FINAL DRAFT                              Chapter 14          IPCC WGII Sixth Assessment Report

1

2                                     Chapter 14: North America

3

4 Coordinating Lead Authors: Jeffrey A. Hicke (USA), Simone Lucatello (Mexico), Linda D. Mortsch
5 (Canada)

6

7 Lead Authors: Jackie Dawson (Canada), Mauricio Dom韓guez Aguilar (Mexico), Carolyn A.F. Enquist
8 (USA), Elisabeth A. Gilmore (USA/Canada), David S. Gutzler (USA), Sherilee Harper (Canada), Kirstin
9 Holsman (USA), Elizabeth B. Jewett (USA), Timothy A. Kohler (USA), Kathleen Miller (USA)

10

11 Contributing Authors: Taylor Armstrong (USA), Craig Brown (Canada), Polly C. Buotte (USA), Malinda
12 Chase (USA), Cecilia Conde (Mexico), Nikki Cooley (USA), Karen Cozzetto (USA), Ashlee Cunsolo
13 (Canada), Dalee Sambo Dorough (USA), Jeffrey Dukes (USA), Emile Elias (USA), Luis Fernandez
14 (Mexico), Halley Froehlich (USA), Elliott Hazen (USA), Blas Perez Henriquez (Mexico), Melissa Kenney
15 (USA), Salvador Lluch (Mexico), Dara Marks-Marino (USA), Deborah McGregor (Canada), Matto
16 Mildenberger (Canada), Alejandro Monterrosso (Mexico), Michelle Rutty (Canada), Silvia Salas (Mexico),
17 Miguel Sioui (Canada), Gustavo Sosa (Mexico), Kyle Whyte (USA)

18

19 Chapter Scientist: Polly C. Buotte (USA)

20

21 Review Editors: Margot Hurlbert (Canada), Linda Mearns (USA)

22

23 Date of Draft: 1 October 2021

24

25 Notes: TSU Compiled Version

26

27

28 Table of Contents

29

30 Executive Summary..........................................................................................................................................3
31 14.1 Introduction and Point of Departure......................................................................................................7

32  14.1.1 Context ............................................................................................................................................9

33 14.2 Current and Future Climate in North America ....................................................................................9

34  14.2.1 Observed Changes in North American Climate............................................................................10

35  14.2.2 Projected Changes in North American Climate ...........................................................................12

36 FAQ 14.1: How has climate change contributed to recent extreme events in North America and their

37  impacts? ...................................................................................................................................................13

38 14.3 Perception of Climate Change Hazards, Risks, and Adaptation in North America........................14

39  14.3.1 Climate Change as a Salient Issue................................................................................................14

40  14.3.2 Public Perceptions, Opinions and Understanding of Climate Change ........................................15

41  14.3.3 Building Consensus on Climate Change.......................................................................................16

42  14.3.4 Factors Influencing Perceptions of Climate Change Risks and Adaptation Action .....................17

43 Box 14.1: Integrating Indigenous `Responsibility-Based Thinking' into Climate Change Adaptation

44  and Mitigation Strategies.......................................................................................................................18

45 14.4 Indigenous Peoples and Climate Change .............................................................................................19

46 14.5 Observed Impacts, Projected Risks, and Adaptation by Sector ........................................................22

47  14.5.1 Terrestrial and Freshwater Ecosystems and Communities ..........................................................22

48 Box 14.2: Wildfire in North America............................................................................................................25

49  14.5.2 Ocean and Coastal Social-Ecological Systems ............................................................................28

50 Box 14.3: Marine Heatwaves .........................................................................................................................30

51  14.5.3 Water Resources ...........................................................................................................................31

52  14.5.4 Food, Fibre, and Other Ecosystem Products................................................................................38

53  14.5.5 Cities, Settlements and Infrastructure...........................................................................................46

54 Box 14.4: Sea Level Rise Risks and Adaptation Responses for Selected North American Cities and

55  Settlements ..............................................................................................................................................50

56  14.5.6 Health and Wellbeing....................................................................................................................53

57  14.5.7 Tourism and Recreation................................................................................................................59

    Do Not Cite, Quote or Distribute         14-1                Total pages: 157
    FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

1   14.5.8 Economic Activities and Sectors in North America ......................................................................62

2 Box 14.5: Climate Change Impacts on Trade Affecting North America...................................................66

3 Box 14.6: The Costs and Economic Consequences of Climate Change in North America......................67

4   14.5.9 Livelihoods ....................................................................................................................................69

5   14.5.10 Violence, Crime, and Security ..............................................................................................71

6 14.6 Key Risks .................................................................................................................................................74

7   14.6.1 Key Risks of Climate Change for North America .........................................................................74

8   14.6.2 Key Risks across Sectors in North America..................................................................................75

9   14.6.3 Cumulative risk, tipping points, thresholds and limits..................................................................77

10 FAQ 14.2: What can we learn from the North American past about adapting to climate change?.......78

11 14.7 Adaptation in North America................................................................................................................81

12  14.7.1 Overview of Observed Adaptation in North America ...................................................................81

13  14.7.2 The Solution Space........................................................................................................................84

14 Box 14.7: Nature-based Solutions to Support Adaptation to Climate Change.........................................90

15 FAQ 14.4: What are some effective strategies for adapting to climate change that have been

16  implemented across North America, and are there limits to our ability to adapt successfully to

17  future change? ........................................................................................................................................93

18 References........................................................................................................................................................95

19

20

    Do Not Cite, Quote or Distribute  14-2        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Executive Summary

 2

 3 Since AR5, climate change impacts have become more frequent, intense, and have affected many
 4 millions of people from every region and sector across North America (Canada, US and Mexico).
 5 Accelerating climate change hazards pose significant risks to the wellbeing of North American
 6 populations and the natural, managed and human systems on which they depend (high confidence1).
 7 Addressing these risks have been made more urgent by delays due to misinformation about climate
 8 science that has sowed uncertainty, and impeded recognition of risk (high confidence). {14.2, 14.3}

 9

10 Without limiting warming to 1.5癈, key risks to North America are expected to intensify rapidly by
11 mid-century (high confidence). These risks will result in irreversible changes to ecosystems, mounting
12 damages to infrastructure and housing, stress on economic sectors, and disruption of livelihoods,
13 mental and physical health, leisure, and safety. Immediate, widespread, and coordinated
14 implementation of adaptation measures aimed at reducing risks and focused on equity have the
15 greatest potential to maintain and improve the quality of life for North Americans, ensure sustainable
16 livelihoods, and protect the long-term biodiversity, and ecological and economic productivity in North
17 America (high confidence). Enhanced sharing of resources and tools for adaptation across economic,
18 social, cultural and national entities enables more effective short- and long-term responses to climate
19 change. {14.2, 14.4, 14.5, 14.6, 14.7}

20

21 Past and Current Impacts and Adaptation

22

23 Over the past 20 years, climate change impacts across North America have become more frequent,
24 intense and affect more of the population (high confidence). Despite scientific certainty of the
25 anthropogenic influence on climate change, misinformation and politicization of climate change science has
26 created polarization in public and policy domains in North America, particularly in the US, limiting climate
27 action (high confidence). Vested interests have generated rhetoric and misinformation that undermines
28 climate science and disregards risk and urgency (medium confidence). Resultant public misperception of
29 climate risks and polarized public support for climate actions is delaying urgent adaptation planning and
30 implementation (high confidence). Including Indigenous knowledge, communication and outreach as well as
31 collaborations to co-create equitable solutions are critical for successful climate action. {Box 14.1, 14.3,
32 14.7}

33

34 Climate change has negatively impacted human health and wellbeing in North America (very high
35 confidence). High temperatures have increased mortality and morbidity (very high confidence), with impacts
36 that vary by age, gender, location, and socioeconomic conditions (very high confidence). Changes in
37 temperature and precipitation have increased risk of vectorborne (very high confidence), waterborne (high
38 confidence), and foodborne diseases (very high confidence). Changes in climate and extreme events have
39 been linked to wide-ranging negative mental health outcomes (high confidence). The loss of access to marine
40 and terrestrial sources of protein has impacted the nutrition of subsistence-dependent communities across
41 North America (high confidence). Climate change has increased the extent of warmer and drier conditions
42 favourable for wildfires (medium confidence) that increase respiratory distress from smoke (very high
43 confidence). {14.5.2, 14.5.6, Box 14.2}

44

45 North American food production is increasingly affected by climate change (high confidence), with
46 immediate impacts on the food and nutritional security of Indigenous Peoples. Climate change and
47 extreme weather events have impacted North American agroecosystems (high confidence), with crop-
48 specific effects that vary in direction and magnitude by event and location. Climate change has generally
49 reduced agricultural productivity by 12.5% since 1961, with progressively greater losses moving south from
50 Canada to Mexico and in drought-prone rainfed systems (high confidence) while favorable conditions
51 increased yields of maize, soybeans in regions like the US Great Plains. Loss of availability and access to
52 marine and terrestrial sources of protein has impaired food security and nutrition of subsistence-dependent

1

 In this Report, the following summary terms are used to describe the available evidence: limited, medium, or robust;
and for the degree of agreement: low, medium, or high. A level of confidence is expressed using five qualifiers: very
low, low, medium, high, and very high, and typeset in italics, e.g., medium confidence. For a given evidence and
agreement statement, different confidence levels can be assigned, but increasing levels of evidence and degrees of
agreement are correlated with increasing confidence.

Do Not Cite, Quote or Distribute  14-3        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Indigenous Peoples across North America (high confidence). Climate change has impacted aquaculture (high
 2 confidence) and induced rapid redistribution of species (very high confidence), and population declines of
 3 multiple key fisheries (high confidence). {14.5.4, 14.5.6, 14.7}

 4

 5 Climate change has impaired North American freshwater resources and reduced supply security (high
 6 confidence). Reduced snowpack and earlier runoff (high confidence) have adversely affected aquatic
 7 ecosystems and freshwater availability for human uses (medium confidence). Recent severe droughts, floods
 8 and harmful algal and pathogen events have caused harm to large populations and key economic sectors
 9 (high confidence). Heavy exploitation of limited water supplies, especially in the western US and northern
10 Mexico, and deteriorating freshwater management infrastructure, have heightened the risks (high
11 confidence). Effective examples of freshwater resource adaptation planning are already underway, but
12 coordinated adaptation implementation across multiple conflicting interests and users is complicated and
13 time consuming (high confidence). {14.5.1, 14.5.2, 14.5.3}

14

15 Extreme events and climate hazards are adversely affecting economic activities across North America
16 and have disrupted supply-chain infrastructure and trade (high confidence). Larger losses and
17 adaptation costs are observed for sectors with high climate exposures, including tourism, fisheries, and
18 agriculture (high confidence) and outdoor labor (medium confidence). Disaster planning and spending,
19 insurance, markets, and individual and household level adaptation have acted to moderate effects to
20 date (medium confidence). Entrenched socioeconomic vulnerabilities have amplified climate impacts for
21 marginalized groups, including Indigenous Peoples due to the impact of colonialism and discrimination
22 (medium confidence). {14.5.4, 14.5.5, 14.5.6, 14.5.7, 14.5.9, Box 14.1, Box 14.5, Box 14.6}

23

24 North American cities and settlements have been affected by increasing severity and frequency of

25 climate hazards and extreme events (high confidence), which has contributed to, infrastructure
26 damage, livelihood losses, damage to heritage resources, and safety concerns. Impacts are particularly
27 apparent for Indigenous Peoples for whom culture, identity, commerce, health and wellbeing are closely
28 connected to a resilient environment (very high confidence). Higher temperatures have been associated with
29 violent and property crime in the US (medium confidence) yet the overall effects of climate change on crime
30 and violence in North America are not well understood. {14.4, 14.5.5, 14.5.6, 14.5.8, 14.5.9, Box 14.1}

31

32 Terrestrial, marine, and freshwater ecosystems are being profoundly altered by climate change across
33 North America (very high confidence). Rising air, water, ocean and ground temperatures have restructured
34 ecosystems and contributed to the redistribution (very high confidence) and mortality (high confidence) of
35 fish, bird, and mammal species. Extreme heat and precipitation trends on land have increased vegetation
36 stress and mortality, reduced soil quality, and altered ecosystem processes including carbon and freshwater
37 cycling (very high confidence). Warm and dry conditions associated with climate change have led to tree die-
38 offs (high confidence) and increased prevalence of catastrophic wildfire (medium confidence) with an
39 increase in the size of severely burned areas in western North America (medium confidence). Nature-based
40 solutions and ecosystem-based management have been effective adaptation approaches in the past but are
41 increasingly exceeded by climate extremes (medium confidence). {14.5.1-3, Box 14.7}

42

43 Climate-driven changes are particularly pronounced within Arctic ecosystems and are unprecedented
44 based on observations from multiple knowledge systems (very high confidence). Climate change has
45 contributed to cascading environmental and socio-cultural impacts in the Arctic (high to very high
46 confidence) that have adversely, and often irreversibly, altered Northern livelihoods, cultural activities,
47 essential services, health, food and nutritional security, community connectivity, and wellbeing (high
48 confidence). {14.5.2, 14.5.4, 14.5.6, 14.5.7, 14.5.8, Box 14.6}

49

50 Future Risks and Adaptation

51

52 Climate hazards are projected to intensify further across North America (very high confidence). Heat
53 waves over land and in the ocean as well as wildfire activity will intensify; sub-Arctic snowpack, glacial
54 mass and sea ice will decline (virtually certain); and sea level rise will increase at geographically differential
55 rates (virtually certain). Humidity-enhanced heat stress, aridification, and extreme precipitation events that
56 lead to severe flooding, erosion, debris flows, and ultimately loss of ecosystem function, life and property,
57 are projected to intensify (high confidence). {14.2}

Do Not Cite, Quote or Distribute  14-4        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Health risks are projected to increase this century under all future emissions scenarios (very high
 3 confidence) but the magnitude and severity of impacts depends on the implementation and
 4 effectiveness of adaptation strategies (very high confidence). Warming is projected to increase heat-
 5 related mortality (very high confidence) and morbidity (medium confidence). Vectorborne disease
 6 transmission, waterborne disease risks, food safety risks and mental health outcomes are projected to
 7 increase this century (high confidence). Available adaptation options will be less effective or unable to
 8 protect human health under high-emission scenarios (high confidence). {14.5.6, Box 14.2}

 9

10 Climate-induced redistribution and declines in North American food production are a risk to future
11 food and nutritional security (very high confidence). Climate change will continue to shift North
12 American agricultural and fishery suitability ranges (high confidence) and intensify production losses of key
13 crops (high confidence), livestock (medium confidence), fisheries (high confidence), and aquaculture
14 products (medium confidence). In the absence of mitigation, incremental adaptation measures may not be
15 sufficient to address rapidly changing conditions and extreme events, increasing the need for cross-sectoral
16 coordination in implementation of mitigation and adaptation measures (high confidence). Combining
17 sustainable intensification, Indigenous knowledge and local knowledge based-approaches, and ecosystem-
18 based methods with inclusive and self-determined decision making will result in more equitable food and
19 nutritional security (high confidence). {14.5.1-4, 14.5.6,14.7, Croxx-Chapter Box INDIG in Chapter 18,
20 Cross-Chapter Box MOVING PLATE in Chapter 5}

21

22 Escalating climate change impacts on marine, freshwater, and terrestrial ecosystems (high confidence)
23 will alter ecological processes (high confidence) and amplify other anthropogenic threats to protected
24 and iconic species and habitats (high confidence). Hotter droughts and progressive loss of seasonal water
25 storage in snow and ice will tend to reduce summer season stream flows in much of western North America,
26 while population growth, extensive irrigated agriculture and the needs of threatened and endangered aquatic
27 species will continue to place high demands on those flows (high confidence). {14.2.2, 14.5.1, 14.5.2, 14.5.3,
28 14.5.4, 14.5.6, Box 14.7.1}

29

30 Market and non-market economic damages are projected to increase to the end of the century from
31 climate impacts (high confidence). Estimates for the costs of climate inaction are substantial across
32 economic sectors, infrastructure, human health and disaster management. Hard limits to adaptation may be
33 reached for outdoor labor (medium confidence) and nature-based winter tourism activities (very high
34 confidence). At higher levels of warming, climate impacts may pose systemic risks to financial markets
35 through impacts on transportation systems, supply-chains, and major infrastructure as well as global scale
36 challenges to trade (medium confidence). {14.2.2, 14.5.4, 14.5.8, 14.5.7, 14.5.9, 14.5.5, Box 14.5, Box 14.6}

37

38 Solution Space, Governance

39

40 Self-determination for Indigenous Peoples is critical for effective adaptation in Indigenous
41 communities (very high confidence). Throughout North America, Indigenous Peoples are actively
42 addressing the compound impacts of climate change, and historical and ongoing forms of colonialism (very
43 high confidence). Indigenous knowledge underpins successful understanding of, responses to, and
44 governance of climate change risks. Western scientific practices and technology may not be sufficient in
45 addressing future natural resource management challenges. Supporting Indigenous self-determination,
46 recognizing Indigenous Peoples' rights, and supporting Indigenous knowledge based-adaptation are critical
47 to reducing climate change risks to achieve adaptation success (very high confidence). {14.7.3, Box 14.1}

48

49 Equitable, inclusive and participatory approaches that integrate climate impact projections into near-
50 term and long-term decision-making reduce future risks (high confidence). Government and private
51 investment are increasingly investing in early warning and rapid response systems, climate and ecological
52 forecasting tools, and integrated climate scenario planning methods. Widespread adoption of these practices
53 and tools for infrastructure planning, disaster risk reduction, ecosystem management, budgeting practices,
54 insurance, and climate risk reporting supports planning for a future with more climate risks (high
55 confidence). Increased capacity to support the equitable resolution of existing and emerging resource
56 disputes (local to international) will reduce climate impacts on livelihoods and improve the effectiveness of
57 resource management (high confidence). {14.5.5, 14.5.10, 14.7}

Do Not Cite, Quote or Distribute  14-5        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Near- and long-term adaptation planning, implementation, and coordination across sectors and
 3 jurisdictions supports equitable and effective climate solutions (high confidence). Recognition of the
 4 need for adaptation across North America is increasing but action has been mostly gradual, incremental, and
 5 reactive (high confidence). Current practices will be increasingly insufficient without coordination and
 6 integration of efforts through equitable policy focused on modifying land use impacts, consumption patterns,
 7 economic activities, and emphasizing nature-based solutions (high confidence). Transformational, long-term
 8 adaptation action that reduces risk and increases resilience can address rapidly escalating impacts in the mid
 9 to latter part of the 21st century, especially if coupled with moderate to high mitigation measures (high
10 confidence). {14.7}

11

12

Do Not Cite, Quote or Distribute  14-6        Total pages: 157
    FINAL DRAFT                             Chapter 14                    IPCC WGII Sixth Assessment Report

1 14.1 Introduction and Point of Departure

2

3 Earth's climate is currently changing in significant ways as a result of human activities, and future

4 projections indicate continued and possibly accelerating change without reductions in greenhouse gas (GHG)

5 emissions (Guti閞rez et al., In Press; IPCC, In Press). Climate change affects human and natural systems; this
6 chapter provides an assessment of present and future climate change impacts, risks, and adaptation for North

7 America, including Mexico, Canada, and the United States (US) and coastal waters within the 370 km

8 exclusive economic zone. We do not consider Hawaii and other island territories of the US in depth as they

9 are assessed in Chapter 15 (Small Islands). Chapter 14 assesses evidence from Arctic Canada and Alaska,

10 which is synthesized in the Polar Regions Cross-Chapter Paper (CCP6).

11

12 Evidence from Indigenous knowledge (IK) systems is included in this chapter to assess climate change risks

13 and solutions in North America following the framing provided in Chapter 1 Special Report on the Ocean

14 and Cryosphere in a Changing Climate (SROCC) (Abram et al., 2019) and Special Report on Climate

15 Change and Land (SRCCL) (IPCC, 2019a). Indigenous Contributing Authors provided this assessment,

16 reflecting the importance of meaningfully including IK in assessment processes (Ford, 2012; Ford et al.,
17 2016; Hill et al., 2020). This addition represents an important advancement since AR5 (IPCC, 2013; IPCC,

18 2014).

19

20 Our main point of departure was the Fifth Assessment Report (AR5) for WGII (IPCC, 2014). Key findings

21 drawn from the Executive Summary for the North America chapter are summarized in Table 14.1.

22 Subsequent IPCC reports such as Special Report on Global Warming of 1.5癈 (SR1.5) (Hoegh-Guldberg et
23 al., 2018; IPCC, 2018), SROCC (IPCC, 2019b), and SRCCL (IPCC, 2019a) also informed the assessment.

24 We additionally incorporated recent national climate assessments of the US (USGCRP, 2018) and Canada

25 (Bush and Lemmen, 2019; Warren and Lulham, 2021) as well as the 6th Mexican national communication of

26 climate change to the United Nations (SEMARNAT and INECC, 2018).

27

28

29 Table 14.1: Key findings from AR5 North America Chapter (Romero-Lankao et al., 2014b).

    General topic       AR5 finding

    Climate hazards     Climate has changed in North America, with some changes attributed to human
                        activities.

    Natural ecosystems  Climate hazards, especially related to heatwaves, heavy precipitation, and
                        snowpack, are expected to change in ways that are adverse to natural and human
                        systems.

                        Warming, increasing carbon dioxide (CO2) concentrations, sea level rise (SLR),
                        and climate extremes are stressing ecosystems.

