savings across multiple sectors, need to be traded off against demand for electricity to operate consumer
19 devices, data centres, and data networks. Measures are required to increase energy efficiency of these
20 technologies (IEA 2017). Section 5.3.4.1 of this report provides further information on energy and
21 emissions benefits and costs of digitalisation.
22 With respect to co-impacts of GPTs, the other focus of this chapter, it is highlighted that assessment of
23 the environmental, social and economic implications of such technologies is challenging and context
24 specific with multiple potential cross-sectoral linkages (de Coninck et al. 2018). Each GPT would need
25 to be explored in context of what it is being used for, and potentially in the geographical context, in
26 order to understand the co-impacts of its use.
27 START BOX 12.5 HERE
28
29 Box 12.5: Hydrogen in the context of cross-sectoral mitigation options
30
31 The interest in hydrogen as an intermediary energy carrier has grown rapidly in the years since the 5th
32 Assessment Report of WGIII (AR5) was published. This is reflected in this WGIII assessment report,
33 where the term ‘hydrogen’ is used more than five times more often than in AR5. In Chapter 6 of this
34 report, it is shown that hydrogen can be produced with low carbon impact from fossil fuels (Section
35 6.4.2.6), renewable electricity and nuclear energy (Section 6.4.5.1), or biomass (Section 6.4.2.5). In the
36 energy sector, hydrogen is one of the options for storage of energy in low-carbon electricity systems
37 (Sections 6.4.4.1 and 6.6.2.2). But, also importantly, hydrogen can be produced to be used as a fuel for
38 sectors that are hard-to-decarbonise; this is possible directly in the form of hydrogen, but also in the
39 form of ammonia or other energy carriers (Section 6.4.5.1). In the transport sector, fuel cell engines
40 (Section 10.3.3) running on hydrogen can become important, especially for heavy duty vehicles
41 (Section 10.4.3). In the industry sector hydrogen already plays an important role in the chemical sector
42 (for ammonia and methanol production (Box 11.1 in Chapter 11) and in the fuel sector (in oil refinery
43 processes and for biofuel production (IEA 2019b). Beyond the production of ammonia and methanol
44 for both established and novel applications, the largest potential industrial application for low-carbon
45 hydrogen is seen in steelmaking (Section 11.4.1.1). Hydrogen and hydrogen-derivatives can play a
Do Not Cite, Quote or Distribute 12-122 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 further role as substitute energy carriers (Section 11.3.5) and for the production of intermediate
2 chemical products such as methanol, ethanol and ethylene when combined with CCU (Section 11.3.6).
3 For the building sector, the exploration of the usefulness of hydrogen is at an early stage (Box 9.4 in
4 Chapter 9).
5
6 An overview report (IEA 2019b) already sees opportunities in 2030 for buildings, road freight and
7 passenger vehicles. This report also suggests a high potential application in iron and steel production,
8 aviation and maritime transport, and for electricity storage. Several industry roadmaps have been
9 published that map out a possible role for hydrogen until 2050. The most well-known and ambitious is
10 the roadmap by the Hydrogen Council (2017), which sketches a global scenario leading to 78 EJ
11 hydrogen use in 2050, mainly for transport, industrial feedstock, industrial energy and to a lesser extent
12 for buildings and power generation. Hydrogen makes up 18% of total final energy use in this vision.
13 An analysis by IRENA on hydrogen from renewable sources comes to a substantially lower number: 8
14 EJ (excluding hydrogen use in power production and feedstock uses). On a regional level, most
15 roadmaps and scenarios have been published for the European Union, e.g. by the Fuel Cell and
16 Hydrogen Joint Undertaking (Blanco et al. 2018; EC 2018; FCH 2019; Navigant 2019). All these
17 reports have scenario variants with hydrogen share in final energy use of 10% to over 20% by 2050.
18 When it comes to the production of low-carbon hydrogen, the focus of the attention is on production
19 from electricity from renewable sources via electrolysis, so-called ‘green hydrogen’. However, ‘blue
20 hydrogen’, produced out of natural gas with CCS is also often considered. Since a significantly
21 increasing role for hydrogen would require considerable infrastructure investments and would affect
22 existing trade flows in raw materials, governments have started to set up national hydrogen strategies,
23 both potential exporting (e.g. Australia) and importing (e.g. Japan) countries (METI 2017; COAG
24 Energy Council 2019).
25
26 As already reported in Chapter 6 (Section 6.2.4.1) production costs of green hydrogen are expected to
27 come down from the current levels of above 100 USD MWh-1. Price expectations are: 40–60 € MWh-
28 1 for both green and blue hydrogen production in the EU by 2050 (Navigant 2019) with production
29 costs already being lower in North Africa; 42–87 USD MWh-1 for green hydrogen in 2030 and 20 – 41
30 USD MWh-1 in 2050 (BNEF 2020); 75 € MWh-1 in 2030 (Glenk and Reichelstein 2019). For fossil-
31 based technologies combined with CCS, prices may range from 33 – 80 USD MWh-1 (Table 6.8 in
32 Chapter 6,). Such prices can make hydrogen competitive for industrial feedstock applications, and
33 probably for several transportation modes in combination with fuel cells, but without further incentives,
34 not necessarily for stationary applications in the coming decades: wholesale natural gas prices are
35 expected to range from 7–31 USD MWh-1 across regions and scenarios, according to the World Energy
36 Outlook (IEA 2020a); coal prices mostly are even lower than natural gas prices (all fossil fuel prices
37 refer to unabated technology and untaxed fuels). The evaluation of macro-economic impacts is
38 relatively rare. A study by (Mayer et al. 2019b) indicated that a shift to hydrogen in iron and steel
39 production would lead to regional GDP losses in the range of 0.4–2.7% in 2050 across EU+3 with some
40 regions making gains under a low-cost electricity scenario.
41
42 The IAM scenarios imply a modest role played by hydrogen, with some scenarios featuring higher
43 levels of penetration. The consumption of hydrogen is projected to increase by 2050 and onwards in
44 scenarios likely limiting global warming to 2°C or below, and the median share of hydrogen in total
45 final energy consumption is 2.1% in 2050 and 5.1% in 2100 (Box 12.4, Figure 1) (Numbers are based
46 on the AR6 scenarios database.) There is large variety in hydrogen shares, but the values of 10% and
47 more of final energy use that occur in many roadmaps are only rarely reached in the scenarios. Hydrogen
48 is predominantly used in the industry and transportation sectors. In the scenarios, hydrogen is produced
49 mostly by electrolysis and by biomass energy conversion with CCS (Box 12.5, Figure 1). Natural gas
Do Not Cite, Quote or Distribute 12-123 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 with CCS is expected to only play a modest role; here a distinct difference between the roadmaps quoted
2 before and the IAM results is observed.
3
4 It is concluded that there is increasing confidence that hydrogen can play a significant role, especially
5 in the transport sector and the industrial sector. However, there is much less agreement on timing and
6 volumes, and there is also a range of perspectives on role of the various production methods of
7 hydrogen.
8
9
10
11 Box 12.5, Figure 1 Fraction of hydrogen (H2, red) in total final energy consumption, and those for each
12 sector. Hinges represent the interquartile ranges and whiskers extend to 5 and 95 percentiles.
13
14 END BOX 12.5 HERE
15
16 12.6.2 Sectoral policy interactions (synergies and trade-offs)
17 A taxonomy of policy types and attributes is provided by Section 13.6. In addition, the sectoral chapters
18 provide an in-depth discussion of important mitigation policy issues such as policy overlaps, policy
19 mixes, and policy interaction as well as policy design considerations and governance. The point of
20 departure for the assessment in this chapter is a focus on cross-sectoral perspectives aiming at
21 maximising policy synergies and minimising policy trade-offs.
22 Synergies and trade-offs resulting from mitigation policies are not clearly discernible from either sector-
23 level studies or global and regional top-down studies. Rather, they would require a cross-sectoral
24 integrated policy framework (von Stechow et al. 2015; Singh et al. 2019; Monier et al. 2018; Pardoe et
25 al. 2018) or multiple-objective-multiple-impact policy assessment framework identifying key co-
26 impacts and avoiding trade-offs (Ürge-Vorsatz et al. 2014) (robust evidence, high agreement).
Do Not Cite, Quote or Distribute 12-124 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Sectoral studies typically cover differentiated response measures while the IAM literature mostly uses
2 uniform efficient market-based measures. This has important implications for understanding the
3 differences in magnitude and distribution of mitigation costs and potentials of Section 12.2 (Rausch and
4 Karplus 2014; Karplus et al. 2013). There is a comprehensive literature on the efficiency of uniform
5 carbon pricing compared to sector-specific mitigation approaches, but relatively less literature on the
6 distributional impacts of carbon taxes and measures to mitigate potential adverse distributional impacts
7 (Åhman et al. 2017; Rausch and Reilly 2015; Mu et al. 2018; Wang et al. 2016b; Rausch and Karplus
8 2014). For example, in terms of cross-sectoral distributional implications, studies find negative
9 competitiveness impacts for the energy intensive industries (Wang et al. 2016b; Åhman et al. 2017;
10 Rausch and Karplus 2014). (robust evidence, medium agreement)
11 Strong inter-dependencies and cross-sectoral linkages create both opportunities for synergies and the
12 need to address trade-offs. This calls for coordinated sectoral approaches to climate change mitigation
13 policies that mainstream these interactions (Pardoe et al. 2018). Such an approach is also called for in
14 the context of cross-sectoral interactions of adaptation and mitigation measures, examples are in the
15 agriculture, biodiversity, forests, urban, and water sectors (Di Gregorio et al. 2017; Arent et al. 2014;
16 Berry et al. 2015). Integrated planning and cross-sectoral alignment of climate change policies are
17 particularly evident in developing countries’ NDCs pledged under the Paris Agreement, where key
18 priority sectors such as agriculture and energy are closely aligned between the proposed mitigation and
19 adaptation actions in the context of sustainable development and the SDGs. An example is the
20 integration between smart agriculture and low carbon energy (Antwi-Agyei et al. 2018; England et al.