    Human systems       Water resources that are already stressed in many parts of North America are
                        expected to become further stressed by climate change. Current adaptation options
                        can address water supply deficits but responses to flooding and water quality
                        concerns are more limited.

                        Climate change has affected yields of major crops, and projections indicate
                        continued declines, although with variability.

                        Extreme climate events have affected human health, although climate change-
                        related trends and attribution to climate change were not confirmed.

                        Multiple aspects of climate change have affected livelihoods, economic activities,
                        infrastructure, and access to services.

                        Much infrastructure is vulnerable to extreme weather events and unless adaptation
                        investments are made, vulnerability to future climate change persists and
                        increases.

    Do Not Cite, Quote or Distribute        14-7                                           Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report
 Adaptation
             Most sectors of the North American economy have been affected by and have
             responded to extreme weather, including hurricanes, flooding, and intense rainfall.

             Technological innovation, institutional capacity-building, economic
             diversification, and infrastructure design are adaptations for reducing current
             climate impacts as well as future risks due to a changing climate.

                                               Predominantly, North American governments have undertaken incremental

                                               adaptation assessment and planning at the municipal level. Limited proactive,

                                               anticipatory adaptation is directed at long-term investment for energy and public

                                               infrastructure.

 1
 2

 3 Chapter 14 sections are organized to address themes and content as contained in the IPCC-approved outline
 4 for regions. Regional climate changes assessed within North America are keyed to Figure 14.1 using
 5 italicized four-letter abbreviations (e.g., CA-ON, US-SE, MX-NW). The assessment addresses recent and
 6 future climate for North America, the impacts, risks and adaptation within sectors, key risks (KR), the nature
 7 of adaptation and sustainable development pathways as well as two additional sections on Indigenous
 8 Peoples and perceptions of climate change. Seven Boxes are used to highlight topics of interdisciplinary
 9 nature while four Frequently Asked Questions (FAQ) were produced in plain language for communication to
10 the public. The chapter utilizes the framework as well as designated terms in the standardized process for
11 evaluating and characterizing the degree of certainty in assessment findings developed through the expert
12 judgment process (Mach et al., 2017) (see WGII 1.3.4) (references to other relevant chapters in this WGII
13 report are abbreviated in this manner). The WGII Glossary provides definitions for terms and concepts used
14 across the report.

15
16

Do Not Cite, Quote or Distribute  14-8        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Figure 14.1: North American regions and subregions, adapted from national climate assessments, and city names,
 3 referred to in discussion of local and regional climate change impacts and adaptation.

 4

 5

 6 14.1.1 Context

 7

 8 With a 2019 total population of over 494 million people (US 329 million, Mexico 128 million and Canada
 9 37 million), North America comprises 6.4% of the global population. Relative to other countries, North
10 America has low population densities (Mexico 64 people/km2; US 35/km2; Canada, 4/km2) (United Nations,
11 2019). Population projections indicate a steady growth in the three countries, which will exert pressure on
12 consumption and increase risks under climate change (United Nations, 2019). North America is also
13 responsible for about a quarter of global greenhouse gas (GHG) emissions. Since 1990, North American
14 GHG emissions have increased by almost 18% (Ritchie and Roser, 2020) and in 2019 the region was
15 responsible for 5.9 MtCO2 emissions worldwide (Friedlingstein et al., 2020). In terms of annual CO2
16 emissions per capita, in 2019 Canada had 15 metric tons CO2 per person (tCO2/person), the US had 16
17 tCO2/person, and Mexico had 3.4 tCO2/person (Friedlingstein et al., 2020).

18

19

20 14.2 Current and Future Climate in North America

21

22 Trends in observed and projected physical climate variables, and changes in extreme weather and climate
23 events, are summarized in this section. Many of the assessments here are adapted from AR6 WGI (IPCC, In
24 Press), especially chapters 11 (Seneviratne et al., 2021) and 12 (Ranasinghe et al., 2021) and the WGI Atlas
25 (Guti閞rez et al., In Press) (references to chapters in WGI are hereafter abbreviated I.11, I.12, I.Atlas, etc.).
26 I.12.4.6 assesses North American climatic impact drivers without assessing their impacts or associated risks.
27 WGI assessments are augmented in this section with regionally specific support from recent national climate
28 assessments or original literature.

Do Not Cite, Quote or Distribute  14-9        Total pages: 157
      FINAL DRAFT                 Chapter 14  IPCC WGII Sixth Assessment Report

1
2

 3

 4 Figure 14.2: Observed and projected climate changes across North America. Black boundary lines delineate North
 5 American subregions (Fig. 14.1). Data were extracted from WGI online atlas, where data set details can be found. A)
 6 recent observations; (B)-(G) from an ensemble of CMIP6 projections. A) Observed annual mean temperature trend over
 7 land for period 1980-2015. B, C) Projected change in annual mean temperature over land relative to 1986-2005 average,
 8 associated with 2癈 or 4癈 global warming. D, E) Like (B, C), but for projected percentage change in annual
 9 precipitation. F, G) Like (B, C), but for projected change in number of days/year with maximum temperature > 40癈
10 ("TX40").

11

12

13 14.2.1 Observed Changes in North American Climate

14

15 Climate changes directly related to increasing mean and extreme temperature, including reduced snowpack,
16 sea and lake ice and glacier extent, and marine heatwaves, can be attributed to human activity and are
17 affecting most of North America (high confidence). Upward trends in annual mean temperature across North
18 America since 1960 are wide-spread (I.Atlas) but nonuniform (Figure 14.2a). Pronounced polar
19 amplification of warming is observed in high latitudes (Figure 14.1a), particularly in winter (Vose et al.,
20 2017; Zhang et al., 2019a ) (I.Atlas). As average temperature rises, extreme high temperature records across
21 North America are being set more frequently than extreme cold records (Meehl et al., 2016) and the
22 probability of cold extreme events is reduced (I.11). Trends in daily maximum and minimum temperature are
23 significant in high latitudes (US-AK, CA-NW, CA-NE). Summertime daily maximum temperature is
24 increasing in southwestern desert regions (US-SW, MX-NW) (Martinez-Austria et al., 2016; Martinez-Austria
25 and Bandala, 2017; Navarro-Estupinan et al., 2018).

26

27 Annual precipitation has increased in recent decades in northern and eastern areas (CA-PR, CA-QU, US-NP,
28 US-SP, US-MW, US-NE, US-AK) (high confidence), and has decreased across the western part of the
29 continent (CA-BC, US-SW, US-NW, MX-NW) (medium confidence), with considerable spatial variability
30 within these regions (Zhang et al., 2019a; Guti閞rez et al., In Press). Elsewhere across North America there is
31 limited evidence and low agreement on detection of observed trends in total precipitation and river flood

Do Not Cite, Quote or Distribute  14-10       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 hazards. The intensity and frequency of one-day heavy precipitation events have very likely2 increased since
 2 the mid-20th Century across most of the US (US-NP, US-MW, US-NE, but not in US-SE) and in Mexico, but
 3 no detectable trend is reported in Canada (Zhang et al., 2019a) (I.11). Recent flooding events along the mid-
 4 latitude Pacific Coast have been attributed to increasingly intense atmospheric river (AR) events (Gershunov
 5 et al., 2019; Vano et al., 2019) (I.8) but there is low confidence in detecting trends in AR activity.

 6

 7 Snowpack and snow extent across much of Canada and the western US have declined as temperatures have
 8 increased (Kunkel et al., 2016; Mote et al., 2018; Mudryk et al., 2018; Derksen et al., 2019) (I.12; I.Atlas)
 9 (very high confidence). Warm "snow droughts", describing a deficit of snowpack available for runoff even in
10 the absence of a winter precipitation deficit (Cooper et al., 2016; Harpold et al., 2017), have become more
11 common in North American mountains (Sproles et al., 2016; Nicholls et al., 2018; Pershing et al., 2018).
12 Glaciers have retreated over the past half-century at high elevation across North America (Frans et al., 2018;
13 Zemp et al., 2019) and in the Arctic (Burgess, 2017; Box et al., 2019; Derksen et al., 2019). Lake ice in
14 Canada, south of the Arctic region delineated in Figure 14.1, has declined (Alexeev et al., 2016; Derksen et
15 al., 2019).

16

17 There is limited evidence of trends in meteorological or hydrologic droughts over the historical record
18 (Wehner et al., 2017) (I.8 and I.11 discuss multiple perspectives on drought), but there is medium confidence
19 in increasing atmospheric evaporative demand acting to intensify surface aridity during recent droughts
20 (Williams et al., 2020) (I.11; US-SW). The ongoing multi-decadal dry period in the Colorado River Basin is
21 as extreme as any drought in the past thousand years (Murphy and Ellis, 2019; Williams et al., 2020).

22

23 The proportion of hurricanes in stronger categories has likely increased globally over the past 40 years, with
24 medium confidence that the onshore propagation speed of hurricanes making landfall in the US has slowed
25 detectably since 1900 (Kossin, 2018) (I.11), contributing to detectable increases in local rainfall and coastal
26 flooding associated with these storms. There is high confidence (I.11) that anthropogenic climate change
27 contributed to extreme precipitation associated with recent intense hurricanes, such as Harvey in 2017.

28

29 North American sea ice extent and volume (thickness) have declined up to 10% per decade since 1981 (Ding
30 et al., 2017; Mudryk et al., 2018; Derksen et al., 2019; IPCC, 2019c)(I.9), with changes accelerating during
31 this time (Schweiger et al., 2019) (robust evidence, high agreement), resulting in longer and larger periods of
32 open water (Wang et al., 2018a). Recent (2018) sea ice extent in the Bering Sea was the lowest in a 5,500 yr
33 record and appears to lag atmospheric CO2 by ~2 decades (Jones et al. 2021). High Arctic sea ice retreat
34 since 1971 and increases in open water duration in the most recent decade are unprecedented (Box et al.,
35 2019) and most pronounced in the Chukchi, Bearing, and Beaufort Seas (US-AK, CA-NW) (high confidence)
36 (Wang and Overland, 2015; Jones et al., 2020).

37

38 Warming of North American offshore waters is significant and attributable to human activities, particularly
39 along the Atlantic coast, contributing to sea level rise (SLR) through thermal expansion (IPCC, 2019c) (I.9)
40 (very high confidence). Rates of SLR have accelerated along most North American coasts during the past
41 three decades, excepting coastlines in southern Alaska and northeastern Canada where land is rising (I.12).
42 Tidal flooding frequency has increased in the North Pacific from once every 1-3 years to every 6-12 mo
43 (Sweet et al., 2014).

44

45 Acidification of North American coastal waters has occurred in conjunction with increased atmospheric CO2
46 concentration (Mathis et al., 2015; Jewett and Romanou, 2017; Claret et al., 2018) combined with other local
47 acidifying inputs such as nitrogen and sulphur deposition (Doney et al., 2007) and freshwater nutrient input
48 (Strong et al., 2014; IPCC, 2019c) (very high confidence). Oxygen minimum zones, particularly in the North
49 Pacific south of US-AK, have expanded in volume and O2 has declined since 1970 (IPCC, 2019c).

50

2

 In this Report, the following terms have been used to indicate the assessed likelihood of an outcome or a result:
Virtually certain 99�100% probability, Very likely 90�100%, Likely 66�100%, About as likely as not 33�66%,
Unlikely 0�33%, Very unlikely 0�10%, and Exceptionally unlikely 0�1%. Additional terms (Extremely likely: 95�
100%, More likely than not >50�100%, and Extremely unlikely 0�5%) may also be used when appropriate. Assessed
likelihood is typeset in italics, e.g., very likely). This Report also uses the term `likely range' to indicate that the assessed
likelihood of an outcome lies within the 17-83% probability range

Do Not Cite, Quote or Distribute  14-11       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 14.2.2 Projected Changes in North American Climate

 2

 3 Climate changes related to warming temperature, including more intense heat waves over land and in the
 4 ocean, diminished snowpack, sea ice reduction and SLR, are projected with high confidence and are strongly
 5 sensitive to future greenhouse gas concentrations (Figure 14.2). Climatic hazards affected by hydrologic
 6 change, including humidity-inclusive heat stress, extreme precipitation, and more intense storms, are
 7 projected to intensify.

 8

 9 Pronounced amplification of warming across the Arctic and continental intensification of warming (Figure
10 14.1bc) is projected with high confidence (Doney et al., 2007; Vose et al., 2017). Extreme heat waves are
11 projected to intensify, particularly in MX-NW, MX-N, MX-NE, US-SW, US-NP and US-SP (Figure 14.2f-g)
12 and become more frequent and longer in duration as average temperature rises across North America (I.11).
13 Extreme cold events are projected to decrease in severity (Wuebbles et al., 2014)(I.12).

14

15 Total precipitation is projected to increase across the northern half of North America (very high confidence)
16 and decrease in southwestern North America (MX-SW, MX-NW, US-SW) (medium confidence) (Fig 14.2d-e,
17 I.Atlas). Further increases in the intensity of locally heavy precipitation are very likely across the continent,
18 as a greater fraction of precipitation falls in intense events (Easterling et al., 2017; Prein et al., 2017a; Zhang
19 et al., 2019a).

20

21 High-humidity hazards are projected to increase (medium confidence) in regions around the Gulf of Mexico
22 and southeastern North America (US-SE, US-SP, MX-NE, MX-SE) (Zhao et al., 2015). In subtropical regions
23 that are less influenced by moisture from the Gulf of Mexico (including US-SW, US-SP, MX-NW and MX-N),
24 the combination of higher temperature and less total precipitation leads to projections of increased aridity:
25 drier surface conditions, higher evaporative demand by plants, and more intense droughts (Jones and
26 Gutzler, 2016; Easterling et al., 2017; Escalante-Sandoval and Nu馿z-Garcia, 2017) (I.12).

27

28 As temperatures rise, snow extent, duration of snow cover and accumulated snowpack are virtually certain to
29 decline in sub-Arctic regions of North America (McCrary and Mearns, 2019; Mudryk et al., 2021) (I.Atlas),
30 with corresponding effects on snow-related hydrologic changes (high confidence). These include declines in
31 snowmelt runoff (Li et al., 2017); increased evaporative losses during snow ablation (Foster et al., 2016;
32 Milly and Dunne, 2020); and increases in the frequency of rain-on-snow events (Jeong and Sushama, 2018a)
33 and consecutive snow drought years in western North America (Marshall et al., 2019a).

34

35 Climate change is projected to magnify the impact of tropical cyclones in US-NE, MX-NE, US-SP, and US-
36 SE by increasing rainfall (Patricola and Wehner, 2018) and extreme wind speed (high confidence) and
37 slowing the speed of land-falling storms (Kossin, 2018)(I.11) (limited evidence, low confidence). The coastal
38 region at severe risk from tropical storms is projected to expand northward within US-NE (Kossin et al.,
39 2017) (medium confidence).

40

41 Additional reduction in polar sea ice is virtually certain (Mudryk et al., 2021)(I.12), with the North
42 American Arctic projected to be seasonally ice-free at least once per decade under 2癈 of global warming
43 (high confidence) (Mudryk et al.; IPCC, 2019b; Mioduszewski et al., 2019). Duration of freshwater lake ice
44 across the northern US and southern Canada is projected to diminish (high confidence) (Dibike et al., 2012;
45 Mudryk et al., 2018; Sharma et al., 2019) (I.12).

46

47 Ocean surface temperature is very likely to increase in future decades in waters around North America
48 (Jewett and Romanou, 2017; Greenan et al., 2019a), but at a slower rate than air temperature over the
49 continent. Rates of change are projected to be relatively higher in northern latitudes, with most rapid
50 warming in summer in the Arctic and Bering Sea (US-AK, CA-NW) (Wang and Overland, 2015; Wang et al.,
51 2018a; Hermann et al., 2019).

52

53 SLR is virtually certain to continue along North American coastlines except for parts of US-AK with
54 geographically variable rates of rise (I.9, I.12, Box 14.4). Relatively greater SLR is projected along the US-
55 SE, CA-AT and MX-SW coastlines and relatively less along CA-BC and US-NW (Fasullo and Nerem, 2018;
56 Greenan et al., 2019a; IPCC, 2019b) (I.9, I.12, Box 14.4).

57

Do Not Cite, Quote or Distribute  14-12       Total pages: 157
    FINAL DRAFT                                      Chapter 14          IPCC WGII Sixth Assessment Report

1 Ocean acidification along North American coastlines is projected to increase (very high confidence) (Jewett

2 and Romanou, 2017). The frequency and extent of oxygen minimum and hypoxic zones are projected to

3 increase, with less confidence, exacerbated by climate-driven eutrophication and increasing stratification

4 (Altieri and Gedan, 2015; IPCC, 2019b).

5

6

7 [START FAQ14.1 HERE]

8

9 FAQ 14.1: How has climate change contributed to recent extreme events in North America and their

10  impacts?

11

12 Multiple lines of evidence indicate that climate change is already contributing to more intense and more

13 frequent extreme events across North America. The impacts resulting from extreme events represent a huge

14 challenge for adapting to future climate change.

15

16 Extreme events are a fundamental part of how we experience weather and climate. Exceptionally hot days,

17 torrential rainfall, and other extreme weather events have a direct impact on people, communities, and

18 ecosystems. Extreme weather can lead to other impactful events such as droughts, floods or wildfires. In a

19 changing climate, people frequently ask whether extreme events are generally becoming more severe or

20 more frequent, and whether an actual extreme event was caused by climate change.

21

22 Because really extreme events occur rarely (by definition), it can be very difficult to assess whether the

23 overall severity or frequency of such events has been affected by changing climate. Nevertheless, careful

24 statistical analysis shows that record-setting hot temperatures in North America are occurring more often

25 than record-setting cold temperatures as the overall climate has gotten warmer in recent decades. The area

26 burned by large wildfires in the western US has increased in recent decades. Observed trends in extreme

27 precipitation events are more difficult to detect with confidence, because the natural variability of

28 precipitation is so large and the observational database is limited.

29

30 Our understanding of how individual extreme weather events have been influenced by climate change has

31 improved greatly in recent years. Climate scientists have developed a formal technique ("event attribution",

32 described in Working Group I FAQ 11.3) for assessing how climate change affects the severity or frequency

33 of a particular extreme event, such as a record-breaking rainfall event or a marine heat wave. This is a

34 challenging task, because any particular event can be caused by a combination of natural variability and

35 climate change. Event attribution is typically carried out using models to compare the probability of a

36 specific event occurring in today's climatic environment, relative to the probability that the same event might

37 have occurred in a modelled climate in which atmospheric greenhouse gases have not risen due to human

38 activities. Using this strategy, multiple studies estimated that the historically extreme rainfall amount that fell

39 across the Houston area from Hurricane Harvey (2017) was 3 to 10 times more likely as the result of climate

40 change.

41

42 The impacts from extreme events depend not just on physical climate system hazards (temperature,

43 precipitation, wind, etc.), but also on the exposure and vulnerability of humans or ecosystems to these

44 events. For example, damage from landfalling hurricanes along the coast of the Gulf of Mexico is expected

45 to increase as very strong hurricanes become more frequent and intense due to climate change. But damage

46 would also increase with additional construction along the shoreline, because coastal development increases

47 exposure to hurricanes. And if some structures are constructed to poor building standards, as was the case

48 when hurricane Andrew made landfall in Florida in 1992, then vulnerability to hurricane-caused impacts is

49 increased.

50

51 Climate change also contributes to impacts from extreme events by making some building codes and zoning

52 restrictions inadequate or obsolete. Many North American communities limit development in areas known to

53 be flood-prone, to minimize exposure to flooding. But as climate change expands the areas at risk of

54 exposure to flooding beyond historical floodplains, the impacts of potential flooding are increased, as

55 Hurricane Harvey demonstrated. Adapting to climate change may require retrofits for existing structures and

56 revised zoning for new construction. Some structures and neighbourhoods may need to be abandoned

57 altogether to accommodate expanded flooding risk.

    Do Not Cite, Quote or Distribute                  14-13                          Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Climate change can be an added stress that increases impacts from extreme events, combined with other
 3 non-climatic stressors. For example, climate change in western North America has contributed to more
 4 extreme fire weather. The devastating impacts of recent wildfire outbreaks, such as occurred across western
 5 Canada in 2016 and 2017, the western United States in 2018 and 2020, and both countries in 2021, are to
 6 some extent associated with expanded development and forest management practices (such as policies to
 7 suppress low-intensity fires, allowing fuel to accumulate). The effects of development and forest
 8 management have dramatically increased the exposure and vulnerability of communities to intense wildfires.
 9 Climate change has added to these stressors: warming temperature leads to more extreme weather conditions
10 that are conducive to increasingly severe wildfires.

11

12 Biodiversity is affected by climate change in this way too. For example, numerous bird populations across
13 North America are estimated to have declined by up to 30% over the past half-century. Multiple human-
14 related factors, including habitat loss and agricultural intensification, contribute to these declines, with
15 climate change as an added stressor. Increasingly extreme events such as severe storms and wildfires can
16 decimate local populations of birds, adding to existing ecological threats.

17

18 [END FAQ 14.1 HERE]

19

20

21 14.3 Perception of Climate Change Hazards, Risks, and Adaptation in North America

22

23 14.3.1 Climate Change as a Salient Issue

24

25 The majority of the climate science community has reached consensus that mean global temperature has
26 increased and human activity is a major cause (Oreskes, 2004; Anderegg et al., 2010; Cook et al., 2013;
27 Cook et al., 2016; IPCC, In Press), setting the context for public policy action. Despite expert scientific
28 consensus on anthropogenic climate change, there is polarization and an ongoing debate over the reality of
29 anthropogenic climate change in the public and policy domains, with attendant risks to society (high
30 confidence) (Doran and Zimmerman, 2009; Ballew et al., 2019; Druckman and McGrath, 2019; Hornsey and
31 Fielding, 2020; Wong-Parodi and Feygina, 2020). Public perception of consensus regarding anthropogenic
32 climate change can be an important gateway belief, which establishes a crucial precondition for public policy
33 action (van der Linden et al., 2015; van der Linden et al., 2019) by influencing the assessment of climate
34 change risks and opportunities, and formulation of appropriate mitigation and adaptation responses (Ding et
35 al., 2011; Bolsen et al., 2015; Drews and Van den Bergh, 2016; Doll et al., 2017; Mase et al., 2017; Morton
36 et al., 2017). Trust in experts, institutions and environmental groups is also important (Cologna and Siegrist,
37 2020; Termini and Kalafatis, 2021).

38

39 Rhetoric and misinformation on climate change and the deliberate undermining of science have contributed
40 to misperceptions of the scientific consensus, uncertainty, disregarded risk and urgency, and dissent (high
41 confidence) (Ding et al., 2011; Oreskes and Conway, 2011; Aklin and Urpelainen, 2014; Cook et al., 2017;
42 van der Linden et al., 2017). Additionally, strong party affiliation and partisan opinion polarization
43 contribute to delayed mitigation and adaptation action, most notably in the US (high confidence) (van der
44 Linden et al., 2015; Cook and Lewandowsky, 2016; Bolsen and Druckman, 2018; Chinn et al., 2020) but
45 with similar patterns in Canada (medium confidence) (Lachapelle et al., 2012; Kevins and Soroka, 2018).
46 Vocal groups can affect public discourse and weaken public support for climate mitigation and adaptation
47 policies (Aklin and Urpelainen, 2014; Lewandowsky et al., 2019) (medium confidence). Vested economic
48 and political interests have organized and financed misinformation and "contrarian" climate change
49 communication (Brulle, 2014; Farrell, 2016b; Farrell, 2016a; Supran and Oreskes, 2017; Bolsen and
50 Druckman, 2018; Brulle, 2018). Traditional media � print and broadcast � frame and transmit climate change
51 information and play a crucial role in shaping public perceptions, understanding, and willingness to act
52 (Happer and Philo, 2013; Schmidt et al., 2013; Hmielowski et al., 2014; Bolsen and Shapiro, 2018; King et
53 al., 2019; Chinn et al., 2020). The journalistic norm of "balance" (giving equal weight to climate scientists
54 and contrarians in climate change reporting) biases coverage by unevenly amplifying certain messages that
55 are not supported by science, contributing to politicization of science, spreading misinformation, and
56 reducing public consensus on action (Boykoff and Boykoff, 2004; Boykoff and Boykoff, 2007; Cook et al.,
57 2017). Much online social media discussion of climate change takes place in ``echo chambers'' � a social

Do Not Cite, Quote or Distribute  14-14       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 network amongst like-minded people in communities dominated by a single view that contributes to
 2 polarization (Williams et al., 2015; Pearce et al., 2019), and the spread of misinformation (Treen et al.,
 3 2020).

 4

 5 14.3.2 Public Perceptions, Opinions and Understanding of Climate Change

 6

 7 In a 2018 survey across 26 nations, people in Canada and Mexico ranked climate change as the top global
 8 threat, whereas in the US climate change ranked third (Poushter and Huang, 2019). The public's responses to
 9 the causes of climate change and risk perceptions in Canada (Mildenberger et al., 2016) and US (Howe et
10 al., 2015) revealed variations among regions (Figure 14.3) and less acceptance of climate change in rural
11 regions than in urban areas. Canadian regions have higher acceptance of climate change (e.g., recognize it is
12 happening and attributable to human activity) than the most liberal areas in the US (Lachapelle et al., 2012;
13 Mildenberger et al., 2016). Western Canadian regions with high carbon intensity economies had lower
14 acceptance of climate change than the rest of Canada, whereas in the US perceptions were more stable across
15 regions (Lachapelle et al., 2012). A recent survey in Mexico found that for 73% of respondents climate
16 change represents a major economic, environmental and social threat, and in the most vulnerable states (MX-
17 SE), the perception is that climate change impacts and extreme events have considerable implications for the
18 way of life in communities (Zamora Saenz, 2018). In a 2017 survey, Az骳ar et al. (2021) found 85% of
19 respondents from Mexico acknowledged anthropogenic climate change. Peoples' experience with extreme
20 events (e.g., hurricanes, high temperatures), socio-demographic characteristics, level of marginalization and
21 economic and social exclusion, as well as education levels were important factors influencing perception of
22 climate change in Mexico (Corona-Jimenez, 2018; Alfie and Cruz-Bello, 2021; Az骳ar et al., 2021).
23 Drawing upon Indigenous knowledge (Box 14.1) as well as lived experience of recent changes in ice,
24 weather patterns, and species' phenology and distribution, Indigenous Peoples recognize that change is
25 occurring in their communities and have effective solutions that are grounded in Indigenous worldviews
26 (Harrington, 2006; Turner and Clifton, 2009; Norton-Smith et al., 2016a; Savo et al., 2016; Maldonado et al.,
27 2017; Chisholm Hatfield et al., 2018).

28

29

Do Not Cite, Quote or Distribute  14-15       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Figure 14.3: Regional distribution of public perception that "the Earth is getting warmer" as a surrogate for public
 3 acceptance that climate change is happening (% of population). Scale is the Canadian federal electoral district or riding
 4 level and US Congressional District. The three northern territories and Labrador, in Canada, did not meet population
 5 thresholds for modelling. The figure updates Mildenberger et al. (2016) and is based on equivalent public surveys in
 6 both countries -- Canadian "Earth is getting warmer" and US "global warming is happening" undertaken in 2019.
 7 Equivalent surveys and modelling for Mexico are not available at this time.

 8

 9

10 14.3.3 Building Consensus on Climate Change

11

12 Building consensus for action on climate change is influenced by individual factors (e.g., ideology,
13 worldview, trust, partisan identity, religion, education, age) and the broader societal context (e.g., culture,
14 media coverage and content, political climate, economic conditions) (high confidence) (McCright and
15 Dunlap, 2011; Brulle et al., 2012; Hornsey et al., 2016; Arbuckle, 2017; Pearson et al., 2017; Bolsen and
16 Shapiro, 2018; Ballew et al., 2020; Cologna and Siegrist, 2020; Goldberg et al., 2020). In a multi-country
17 assessment of acceptance of global warming influenced by ideology (e.g., conspiratorial ideation,
18 individualism, hierarchy, and left璻ight and liberal璫onservative political orientation), the US uniquely had
19 the strongest link to doubt out of 25 countries for all factors, while Canada's dominant influence on non-
20 acceptance was conservative political ideology, and for Mexico, there were no ideological effects (Hornsey
21 et al., 2018).

22

23 Political affiliation and partisan group identity contribute to polarization on the causes and state of climate
24 change, most notably in the US (medium confidence). Fewer US Republicans hold the belief that human
25 activity causes climate change than Democrats (Bolsen and Druckman, 2018; Druckman and McGrath,
26 2019). Partisanship in the US with respect to climate change has evolved over the period 1997 to 2016;
27 initially, it was limited, but since 2008, there has been a widening, more entrenched partisan "divide"
28 (Dunlap et al., 2016). The millennial generation (born 1980s, 1990s), emerging as the largest US population

Do Not Cite, Quote or Distribute  14-16       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 cohort, has a potentially important political influence � reduction in polarization � as they show relatively
 2 higher levels of concern and acceptance of climate change science than older age groups. Political affiliation
 3 does not have as strong an effect on their climate change beliefs (Corner et al., 2015; Ross et al., 2019).

 4

 5 Communicating to educate or enhance knowledge on climate change science or consensus does not
 6 necessarily lead individuals to revise their beliefs (Bolsen et al., 2015; Druckman and McGrath, 2019)
 7 (medium confidence). People may reject new information that conflicts with their beliefs or not consider it
 8 credible, as political ideology and partisan affiliation are strong influences (Arbuckle, 2017). The climate
 9 change issue may create resistance from individuals with conservative political ideologies and hierarchical,
10 individualistic worldviews because it ascribes responsibility to developed, industrialized countries for
11 emissions and brings about more environmental regulation (Stevenson et al., 2015). Lack of trust in scientific
12 consensus on climate change may actually originate from opposition by US conservatives to the perceived
13 advocacy for different climate change policy approaches that challenge their worldviews (Bolsen and
14 Druckman, 2018).

15

16 14.3.4 Factors Influencing Perceptions of Climate Change Risks and Adaptation Action

17

18 Projected climate change risk, urgency and necessary adaptations are perceived and understood differently
19 by the public, communities, professional groups, climate scientists, and public policy makers (high
20 confidence) (Bolsen et al., 2015; Drews and Van den Bergh, 2016; Morton et al., 2017; Treuer et al., 2018).
21 People can engage with climate change across three dimensions: cognitive - knowledge, affective - feelings,
22 and behavioural - responses and actions (Galway, 2019; Brosch, 2021). Risk assessment can be influenced
23 by values regarding the subject under evaluation (Allison and Bassett; Stevenson et al., 2015) and can
24 interact with other risks and change over time (Mach et al., 2016). Communities and practitioners (e.g.,
25 farmers, foresters, water managers) are influenced in their willingness to modify current practices and adopt
26 new measures based on how they perceive, understand, and experience climate change uncertainty, risk and
27 urgency as well as political and social norms (van Putten et al., 2015; Doll et al., 2017; Mase et al., 2017;
28 Morton et al., 2017; Zanocco et al., 2018). Place-based and local-focused assessments allow individuals to
29 more readily assess and adapt to risks as well as identify roles and responsibilities in the face of multiple,
30 interacting, and often unequally distributed climate change impacts (Khan et al., 2018; Galway, 2019).
31 Interest in preserving local archaeological sites threatened by SLR initiated collaboration and co-production
32 of knowledge among disparate US communities -- citizens, archaeologists, preservationists, planners, land
33 managers, and Indigenous Peoples (Fatori and Seekamp, 2019; Dawson et al., 2020).

34

35 Psychological distancing -- the perception that the greatest impacts occur sometime in the distant future and
36 to people and places far away � can lead to discounting of risk and the need for adaptation (Leviston et al.,
37 2014; Mildenberger et al., 2019) (medium confidence). Communication directed at local and personal
38 framing of climate change impact and risk information is one option for addressing low salience (Bolsen et
39 al., 2019), particularly related to established risks such as SLR, flooding, and wildfires in North America
40 (Mildenberger et al., 2019). "Personalized" risk communications have had mixed results creating
41 behavioural change and policy support, and even caused resistance (Schoenefeld and McCauley, 2016).
42 Communication focused extensively on risks and dangers of climate change can produce fear or dread,
43 lessen agency and create fatalism that hinders action (Giddens, 2015; Mayer and Smith, 2019); it also can be
44 labelled alarmist (Leiserowitz, 2005). Detailed SLR flooding maps for the San Francisco Bay area did not
45 increase climate risk assessment but lessened personal risk perception of those with a strong belief in climate
46 change although policy preferences and support for adaptation did not change (Mildenberger et al., 2019).
47 Defining coherent groups based on variations in beliefs, risk perceptions, and policy preferences offers
48 opportunities for effectively engaging with segments of the population instead of using the same approach
49 for everyone (low confidence) (Maibach et al., 2011; Chryst et al., 2018). As an example, the US population
50 was segmented into a continuum ranging from the "Alarmed", the dominant group who were "Concerned",
51 then the Cautious, Disengaged, Doubtful, and least prevalent, the Dismissive (Chryst et al., 2018).

52

53

54 [START BOX 14.1 HERE]

55

Do Not Cite, Quote or Distribute  14-17       Total pages: 157
    FINAL DRAFT                       Chapter 14                      IPCC WGII Sixth Assessment Report

1 Box 14.1: Integrating Indigenous `Responsibility-Based Thinking' into Climate Change Adaptation

2   and Mitigation Strategies

3

4 Indigenous Peoples throughout North America have experienced five centuries of territorial expropriation,

5 loss of access to natural resources and in many cases, barriers to the use of their sacred sites (Gabbert, 2004;

6 Louis, 2007). The history of Indigenous struggles to preserve distinct cultural knowledges and assert

7 autonomy in the face of colonialism has shaped land-use patterns and relationships with traditional territories

8 (Alfred and Corntassel, 2005; Tuhiwai Smith, 2021) (Cross-Chapter Box INDIG, Chapter 18). Climate

9 change is now creating additional challenges for Indigenous Peoples. For example, increased water scarcity

10 due to higher temperatures and diminished precipitation have led to reduced crop yields for Maya farmers in

11 the Yucatan (Sioui, 2019). Thawing permafrost in subarctic Canada (Quinton et al., 2019) has interfered with

12 the land-based livelihoods of the Indigenous Dene Peoples (CCP6).

13

14 Recent climate-related changes represent cultural threats similar to the ones that occurred when European

15 settlement began in the Americas over 500 years ago (Whyte, 2016; Whyte, 2017). Thus, for Indigenous

16 Peoples, who often disproportionately bear the impacts of climate change, such changes are not novel, but

17 seen as `d閖� vu' (Whyte, 2016). Since livelihoods and subsistence are often directly dependent on the land

18 and water, Indigenous Peoples have direct insights into the localized impacts of global environmental

19 change. Indeed, Indigenous Peoples consider themselves stewards of the land (and water), and have a

20 spiritual duty to care for the land and its flora, fauna, and aquatic community, or `Circle,' of beings.

21 Indigenous knowledge (IK) has gained recognition for its potential to bolster western scientific research

22 about climate change. Many recent examples demonstrate the scientific value of IK for resource

23 management in climate change adaptation and mitigation (e.g. Kronik and Verner, 2010; Maldonado et al.,

24 2013; Wildcat, 2013; Etchart, 2017; Nursey-Bray et al., 2019). For example, Indigenous practices have not

25 only contributed to the present understanding of North American forest fires, but also that the practice of

26 frequent small-scale anthropogenic fires, also called cultural burns, is a key method to prevent large-scale

27 destructive fires (14.7.1). The growing interest and recognized value in these practices, particularly in

28 California, has led to formal agreements with state and federal agencies (Long et al., 2020a; Lake, 2021).

29

30 Indigenous relationships with the land are commonly informed and guided by a cultural ethic of

31 `responsibility-based thinking' (Sioui and McLeman, 2014). The Indigenous cultural ethic informs and

32 mediates personal and collective conduct with a sense of duty or responsibility toward human and other-

33 than-human relations (see Sioui, 2020). The Indigenous responsibility-based outlook stems from a cultural

34 paradigm that understands that it is human beings who must learn to live with the land (Cajete, 1999; Pierotti

35 and Wildcat, 2000; McGregor et al., 2010a; McGregor, 2014). This way of thinking instils in its adherents an

36 inherent awareness that the other-than-human realm is capable of existing and thriving without humans.

37 Thus, it is for our own sake (as humans) that we learn to live according to certain, ever-shifting, parameters,

38 requiring us to remain acutely attuned to our physical surroundings. This Indigenous cultural precept is

39 perhaps among the most significant contributions of Indigenous Peoples to the rest of humanity in the face of

40 climate change.

41

42 Indigenous relationships with natural systems continue to be mediated by cultural orders of governance and

43 legal systems that pre-date, by several millennia, European traditions in North America. Napolean (2012)

44 describes Indigenous legal orders as dynamic and encompassing knowledge that is simultaneously legal,

45 religious, philosophical, social, and scientific. Customary Indigenous legal orders (e.g. Borrows, 2002;

46 Napolean, 2012) stand in contrast to Eurocentric understandings of law, which are closely related to, and

47 founded on, the Western principles of rights. Indigenous legal orders are based on duties, obligations and

48 responsibilities to the land and all beings, including humans, animals, plants, future generations and the

49 departed/ancestors (Borrows, 2002; Borrows, 2010a; Borrows, 2010b; Borrows, 2016). Indigenous spiritual

50 laws centred on the values of responsibility and accountability to the land, and how these differ, in theory

51 and in practice, from Western law, which is based on "universal" principles, with little consideration for the

52 local environmental context (Craft, 2014). Research has elucidated these Indigenous understandings about

53 how their land-based responsibilities act as the foundation of how humans must operate according to the land

54 on which they live and depend.

55

56 With increasing climate change threats to land-based subsistence and cultural practices, Indigenous Peoples

57 are increasingly taking their rightful leadership roles in resource co-management arrangements and other

    Do Not Cite, Quote or Distribute  14-18                           Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 stewardship activities (14.5.2.2). Indeed, Indigenous Peoples are increasingly assuming leadership positions
 2 with regard to land governance and climate change action, as the stewards of their traditional territories since
 3 time immemorial. Therefore, it is imperative for Indigenous scholars, Elders, and knowledge holders to
 4 occupy leadership roles in climate change adaptation and mitigation, especially when their territories are
 5 concerned (14.7; CCP6). For instance, Indigenous "resurgence" paradigms draw on the strengths of
 6 traditional land-based culture and knowledge with regard to Indigenous leadership in land governance and
 7 stewardship (Alfred and Corntassel, 2005; Alfred, 2009; Simpson, 2011; Corntassel and Bryce, 2012;
 8 Coulthard, 2014; Alfred, 2015). Indigenous leadership in climate change policy, therefore, can ensure that
 9 Indigenous right to self-determination is respected and upheld to allow Indigenous Peoples to continue to
10 carry out their cultural responsibilities to the land, for the benefit of all North Americans (Powless, 2012;
11 Etchart, 2017).

12

13 In Northern Canada, a fusion of leading-edge western science and IK on permafrost informed the co-
14 development of predictive decision support tools and risk management strategies to inventory and manage
15 permafrost and adapt to permafrost thaw (CCP6). Permafrost thaw in the Dehcho region of Canada is
16 widespread and occurring at unprecedented rates (WGI). The Dehcho Collaborative on Permafrost (DCoP)
17 aims to improve the understanding of and ability to predict and adapt to permafrost thaw
18 (http://scottycreek.com/DCoP/). DCoP's collaborative approach, which places Indigenous Peoples in
19 leadership positions, generates the new knowledge, predictive capacity and decision-support tools to manage
20 natural resources that support Indigenous Dene Peoples' ways of life. Indigenous-academic partnerships can
21 enhance climate change adaptation and mitigation capacity and provide openings for more holistic co-
22 management approaches that recognize and affirm the central role of Indigenous Peoples as stewards of their
23 ancestral territories, especially as they face accelerating climate change impacts. Academic researchers and
24 their Indigenous partners can support climate change resilience via mobilizing IK in stewardship and
25 adaptation; researching governance arrangements, economic relationships and other factors that hinder
26 Indigenous efforts in these areas; proposing evidence-based policy solutions at international and national
27 scales; and outlining culturally relevant tools for assessing vulnerability and building capacity will also
28 support climate change resilience. IK underpins successful climate change adaptation and mitigation (very
29 high confidence) (see Green and Raygorodetsky, 2010; Kronik and Verner, 2010; Alexander et al., 2011;
30 Powless, 2012; Ford et al., 2016; Nakashima et al., 2018). The inclusion of IK in adaptation and mitigation
31 not only supports Indigenous cultural survival but also enables governments to recognize the territorial
32 sovereignty of Indigenous Peoples.

33

34 Responsibility-based philosophies of Indigenous Peoples from across the continent support the development
35 of climate change adaptation and mitigation strategies that promote responsible and respectful relationships
36 with the environment over the long term. Adapting to change, in all its forms, has since time immemorial
37 been one of the defining characteristics of Indigenous cultures on Turtle Island (the American continent). In
38 the Yucatan, one Elder explained that with regards to climate change impacts in the region, the Maya have
39 always dealt with "k'ech", or change, and that accepting and responding to change is part of the Maya
40 identity and responsibility (Sioui, 2020). Given successive failures in adequately and effectively responding
41 to climate change, it has become urgent for the rest of the human collective to (re)learn from Indigenous
42 cultures to (re)consider our responsibility/ies to the land--the world over--and to reorient our societal
43 imperatives to better respond and react to change. Such a process of learning from IK could foster the
44 development of climate change policies that promote responsible and respectful relationships with the
45 environment over the long term, and prove to be more effective and holistic. Although most inhabitants of
46 North America are non-Indigenous, it is possible and beneficial for our societies to learn to think and act in a
47 more responsibility-based way about our relations to the land, and, by extension, about climate change
48 policy. A collective commitment to protecting and advancing Indigenous territorial rights, so Indigenous
49 Peoples can continue to reassert their spiritual duty and role as stewards of their traditional territories,
50 benefits of all human and other-than-human `Peoples'.