21 2018). Yet, there appear to be significant challenges relating to institutional capacity and resources to
22 coordinate and implement such cross-sectoral policy alignment, particularly in developing country
23 contexts (Antwi-Agyei et al. 2018) (robust evidence, high agreement).
24 Another dimension of climate change policy interactions in the literature is related to trade-offs and
25 synergies between climate change mitigation and other societal objectives. For example, in mitigation
26 policies related to energy, trade-offs and synergies between universal electricity access and climate
27 change mitigation would call for complementary policies such as pro-poor tariffs, fuel subsidies, and
28 broadly integrated policy packages (Dagnachew et al. 2018). In agriculture and forestry, research
29 suggests that integrated policy programs enhance mitigation potentials across the land-use-agriculture-
30 forestry nexus and lead to synergies and positive spill-overs (Galik et al. 2019). To maximise synergies
31 and deal with trade-offs in such a cross-sectoral context, evidence-based/informed and holistic policy
32 analysis approaches like nexus approaches and multi-target back-casting approaches that take into
33 account unanticipated outcomes and indirect consequences would be needed (Klausbruckner et al.
34 2016; van der Voorn et al. 2020; Hoff et al. 2019; see Box 12.6) (robust evidence, high agreement).
35 The consequences of large-scale land-based mitigation for food security, biodiversity, (Dasgupta 2021)
36 the state of soil, water resources, etc. can be significant depending on many factors, such as economic
37 development (including distributional aspects), international trade patterns, agronomic development,
38 diets, land use governance and policy design, and not least climate change itself (Fujimori et al. 2018;
39 Hasegawa et al. 2018; Van Meijl et al. 2018; Winchester and Reilly 2015). Policies and regulations that
40 address other aspects apart from climate change can indirectly influence the attractiveness of land-
41 based mitigation options. For example, farmers may find it attractive to shift from annual food/feed
42 crops to perennial grasses and short rotation woody crops (suitable for bioenergy) if the previous land
43 uses become increasingly restricted due to impacts on groundwater quality and eutrophication of water
44 bodies (Sections 12.4 and 12.5) (robust evidence, medium agreement).
Do Not Cite, Quote or Distribute 12-125 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Finally, there are knowledge gaps in the literature particularly in relation to policy scalability and in
2 relation to the extent and magnitude of policy interactions when scaling the policy to a level consistent
3 with low GHG emissions pathways such as 2ºC and 1.5ºC.
4
5 START BOX 12.6 HERE
6
7 Box 12.6: Case Study, Sahara Forest Project in Aqaba, Jordan
8
9 Nexus Framing
10 Shifting to renewable (in particular solar) energy reduces dependency on fossil fuel imports and
11 greenhouse gas emissions, which is crucial for mitigating climate change. Employing renewable energy
12 for desalination of seawater and for cooling of greenhouses in integrated production systems can
13 enhance water availability, increase crop productivity and generate co-products and co-benefits (e.g.,
14 algae, fish, dryland restoration, greening of the desert).
15
16 Nexus Opportunities
17 The Sahara Forest project integrated production system uses amply available natural resources, namely
18 solar energy and seawater, for improving water availability and agricultural/biomass production, while
19 simultaneously providing new employment opportunities. Using hydroponic systems and humidity in
20 the air, water needs for food production are 50% lower compared to other greenhouses.
21
22 Technical and Economic Nexus Solutions
23 Several major technologies are combined in the Sahara Forest Project, namely electricity production
24 through the use of solar power (PV or CSP), freshwater production through seawater desalination using
25 renewable energy, seawater-cooled greenhouses for food production, and outdoor revegetation using
26 run-off from the greenhouses.
27
28 Stakeholders Involved
29 The key stakeholders which benefit from such an integrated production system are from the water sector
30 which urgently requires an augmentation of irrigation (and other) water, as well as from the agricultural
31 sector, which relies on the additional desalinated water to maintain and increase agricultural production.
32 The project also involves public and private sector partners from Jordan and abroad, with little
33 engagement of civil society so far.
34
35 Framework Conditions
36 The Sahara Forest Project has been implemented at pilot scale so far, including the first pilot with one
37 hectare and one greenhouse pilot in Qatar and a larger “launch station” with three hectares and two
38 greenhouses in Jordan). These pilots have been funded by international organisations such as the
39 Norwegian Ministry of Climate and Environment, Norwegian Ministry of Foreign Affairs and the
40 European Union. Alignment with national policies, institutions and funding as well as upscaling of the
41 project is underway or planned.