51

52 [END BOX 14.1 HERE]

53

54

55 14.4 Indigenous Peoples and Climate Change

56

Do Not Cite, Quote or Distribute  14-19       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Indigenous knowledge and science are resources for understanding climate change impacts and
 2 adaptive strategies (very high confidence) (SM14.1, Table SM14.1). The Indigenous Peoples of North
 3 America have and continue to contribute substantially to the growing literature, scholarship, and research on
 4 climate change (Barreiro, 1999; Houser et al., 2001; Mustonen, 2005; Bennett et al., 2014; Maynard, 2014;
 5 Merculieff et al., 2017; FAQI, 2019; Ijaz, 2019; BIA, 2021). For thousands of years, Indigenous Peoples
 6 have developed and relied on their own knowledge systems for sustaining their health, cultures and arts,
 7 livelihoods, and political security (Battiste and Henderson, 2000; Colombi, 2012; Nelson and Shilling,
 8 2018). Diverse Indigenous knowledge systems in North America consider weather and climate as major
 9 dimensions of understanding the relationship between society and the environment. Indigenous Peoples have
10 distinct knowledge of climate change, over extensive temporal measures (Trosper, 2002; Barrera-Bassols
11 and Toledo, 2005; Gearheard et al., 2013). The basis of this knowledge is often Indigenous Peoples' long
12 and profound relationships to the environment, that is to the ecosystems, waters, ice, lands, territories, and
13 resources in their homelands. The relationships were forged by adaptation to a particular environment and
14 involve systematic activities. Indigenous harvesters, including hunters, fishers, agriculturalists, and plant
15 gatherers, observe and monitor environmental change, and engage in systematic reflection with one another
16 about trends over short term and long-term periods (Sakakibara, 2010; S醤chez-Cort閟 and Chavero, 2011;
17 Kermoal and Altamirano-Jim閚ez, 2016; Metcalfe et al., 2020b). The holistic perspective of the interrelated
18 and interdependent nature of ecosystems is a distinct characteristic of Indigenous knowledge and often
19 contrasts with findings and results of science alone. Indigenous harvesters, agriculturalists, leaders, culture-
20 bearers, educators, and government employees develop theoretical and practical knowledge of seasonal and
21 climate change that seeks to furnish the best available knowledge and information to inform climate change
22 policy and decisions (Barrera-Bassols and Toledo, 2005; McNeeley and Shulski, 2011). Examples of
23 theoretical knowledge systems include Indigenous calendars of seasonal change and systems of laws and
24 protocols for environmental stewardship (Kootenai Culture Committee, 2015; Donatuto et al., 2020) (Box
25 14.1).

26

27 The practice and use of Indigenous knowledge systems is recognized and affirmed by the United Nations
28 Declaration on the Rights of Indigenous Peoples (UNDRIP) (UNGA, 2007), and consistent with reports and
29 guidance from UN bodies including the High Commissioner for Human Rights (Bachelet, 2019), Expert
30 Mechanism on the Rights of Indigenous Peoples (UNGA, 2015; UNGA, 2018), the Permanent Forum of
31 Indigenous Issues (Dodson, 2007; Cunningham Kain et al., 2013; Sena and UNPFII, 2013; Sena, 2014;
32 Quispe and UNPFII, 2015), and the Special Rapporteur on the Rights of Indigenous Peoples (Toledo, 2013;
33 UNGA, 2017)(Cross-Chapter Box INDIG in Chapter 18). Rights to self-determination, to control over
34 territorial development, and cultural integrity, make it important that climate scientists practice equitable
35 engagement of Indigenous knowledge and Indigenous knowledge holders. There is a growing literature of
36 success and lessons learned from co-production of knowledge between Indigenous knowledge systems and
37 diverse scientific traditions relating to climate change (Behe et al., 2018; Latulippe and Klenk, 2020;
38 Camacho-Villa et al., 2021).

39

40 Current and projected climate change impacts disproportionately harm Indigenous Peoples'
41 livelihoods and economies (very high confidence). Indigenous Peoples' livelihoods in North America
42 include a range of activities closely tied to traditional lands, waters, and territories. These activities support a
43 core economic base and an array of sustenance, including financial stability, food security, health and
44 nutrition, safety, and adequate provisions and reserves of important supplies and resources and the passing
45 down of traditional knowledge. Indigenous lives and livelihoods are at risk in the following ways.
46 Indigenous persons are more at risk of losing their lives due to factors that are exacerbated by climate change
47 impacts (Ford et al., 2006; Barbaras, 2014; Khalafzai et al., 2019). Indigenous Peoples' livelihood practices
48 are being distressed, interrupted, and in some cases, made entirely inaccessible. Livelihood activities known
49 and anticipated to be impacted by climate change are food security (Meakin and Kurtvits, 2009; Wesche and
50 Chan, 2010; Nyland et al., 2017), harvesting of fish, plants, and wildlife (Dittmer, 2013; Parlee et al., 2014;
51 Jantarasami et al., 2018b; ICC Alaska, 2020), agriculture (St. Regis Mohawk Tribe, 2013; Shinbrot et al.,
52 2019; Settee, 2020), transportation (Swinomish Indian Tribe Community, 2010; Hori et al., 2018a; Hori et
53 al., 2018b), and tourism and recreation (ICC Canada, 2008). Indigenous Peoples have been active in
54 gathering to assess the impacts of climate change on their livelihoods, one example being the Bering Sea
55 Elders Advisory Group (Bering Sea Elders Advisory Group and Alaska Marine Conservation Council, 2011;
56 Bering Sea Elders Group, 2016).

57

Do Not Cite, Quote or Distribute  14-20       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Climate change impacts have harmful effects on Indigenous Peoples' public health, physical health,

 2 and mental health, including harmful effects connected to the cultural and community foundations of
 3 health (very high confidence). Health and climate change is a major issue for Indigenous Peoples (Ford,
 4 2012; Ford et al., 2014; Gamble et al., 2016; Jantarasami et al., 2018b; Middleton et al., 2020a; Donatuto et
 5 al., 2021)(14.5.6). Climate change impacts and risks affect Indigenous Peoples' health negatively in different
 6 ways. Indigenous health, as tied to nutrition and exercise, is threatened when local foods are less available
 7 and harvesting activities are less possible to practice (Norton-Smith et al., 2016b; Rosol et al., 2016;
 8 Gonzalez et al., 2018). Indigenous Peoples experience widespread public health concerns from severe
 9 droughts (Stewart et al., 2020; Schlinger et al., 2021; Wiecks et al., 2021), extreme heat (Doyle et al., 2013;
10 Campo Caap, 2018; Kloesel et al., 2018a; Meadow et al., 2018; ITK, 2019; Ute Mountain Ute Tribe and
11 Wood Environment Infrastructure Solutions Inc, 2019; Whyte et al., 2021), unpredictable precipitation
12 patterns (Chavarria and Gutzler, 2018; Tom et al., 2018; Tlingit and Haida, 2019; Schlinger et al., 2021),
13 flooding and coastal erosion (Jamestown S'klallam Tribe, 2016; Norton-Smith et al., 2016b; Puyallup Tribe
14 of Indians, 2016; Marks-Marino, 2019; Ristroph, 2019; Marks-Marino, 2020b; Schlinger et al., 2021),
15 wildfires and wildfire smoke (Edwin and M鰈ders, 2018; USEPA, 2018; Christianson et al., 2019a; ITK,
16 2019; Marks-Marino, 2020a; Mottershead et al., 2020; Woo et al., 2020; Wiecks et al., 2021), algal blooms
17 (Peacock et al., 2018; Gobler, 2020; Donatuto et al., 2021; Preece et al., 2021; Schlinger et al., 2021), storms
18 and hurricanes (Rioja-Rodr韌uez et al., 2018), influxes of invasive species (Pfeiffer and Huerta Ortiz, 2007;
19 Pfeiffer and Voeks, 2008; Voggesser et al., 2013; Bad River Band of Lake Superior Tribe of Chippewa
20 Indians and Abt Associates Inc., 2016; Scott et al., 2017; Reo and Ogden, 2018; Middleton et al., 2020a),
21 and changing production systems (Rioja-Rodr韌uez et al., 2018). Indigenous Peoples' mental health is at risk
22 and has already been affected negatively by climate change (Donatuto et al., 2021). Water security is one of
23 the most serious concerns to Indigenous Peoples' health and wellbeing (Vanderslice, 2011; Cozzetto et al.,
24 2013a; Redsteer et al., 2013; Hanrahan et al., 2014; Chief et al., 2016; Gamble et al., 2016; Jantarasami et
25 al., 2018b; Kloesel et al., 2018a; Tom et al., 2018; Martin et al., 2020a; Arsenault, 2021). When some people
26 are less able to practice traditional, cultural, social, and family activities, they can become alienated,
27 compounding the negative effects of traumas Indigenous persons already experience. Traumas include
28 historic and continuing land dispossession, assimilation, social marginalization and discrimination, and food
29 and financial insecurities. The practice of cultural traditions are associated with education, harvesting and
30 agriculture, exercise, positive social relationships, and family life, which play foundational roles in the
31 achievement of physical, public, and mental health (Bell et al., 2010; Cunsolo Willox et al., 2015;
32 Jantarasami et al., 2018b; Norgaard and Tripp, 2019; Billiot et al., 2020b; Adams et al., 2021; Donatuto et
33 al., 2021).

34

35 Indigenous Peoples are affected dramatically by climate-related disasters and other climate-related
36 extreme environmental events (very high confidence). Indigenous Peoples face numerous threats and have
37 already been harmed by and are planning for extreme weather events with associations to climate change,
38 including hurricanes and tornadoes (Oneida Nation Pre-Disaster Mitigation Plan Steering Committee and
39 Bay-Lake Regional Planning Commission, 2016; Emanuel, 2019; Cooley, 2021; Marks-Marino, 2021;
40 Zambrano et al., 2021), heat waves (Confederated Tribes of the Umatilla Indian Reservation, 2016; Wall,
41 2017; La Jolla Band of Luiseno Indians, 2019; Mashpee Wampanoag, 2019; Wiecks et al., 2021), ocean
42 warming and marine heat waves (Hoh Indian Tribe, 2016; Port Gamble S'klallam Tribe, 2016; Port Gamble
43 S'klallam Tribe, 2020; State of Alaska, 2020; Muckleshoot Tribal Council, 2021; Port Gamble S'klallam
44 Tribe, 2021), wildfires (Voggesser et al., 2013; Billiot et al., 2020a; Cozzetto et al., 2021b; Gaughen et al.,
45 2021; Morales et al., 2021; National Tribal Air Association, 2021; Zambrano et al., 2021), permafrost thaw
46 (Haynes et al., 2018; Low, 2020), flooding (Riley et al., 2011; Ballard and Thompson, 2013; Brubaker et al.,
47 2014; Thompson et al., 2014; Burkett et al., 2017; Quinault Indian Nation, 2017; Ristroph, 2019; Sharp,
48 2019; Thistlethwaite et al., 2020b), and drought (Knutson et al., 2007; Chief et al., 2016; Redsteer et al.,
49 2018; Sioui, 2019; Bamford et al., 2020; Sauchyn et al., 2020). Some Indigenous Peoples are facing climate
50 change impacts that generate community-led permanent relocation and resettlement as an adaptation option
51 (Maldonado et al., 2021). Coastal erosion is one climate change issue that is often connected to Indigenous
52 Peoples planning to resettle, including vulnerability connected to higher sea levels and storm surges
53 (Quinault Indian Nation, 2017; Bronen et al., 2018; Affiliated Tribes of Northwest Indians, 2020). Adapting
54 to new settlement areas threatens the continuity of communities. In a number of cases, Indigenous Peoples'
55 having less access to adequate infrastructure is a driver of vulnerability to climate related disasters and
56 extreme weather events (Doyle et al., 2018; Patrick, 2018; Cozzetto et al., 2021a; Indigenous Climate Action
57 et al., 2021). Disasters and extreme events are particularly severe when their impacts are compounded by

Do Not Cite, Quote or Distribute  14-21       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 inadequate infrastructure. Lack of flood protection infrastructure on Indigenous reserve communities, leads
 2 to displacement, loss of homes, and perpetuates disproportionate levels of risk to extreme weather events
 3 (Cunsolo et al., 2020; Fayazi et al., 2020; Yellow Old Woman-Munro et al., 2021).

 4

 5 Indigenous self-determination and self-governance are the foundations of adaptive strategies that
 6 improve understanding and research on climate change, develop actionable community plans and

 7 policies on climate change, and have demonstrable influence in improving the design and allocation of

 8 national, regional, and international programs relating to climate change (very high confidence).
 9 Historical and contemporary developments have crystallized international norms recognizing the distinct
10 status, role, and rights of Indigenous Peoples in the form of significant international human rights
11 instruments. Premier among them is the UNDRIP (UNGA A/RES/61/295), which has received universal
12 consensus since its adoption by the UN General Assembly. UN member States have affirmed the right of
13 self-determination (Article 3, UNDRIP) regarded as the prerequisite to the exercise and enjoyment of all
14 other human rights.

15

16 The integrity of the environment is impacting all of humanity, including Indigenous Peoples, their lands,
17 territories, resources and their communities. Through self-determination, durable, sustainable, and robust
18 contributions from those with close, symbiotic relationships with the environment can be revealed in favor of
19 all humanity. Indigenous Peoples of North America have been engaged in wide-ranging activities to address
20 climate change (Doolittle, 2010; Parker and Grossman, 2012; Abate and Kronk, 2013; STACCWG, 2021).
21 They include actions in the spheres of education (Donatuto et al., 2020; McClain, 2021; Morales et al.,
22 2021), development of Indigenous knowledge and science (Maldonado et al., 2016; AFN, 2020; Ferguson
23 and Weaselboy, 2020; Huntington et al., 2021a; Jones et al., 2021; Sawatzky et al., 2021), adaptation
24 planning and implementation (Angel et al., 2018a; Tribal Climate Adaptation Guidebook Writing Team et
25 al., 2018; Hepler and Kronk Warner, 2019; Tribal Adaptation Menu Team, 2019; Metcalfe et al., 2020b),
26 and political action and diplomacy (including treaty-based diplomacy) (Grossman, 2008; Kronk Warner and
27 Abate, 2013; Callison, 2015).

28

29

30 14.5 Observed Impacts, Projected Risks, and Adaptation by Sector

31

32 14.5.1 Terrestrial and Freshwater Ecosystems and Communities

33

34 14.5.1.1 Terrestrial Ecosystems: Observed Impacts and Projected Risks

35

36 Evidence continues to mount about the impacts of recent climate change on species and ecosystems
37 (Weiskopf et al., 2020) (Table 14.2) (very high confidence). Ranges and abundances of species continue to
38 shift in response to warming throughout North America (Cavanaugh et al., 2014; Molina-Mart韓ez et al.,
39 2016; Tape et al., 2016; Miller et al., 2017; Pecl et al., 2017; Zhang et al., 2018a) (Cross-Chapter Box
40 MOVING PLATE in Chapter 5) (very high confidence). Future climate change will continue to affect
41 species and ecosystems (IPBES, 2018) (high confidence), with differential responses related to species
42 characteristics and ecology (D'Orangeville et al., 2016; Weiskopf et al., 2019). Climate change is projected
43 to adversely affect the range, migration, and habitat of caribou, an important food and cultural resource in the
44 Arctic (Leblond et al., 2016; Masood et al., 2017; Barber et al., 2018b; Borish, Accepted) (CCP6).

45

46 Climate-induced shifts in the timing of biological events (phenology) continue to be a well-documented
47 ecological response (Vose et al., 2017; Lipton et al., 2018; Vose et al., 2018; Molnar et al., 2021) (Table
48 14.2) (very high confidence). Reduced snow season length may potentially lead to adverse camouflage
49 effects on animals that change coat colour (Mills et al., 2013; Mills et al., 2018). Human conflicts with bears
50 are expected to increase in response to shifts in hibernation patterns (Johnson et al., 2018) and food resources
51 (Wilder et al., 2017; Wilson et al., 2017).

52

53 Severe ecosystem consequences of warming and drying are well documented (very high confidence) (Table
54 14.2). Significant ecosystem changes are expected from projected climate change (high confidence), such as
55 in Mexican cloud forests (Helmer et al., 2019), North American rangelands (Polley et al., 2013; Reeves et
56 al., 2014), and montane forests (Stewart et al.; Wright et al., 2021). Permafrost thaw is projected to increase
57 in Alaska and Canada (DeBeer et al., 2016) (see also AR6, WG I, Chapter 12), accelerating carbon release

Do Not Cite, Quote or Distribute  14-22       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 (Schaefer, 2104) (CCP6, see also AR6, WG I, Chapter 5) and affecting hydrology. Predicting which species
 2 or ecosystems are vulnerable is challenging (Stephenson et al., 2019), although paleoecological data (e.g.,
 3 pollen, tree rings) provide context from past events to better understand current and future transformations
 4 (Nolan et al., 2018).

 5

 6 Climate change impacts on natural disturbances have affected ecosystems (very high confidence) (Table 14.2
 7 and Box 14.2), and these impacts will increase with future climate change (medium confidence). Facilitated
 8 by warm, dry conditions, "mega-disturbances" and synergies between disturbances that include wildfires,
 9 insect and disease outbreaks, and drought-induced tree mortality continue to affect large areas of North
10 America (Cohen et al., 2016; Young et al., 2017a; Hicke et al., 2020), overwhelming adaptive capacities of
11 species and degrading ecosystem services (Millar and Stephenson, 2015; Stewart et al., 2021). This era of
12 mega-disturbances is expected to become more widespread and severe in coming decades (Cook et al., 2015;
13 Seidl et al., 2017; Buotte et al., 2019), with potentially significant impacts on ecosystems (Allen et al., 2015;
14 Crausbay et al., 2017; Schwalm et al., 2017; Coop et al., 2020; Dove et al., 2020 Thompson et al. 2020,
15 Stewart et al. 2021). Effects include widespread tree mortality (Allen et al., 2015; Kane et al., 2017; van
16 Mantgem et al., 2018) and accelerated ecosystem transformation (Guiterman et al., 2018; Crausbay et al.,
17 2020; Munson et al., 2020) (medium confidence).

18

19 14.5.1.2 Freshwater Ecosystems: Observed Impacts and Projected Risks

20

21 Climate change, either directly (warming water) or indirectly (glacier and snow inputs), has affected
22 biogeochemical cycling and species composition in North American aquatic ecosystems (Moser et al., 2005;
23 Saros et al., 2010; Preston et al., 2016) (Table 14.2) (very high confidence), possibly amplifying other
24 human-caused stresses on these systems (Richter et al., 2016). Excess nutrients associated with high farm
25 animal density can be transported during intense rainfall events (expected to increase with climate change)
26 causing algal blooms, fish kills, and other detrimental ecological effects (Huisman et al., 2017; Coffey et al.,
27 2019).

28

29 Projected climate change will cause habitat loss, alter physical and biological processes, and decrease water
30 quality in freshwater ecosystems (Poesch et al., 2016; Crozier et al., 2019) (high confidence). Projected river
31 warming of 1-3篊 is expected to reduce thermal habitat for important salmon and trout species in the
32 northwestern US by 5�31% (Isaak et al., 2018) and in Mexico (Meza-Matty et al., 2021), and for multiple
33 fish species in Canada (Poesch et al., 2016). Cold-water streams at higher elevations will warm less and
34 therefore may become climate refugia (Isaak et al., 2016). Projected warming of mountain lake ecosystems
35 (Roberts et al., 2017b; Redmond, 2018) will affect ecosystem processes (Preston et al., 2016; Redmond,
36 2018; Moser et al., 2019). Loss of cold water inputs from retreating glaciers are expected to adversely affect
37 alpine stream ecosystems (Fell et al., 2017; Giersch et al., 2017). For anadromous fish species (e.g., Chinook
38 salmon), future warming will reduce habitat suitability from river headwaters to oceans (Crozier et al.,
39 2021).

40

41 Freshwater ecosystems across North America are increasingly at risk from extreme drought, compounded by
42 human demands for water (14.5.3) (Kovach et al., 2019). Implications for aquatic and riparian species can
43 vary, but it is widely agreed that these systems are highly sensitive to fluctuations in the hydrologic cycle,
44 which can increase competition by invasive species and compromise connectivity between potential cold-
45 water refugia (Melis et al., 2016; Poff, 2019).

46

47 14.5.1.3 Adaptation in Terrestrial and Freshwater Ecosystems

48

49 Adaptation efforts to assess vulnerability of species and ecosystems, predict adaptive capacity, and identify
50 conservation-oriented options have increased markedly across North America (e.g., Hagerman and Pelai,
51 2018; Keeley et al., 2018; Thurman et al., 2020; Peterson St-Laurent et al., 2021; Thompson et al., 2021).
52 Scenario-based planning, an approach for addressing uncertainty, continues to gain traction and is regularly
53 applied by the US National Park Service (Star et al., 2016). Nonetheless, barriers to implementation of
54 specific actions often exist (e.g., inflexible policies, lack of resources and stakeholder buy-in, political will),
55 hampering progress (Stein et al., 2013; Shi and Moser, 2021). Efforts to evaluate the efficacy of
56 implemented adaptation actions are also lacking (Prober et al., 2019), but some cases show progress. For
57 example, ongoing efforts are quantifying how variable water releases from the Colorado River's Glen

Do Not Cite, Quote or Distribute  14-23       Total pages: 157
    FINAL DRAFT                                          Chapter 14    IPCC WGII Sixth Assessment Report

1 Canyon Dam affect endangered fish species (Melis et al., 2016). Nature-based solutions (NbS) for adaptation

2 (Box 14.7) are increasingly evaluated, especially at larger scales.