42
43 Monitoring and Evaluation and Next Steps
44 The multi-sectoral planning and investments that are needed to up-scale the project require cooperation
45 among the water, agriculture, and energy sectors and an active involvement of local actors, private
46 companies, and investors. These cooperation and involvement mechanisms are currently being
47 established in Jordan. Given the emphasis on the economic value of the project, public-private
48 partnerships are considered as the appropriate business and governance model, when the project is up-
49 scaled. Scenarios for upscaling (seawater use primarily in low lying areas close to the sea, to avoid
50 energy-intensive pumping) include 50MW of CSP, 50 hectares of greenhouses, which would produce
51 34,000 tons of vegetables annually, provide employment for over 800 people, and sequester more than
52 8,000 tons of CO2 annually.
53
Do Not Cite, Quote or Distribute 12-126 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 Source: SFP Foundation; Hoff et al. 2019
2
3 END BOX 12.6 HERE
4
5 12.6.3 International trade spill-over effects and competitiveness
6 International spill-overs of mitigation policies are effects that carbon-abatement measures implemented
7 in one country have on sectors in other countries. These effects include 1) carbon leakage in
8 manufacture, 2) the effects on energy trade flows and incomes related to fossil fuel exports from major
9 exporters, 3) technology and knowledge spill-overs; 4) transfer of norms and preferences via various
10 approaches to establish sustainability requirements on traded goods, e.g., EU-RED and environmental
11 labelling systems to guide consumer choices (robust evidence, medium agreement). This section focuses
12 on cross-sectoral aspects of international spill-overs related to the first two effects.
13 12.6.3.1 Cross-sectoral aspects of carbon leakage
14 Carbon leakage occurs when mitigation measures implemented in one country/sector lead to the rise in
15 emissions in other countries/sectors. Three types of spill-overs are possible: 1) domestic cross-sectoral
16 spill-overs when mitigation policy in one sector leads to the re-allocation of labour and capital towards
17 the other sectors of the same country; 2) international spill-overs within a single sector when mitigation
18 policy leads to substitution of domestic production of carbon-intensive goods with their imports from
19 abroad; 3) international cross-sectoral spill-overs when mitigation policy in one sector in one country
20 leads to the rise in emissions in other sectors in other countries. While the first two are described in
21 Section 13.6, this section focuses on the third. Though some papers address this type of leakage, there
22 is still significant lack of knowledge on this topic.
23 One possible channel of cross-sectoral international carbon leakage is through global value chains.
24 Mitigation policy in one country not only leads to shifts in competitiveness across industries producing
25 final goods but also across those producing raw materials and intermediary goods all over the world.
26 This type of leakage is especially important because the countries that provide basic materials are
27 usually emerging or developing economies, many of which have no or limited regulation of GHG
28 emissions. For this reason, foreign direct investment in developing economies usually leads to an
29 increase in emissions (Bakhsh et al. 2017; Shahbaz et al. 2015; Kivyiro and Arminen 2014): in case of
30 basic materials the effect of expansion of economic activity on emissions exceeds the effect of
31 technological spill-overs, while for developed countries the effect is opposite (Pazienza 2019; Shahbaz
32 et al. 2015). Meng et al. (2018) calculated that environmental costs for generating one unit of GDP
33 through international trade was 1.4 times higher than that through domestic production in 1995. By
34 2009, this difference increased to 1.8 times. Carbon leakage due to the differences in environmental
35 regulation was the main driver of this increase.
36 In order to address emissions leakage through global value chains, Liu and Fan (2017) propose the
37 value-added-based emissions accounting principle, that makes possible to account for GHG emissions
38 within the context of the economic benefit principle. Davis et al. (2011) notice that the analysis of value
39 chains gives an opportunity to find the point where regulation would be the most efficient and the least
40 vulnerable to leakage. For instance, transaction costs of global climate policy and the risks of leakage
41 may be reduced if emissions are regulated at the extraction stage as there are far fewer agents involved
42 in this process than in burning of fossil fuels or consumption of energy-intensive goods. Li et al. (2020)
Do Not Cite, Quote or Distribute 12-127 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 calls for coordinated efforts to reduce emissions in trade flows in pairs of the economies with the highest
2 leakage such as China and the United States, China and Germany, China and Japan, Russia and
3 Germany.
4 Unfortunately, these proposals either face difficulties in collection and verification of data on emissions
5 along value chains or require a high level of international cooperation which is hardly achievable at the
6 moment. (Neuhoff et al. 2016; Pollitt et al. 2020) focus on the regulation of emissions embodied in
7 global value chains through national policy instruments. They propose implementation of a charge on
8 consumption of imported basic materials into the European emissions trading system. Such a charge,
9 equivalent to around €80 tCO2-1, could reduce the EU’s total CO2 emissions by up to 10% by 2050
10 (Pollitt et al. 2020) without significant effects on competitiveness. This proposal is very close to border
11 carbon adjustment introduced in the EU and described in more detail in Sections 13.2 and 13.6.
12 Cross-sectoral effects of carbon leakage also occur through the multiplier effect, when the mitigation
13 policy in any sector in country A leads to the increase of relative competitiveness and therefore
14 production of the same sector in country B that automatically leads to the expansion of economic
15 activity in other sectors of country B. This expansion may in turn lead to the rise of production and
16 emissions in country A as a result of feedback effects. These spill-overs should be taken into
17 consideration while designing climate policy, along with potential synergies that may appear due to
18 joint efforts. However, the scale of these effects with regards to leakage shouldn’t be overestimated.
19 Even for intrasectoral leakage, many ex-ante modelling studies generally suggest limited carbon
20 leakage rates (Chapter 13). Intersectoral leakage should be even less significant. Interregional spill-over
21 and feedback effects are well-studied in China (Zhang 2017; Ning et al. 2019). Even within a single
22 country, interregional spill-over effects are much lower than intraregional effects, and feedback effects
23 are even less intense. Cross-sectoral spill-overs across national borders as a result of mitigation policy
24 should be even smaller, although these are less well-studied. In future, if the differences in carbon price
25 between regions increase, leakage through cross-sectoral multipliers may play a more important role.
26 Another important cross-sectoral aspect of carbon leakage concerns the transport sector. If mitigation
27 policy leads to the substitution of domestic carbon-intensive production with imports, one of the side
28 effects of this substitution is the rise of emissions from transportation of imported goods. International
29 transport is responsible for about a third of worldwide trade-related emissions, and over 75 percent of
30 emissions for major manufacturing categories (Cristea et al. 2013). Carbon leakage would potentially
31 increase the emissions from transportation significantly as the trade of major consuming economies of
32 the EU and US would shift towards distant trading partners in East and South Asia. Meng et al. (2018)
33 consider more distant transportation as one of the major contributors to the rise in emissions embodied
34 in international trade from 1995 to 2009.
35 Emissions leakage due to international trade, investment and value chains is a significant obstacle to
36 more ambitious climate policies in many regions. However, it doesn’t mean that disruption of trade
37 would reduce global emissions. Zhang et al. (2020) show that deglobalisation and the drop in
38 international trade may result in emissions reductions in the short term, but in the longer term it will
39 make each country build more complete industrial systems to satisfy their final demand, although they
40 have comparative disadvantages in some production stages. As a result, emissions would increase.
41 According to Zhang et al. (2020) for China, the decrease of the degree of global value chain participation
42 (which ranges from 0 to 1) by 0.1 would lead to an increase in gross carbon intensity of China’s exports
43 of 11.7%. On distributional implications, Parrado and De Cian (2014) report that trade-driven spill-
44 overs effects transmitted through imports of materials and equipment result in significant inter-sectoral
45 distributional effects with some sectors witnessing substantial expansion in activity and emissions and
46 others witnessing a decline in activities and emissions.
Do Not Cite, Quote or Distribute 12-128 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 It should also be mentioned that international trade leads to important knowledge and technology spill-
2 overs (Sections 16.3 and 16.5) and is critically important for achieving other Sustainable Development
3 Goals (Section 12.6.1). Any policies imposing additional barriers to international trade should be
4 therefore implemented with great caution and require comprehensive evaluation of various economic,
5 social and environmental effects.
6 12.6.3.2 The spill-over effects on the energy sector
7 Cross-sectoral trade-related spill-overs of mitigation policies include their effect on energy prices. Other
8 things being equal, regulation of emissions of industrial producers decreases the demand for fossil fuels
9 that would reduce prices and encourage the rise of fossil fuel consumption in regions with no or weaker
10 climate policies (robust evidence, medium agreement).
11 Arroyo-Currás et al. (2015) study the energy channel of carbon leakage with the REMIND IAM of the
12 global economy. They come to the conclusion that the leakage rate through the energy channel is less
13 than 16% of the emission reductions of regions who introduce climate policies first. This result doesn’t
14 differ much for different sizes and compositions of the early mover coalition.
15 Bauer et al. (2015) built a multi-model scenario ensemble for the analysis of energy-related spill-overs
16 of mitigation policies and reveal huge uncertainty: energy-related carbon leakage rates vary from
17 negative values to 50%, primarily depending on the trends in inter-fuel substitution.
18 Another kind of spill-over in energy sector concerns the “green paradox”; announcement of future
19 climate policies causes an increase in production and trade in fossil-fuels in the short term (Jensen et al.
20 2015; Kotlikoff et al. 2016). The delayed carbon tax should therefore be higher than an immediately
21 implemented carbon tax in order to achieve the same temperature target (van der Ploeg 2016). Studies
22 also make a distinction between a “weak” and “strong” green paradox (Gerlagh 2011). The former
23 refers to a short-term rise in emissions in response to climate policy, while the latter refers to rising
24 cumulative damage.