3

4 Effective climate-informed ecosystem management requires a well-coordinated suite of adaptation efforts

5 (e.g., assessment, planning, funding, implementation, and evaluation) that is co-produced among
6 stakeholders, Indigenous Peoples, and across sectors (Millar and Stephenson, 2015; Dilling et al., 2019)

7 (high confidence). New applications of conventional strategies can be modified to achieve conservation goals
8 under climate change (USGCRP, 2019). For example, mechanical thinning and prescribed burning (to reduce

9 fuel loads and benefit ecosystems) could be used in combination with planting species better suited to new

10 conditions to build resilience in western US forests to longer and hotter drought conditions (Bradford and

11 Bell, 2017; Vernon et al., 2018). Protection of buffer areas, such as riparian strips in arid regions and boreal

12 ecosystems, reduces water temperature, builds resistance to invasive species, increases suitable habitat

13 (Johnson and Almlof, 2016), and facilitates protection of freshwater systems from runoff during and after

14 intense rain events (National Research Council, 2002).

15

16 Innovative approaches may facilitate species' responses to climate change, particularly when vulnerability is
17 exacerbated by habitat loss and fragmentation. Strategies include improved landscape connectivity for

18 species dispersal (Carroll et al., 2018; Littlefield et al., 2019; Lawler et al., 2020; Thomas, 2020) or assisted

19 migration (also called managed relocation) to climatically suitable locations (Schwartz et al., 2012;
20 Dobrowski et al., 2015). Examples include translocation of salmon in the Columbia River (Holsman et al.,

21 2012), genetic rescue (assisted gene flow increases genetic diversity to address local maladaptation) (Aitken

22 and Whitlock, 2013), and locating and conserving climate refugia, such as in alpine meadows of the Sierra
23 Nevada (Javeline et al., 2015; Morelli et al., 2016). Maintaining diverse spawning habitats and salmon runs

24 can increase resilience of salmonid populations to climate change (Schoen et al., 2017; Crozier et al., 2021).

25 Newer modelling approaches can facilitate the visualization of future management scenarios, per a recent

26 study of fires in the southwestern US (Loehman et al., 2018), in addition to technologies in genomics for

27 monitoring species and modifying adaptive traits (Phelps, 2019).

28

29 Adaptation actions have important limitations (Dow et al., 2013), particularly in the context of biodiversity

30 conservation goals. "Hard" limits include species extinctions and vegetation mortality events, despite

31 conservation action (i.e., besides significant emissions reductions to mitigate warming, few if any

32 interventions could have prevented these losses). In contrast, "soft" adaptation limits exist primarily as a
33 function of the social-ecological value systems of local communities and government entities that are
34 reflected as goals and objectives in their management plans for ecosystems and species across North

35 America. Soft limits are often mutable or can be removed altogether (Dow et al., 2013). In contrast, human

36 modifications of landscapes that change or irreparably damage can limit adaptation by reducing connectivity
37 and therefore range shifts (Parks and Abatzoglou, 2020).

38

39

40 Table 14.2: Examples of observed climate change impacts on terrestrial and freshwater ecosystems.

    Impact                                                             References

    local extinctions                                                  (Pomara et al., 2014; Wiens, 2016)

    greening and increased productivity of North American vegetation from (Smith et al., 2016b; Zhu et al., 2016;

    CO2 fertilization                                                  Huang et al., 2018).

    changes in phenology, including migration as well as mismatches between (Mayor et al., 2017; Zaifman et al.,

    species and with human visitation                                  2017; Breckheimer et al., 2020)

    vegetation conversions, including

            shifts to denser forests with smaller trees                (McIntyre et al., 2015)

            trees to savannas and grasslands                           (Bendixsen et al., 2015)

            woody plant encroachment into grasslands                   (Archer et al., 2017)

            changes in tundra plant phenology and abundance

    Do Not Cite, Quote or Distribute                     14-24                                        Total pages: 157
FINAL DRAFT                                   Chapter 14              IPCC WGII Sixth Assessment Report

expansion of boreal and subalpine forests into tundra, meadows        (Myers-Smith et al., 2019)
reduced or lack of recovery following severe fire
                                                                      (Juday et al., 2015; Lubetkin et al.,
                                                                      2017)

                                                                      (Coop et al., 2020; O'Connor et al.,
                                                                      2020), Box 14.2

warmer droughts reducing plant productivity and carbon sequestration  (Mekonnen et al., 2017; Gampe et al.,
                                                                      2021)

slowing ecosystem function recovery of vegetation to pre-disturbance  (Schwalm et al., 2017; Crausbay et al.,
conditions following droughts                                         2020)

warming streams and lakes and changes in seasonal flows that have     (O'Reilly et al., 2015; Lynch et al.,
affected freshwater fish distributions and populations                2016; Poesch et al., 2016; Roberts et
                                                                      al., 2017b; Isaak et al., 2018;
                                                                      Christianson et al., 2019b; Zhong et al.,
                                                                      2019)

upstream expansion of human-mediated invasive hybridization and       (Muhlfeld et al., 2014)
enhanced the risk of extinction of native salmonid species

declining wetlands in western North America important for bird        (Donnelly et al., 2020)
migrations

increases in harmful freshwater algal blooms                          Section 14.5.3

 1
 2

 3 [START BOX 14.2 HERE]

 4

 5 Box 14.2: Wildfire in North America

 6

 7 Recent Observations, Attribution to Climate Change, and Projections

 8

 9 Anthropogenic climate change has led to warmer and drier conditions (i.e., fire weather) that favour wildland
10 fires in North America (see AR6, WGI, Chapter 12; high confidence). In response, increased burned area in
11 recent decades in western North America has been facilitated by anthropogenic climate change (medium
12 confidence). Annual numbers of large wildland fires and area burned have risen in the last several decades in
13 the western US (USGCRP, 2017; USGCRP, 2018), and area burned has increased in Canada (the number of
14 large fires has declined slightly recently) (Gauthier et al., 2014; Natural Resources Canada, 2018; Hanes et
15 al., 2019). Attribution studies have reported that climate change increased burned area in Canada (1959-
16 1999) (Gillett et al., 2004) as well as the western US (1984�2015) (Abatzoglou and Williams, 2016) and
17 California (1972-2018) (Williams et al., 2019a). Decreased precipitation was the primary climate change
18 cause of increased burned area in the western US, with warming a secondary influence (Holden et al. 2018),
19 whereas warming (through aridity) was most important in a California study (Williams et al., 2019a). A drier
20 atmosphere (including reduced precipitation) has been linked to climate change through altered large-scale
21 atmospheric circulation, which then facilitated greater burned area in the western US (Zhang et al., 2019c).
22 Through anomalous warm and dry conditions, anthropogenic climate change contributed to the extreme fires
23 of 2016 (Kirchmeier-Young et al., 2019; Tan et al., 2019) in western Canada and the extreme fire season in
24 2015 in Alaska (Partain et al., 2017). These studies did not include human activities that influence fire-
25 climate relationships (Syphard et al., 2017).

26

27 Warming has led to longer fire seasons (Westerling, 2016) and drier fuels (Williams et al., 2019a). Warmer
28 and drier fire seasons in the western US during 1985-2017 have contributed to greater burned area of severe
29 fires (Parks and Abatzoglou, 2020). Simultaneity in fires increased during 1984-2015 (Podschwit and Cullen,
30 2020), challenging firefighting effectiveness and resource sharing. In Mexico, fires have been correlated with
31 dry conditions (Kent et al., 2017; Marin et al., 2018; Zuniga-Vasquez et al., 2019). Wildland fire activity in
32 the grasslands of the US Great Plains has increased during the last several decades (Donovan et al., 2017)
33 related to antecedent precipitation or aridity that affected fuel quantity (Littell et al., 2009).

Do Not Cite, Quote or Distribute              14-25                                            Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Climate change is projected to increase fire activity in many places in North America during the coming
 3 decades (see also AR6, WGI, Chapter 12) (Boulanger et al., 2014; Williams et al., 2016; Halofsky et al.,
 4 2020), via longer fire seasons (Wotton and Flannigan, 1993; USGCRP, 2017), long-term warming (Villarreal
 5 et al., 2019; Wahl et al., 2019), and increased lightning frequency in some areas of the US and Canada
 6 (Romps et al., 2014; Finney et al., 2018; Chen et al., 2021) (medium confidence). Unusually extensive and
 7 severe fires have occurred in the Arctic tundra during recent extremely warm and dry years, suggesting that
 8 continued warming may increase the probability of such fires in the future (Hu et al., 2015). In drier non-
 9 forest ecosystems in the western US, fires are limited by fuel availability and vegetation productivity;
10 warming will decrease productivity, leading to lower burned area (Littell et al., 2018).

11

12 Impacts on Natural Systems

13

14 Although fire is a natural process in many North American ecosystems, increases in burned area and severity
15 of wildland fires have had significant impacts on natural ecosystems (medium confidence). The length of
16 streams and rivers impacted by fire has increased in the US along with burned area (Ball et al. 2021). Mega-
17 fires can cause major changes in the structure and composition of ecosystems, particularly where human
18 alterations are significant (Stephens et al., 2014; Loehman et al., 2020). Unusually severe fires may have led
19 to the conversion of forest to grassland in the US Southwest (Haffey et al., 2018). Recent warming and
20 drying have limited post-fire tree seedling and shrub establishment, limiting ecosystem recovery (Davis et
21 al., 2019; O'Connor et al., 2020; Rodman et al., 2020). In boreal forests, soil carbon is being lost through
22 increasingly severe or frequent fires (Walker et al., 2019).

23

24 Projected future fire activity will continue to affect ecosystems and alter their structure and function (medium
25 confidence) (Coop et al., 2020; Loehman et al., 2020). Increased fire activity (Stevens-Rumann et al., 2018;
26 Stevens-Rumann and Morgan, 2019; Turner et al., 2019a; Cadieux et al., 2020), further warming and drying
27 that stresses tree seedlings, and model projections of stand-replacing fires at the forest-non-forest boundary
28 in the western US (Parks et al., 2019) have raised the possibility of shifts in species composition or
29 vegetation type (Halofsky et al., 2020). These projections suggest high variability in ecosystem responses
30 depending on interactions between vegetation type, moisture stress, disturbances regimes, and human
31 alterations (Hurteau et al., 2008; Kitzberger et al., 2017; Littell et al., 2018; Hurteau et al., 2019; Loehman et
32 al., 2020; O'Connor et al., 2020).

33

34 Impacts on Human Systems

35

36 Increased fire activity, partly attributable to anthropogenic climate change, has had direct and indirect effects
37 on mortality and morbidity, economic losses and costs, key infrastructure, cultural resources, and water
38 resources (medium confidence), although other factors, such as increasing populations in the wildland-urban
39 interface, also contributed. During 2000�2018, significant fire events claimed 315 lives in the US (NOAA,
40 2019); the economic impacts (capital, health, indirect losses from economic disruption) from the 2018
41 California fires were US$149 billion (Wang et al., 2021). Poor air quality from fires caused increased
42 respiratory distress (very high confidence); exposure extends long distances from the fire source (Section
43 14.5.6.3). In addition to public and private property damage and loss, fires have caused irretrievable losses
44 from archaeological and historical sites (Ryan et al., 2012). Post-fire conditions have created unanticipated
45 challenges for communities' water supply operations (Bladon et al., 2014; N醰ar, 2015; Martin, 2016) by
46 altering water quality and availability (Smith et al., 2011; Bladon et al., 2014; Robinne et al., 2020) or public
47 safety by increasing exposure to mass wasting events after extreme rainfall events (Cui et al., 2019; Kean et
48 al., 2019). California utilities have proactively shut down parts of their electricity grid to reduce risk of fire
49 during extreme weather, and substantial numbers of people will be increasingly vulnerable to this action in
50 the coming decades (Abatzoglou et al., 2020).

51

52 In the US, annual costs of federal wildland fire suppression have increased by a factor of 4 since 1985
53 (USGCRP, 2018) and were US$1.5-3B during 2016-202 (NIFC, 2021). Annual costs of fire protection in
54 Canada have risen 2�3 fold from 1970�2017, to CAD$1.0-1.4B during 2015-2017 (2017 dollars) (Natural
55 Resources Canada, 2021). In one of its worst fire seasons, British Columbia expended over CAD$500M in
56 2017 for fire suppression (Natural Resources Canada, 2018). The number of days of synchronous fire danger

Do Not Cite, Quote or Distribute  14-26       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 is expected to double in the western US by 2051-2080, thereby increasing demands on fire suppression
 2 resources (Abatzoglou et al., 2021).

 3

 4 The 2016 Fort McMurray fire ranks as the costliest natural disaster in Canada to date (CAD$3B in insured
 5 damages) (Mamuji and Rozdilsky, 2018; IBC, 2020). More than 88,000 people were evacuated; many were
 6 not aware of the high pre-existing fire risk and had limited warning to prepare and leave (McGee, 2019). The
 7 community subsequently required extensive social support and experienced mental health challenges
 8 (Government of Alberta, 2016; Cherry and Haynes, 2017; Mamuji and Rozdilsky, 2018; Brown et al., 2019a;
 9 McGee, 2019). Although a broad recovery plan was developed (Regional Municipality of Wood Buffalo,
10 2016), reconstruction and economic recovery has been slow (Mamuji and Rozdilsky, 2018).

11

12 Wildland fire was identified as a top climate change risk facing Canada (Council of Canadian Academies,
13 2019) and poses a challenge to communities and fire management (Coogan et al., 2019). Projected area
14 burned in Canada using RCP2.6 will increase annual fire suppression costs to CAD$1B by end of century
15 (60% increase relative to 1980-2009) and to CAD$1.4B using RCP8.5 (119% increase) (Hope et al., 2016).
16 In the US, cumulative costs of fire response through 2100 are projected to be US$23B (2015 dollars) per
17 year under RCP8.5 (EPA, 2017). Lower emissions scenarios reduce these future cumulative costs by
18 US$55M (EPA, 2017) to US$7-9B (2005 dollars) (Mills et al., 2015a). Fire increases from future warming
19 will reduce timber supply in eastern Canada (Gauthier et al., 2015; Chaste et al., 2019) and increase post-fire
20 sedimentation in watersheds of the western US (Sankey et al., 2017).

21

22 Adaptation

23

24 Wildland fire risks are not equitably distributed as they intersect with exposure and socioeconomic attributes
25 (e.g., age, income, ethnicity) to influence vulnerability and adaptive capacity (Wigtil et al., 2016; Davies et
26 al., 2018; Palaiologou et al., 2019) (medium confidence). Individuals in rural areas, low-income
27 neighbourhoods, and immigrant communities as well as renters in California had less capacity to prepare for
28 and recover from fire (Davies et al., 2018). In the US, 29 million people live in areas with significant
29 potential for wildfires and 12 million are socially vulnerable (Davies et al., 2018). In Canada, there are 117
30 million ha of wildland-human interface (14% of total land area), and 96% of populated places have some
31 wildland-urban interface within 5 km (Johnston and Flannigan, 2018).

32

33 There is growing recognition of the need to shift fire management and suppression activities to co-exist with
34 more fire on the landscape. This includes widespread use of prescribed fire across landscapes to increase
35 ecological and community-based resilience (Schoennagel et al., 2017; McWethy et al., 2019; Tymstra et al.,
36 2020) (high agreement, medium evidence). Otherwise, the unprecedented combination of increased human
37 exposure and size of recent megafires creates community risks that may exceed conventional operational and
38 forest management response capacity and budgets (Podur and Wotton, 2010; Wotton et al., 2017; Loehman
39 et al., 2020; Moreira et al., 2020; Parisien et al., 2020) particularly with ongoing population and
40 infrastructure expansion into the wildland-urban interface (Canadian Council of Forest Ministers, 2016;
41 Coogan et al., 2019).

42

43 Climate-informed post-fire ecosystem recovery measures (e.g., strategic seeding, planting, natural
44 regeneration), restoration of habitat connectivity, and managing for carbon sequestration (e.g., soil
45 conservation through erosion control, preservation of old growth forests, sustainable agro-forestry) are
46 critical to maximize long-term adaptation potential and reduces future risk through co-benefits with carbon
47 mitigation (Davis et al., 2019; Hurteau et al., 2019; Coop et al., 2020; Stewart et al., 2021). Innovation in and
48 scaling up the use of prescribed fire and thinning approaches are contributing to pre-and post-fire resilience
49 goals, including use of Indigenous Peoples burning practices that are receiving a new level of awareness
50 (Kolden, 2019; Marks-Block et al., 2019; Long et al., 2020b) (Box 14.1).

51

52 The tools FireSmart Canada (https://www.firesmartcanada.ca), Firewise USA (https://www.nfpa.org/) and
53 Think-Hazard Mexico (https://thinkhazard.org) were devised to reduce fire risks and create fire-resilient
54 communities. They provide design guidance at building, lot, subdivision and community scales, and instruct
55 citizens on creating defensible space (National Fire Protection Association, 2013; Firesmart Canada, 2018).
56 Implementation has been fragmented and variable as it depends on voluntary uptake by individuals, business
57 and communities across a range of adaptive capacities and fire-exposed landscapes (Smith et al., 2016a).

Do Not Cite, Quote or Distribute  14-27       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Many vulnerable groups do not have access to financial or physical resources to reduce fire risk (Collins and
 2 Bolin, 2009; Palaiologou et al., 2019).

 3

 4 Although innovative, holistic approaches to wildland fire management are becoming more common across
 5 North America, broader application is necessary to address the growing risks (medium confidence). A social-
 6 ecological perspective blends ecosystem complexity, scale and processes into land use planning along with
 7 community values, perception, and capacities as well as institutional arrangements (Smith et al., 2016a;
 8 Spies et al., 2018). A risk assessment perspective expands from short-term, reactive fire response to
 9 landscape-scale, long-term prevention, mitigation, and preparedness with community and practitioner
10 engagement (Coogan et al., 2019; Sherry et al., 2019; Johnston et al., 2020; Tymstra et al., 2020).

11

12 [END BOX 14.2 HERE]

13

14

15 14.5.2 Ocean and Coastal Social-Ecological Systems

16

17 14.5.2.1 Observed Impacts and Projected Risks of Climate Change

18

19 Warming of surface and subsurface ocean waters has been broadly observed across all North American
20 marine ecosystems from the polar Arctic to the subtropics of Mexico (virtually certain) (Hobday et al., 2016;
21 Jewett and Romanou, 2017; Pershing et al., 2018; Smale et al., 2019a). Higher ocean temperatures have
22 directly affected food-web structure (Gibert, 2019) and altered physiological rates, distribution, phenology,
23 and behaviour of marine species with cascading effects on food-web dynamics (very high confidence)
24 (Gattuso et al., 2015; Pinsky and Byler, 2015; Sydeman et al., 2015; Poloczanska et al., 2016; Fr鰈icher et
25 al., 2018; Le Bris et al., 2018; Free et al., 2019; Stevenson and Lauth, 2019; Barbeaux et al., 2020; Dahlke et
26 al., 2020). Pacific coastal waters from Mexico to Canada and US mid-Atlantic coastal waters have a high
27 proportion of species (>5% of all marine species) near their upper thermal limit, representing hotspots of risk
28 from marine heatwaves (medium confidence) (Smale et al., 2019a; Dahlke et al., 2020). Kelp, a macro-algae,
29 forms important habitat for other marine species, and its biomass has decreased 85�99% in the past 4�6
30 decades off Nova Scotia, Canada, replaced by invasive and turf algae; this is associated directly with
31 warming waters (Filbee-Dexter et al., 2016).

32

33 Climate change has induced phenological and spatial shifts in primary productivity with cascading impacts
34 on foodwebs (high confidence) (Siddon et al., 2013; Stortini et al., 2015; Sydeman et al., 2015; Stanley et al.,
35 2018). This includes widespread starvation events of fish, birds (e.g., tufted puffins in Bering Sea in
36 2016/2017 and Cassin's Auklets in British Columbia in 2014/2015) and marine mammals (gray whales along
37 both coasts of North America) (Sydeman et al., 2015; Duffy-Anderson et al., 2019; Jones et al., 2019b;
38 Cheung and Fr鰈icher, 2020; Piatt et al., 2020), which challenge protected species and fisheries management
39 (section 14.5.4) (Chasco et al., 2017; Wilson et al., 2018; Barbeaux et al., 2020; Free et al., 2020; Holsman et
40 al., 2020). Climate change has altered foraging behaviour and distribution of North Atlantic right whales and
41 their target copepod prey (Record et al., 2019) increasing entanglement rates in lobster and snow crab fishing
42 gear on the East coast of the US and Canada as lobster and crab distributions also shift due to changing water
43 temperatures (Meyer-Gutbrod et al., 2018; Davies and Brillant, 2019). Similarly, whale entanglements in
44 fishing gear along the Pacific coast has increased 20 fold (Hazen et al., 2018). Projected shifts in the North
45 Pacific Transition Zone (NPTZ) by up to 1000 km northward (by the end of the century under RCP8.5)
46 combined with changes in coastal upwelling (Polovina et al., 2011; Hazen et al., 2013; Rykaczewski et al.,
47 2015) could alter up to 35% of elephant seal and bluefin tuna foraging habitat (Robinson et al., 2009; Kappes
48 et al., 2010).