25 The green paradox may work in different ways for different kinds of fossil fuels. For instance, Coulomb
26 and Henriet (2018) show that climate policies in the transport and power-generation sectors increase
27 the discounted profits of the owners of conventional oil and gas, compared to the no-regulation baseline,
28 but will decrease these profits for coal and unconventional oil and gas producers.
29 Many studies also distinguish different policy measures by the scale of green paradox they provide. The
30 immediate carbon tax is the first-best instrument from the perspective of the global welfare. Delayed
31 carbon tax leads to some green paradox but it is less than in the case of the support of renewables
32 (Michielsen 2014; van der Ploeg and Rezai 2019). With respect to the latter, support of renewable
33 electricity has a lower green paradox than the support of biofuels (Gronwald et al. 2017; Michielsen
34 2014). The existence of the green paradox is an additional argument in favour of more decisive climate
35 policy now: any postponements will lead to additional consumption of fossil fuels and consequently the
36 need for more ambitious and costly efforts in future.
37 The effect of fossil fuel production expansion as a result of anticipated climate policy may be
38 compensated by the effect of divestment. Delayed climate policy creates incentives for investors to
39 divest from fossil fuels. Bauer et al. (2018) show that this divestment effect is stronger and thus
40 announcing of climate policies leads to the reduction of energy-related emissions.
41 The implication of the effects of mitigation policies through the energy related spill-overs channel is of
42 particular significance to oil-exporting countries (medium evidence, medium agreement). Emissions
Do Not Cite, Quote or Distribute 12-129 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 reduction-measures lead to the decreasing demand for fossil fuels and consequently to the decrease in
2 its exports from major oil- and gas- exporting countries. The case of Russia is one of the most
3 illustrative. Makarov et al. (2020) show that the fulfilment of Paris Agreement parties of their NDCs
4 would lead to 25% reduction of Russia’s energy exports by 2030 with significant reduction of its
5 economic growth rates. At the same time, the domestic consumption of fossil fuels is anticipated to
6 increase in response to the drop of external demand that would provoke carbon leakage (Orlov and
7 Aaheim 2017). Such spill-overs demonstrate the need for the dialogue between exporters and importers
8 of fossil fuels while implementing the mitigation policies.
9
10 12.6.4 Implications of finance for cross-sectoral mitigation synergies and trade-offs
11 Finance is a principal enabler of GHG mitigation and an essential component of countries’ NDC
12 packages submitted under the Paris climate agreement (UNFCCC 2016). The assessment of investment
13 requirements for mitigation along with their financing at sectoral levels are addressed in detail by
14 sectoral chapters while the assessment of financial sources, instruments, and the overall mitigation
15 financing gap is addressed by Chapter 15 (Sections 15.3, 15.4, and 15.5). The focus in this chapter with
16 respect to finance is on the scope and potential for financing integrated solutions that create synergies
17 between and among sectors.
18 Cross-sectoral considerations in mitigation finance are critical for the effectiveness of mitigation action
19 as well as for balancing the often conflicting social, developmental and environmental policy goals at
20 the sectoral level. True measures of mitigation policy impacts and hence plans for resource mobilisation
21 that properly address costs and benefits cannot be developed in isolation of their cross-
22 sectoral implications. Unaddressed cross-sectoral coordination and interdependency issues are
23 identified as major constraints in raising the necessary financial resources for mitigation in a number of
24 countries (Bazilian et al. 2011; Welsch et al. 2014; Hoff et al. 2019a).
25 Integrated financial solutions to leverage synergies between sectors, as opposed to purely sector-based
26 financing at international, national, and local levels are needed to scale up GHG mitigation
27 potentials. At the international level, Finance from Multilateral Development Banks (MDBs) is a major
28 source of GHG mitigation finance in developing countries (World Bank Group 2015; Ha et al. 2016;
29 Bhattacharya et al. 2016, 2018) (medium evidence, medium agreement). In 2018, MDBs reported a total
30 of USD 30,165 million in financial commitments to climate change mitigation, with 71% of total
31 mitigation finance being committed through investment loans and the rest in the form of equity,
32 guarantees, and other instruments. GHG reduction activities eligible for MDB finance are limited to
33 those compatible with low-emission pathways recognising the importance of long-term structural
34 changes, such as the shift in energy production to low-carbon energy technologies and the modal shift
35 to low-carbon modes of transport leveraging both greenfield and energy efficiency projects. Sector-
36 wise, the MDBs mitigation finance for 2018 is allocated to renewable energy (29%), transport (18%),
37 energy efficiency (18%), lower-carbon and efficient energy generation (7%), agriculture, forestry and
38 land use (8%), waste and waste-water (8%), and other sectors (12%) (MDB 2019). Unfortunately, due
39 to institutional and incentives issues MDBs finance has mostly focused on sectoral solutions and has
40 not been able to properly leverage cross-sectoral synergies. At the national level, applied research has
41 shown that integrated modelling of land, energy and water resources not only has the potential to
42 identify superior solutions, but also reveals important differences in terms of investment requirements
43 and required financing arrangements compared to the traditional sectoral financing toolkits (Welsch et
44 al. 2014). Agriculture, forestry, nature-based-solutions (NBS) and other forms of land use are promising
45 sectors for leveraging financing solutions to scale up GHG mitigation efforts (Section 15.4). Moving to
Do Not Cite, Quote or Distribute 12-130 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 more productive and resilient forms of land use is a complex task given the crosscutting nature of land-
2 use that necessarily results in apparent trade-offs between mitigation, adaptation, and development
3 objectives. Finance is one area to manage these trade-offs where there may be opportunities to redirect
4 the hundreds of billions spent annually on land use around the world towards green activities, without
5 sacrificing either productivity or economic development (Falconer et al. 2015). Nonetheless, that would
6 require active public support in design of land use mitigation and adaptation strategies, coordination
7 between public and private instruments across land-use sectors, and leveraging of policy and financial
8 instruments to redirect finance toward greener land-use practices (limited evidence, medium
9 agreement). For example, the Welsch et al. (2014) study on Mauritius shows that the promotion of a
10 local biofuel industry from sugar cane could be economically favourable in the absence of water
11 constraints, leading to a reduction in petroleum imports and GHG emissions while enhancing energy
12 security. Yet, under a water-constrained scenario as a result of climate change, the need for additional
13 energy to expand irrigation to previously rain-fed sugar plantations and to power desalination plants
14 yields the opposite result in terms of GHG emissions and energy costs, making biofuels a sub-optimal
15 option, and negatively affects their economics and the prospects for financing.
16 At the local level, integrated planning and financing are needed to achieve more sustainable
17 outcomes. For example, at a city level integration is needed across sectors such as transport, energy
18 systems, buildings, sewage and solid waste to optimise emissions footprints. How a city is designed
19 will affect transportation demands, which makes it either more or less difficult to implement efficient
20 public transportation, leading in turn to more or less emissions. Under such cases, solutions in terms of
21 public and private investment paths and financing policies based on purely internal sector
22 considerations are bound to cause adverse impacts on other sectors and poor overall
23 outcomes (Gouldson et al. 2016).
24 Availability and access to finance are among the major barriers to GHG emissions mitigation across
25 various sectors and technology options (robust evidence, high agreement). Resource maturity
26 mismatches and risk exposure are two main factors limiting ability of commercial banks and other
27 private lenders to contribute to green finance (Mazzucato and Semieniuk 2018). At all levels,
28 mobilising the necessary resources to leverage cross-sectoral mitigation synergies would require the
29 combination of public and private financial sources (Jensen and Dowlatabadi 2018). Traditional public
30 financing would be required to synergise mitigation across sectors where the risk-return and time
31 profiles of investment are not sufficiently attractive for the business sector. Over the years, private
32 development financing through public-private partnerships (PPP) and other related variants has been a
33 growing source of finance to leverage cross-sectoral synergies and manage trade-offs (Ishiwatari et al.
34 2019; Attridge and Engen 2019; Anbumozhi and Timilsina 2018). Promoting such blended approaches
35 to finance along with result-based financing architectures to strengthen delivery institutions are
36 advocated as effective means to mainstream cross-sectoral mitigation finance (Ishiwatari et al. 2019;
37 Attridge and Engen 2019) (limited evidence, high agreement). The World Bank group and the
38 International Financial Corporation (IFC) have used the blended finance results-based approach to
39 climate financing that addresses institutional, infrastructure, and service needs across sectors
40 targeting developing countries and marginalised communities (GPRBA 2019; IDA 2019).
41
42 12.7 Knowledge Gaps
43 Finally, the literature review and analysis in Chapter 12 has taken account of the post-AR5 literature
44 available and accessible to the chapter authors. Nonetheless, the assessment of the chapter is incomplete
Do Not Cite, Quote or Distribute 12-131 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 without mentioning knowledge gaps encountered during the assessment. These knowledge gaps
2 include:
3 1) Interactions (synergies and trade-offs) between different CDR methods when deployed together are
4 under-researched.
5 • Co-benefits and trade-offs with biodiversity and ecosystem services associated with the
6 implementation of CDR methods.
7 • Constraining technical costs and potentials for CDR methods to define realistically achievable
8 costs and potentials. Such research is useful for improving the representation of CDR methods
9 in IAMs and country-level mitigation pathway modelling.
10 2) More work is required on how framing and communication of mitigation actions in terms of
11 mitigation versus co-benefits potential affects public support in different contexts.