49

50 In North American Arctic marine systems, rapid warming is significant, with cascading impacts beyond
51 polar regions (CCP6), and presents limited opportunities (tourism, shipping, extractive) but high risks
52 (shipping, and fishing industries, Indigenous subsistence and cultural activities) (high confidence )(Gaines et
53 al., 2018; IPCC, 2019b; Samhouri et al., 2019; Free et al., 2020; Holsman et al., 2020) (see sections 14.5.4;
54 14.5.9; 14.5.11;CCP6). Both direct hazards and indirect food web alterations from sea ice loss have
55 imperilled seabirds, marine mammals, small boat operators, subsistence hunters and coastal communities
56 (Sigler et al., 2014; Allison and Bassett, 2015; Huntington et al., 2015; Hauser et al., 2018; Raymond-
57 Yakoubian and Daniel, 2018; Dezutter et al.) (CCP6). Increasingly favourable environmental conditions due

Do Not Cite, Quote or Distribute  14-28       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 to warming combined with shipping and other activities has raised the rate of invasive species movement
 2 into the Arctic (Mueter et al., 2011). Sea ice loss due to climate change is expected to accelerate over the
 3 next century (14.2, WG1 9.3.1).

 4

 5 Coral reefs in Gulf of Mexico and along the coasts of Florida and Yucatan Peninsula are facing increasing
 6 risk of bleaching and mortality from warming ocean waters interacting with non-climate stressors (very high
 7 confidence) (Cinner et al., 2016; Hughes et al., 2018; Sully et al., 2019; Williams et al., 2019b). Coral reefs
 8 are contracting in equatorial regions and expanding poleward (Lluch-Cota et al., 2010; Jones et al., 2019a).
 9 Loss of coral habitat leads to loss of ecosystem structure, fish habitat, and food for coastal communities and
10 impacts tourism opportunities (14.5.7) (Weijerman et al., 2015a; Weijerman et al., 2015b). Without
11 mitigation to keep surface temperatures below a 2.0篊 increase by the end of the century, up to 99% of coral
12 reefs will be lost. However, 95% of reefs will still be lost even if warming is kept below 1.5篊 (high
13 confidence) (Hoegh-Guldberg et al.; Hoegh-Guldberg et al., 2019a). In Florida, by 2100, an estimated
14 US$24�55B may be lost in recreational use and value derived by people knowing the reef exists and is
15 healthy (Lane et al., 2013; Hoegh-Guldberg et al., 2019b) as coral reefs decline (14.5.9).

16

17 SLR has led to flooding, erosion and damage to infrastructure along the western Gulf of Mexico, the
18 southeast US coasts, and the southern coast of the Gulf of St Lawrence (14.2) (Daigle, 2006; Lemmen et al.,
19 2016; Frederikse et al., 2020) (very high confidence). Mangroves, important nurseries for fish and climate
20 refugia for corals (Yates et al., 2014), are under threat from climate change along the east coast of Mexico
21 (Pedrozo Acu馻, 2012). SLR, storm surge and attendant erosion of coastlines and barrier habitats are
22 projected to have large impacts on coastal ecosystems, maritime industries (14.5.9), urban centres and cities
23 (14.5.5) along the Gulf of Mexico, Caribbean Sea, Southeast US, the southern Gulf of St Lawrence and the
24 Pacific Coast of Mexico (Box 14.4) (Semarnat, 2014; Sweet et al., 2017; Vousdoukas et al., 2020). Coastal
25 archaeological and historical sites are especially vulnerable to SLR (Anderson et al., 2017; Hestetune et al.,
26 2018; Hollesen et al., 2018).

27

28 Future seawater CO2 levels have been shown in laboratory studies to negatively impact Pacific and Atlantic
29 squid, bivalve, crab, and fish species (Pacific cod), and indirectly alter food-web dynamics (Kaplan et al.,
30 2013; Long et al., 2013b; Gledhill et al., 2015; Seung et al., 2015; Punt et al., 2016; Swiney et al., 2017;
31 Hurst et al., 2019; Wilson et al., 2020) (high confidence). Long-term exposure to CO2 reduced growth of
32 Atlantic halibut (Gr鋘s et al., 2014), whereas some cultured oysters (Fitzer et al., 2019) and key Alaskan
33 commercial fish species show tolerance for high CO2 waters (i.e., juvenile walleye pollock) (Hurst et al.,
34 2012). Ocean acidification has already caused shellfish growers in the US and Canada to modify hatchery
35 procedures and farming locations to protect the most vulnerable life-stages (Cross et al., 2016) and is
36 projected to increasingly impact shellfish resources in the central and NE Pacific and Atlantic coasts (Seung
37 et al., 2015; Punt et al., 2016) (Section 14.5.4).

38

39 Open ocean oxygen minimum zones (OMZ) are expanding in the North Atlantic, the North Pacific
40 California Current and tropical oceans due to warming waters, stratification, and changes in precipitation
41 (medium confidence) (Deutsch et al., 2015b; Breitburg et al., 2018; Claret et al., 2018; Ito et al., 2019) (WGI,
42 3.6.2). Hypoxic (extreme low oxygen) events along coasts, which are partially influenced by climate change,
43 have been documented for all three countries, with events more prevalent on the east coast and around the
44 Gulf of Mexico due to a regional oceanography dominated by rivers and estuaries carrying land-based
45 nutrients (Breitburg et al., 2018). Hypoxia has directly caused large mortality events for fish and crabs in US
46 estuaries in the Northwest Atlantic (Chesapeake Bay), Northeast Pacific (Puget Sound) and the Gulf of
47 Mexico (Froehlich et al., 2015; Rakocinski and Menke, 2016; Sato et al., 2016; Kolesar et al., 2017). OMZs
48 and hypoxic events are projected to increase over the next century and may limit where fish can move
49 (medium confidence) (Deutsch et al., 2015b; Stortini et al., 2015; Bianucci et al., 2016; Li et al., 2016).

50

51 Favourable conditions for harmful algal blooms (HABs) have expanded due to warming, more frequent
52 extreme weather events (Gobler et al., 2017; Pershing et al., 2018; Trainer et al., 2019) and increased
53 stratification, CO2 concentration, and nutrient inputs (Wells et al., 2015; Gobler et al., 2017; Griffith and
54 Gobler, 2019) (high confidence). Increased occurrence of HABs (McCabe et al., 2016; Yang et al., 2016;
55 Gobler et al., 2017; USGCRP, 2018) has induced ecological impacts and societal costs (see 14.5.4 for fishery
56 closures). During the 2013�2016 Pacific Marine Heat Wave (MHW; Box 14.3), a Pseudo-nitzschia diatom

Do Not Cite, Quote or Distribute  14-29       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 bloom off the US West Coast caused extensive closures of crab and razor clam fisheries (Trainer et al.,
 2 2019), with economic and socio-cultural impacts beyond those in the fisheries sector (Ritzman et al., 2018).

 3

 4 Beaching of massive Sargassum seaweed mats (Sargassum natans and S. fluitans) have been reported across
 5 the Caribbean and Gulf of Mexico from 2011-present day, affecting US and Mexico nearshore ecosystems,
 6 human health and the tourism industry (Franks et al., 2016; Resiere et al.; Wang et al., 2019). Costs of beach
 7 clean-up is high, with Texas spending over USD$2.9 million annually (Webster and Linton, 2013).
 8 Attribution of Sargassum blooms to climate change is still tenuous and complicated by multiple drivers and
 9 few observational data sources (Wang et al., 2019) (low confidence).

10
11

12 [START BOX 14.3 HERE]

13

14 Box 14.3: Marine Heatwaves

15

16 Marine heat waves (MHWs) are periods of discrete anomalously high (compared to 30-year history) sea
17 surface temperatures that persist for a minimum 5 days but up to several months (Hobday et al., 2016;
18 Fr鰈icher et al., 2018; Holbrook et al., 2019; Laufk鰐ter et al., 2020). There have been MHWs attributed to
19 climate change in every marine system of North America including large areas of the Northwest Atlantic
20 (2012), Caribbean Sea (2015), Bering Sea (2016-2018), and central through Northeast Pacific (2013-2016)
21 (NOAA, 2018; Holbrook et al., 2019; Smale et al.). MHW events have affected kelp forests (Arafeh-
22 Dalmau et al., 2019), corals (Eakin et al., 2018), seagrasses, bottom-dwelling organisms, marine birds
23 (Loredo et al., 2019; Smale et al., 2019a) mammals (Suryan et al., 2021), fish and shellfish and marine
24 dependent human communities (Huntington et al., 2020; Fisher et al., 2021; Suryan et al., 2021). Increased
25 sea temperatures directly increase metabolic demand and change productivity and behaviour of fish species
26 (Stock et al., 2017; Free et al., 2019) as well as inducing rapid redistribution of species poleward and to
27 deeper colder waters (Pecl et al., 2017; Rheuban et al., 2017; Crozier et al., 2019; Stevenson and Lauth,
28 2019; Yang et al., 2019; Barbeaux et al., 2020; Cheung and Fr鰈icher, 2020). In the Pacific, from the Baja
29 Peninsula to the Bering Sea, there is evidence of widespread shifts in coastal biota and multi-trophic level
30 starvation of seabirds and whales from combined metabolic demand and reduced prey quality associated
31 with protracted MHWs across multiple regions (CCP6)(Sydeman et al., 2015; Duffy-Anderson et al., 2019;
32 Sanford et al., 2019; Smale et al., 2019a) (Suryan et al. 2021). The distribution of two economically
33 important North American species, Bering Sea Pacific cod (Pinsky et al., 2013b; Stevenson and Lauth, 2019;
34 Barbeaux et al., 2020; Spies et al., 2020) and American Lobster (Rheuban et al., 2017), have shifted north.
35 MHW-induced loss of coral reefs across tropical North American waters has varied in severity regionally.
36 For instance, in 2015 and 2016, extensive, severe bleaching affected more than 30% of corals off the
37 southeast US and a large proportion of US Hawaiian Islands, but had moderate to no impact off the Mexican
38 Yucatan Peninsula (Frieler et al., 2013; Weijerman et al., 2015a; Weijerman et al., 2015b; Cinner et al.,
39 2016; van Hooidonk et al., 2016; Hughes et al., 2018; Sully et al., 2019; Williams et al., 2019b). Some reefs
40 are exhibiting recovery following efforts focused at reducing non-climate stressors (e.g. overfishing, nutrient
41 pollution and tourism use). MHWs are increasing in intensity and frequency(Hobday et al., 2016; Smale et
42 al., 2019a) with largest increases in frequency and spatial coverage projected for the Gulf of Mexico, US
43 southern East Coast and US Pacific Northwest (Ranasinghe et al., 2021) and pose a key risk to marine
44 systems in North America (14.5.2, Ch 3, 16,).

45

46 [END BOX 14.3 HERE]

47

48

49 14.5.2.2 Adaptation: Current State, Barriers and Opportunities

50

51 Emerging technologies and cooperative marine management are approaches to facilitate adaptation but
52 require coordination and investment for implementation.(Gattuso et al., 2018; Miller et al., 2018; Holsman et
53 al., 2019; Karp et al., 2019) (high confidence). Advancements in oceanographic and ecological nowcasting
54 and forecasting tools (i.e., O2, pH, temperature, aragonite saturation state, sea ice conditions) can reduce
55 climate impacts by supporting fisheries and aquaculture adaptation along US coasts (Section 14.5.4) (Cooley
56 et al., 2015; Irby et al., 2015; Siedlecki et al., 2015; Siedlecki et al., 2016; Siddon and Zador, 2017).

Do Not Cite, Quote or Distribute  14-30       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Forecasts and warnings reduce human exposure to HAB toxins in the Great Lakes, the west coast of Florida,
 2 east coast of Texas and the Gulf of Maine (Anderson et al., 2019).

 3

 4 Ocean management that utilizes a portfolio of nested, multi-scale, climate-informed and ecosystem-based
 5 management approaches in North American waters can increase the resilience of marine ecosystems by
 6 addressing multiple stressors simultaneously (Marshall et al., 2018; Holsman et al., 2019; Smale et al.,
 7 2019a; Holsman et al., 2020) (high confidence). Integrated Ecosystem Assessments (Foley et al., 2013;
 8 Levin et al., 2014) are increasingly used to provide strategic advice and context for harvest allocations and
 9 bycatch avoidance (Zador et al., 2017) and early warnings of ecosystem-wide change (e.g., sentinel species,
10 ecological indicators) (Cavole et al., 2016; Hazen et al., 2019; Moore and Kuletz, 2019). Dynamic ocean
11 management policies may improve resilience of marine species and ecosystems to climate (Hyrenbach et al.,
12 2000; Maxwell et al., 2015; Dunn et al., 2016; Tommasi et al., 2017a; Tommasi et al., 2017b; Hazen et al.,
13 2018; Wilson et al., 2018; Holsman et al., 2019; Karp et al., 2019) (medium confidence). New proactive and
14 rapid management approaches have been developed to minimize impacts of increasingly frequent
15 entanglements of protected species, caused by climate-driven changes in prey and fishery activities
16 (Corkeron et al., 2018; Meyer-Gutbrod et al., 2018). Dynamic closure areas are being used to address these
17 issues and reduce loggerhead turtle bycatch in Hawaiian shallow-set longline fisheries (Howell et al., 2015;
18 Lewison et al., 2015), blue whale ship-strike risk in near-real time (Hazen et al., 2017; Abrahms et al.,
19 2019b), and bycatch of multiple top predator species in a West Coast drift gillnet fishery (Hazen et al.,
20 2018).

21

22 Improved coordination and planning at multiple scales will be important for marine species conservation and
23 recovery as species redistribute across fishery areas, marine protected zones, and international and
24 jurisdictional boundaries (Cross-Chapter Box MOVING PLATE in Chapter 5) (Pinsky et al., 2018; Karp et
25 al., 2019) (Section 14.5.4). Indigenous Peoples' co-management with federal and state partners of marine
26 resources and protected species is an important approach (see Section 14.5.4, Chapter 5, Chapter 6, and
27 CCP6) (Galappaththi et al., 2019).

28

29 Securing broodstocks for rebuilding and supplementation can be challenging for marine populations already
30 in decline (e.g., blue king crab in Alaska, steelhead salmon in Puget Sound, white abalone in California,
31 most groundfish in Northeast US and Canada) (14.5.4; Table SM14.8). Marine protected areas can attenuate
32 climate impacts through trophic redundancy, preserving ecological processes, biodiversity, and climate
33 refugia (Roberts et al., 2017a; Schoen et al.), although benefits decrease after mid-century (or sooner for high
34 latitude marine protected areas) as species reach their thermal limit, unless coupled with greenhouse gas
35 (GHG) mitigation (Bruno et al., 2018). Transport, relocation and cultivation of resistant breeds of salmon,
36 oysters, corals, marine mammals, and other keystone species as well as hatchery supplementation of
37 impaired populations of fish and shellfish are species conservation and recovery methods that will be in
38 greater demand under climate change, although unintended environmental impacts must be considered.
39 Options for protecting and restoring coral reefs to prevent loss of ecosystem function are under development
40 with Florida reef species (Gattuso et al., 2018; National Academies of Sciences, 2019). An emerging
41 approach for financing the protection of reefs involves re-categorizing reefs as "natural infrastructure" which
42 has allowed for use of insurance to rebuild lost reefs (Storlazzi et al., 2019).

43

44 14.5.3 Water Resources

45

46 Climate change poses increasing threats to North American aquatic ecology, water quality, water availability
47 for human uses, and flood exposure, through reductions in snow and ice, increases in extreme precipitation,
48 and hotter droughts. Adaptation will be impeded in cases where there are conflicts over competing interests
49 or unintended consequences of uncoordinated efforts, heightening the importance of cooperative, scenario-
50 based water resource planning and governance (high confidence).

51

52 14.5.3.1 Observed Impacts

53

54 North American water resources continue to be affected by ongoing warming, with impacts driven by
55 reductions in snow and ice, increases in extreme precipitation, and hotter droughts (Section 14.2) (Fleming
56 and Dahlke, 2014; Mortsch et al., 2015; Dudley et al., 2017; Fyfe et al., 2017; McCabe et al., 2017;
57 Chavarria and Gutzler, 2018; Lall et al., 2018; Bonsal et al., 2019; USGCRP, 2019) (high confidence). The

Do Not Cite, Quote or Distribute  14-31       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 cascading effects of severe droughts, floods, sediment mobilization, harmful algal blooms (HABs) and
 2 pathogen contamination episodes have revealed the vulnerability and exposure of large numbers of people
 3 and economic activities to those hazards.

 4

 5 North America's dams, levees, wastewater-management and water conveyance facilities have improved
 6 water supply safety and have reduced flood and drought risks, but a substantial portion of that infrastructure
 7 is aging and inadequate for modern conditions (Ho et al., 2017; Tellman et al., 2018; Carlisle et al., 2019;
 8 FEMA, 2019; ASCE, 2021). Increasingly heavy precipitation from a variety of storm types has affected parts
 9 of North America (Feng et al., 2016; Prein et al., 2017a; Kunkel and Champion, 2019; Kunkel et al., 2020),
10 contributing to contamination from combined sewer overflows (Olds et al., 2018) and increased flood
11 damages that are partially attributed to anthropogenic climate change (van der Wiel et al.; Davenport, 2021).
12 Extreme precipitation events have overwhelmed water control infrastructure, imperilling public safety and
13 contributing to extensive damages in parts of North America(Kytomaa et al., 2019; Vano et al., 2019; White
14 et al., 2019). Damages stem from extremity of the event and prior land use and infrastructure decisions (high
15 confidence).

16

17 In South Carolina, five days of heavy rainfall in October 2015 caused the failure of more than 50 dams and
18 some levees, significantly magnifying destruction from the floodwaters (FEMA, 2016). Slow-moving,
19 destructive storms like hurricanes Harvey (2017) and Florence (2018) have caused significant flooding (van
20 Oldenborgh et al., 2017; Paul et al., 2019b). In those cases, urban sprawl may have altered storm dynamics
21 (Zhang et al.), while increased asset exposure to the flood hazard amplified the multi-billion dollar losses
22 (Klotzbach et al., 2018; Trenberth et al., 2018). A substantial fraction of the damage from hurricane
23 Harvey's extreme rainfall has been attributed to anthropogenic climate change (Emanuel, 2017; Risser and
24 Wehner, 2017) (Box 14.5). A near-disaster at California's Oroville dam in 2017 was caused by inadequate
25 infrastructure design and maintenance together with an unusually large number of atmospheric river (AR)
26 storms. The event required emergency reservoir spills while the state was beginning recovery from the
27 extreme 2012�2016 drought (Vano et al., 2019; White et al., 2019).

28

29 In Mexico, some poor neighbourhoods and informal settlements are located in areas exposed to recurrent
30 flooding. Residents often lack access to public services and technical resources for risk reduction, which
31 heightens their vulnerability (Castro and De Robles, 2019).

32

33 Population growth and urban development have increased the exposure and vulnerability of Canadian
34 communities to flood damages, with cumulative damages (including uninsured losses) exceeding US $10B
35 in the past decade (The Geneva Association et al., 2020). Recurring floods are particularly costly (e.g., New
36 Brunswick (Beltaos and Burrell, 2015; Kovachis et al., 2017)). Floods in High River, AB (2013) and
37 Gatineau, QC (2017, 2019) initiated considerations of building flood resilience including planned retreat
38 (Saunders-Hastings et al., 2020).

39

40 Extended and severe droughts in the western US, northern Mexico and Canadian Prairies, exacerbated by
41 higher temperatures, have caused economic and environmental damage (Williams et al., 2013;
42 AghaKouchak et al., 2015; Diaz et al., 2016; Bain and Acker, 2018; Lopez-Perez et al., 2018; Ortega-Gaucin
43 et al., 2018; Xiao et al., 2018; Martinez-Austria et al., 2019; Bonsal et al., 2020; Martin et al., 2020b; Milly
44 and Dunne, 2020; Overpeck and Udall, 2020). Droughts have intensified tensions among competing water
45 use interests and accelerated depletion of groundwater resources (14.5.4) (Pauloo et al., 2020) (high
46 confidence).

47

48 Climate trends are affecting riverine, lake and reservoir water quality (medium confidence). Droughts and
49 increased evapotranspiration have impaired water quality by concentrating pollutants in diminished water
50 volumes (Paul et al., 2019a). Cyanobacterial blooms and pathogen exposure events are increasing in
51 frequency, intensity, and duration in North America (Taranu et al., 2015). They are closely associated with
52 observed changes in precipitation intensity and associated nutrient loading (e.g., agricultural runoff, sanitary
53 sewer overflows), elevated water temperatures and eutrophication (Michalak et al., 2013; Michalak, 2016;
54 Trtanj et al., 2016; Chapra et al., 2017; IBWC, 2017; Williamson et al., 2017; Olds et al., 2018; Coffey et al.,
55 2019). These events endanger human and animal health, recreational and drinking water uses, aquatic
56 ecosystem functioning, and cause economic losses (Michalak et al., 2013; Bullerjahn et al., 2016; Chapra et
57 al., 2017; Huisman et al., 2018). Households and communities dependent on substandard wells, unimproved

Do Not Cite, Quote or Distribute  14-32       Total pages: 157
FINAL DRAFT                                         Chapter 14               IPCC WGII Sixth Assessment Report

 1 water sources, or deficient water provision systems are more likely than others to experience climate-related
 2 impairment of drinking water quality (14.5.6.5) (Allaire et al.; Baeza et al., 2018; California State Water
 3 Resources Control Board, 2021; Navarro-Espinoza et al., 2021; Water and Tribes Initiative, 2021).

 4

 5 14.5.3.2 Projected Impacts and Risks

 6

 7 Climate change is projected to amplify current trends in water resource impacts, potentially reducing water
 8 supply security, impairing water quality, and increasing flood hazards to varying degrees across North
 9 America (high confidence). Examples are presented in Table 14.3.

10

11

12 Table 14.3: Selected Projected Water Resource Impacts in North America.

Climatic Drivers and        Examples of Future                Location (see Figure 14.1) References
Processes                   Risks/Impacts

Warming-induced             Projected decreases in annual     US-NW, US-SW   (Jost et al., 2012;
reductions in mountain      and late-summer streamflow        CA-BC, CA-PR   Solander et al., 2018;
snow and glacial mass       from high-elevation reaches                      Bonsal et al., 2019; Milly
                            of snow-fed rivers, affecting                    and Dunne, 2020)
                            stream ecology and water
                            supplies, (high confidence)

Earlier seasonal snowmelt   Greater winter/early spring       US-NW, US-SW   (Cohen et al., 2015;
runoff                                                        CA-BC, CA-PR,  Dettinger et al., 2015;
                            flooding risks and reduced                       Bonsal et al., 2019;
                                                                             Bonsal et al., 2020;
                            summer surface water                             RMJOC, 2020; Bureau of
                                                                             Reclamation, 2021d)
                            availability, intensifying

                            seasonal mismatch with water
                            demands (high confidence),

                            Increased challenges for

                            balancing multi-purpose

                            reservoir objectives (e.g.

                            flood-management, water

                            supply, ecological protection
                            and hydropower) (high
                            confidence)

Earlier seasonal snowmelt   Possible reductions in water      US-SW          (Medellin-Azuara et al.,
runoff                      supply security (medium                          2015; Ullrich et al.,
                            confidence); Reduced                             2018; Bai et al., 2019;
                                                                             Milly and Dunne, 2020;
                            viability of some small-scale                    Ray et al., 2020;
                            irrigation systems (medium                       Bureau of Reclamation,
                            confidence)                                      2021b; Bureau of
                                                                             Reclamation, 2021a;
                                                                             Bureau of Reclamation,
                                                                             2021c)

Changes in seasonal         Impacts on electric power         US-SW, US-NW   (Haguma et al., 2014;
timing and/or total annual  generation (medium                CA-QC          Bartos and Chester, 2015;
runoff                      confidence) varying by                           Guay et al., 2015; Turner
                                                                             et al., 2019b; RMJOC,
                            location and type of                             2020; Bureau of
                                                                             Reclamation, 2021d)
                            generation

Changes in seasonal         Impacts on urban water            CA-QC          (Foulon and Rousseau,
                                                                             2019)
timing and/or total annual supplies
                                                                                          Total pages: 157
runoff

Do Not Cite, Quote or Distribute                       14-33
FINAL DRAFT                       Chapter 14                  IPCC WGII Sixth Assessment Report

Warming-related               Greater pressures on            US-SW, US-SP, US-SE,     (Bauer et al., 2015;
increased imbalance                                           MX-N, MX-NW              Molina-Navarro et al.,
between renewable             groundwater resources,                                   2016; Russo and Lall,
surface water supplies and                                                             2017; Brown et al.;
consumptive water             possible increased aquifer                               Nielsen-Gammon et al.,
demands                                                                                2020; Bureau of
                              depletion, reduced baseflow                              Reclamation, 2021b)

                              into surface streams and

                              reduced long-term water
                              supply sustainability (medium
                              confidence)

Warming-related drought       Reduced water availability for  Widespread especially:   (Prein et al., 2016; Dibike
amplification                                                 US-SW, US-NP, US-SP      et al., 2017; Lall et al.,
                              human uses and ecological       CA-PR                    2018; Paredes-Tavares et
                              functioning (medium to high     MX-NW, MX-N              al., 2018; Martinez-
                              confidence) varying by                                   Austria et al., 2019; Tam
                                                                                       et al., 2019; Martin et al.,
                              location; increased                                      2020b; Milly and Dunne,
                                                                                       2020; Overpeck and
                              evaporative losses from                                  Udall, 2020; Williams et
                                                                                       al., 2020; Bureau of
                              reservoirs.                                              Reclamation, 2021b)

Heavier and/or prolonged      Flooding, infrastructure and    Widespread; especially:  (Feng et al., 2016;
rainfall events               property damage (medium to      US-SE, US-NE, US-NP,     Emanuel, 2017; Prein et
                              high confidence) varying by     US-SP, US-SW             al., 2017a; Prein et al.,
                              location; increased erosion     CA-BC                    2017b; Haer et al., 2018;
                              and debris flows with impacts   MX-CE; MX-NE; MX-        Kossin, 2018; Mahoney et
                              on public safety, reservoir     SE                       al., 2018; Thistlethwaite
                              sedimentation and stream                                 et al., 2018; Curry et al.,
                              ecology -- hazards amplified                             2019; Larrauri and Lall,
                              in watersheds affected by                                2019; Wobus et al., 2019;
                              wildfires.                                               Ball et al., 2021)

Heavier and/or prolonged      Water quality impairment,       US-MW, US-NE, US-SE,     (Alam et al., 2017;
rainfall events                                               US-NP, US-SP             Chapra et al., 2017; Sinha
                              increasing HAB events due to    CA-ON, CA-AT, MX-        et al., 2017; Ballard et al.,
                                                              NE, MX-NW                2019)
                              increased sediment and

                              nutrient loading together with

                              warming. Greatest impacts in

                              humid areas with extensive
                              agriculture (medium
                              confidence to high
                              confidence) varying by

                              location.

Increasingly variable         Highly variable precipitation   US-SW, US-NW, CA-BC (Gershunov et al., 2019;
precipitation,                poses challenges for water                                        Huang et al., 2020)
                              management, worsening
                              water supply and flooding
                              risks. Atmospheric River
                              (AR) events are projected to
                              increase variability by
                              dominating future North
                              American west coast
                              precipitation (medium
                              confidence)

        Hotter summer season  Evaporative losses from         US-SW, US-NW, US-NP (Bureau of Reclamation,
                                                                                                2021b)
1                             reservoirs are projected to
                              increase significantly (very
                              high confidence)

Do Not Cite, Quote or Distribute  14-34                                                Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Projected long-term reduction in water availability in the US Southwest and northern Mexico, e.g., from the
 3 Colorado and Rio Grande Rivers, will have substantial ecological and economic impacts given the region's
 4 heavy water demands (high confidence) (Lall et al., 2018; Paredes-Tavares et al., 2018; Martinez-Austria et
 5 al., 2019; Milly and Dunne, 2020; Williams et al., 2020). Increased water scarcity will intensify the need to
 6 address competing interests across state and national boundaries, including honouring commitments to
 7 Indigenous Peoples who have long struggled with inadequate access to their water entitlements and
 8 marginalization in water resource planning (Mumme, 1999; Cozzetto et al., 2013b; Mumme, 2016;
 9 McNeeley, 2017; Radonic, 2017; Robison et al., 2018; Curley, 2019; Water and Tribes Initiative, 2020;
10 Wilder et al., 2020).

11

12 Increased scarcity of renewable water relative to legally allocated or desired uses may develop in many parts
13 of North America. A detailed analysis of projected water demands (consumptive uses) and availability found
14 increasingly frequent shortages in several watersheds across the United States (Brown et al., 2019b). This
15 might lead to maladaptive increased groundwater mining, or alternatively to policies promoting sustainable
16 balancing of water consumption with renewable supplies, for example by facilitating voluntary water
17 transfers or improving enforcement of groundwater rights (Colorado River Basin Stakeholders, 2015;
18 California Natural Resources Agency et al., 2020; Colorado Water Conservation Board, 2020; Pauloo et al.,
19 2020).

20

21 Climate change is projected to reduce groundwater recharge in major US Southwest aquifers (e.g., Southern
22 High Plains, San Pedro and Wasatch Front), exacerbating their ongoing depletion due to unsustainable
23 pumping. Other aquifers, especially those farther north, face uncertain or possibly increasing recharge
24 (Meixner et al., 2016) (medium confidence).

25

26 Projected changes in temperature and precipitation present direct risks to North American water quality,
27 varying with regional and watershed contexts (Chapra et al., 2017; Coffey et al., 2019; Paul et al., 2019a),
28 and related to streamflow, population growth (Duran-Encalada et al., 2017) and land use practices (Mehdi et
29 al., 2015) (medium confidence). HABs increase in frequency across the US (Wells et al., 2015) with the
30 highest risk projected for the Great Plains and Northeast US, and greatest economic impacts from lost
31 recreation value in Southeast US (Chapra et al., 2017).

32

33 The diversity of climate regimes across North America results in regional differences in water-related
34 climate change risks (Figure 14.4).

35

36

Do Not Cite, Quote or Distribute  14-35       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Figure 14.4: Freshwater resource risks as a function of global mean surface temperature increase relative to
 3 preindustrial (1850-1900). Estimated sensitivities are based on references cited in Table 14.3, see SM14.4.

 4

 5

 6 14.5.3.3 Adaptation

 7

 8 North American water planners and policy makers have abandoned stationarity assumptions (Milly et al.,
 9 2015) to address climate change. Transboundary institutions, government agencies, and professional
10 organizations are taking the lead on adaptation planning and implementation. (ASCE, 2018b; Clamen and
11 Macfarlane, 2018; International Joint Commission (IJC), 2018). Major water agencies are using climate
12 scenarios to identify vulnerabilities and evaluate adaptation options (Yates et al., 2015; Vogel et al., 2016;
13 California Department of Water Resources, 2019; Ray et al., 2020; Bureau of Reclamation, 2021d).
14 The Water Utility Climate Alliance advises municipal water providers, to address uncertainty by considering
15 a wide range of plausible future climate conditions (WUCA, 2010). In some areas, the impacts of wildfires
16 on water supply resiliency are being considered (Martin, 2016). Many North American Indigenous Peoples
17 are engaged in climate change adaptation planning although these efforts may be hampered by the
18 complicated legal and administrative setting in which they must operate (Norton-Smith et al., 2016a;
19 McNeeley, 2017).

20

21 Recent climate extremes have heightened governmental attention to climate change impacts (e.g., (California
22 Natural Resources Agency et al., 2020). Droughts have exposed shortcomings in water management and
23 governance (Gray et al., 2015; Xiao et al., 2017b; Lopez-Perez et al., 2018) spurring legislation and
24 administrative changes to improve groundwater regulation and documentation of water rights (California
25 Department of Food and Agriculture, 2017; Miller, 2017; Lund et al., 2018; Hanak et al., 2019).Water
26 allocation policies are being reassessed to enhance equity, sustainability and flexibility through shortage
27 sharing agreements, improved groundwater regulation and voluntary water transfers. Developments include
28 an interstate drought management agreement for the Colorado River (US Law, 2019), and agreements
29 between the US and Mexico to provide pulse flows to benefit the ecology of the Colorado River Delta (Pitt
30 and Kendy, 2017). State-wide water planning in Colorado has emphasized building drought resilience (e.g.
31 by facilitating temporary water transfers) (Colorado State Government, 2015; Yates et al., 2015). At local
32 scales there have been innovations in cooperative watershed protection and water resource planning (Cant�,
33 2016). Indigenous Peoples are playing an increasing role in identifying equitable and resilient options for
34 adaptation by contributing their knowledge and voicing their perspectives on the importance of healthy water
35 bodies for human and environmental well-being (Norton-Smith et al., 2016a; Water and Tribes Initiative,

Do Not Cite, Quote or Distribute  14-36       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 2020). Collaboration between stakeholders, policymakers and scientists is increasingly common in water
 2 resources adaptation planning and assessment.

 3

 4 Examples of adaptation include increasing adoption of water-saving irrigation methods in California
 5 (Cooley, 2016), experimentation with using flood waters to enhance groundwater recharge (Kocis and
 6 Dahlke, 2017; California Department of Water Resources, 2018), and agricultural land management
 7 programs, including developing riparian buffers to protect water quality (14.5.4) (Mehdi et al., 2015)
 8 (Schoeneberger et al., 2017). Indigenous Peoples are building upon traditional practices to adapt to the
 9 effects of climate change, for example by working jointly to recharge local aquifers (Basel et al., 2020).

10

11 Water right laws, interstate compacts and international treaties regarding transboundary water shape the
12 context for climate change adaptation, but the possibility of long-term climate change typically was not
13 contemplated at their inception. Gaps in coverage and vaguely defined terms can lead to tensions and
14 disputes, especially in areas facing increased aridity, creating difficulties for adaptation. For example,
15 unregulated pumping of groundwater for irrigation during short-term droughts can serve as an adaptation to
16 acute conditions, (14.5.4) but if persisting long-term, can deplete finite groundwater resources and dewater
17 hydrologically connected rivers. Such outcomes have engendered bitter and costly interstate conflicts in the
18 US, some reaching the US Supreme Court including Texas v New Mexico (Rio Grande) and Florida v.
19 Georgia (Apalachicola-Chattahoochee-Flint).

20

21 Trans-boundary rivers that exemplify the need to address climate impacts include the Colorado (Gerlak et
22 al., 2013), Columbia (Cosens et al., 2016), and Rio Grande/Rio Bravo (Mumme, 1999; Mumme, 2016;
23 Garrick et al., 2018; Payne, 2020). Drought emergencies can open opportunities for progress on collaborative
24 adaptive governance, but such windows may quickly close when wetter conditions return (Sullivan, (2019).

25

26 Water serves a wide variety of environmental functions and human uses as it moves through North
27 America's river basins, so the impacts of climate change are expected to be widespread and multifaceted.
28 This increases the importance of collaborative adaptation efforts that are equitable, transparent and give
29 voice to differing values, perspectives, and entitlements across a broad socioeconomic spectrum of urban and
30 rural, Indigenous and non-Indigenous participants (Miller et al., 2016; Cosens et al., 2018). Adaptation
31 planning may be hampered by conflicting interests, jurisdictional boundaries, and inherent interconnections
32 between actions and impacts at different points throughout a watershed or river basin. Differential power
33 relationships, decision-making authority and access to information also can interfere with effective adaptive
34 governance, while equitable processes for decision-making bolstered by reliable shared information can help
35 to overcome those impediments (Cosens et al., 2016; Arnold et al., 2017; Cosens et al., 2018; Porter and
36 Birdi, 2018).

37

38 Across North America, there are growing signs of progress toward adaptive water governance and
39 implementation of climate-resilient, and ecosystem-based, water management solutions (Colorado River
40 Basin Stakeholders, 2015). California's approach to groundwater sustainability regulation intends to foster
41 such collaborative problem-solving by giving local Groundwater Sustainability Agencies the authority to
42 design locally appropriate plans to meet state-defined sustainability goals (State of California, 2014; Miller,
43 2017). As evidenced by the US interstate disputes, the greatest difficulties arise in cases where stark
44 upstream-downstream differences in interests leave little room for mutual benefit. Severe aridification may
45 test the limits of adaptive capacity.

46

47 Research on water diplomacy recommends broadening negotiations beyond a narrow focus on zero-sum
48 issues, like rigid water allocations, to embrace a more diverse set of shared interests including the need for
49 flexibility to respond to changing conditions. A process for ongoing inclusive engagement of a watershed's
50 stakeholders in mutual social, policy and science learning is important. Such mutual learning can build trust
51 and establish a common platform of credible information for co-creation of adaptation solutions. In addition,
52 better understanding of the policy positions and constraints of others can help stakeholders to identify
53 workable solutions to contentious water management issues (Payne, 2020; Wilder et al., 2020). Cooperation
54 between Mexico and the US on mapping and assessment of transboundary aquifers is a product of such
55 ongoing engagement (Callegary et al., 2018; Sanchez et al., 2018). Other examples of the benefits of
56 sustained engagement are provided by a set of co-management arrangements between state, federal and
57 Indigenous authorities on water management for fishery restoration in the US Pacific Northwest (Tsatsaros et

Do Not Cite, Quote or Distribute  14-37       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 al., 2018), and Indigenous involvement in multi-level co-management of water resources in Canada's
 2 Northwest Territories (Latta, 2018).

 3

 4 14.5.4 Food, Fibre, and Other Ecosystem Products

 5

 6 14.5.4.1 Observed Impacts and Projected Risks: Agriculture, livestock, and forestry

 7

 8 Climate change has affected crops across North America through changes in growing seasons and regions,
 9 extreme heat, precipitation, water stress, and soil quality (Table 14.1, 5.4.1; Figure 5.3) (Mann and Gleick,
10 2015; Galloza et al., 2017; Otkin et al., 2018). These changes directly influence crop productivity, quality
11 and market price (high confidence) (Kistner et al., 2018; Reyes and Elias, 2019). Effects of historical climate
12 change on maize, soybean, barley and wheat crop yields vary from strong increases to strong decreases (e.g.
13 >-0.5 to >+0.5 t ha-1yr-1 for maize) within North America's agroecological regions, even for the same crop
14 (Ray et al., 2019). Across North America, climate change has generally reduced agricultural productivity by
15 12.5% since 1961, with progressively greater losses moving south from Canada to Mexico (Ortiz-Bobea et
16 al., 2021), yet responses are highly differential across regions and crops. Some crop loss events are partially
17 attributed to climate change (high confidence) such as the 2012 Midwest and Great Plains drought, which
18 cost agriculture USD$30B (Smith and Matthews, 2015; Rupp et al., 2017). Aridity is extending northward,
19 altering crop suitability ranges (Fig 14.4); up to 50% of distributional shifts in growing regions for US crops
20 between 1970-2010 may be related to climate change (Lant et al., 2016; Cho and McCarl, 2017). Irrigation is
21 expanding to areas formerly largely dependent on rainfall (Wang et al., 2018b).

22

23 Without adaptation, climate change is projected to reduce overall yields of important North American crops
24 (e.g., wheat, maize, soybeans) (high confidence) (Chen et al., 2017; Levis et al., 2018) (Tables SM14.3-4).
25 For example, projected heat stress (RCP8.5) reduced midcentury (2040�2069) maize and cotton yields by
26 12-15% of historical yields (1950�2005), with the US-SW suffering the largest impacts (Elias et al., 2018)
27 (Table SM14.5). Warming and heat extremes will delay or prevent chill accumulation, affecting perennial
28 crop development (e.g. fruit set failure), yield (e.g., walnuts, pistachios, stone fruit), and quality (e.g. grapes)
29 (medium confidence) (Parker et al., 2020). Warming will alter the length of growing seasons of cold-season
30 crops (e.g., broccoli, lettuce) and will shift suitability ranges of warm-season California crops (e.g.,
31 tomatoes) (medium confidence) (Marklein et al., 2020). Increasing atmospheric CO2 will enhance yields yet
32 reduce nutrient content of many crops (high confidence); a CO2 concentration of 541 ppm (seen by 2050 in
33 RCP 8.5) would reduce per capita nutrient availability in North American diets by 2.5�4.0% (Beach et al.,
34 2019). Crop pest and pathogen outbreaks are expected to worsen under climate change (high confidence)
35 (Deutsch et al., 2018; Wolfe et al., 2018; Zhang et al., 2019a).

36

37 Climate change is anticipated to cause declines in livestock production across North America (high
38 confidence; Table 14.4 & SM14.6) (Havstad et al., 2018; Murray-Tortarolo et al., 2018); increases in
39 extreme temperature raise the risk of livestock heat stress, disease, and pest impacts (Rojas-Downing et al.,
40 2017). Projected aridification reduces forage production in the Southwest US and Northern Mexico (high
41 confidence) (Polley et al., 2013; Reeves et al., 2014; Cooley, 2016; Bradford et al., 2020) and transforms
42 grasslands to woody shrublands (Briske et al., 2015; Murray-Tortarolo et al., 2018), while warmer and wetter
43 conditions in the northern regions (CA-PR, US-NW, US-NP) may enhance rangeland production by
44 extending growing seasons (high confidence) (Hufkens et al., 2016; Derner et al., 2018; Zhang et al., 2019a).
45 Increased CO2 will enhance production (medium confidence), but reduce forage quality (high confidence) in
46 US-NP and US-NW (Table SM14.6) (Derner et al., 2018).

47

48 Climate change impacts on forests (14.5.1, Box 14.2) may affect timber production by altering tree species
49 distributions, productivity, and wildfire and insect disturbances (medium confidence). Southern or drier
50 locations may shift from forests to other vegetation types, whereas higher latitude areas may experience
51 forest expansion (Brecka et al., 2018). Tree species composition is projected to change with climate change
52 (Wang et al., 2015; Bose et al., 2017). Tree growth may increase or decrease from changes in temperature or
53 moisture depending on location, with lower growth expected from warming in water-limited areas (Littell et
54 al., 2010). Increased productivity associated with more favourable climate conditions is projected for boreal
55 forests (Brecka et al., 2018), although in some regions, growth will reverse and decline with additional
56 warming (D'Orangeville et al., 2018; Chaste et al., 2019). As a result of these changes, timber yields in
57 North America may increase in the future (Beach et al., 2015; EPA, 2015a) or decrease (Boulanger et al.,

Do Not Cite, Quote or Distribute  14-38       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

1 2014; McKenney et al., 2016; D'Orangeville et al., 2018; Thorne et al., 2018; Chaste et al., 2019) depending
2 on location and mechanisms included. Wildfires and insect outbreaks are projected to increase with future
3 climate change, thereby limiting biomass (Gauthier et al., 2015; Bentz et al., 2019; Chaste et al., 2019).