12 3) Additional research work is required to determine the cross-sectoral mitigation potential of emerging
13 General Purpose Technologies.
14 4) Lack of literature on mitigation finance frameworks promoting cross-sectoral mitigation linkages.
15 5) Additional research is needed to better quantify the net GHG emissions and co-benefits and adverse
16 effects of emerging food technologies.
17 • Research in social and behavioural sciences should invest in assessing effectiveness of
18 instrument aiming at shifting food choices in different national contexts.
19 • A better evidence basis is required to understand synergistic effects of policies in food system
20 policy packages.
21 6) Literature on regional/global mitigation potential of biomass production systems that are strategically
22 deployed in agriculture/forestry landscapes, to achieve specific co-benefits.
23 7) Knowledge on land occupation and associated co-benefits and adverse side-effects from large-scale
24 deployment of non-AFOLU mitigation options, and how such options can be integrated with agriculture
25 and forestry to maximise synergies and minimise trade-offs.
26 Frequently Asked Questions (FAQs)
27 FAQ 12.1 How could new technologies to remove carbon dioxide from the atmosphere contribute
28 to climate change mitigation?
29 Limiting the increase in warming to well below 2°C, and achieving net zero CO2 or GHG emissions,
30 will require anthropogenic CO2 removal (CDR) from the atmosphere.
31 The CDR methods studied so far have different removal potentials, costs, co-benefits and side effects.
32 Some biological methods for achieving CDR, like A/R or wetland restoration, have long been practiced.
33 If implemented well, these practices can provide a range of co-benefits, but they can also have adverse
34 side effects such as biodiversity loss or food price increases. Other chemical and geochemical
35 approaches to CDR include Direct Air Carbon Capture and Storage (DACCS), Enhanced Weathering
36 or Ocean Alkalinity Enhancement. They are generally less vulnerable to reversal than biological
37 methods.
Do Not Cite, Quote or Distribute 12-132 Total pages: 220
Final Government Draft Chapter 12 IPCC AR6 WGIII
1 DACCS uses chemicals that bind to CO2 directly from the air; the CO2 is then removed from the sorbent
2 and stored underground or mineralised. Enhanced Weathering involves the mining of rocks containing
3 minerals that naturally absorb CO2 from the atmosphere over geological timescales, which are crushed
4 to increase the surface area and spread on soils (or elsewhere) where they absorb atmospheric CO2.
5 Ocean Alkalinity Enhancement involves the extraction, processing, and dissolution of minerals and
6 addition to the ocean where they enhance sequestration of CO2 as bicarbonate and carbonate ions in the
7 ocean.
8 FAQ 12.2 Why is it important to assess mitigation measures from a systemic perspective, rather
9 than only looking at their potential to reduce Greenhouse Gas (GHG) emissions?
10 Mitigation measures do not only reduce GHGs, but have wider impacts. They can result in decreases or
11 increases in GHG emissions in another sector or part of the value chain to where they are applied. They
12 can have wider environmental (e.g., air and water pollution, biodiversity), social (e.g., employment
13 creation, health) and economic (e.g., growth, investment) co-benefits or adverse side effects. Mitigation
14 and adaptation can also be linked. Taking these considerations into account can help to enhance the
15 benefits of mitigation action, and avoid unintended consequences, as well as provide a stronger case for
16 achieving political and societal support and raising the finances required for implementation.
17 FAQ 12.3 Why do we need a food systems approach for assessing GHG emissions and mitigation
18 opportunities from food systems?
19 Activities associated with the food system caused about one-third of total anthropogenic GHG
20 emissions in 2015, distributed across all sectors. Agriculture and fisheries produce crops and animal-
21 source food, which are partly processed in the food industry, packed, distributed, retailed, cooked, and
22 finally eaten. Each step is associated with resource use, waste generation, and GHG emissions.
23 A food systems approach helps identify critical areas as well as novel and alternative approaches to
24 mitigation on both supply side and demand side of the food system. But complex co-impacts need to be
25 considered and mitigation measures tailored to the specific context. International cooperation and
26 governance of global food trade can support both mitigation and adaptation.
27 There is large scope for emissions reduction in both cropland and grazing production, and also in food
28 processing, storage and distribution. Emerging options such as plant-based alternatives to animal food
29 products and food from cellular agriculture are receiving increasing attention, but their mitigation
30 potential is still uncertain and depends on the GHG intensity of associated energy systems due to
31 relatively high energy needs. Diet changes can reduce GHG emissions and also improve health in
32 groups with excess consumption of calories and animal food products, which is mainly prevalent in
33 developed countries. Reductions in food loss and waste can help reduce GHG emissions further.
34 Recommendations of buying local food and avoiding packaging can contribute to reducing GHG
35 emissions but should not be generalised as trade-offs exist with food waste, GHG footprint at farm gate,
36 and accessibility to diverse healthy diets.
37
Do Not Cite, Quote or Distribute 12-133 Total pages: 220