4

5

 6

 7 Figure 14.5: Crop responses to climate change will depend on existing mean climate, the type of climate change, and
 8 characteristics of crop types. Hypothesized responses for Crop Types A, B, C, and D include changing crop yields or
 9 changing crop area. Adaptation actions may alter hypothesized responses; maps from Matthews et al. (2019)

10
11

12 14.5.4.2 Observed Impacts and Projected Risks: Fisheries and Aquaculture

13

14 Climate impacts outlined in Section 14.5.2 have induced yield losses for multiple subsistence, recreational,
15 and commercial fisheries (very high confidence) and contributed to commercial fishery closures across North
16 America (Figure 14.6, Table SM14.7, 14.5.1, 14.5.3) (Lynn et al., 2014; Barbeaux et al., 2020; Fisher et al.,
17 2021). Climate-driven declines in productivity are widespread (high confidence) (Figure 14.6), although a
18 few increases are observed in northern regions (medium confidence) (Cunningham et al., 2018; Crozier et al.,
19 2019; Zhang et al., 2019b). Redistribution of species has increased travel distance to fishing grounds, shifted
20 stocks across regulatory and international boundaries, and increased interactions with protected species (very
21 high confidence) (Cross-Chapter Box MOVING PLATE in Chapter 5) (14.5.2) (Morley et al., 2018; Free et
22 al., 2019; IPCC, 2019c; Rogers et al., 2019; Stevenson and Lauth, 2019; Young et al., 2019) (Figure 14.6,
23 Table SM14.7). Climate shocks have reduced yield and increased instability in fishery revenue (high
24 confidence) (Fisher et al., 2021).

25

26 Declines in yield and poleward stock redistributions (avg. ~20.6 km decade-1) are expected to continue under
27 climate change, and increase in magnitude with atmospheric carbon (high confidence) (Table 14.4) (Hare et
28 al., 2016; Pecl et al., 2017; Rheuban et al., 2017; Morley et al., 2018; Smale et al., 2019a; Szuwalski et al.,
29 2021). For example, without adaptation, end of century losses of Bering sea pollock yield (relative to
30 persistence scenarios) is likely to reach 50% under moderate (RCP4.5) and 80% under low (RCP8.5)
31 mitigation scenarios, respectively (Holsman et al., 2020). Expanding HABs, pathogens, and altered ocean
32 chemistry (OA and dissolved oxygen) will reduce yields and increase closures of fisheries along all North
33 American coasts (medium confidence) (14.5.2) (Deutsch et al., 2015a; Ekstrom et al., 2015; Seung et al.,
34 2015; Punt et al., 2016; Howard et al., 2020). For fisheries that represent 56% of current US fishing revenue,
35 projected annual net losses under high emission scenarios (RCP 8.5; 2021-2100) may reach double that of
36 low emission scenarios (RCP2.6) (Moore et al., 2021).

37

Do Not Cite, Quote or Distribute  14-39       Total pages: 157
      FINAL DRAFT                 Chapter 14  IPCC WGII Sixth Assessment Report

1

 2

 3 Figure 14.6: Climate change impacts on North American fisheries and aquaculture

 4

 5

 6 Warming waters and OA have impacted aquaculture production in North America (high confidence) (Figure
 7 14.6) (Clements and Chopin, 2017; Reid et al., 2019; Stewart-Sinclair et al., 2020). Under climate change
 8 (RCP8.5), declines in marine finfish and bivalve aquaculture become likely by mid-century (Froehlich et al.,
 9 2018; Stewart-Sinclair et al., 2020). Adaptation is possible but uncertain (Bitter et al., 2019; Fitzer et al.,
10 2019; Reid et al., 2019), especially with increasing extreme events. Nature-based aquaculture solutions (e.g.,
11 conservation aquaculture, restorative aquaculture) could aid carbon mitigation and local-level adaptation,
12 especially for seaweed and bivalve culture (Box 14.7) (Froehlich et al., 2017; Froehlich et al., 2019; Reid et
13 al., 2019; Theuerkauf et al., 2019).

14

Do Not Cite, Quote or Distribute  14-40       Total pages: 157
FINAL DRAFT                                  Chapter 14                      IPCC WGII Sixth Assessment Report

1 Table 14.4: Observed and projected impacts to food and fibre resources.

Climate Driver Observed Change1                                   Reference                     Projected change                                    Reference

Agriculture and livestock (Tables SM14.2-5)

                                                                                                Projected heat stress (RCP8.5) reduces

               Estimates of yield reduction from heat stress for                                midcentury (2040�2069) maize and cotton

               both maize and cotton indicate that historically,                                yields by 12-15% of historical yields (1950�

               US-SW heat stress reduced cotton yield by 26%                                    2005) with largest impacts in US-SW additional

               and maize yield by 18% compared to potential (Altieri and Nicholls,              drought-related stress in US-MW could reduce

               yield. Extreme heat was associated with            2009; Mastachi-Loza et maize and soybean yields by ~5% and ~10%, (Jin et al.; Rojas-Downing

Extreme events increased crop failure in MX-CE, US-SW;            al., 2016; Elias et al.,      respectively, by late century under RCP 4.5 ; et al., 2017; Elias et al.,

               Hailstorm increased frequency observed in MX 2018; Kistner et al., 2018; warming and extreme heat (>35%) will delay 2018; Parker et al., 2020)

               coinciding with the most vulnerable stage or       Reyes and Elias, 2019)        (or prevent) chill accumulation, impacting

               flowering period of maize; extreme precipitation                                 perennial crop development, yields, and quality

               damages to soil, increased erosion, and reduced                                  (US-SW). Increases in extreme temperature

               crop yields observed in MX and US-MW                                             raise the risk of livestock heat stress, disease,

                                                                                                and pest impacts.

Mean growing   Across the US Great Plains (US-SP, US-NP)          (Frisvold and Konyar,         Warming alters the length of growing seasons of     (St-Pierre et al., 2003;
season         between 1968-2013 climate change induced           2012; Leskovar et al.,        cold-season crops and shifts suitability ranges of  Polley et al., 2013; Key and
precipitation  3.55%, -0.55%, and 0.94% change in yield for       2012; Aladenola and           warm-season California crops; aridification         Sneeringer, 2014; Reeves et
decline, mean  (irrigated and non-irrigated) maize, sorghum       Madramootoo, 2014;            reduces forage production US-SW, MX-N;              al., 2014; Cooley, 2016;
temperature    and soybeans (respectively); Droughts and          Galloza et al., 2017;         warming is associated with reduced livestock        Hristov et al., 2018; Ortiz-
increase,      increasing temperatures reduced soil fertility in  Havstad et al., 2018;         growth and fertility, increased pathogens in US-    Col髇 et al., 2018; Bowling
drought        MX and contributed to soil erosion and             Kukal and Irmak, 2018)        SE, US-SP, US-MW, US-NE, and reduced milk           et al.; Bradford et al., 2020;
               degradation and suitability loss of 18-22%;                                      production in US-MW.                                Marklein et al., 2020)
               Experimental and simulated reductions in water                                                                                       (Hufkens et al., 2016;
               supply of 25-50% result in similar magnitude                                                                                         Derner et al., 2018; Zhang
               declines in yield for multiple food and forage                                                                                       et al., 2019b)
               crops (e.g., wheat, maize)

Multiple       Climate change reduced total factor productivity   (Garru馻-Hern醤dez et al.;    Projected declines in yield and changes of in       (Calder髇-Garc韆 et al.;
drivers        of agriculture and livestock in North America      Loreto et al.; Wolfe et al.,                                                      Herrera-Pantoja and
               by 12.5% (ranging from approx. -35% to +8%)        2018; Torres Castillo et      suitability ranges for maize (-18%-+5%),            Hiscock; Lant et al., 2016;
               between 2016-2015; losses have been greatest       al., 2020; Ortiz-Bobea et                                                         Chen et al., 2017; Montiel-
               in Mexico (-30% to -25% Figure5), and lowest       al., 2021)                    sorghum (-16 to +12%), and wheat (-38 to -          Gonz醠ez et al.; Reyer et al.;
               in Canada (>0% );Reduced yield in MX, US;                                                                                            Derner et al., 2018; Deutsch
               increased weed, pest pressure in US-NE, US-                                      15%) in MX (RCP 4.5, 8.5; 2040-2099);               et al., 2018; Levis et al.,
               MW, US-NP, US-NW                                                                                                                     2018; L髉ez-Blanco et al.;
                                                                                                northward shifts in the suitable area for 6 crops   Murray-Tortarolo et al.;
                                                                                                                                                    Wolfe et al., 2018; Gomez
                                                                                                from the central US (2100); Warming                 Diaz et al.; Qian et al.;
                                                                                                                                                    Zhang et al., 2019b; Arce
                                                                                                accompanied by increased CO2 may benefit            Romero et al.)

                                                                                                crop production of small grains in southern
                                                                                                Canada up to 3 oC GWL, although benefits
                                                                                                decline after 2.5oC GWL. Increased CO2

                                                                                                enhances production but reduces forage quality
                                                                                                US-NP,US-NW. Without adaptation, 2oC GWL

Do Not Cite, Quote or Distribute             14-41                                              Total pages: 157
FINAL DRAFT                                        Chapter 14       IPCC WGII Sixth Assessment Report

Aquaculture and fisheries (Tables SM14.6, SM14.8)                                                increased insect-caused production losses ~36%
                                                                                                 and~44% for maize and wheat, respectively.
                MHW and HAB event of 2014-2016 resulted in
                                                                                                 Projected doubling of MHW impact levels by
                multiple fishery closures along the west coast                                   2050 amongst the most important fisheries
                                                                                                 species (over previous assessments that focus
                (US-NW, US-SW); disparate impacts observed                                       only on long-term climate change)

                between small and large vessels with greatest

                impacts on small vessel revenue and fishery

                participation; impacts were highest for ports in    (Handisyde et al., 2017;
                                                                    Food Agriculture
                the CC-N and least for fishing communities with     Organization of the United
                                                                    Nations; Froehlich et al.,
                diverse livelihoods and harvest portfolios; In the  2019; Reid et al., 2019;
                                                                    Bertrand et al., 2020;
                EBS, GOA, and N-CC, declines in fish biomass        Cheung and Fr鰈icher,
                                                                    2020; Jardine et al., 2020;
Extreme events  and shifts in distribution were 4 times higher      Sippel et al., 2020; Fisher                                                    (Cheung and Fr鰈icher,
                and greater during MHWs than that of general        et al., 2021)                                                                  2020)

                warming over the same period; pelagic fish

                showed largest decrease in biomass (7%), as did

                Sockeye salmon and California anchovy;

                Increased risk to hatcheries and low lying pond

                systems from severe storms. Extreme heat is

                associated with reduced productivity of

                aquaculture species.

Multiple        Climate shocks reduce catch, revenue and            (Marushka et al., 2019;      Declines in North American catch potential of     (Weatherdon et al., 2016;
drivers         county-level wages and employment among             Oremus, 2019) (14.5.6        flatfish are projected under RCP8.5 for the EBS,  Cheung, 2018; Carozza et
                commercial harvesters in US-NE; climate             Health)                      GOA, GOMX, US-SE, and US-NE; declines in          al., 2019; Cisneros-Mata et
                variability 1996 - 2017 is responsible for a 16%                                 productivity projected for multiple species in    al.; Reum et al., 2019; Tai
                (95% CI: 10% to 22%) decline in county-level                                     MX, with largest declines in productivity         et al., 2019; Mendenhall et
                fishing employment in New England; impacts                                       (>35%) for abalone and pacific sardine; Impacts   al., 2020; Wilson et al.,
                mediated by local biology and institutions;                                      are greatest for artisanal species ; projected    2020)
                Seafood is an important source of nutrients and                                  declines in fish community biomass for all
                protein for Indigenous Peoples in CA-BC;                                         North American coasts except US-SW and the
                polices that incorporate nutrition in fisheries                                  Canadian Arctic; declines are greater under
                management are limited in North America                                          RCP8.5 than RCP2.6. Modest increases (up to
                                                                                                 10%) in landings of CA-QC and CA-AT surf
                                                                                                 clams and shrimp are projected under RCP2.6
                                                                                                 by 2100 while declines in snow crab up to 16%
                                                                                                 are expected (RCP2.6,8.5); Mussel landings
                                                                                                 projected to increase 21%., while declines in
                                                                                                 shellfish and lobster landings (2090) are twice
                                                                                                 as high under RCP 8.5 (42%-54%) as RCP 2.6.

Do Not Cite, Quote or Distribute                   14-42                                         Total pages: 157
FINAL DRAFT                       Chapter 14                      IPCC WGII Sixth Assessment Report

Ocean and lake  OA reduced maximum sustainable yield, catch       (Long et al., 2013a; Seung    Shellfish, snow crab landings projected to         (Ekstrom et al., 2015; Reum
acidification   and profits of EBS Tanner crab in simulations;    et al., 2015; Punt et al.,    decline in CA-QC and CA-QT; declines under         et al., 2019; Steiner et al.,
                survival of larval and juvenile red king crab     2016; Clements and            RCP 8.5 are double that of RCP 2.6; Climate        2019; Wilson et al., 2020;
Mean            (RKC) in the lab decreased 97-100% with           Chopin; Handisyde et al.,     change reduces EBS blue king crab recovery in      Punt et al., 2021)
temperature     decreasing pH;No appreciable effects of pH on     2017; Swiney et al., 2017;    simulations; Relative to US and CA, MX has
increase        larval growth of walleye pollock in the lab       Food Agriculture              strongest benefits in net catch under RCP2.6       (Weatherdon et al., 2016;
                (Hurst, 2013); Mixed evidence of impacts of       Organization of the United    relative to RCP8.5 ( >30% increase in catch);      Cheung, 2018; Froehlich et
                changes in pH on freshwater or saltwater finfish  Nations; Froehlich et al.;    increases of 70% in catch potential projected for  al., 2018; Morley et al.,
                aquaculture; OA reduced growth, calcification,    Reid et al., 2019; Stewart-   the Canadian Arctic (CA-NE,CA-NW) under            2018; Greenan et al.,
                attachment and increased mortality in calcifying  Sinclair et al.)              RCP 8.5 (versus minimal changes under              2019b; Steiner et al., 2019;
                molluscs and seaweeds in US, CA; OA may                                         RCP2.6); high resolution and size spectrum         Sumaila et al., 2019;
                benefit non-calcifying seaweeds.                  (Poloczanska et al., 2016;    models project declines in groundfish catch and    Bryndum-Buchholz et al.,
                Species distributions have shifted poleward and   McCoy et al., 2017;           biomass in S-EBS; shifting transboundary           2020; Holsman et al., 2020;
                phenology has shifted earlier with strongest      Swiney et al., 2017; Le       stocks may increase challenges.                    Palacios-Abrantes et al.,
                effects on bony fish. Warming over the last       Bris et al., 2018; Miller et                                                     2020; Reum et al., 2020;
                century (2001-2010) - (1930-1939) is associated   al., 2018; Food               Projected declines for some shellfisheries and     Sumaila and Zwaag, 2020;
                with declines in MSY along the entire west        Agriculture Organization      flatfish due to OA and temperature; OA             Whitehouse and Aydin,
                coast of North America that range from -14% in    of the United Nations;        conditions under RCP 8.5 reach critical risk       2020; Wilson et al., 2020)
                the EBS to -29% in the CC-S. Along the east       Free et al., 2019;            thresholds for mollusc harvests earlier in
                coast, declines of -3% to -9% were observed in    Froehlich et al.; Reid et     northern regions than southern areas; OA risk to
                the GOMX and US-SE, while increased of 8-         al., 2019; Weiskerger et      shellfisheries is highest in N-CC;OA caused 1%
                15% were observed in the US-NE and CA-CQ.         al., 2019; Bertrand et al.,   additional decline in Arctic cod populations by
                Mixed positive and negative growth and            2020; Le et al., 2020)        2100 under RCP8.5;OA influences management
                mortality responses for aquaculture species in                                  reference points of Northern Rock sole. OA and
                North America; Juvenile red king crab survival                                  temperature reduce probability of recovery in
                decreases as temperatures increase in lab                                       simulations of EBS blue king crab.
                experiments . American Lobster abundances
                                                                                                By end of century, North America fish biomass,
                                                                                                catch potential and revenue are ~9% higher in
                                                                                                RCP 2.6 than RCP 8.5 and differences are
                                                                                                greatest for US fisheries (relative to CA, MX;
                                                                                                Projected poleward redistributions (reported
                                                                                                ranges of 10.3 to 39.1 km decade-1) and to depth
                                                                                                decrease access to shellfisheries in CA-QC and
                                                                                                subsistence species in CA-BC (-28% by 2100),
                                                                                                with impacts increasing N to S and under RCP
                                                                                                8.5 as compared to RCP 2.6.; Climate change
                                                                                                (RCP8.5) is projected to shift the relative % of
                                                                                                catch and profits for US - Canada transboundary
                                                                                                stocks under RCP8.5 (but not RCP2.6).
                                                                                                Projected decreases in biomass of historically
                                                                                                large fisheries US-NA and CA-QC, and US-AK

Do Not Cite, Quote or Distribute  14-43                                                         Total pages: 157
   FINAL DRAFT                       Chapter 14               IPCC WGII Sixth Assessment Report

                declined (78%) in South New England and have                        and important subsistence species in CA-WA
                increased (515%) in the Gulf of Maine due to                        and CA-BC, while some increases in the North
                water temperature changes and differing                             Atlantic; Declines are greater under RCP 8.5
                conservation measures (between 1985 and 2014                        relative to RCP 2.6; in EBS (US-AK)
                for GOM and 1997 and 2014 for Southern New                          community biomass, catches, and mean body
                England).                                                           size decreased by 36%, 61%, and 38%,
                                                                                    respectively under RCP 8.5 (2100). Climate
                                                                                    change causes projected declines in global
                                                                                    marine aquaculture production under RCP 8.5
                                                                                    with impacts greater for bivalve than finfish and
                                                                                    with significant disparities among regions in
                                                                                    direction and magnitude of changes; greatest
                                                                                    declines for finfish aquaculture expected in
                                                                                    Northern regions (GOA, CA-BC, CA-CQ), and
                                                                                    large declines for bivalve production (declines
                                                                                    of 20-100%) for Canada. Declines become more
                                                                                    probable by 2050-2070.

1
2

   Do Not Cite, Quote or Distribute  14-44                    Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 14.5.4.3 Food and Fibre Adaptation: Cross Cutting Themes

 2

 3 Across food and fibre systems, climate resilience is enhanced through diversifying income and harvest
 4 portfolios and increasing local biodiversity and functional redundancy (high confidence) (Messier et al.,
 5 2019; Rogers et al., 2019; Young et al., 2019; Aquilu� et al., 2020; Fisher et al., 2021). Ecosystem-based
 6 practices and sustainable intensification (increasing yields while minimizing resource demand and ecosystem
 7 impacts) (Cassman and Grassini; Rockstr鰉 et al., 2021) will help the sector meet food production demands
 8 under climate change (medium confidence), but effectiveness generally declines and is less certain after 2050
 9 in scenarios without carbon mitigation (high confidence) (Bermeo et al., 2014; Gaines et al., 2018; Costello
10 et al., 2020; Free et al., 2020; Holsman et al., 2020). Across the sector, successful adaptation is underpinned
11 by approaches that meaningfully consider the coupled social-ecological networks around food and fibre
12 production and value Indigenous Knowledge (very high confidence) (Box 14.1) (FAO, 2018; Steele et al.,
13 2018; Calliari et al.). Integrated modeling, participatory planning and inclusive decision making promote
14 effective and equitable adaptation responses (very high confidence) (Figure 14.7, 14.7) (Toledo-Hern醤dez et
15 al., 2017; Eakin et al., 2018; Monterosso and Conde, 2018; Alexander et al., 2019; Hodgson and Halpern,
16 2019; Holsman et al., 2019; Samhouri et al., 2019; Barbeaux et al., 2020; Hollowed et al., 2020), while a
17 paucity of high resolution and locally tailored climate change information remains a barrier to adaptation
18 (Ekstrom et al., 2015; Donatti et al., 2017; Young et al., 2019).

19
20

21

22 Figure 14.7: Adaptation in North American food sectors modified from Cottrell et al. (2019).

23

24

25 14.5.4.4 Food and Fibre Adaptation: Agriculture, Livestock, and Forestry

26

27 Land management and horticulture approaches that preserve and improve soil structure and organic matter
28 can reduce erosion (high confidence) (Section 14.5.1, 3) (Lal et al., 2011; Bisbis et al., 2018), and preserving
29 biodiversity and water, changing planting dates, and double cropping are effective climate adaptation
30 strategies (Bisbis et al., 2018; Hernandez-Ochoa et al., 2018; Monterroso-Rivas et al., 2018; Wolfe et al.,
31 2018). Traditional agriculture inherently includes climate adaptive practices that enhance biodiversity, soil
32 quality and agricultural production (e.g., multiple cultivars, heat-tolerant heritage cattle breeds) (Bermeo et
33 al.; Gomez-Aiza et al., 2017; Ortiz-Col髇 et al., 2018). Agroecology and agroforestry (Box 14.7) in North
34 America has expanded from (but not replaced) traditional and rural practices in Mexico (Metcalfe et al.,

Do Not Cite, Quote or Distribute  14-45       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 2020a) as a sustainable and climate-resilient alternative to industrial agriculture (Schoeneberger et al., 2017)
 2 that increases productivity (by 6-65% depending on the crop), enhances microclimatesand provides co-
 3 benefits for GHG mitigation (Abbas et al., 2017; Cardinael et al., 2017; Schoeneberger et al., 2017; Snapp et
 4 al., 2021). Irrigation is an effective adaptation strategy in key agricultural areas (Miller, 2017; Lund et al.,
 5 2018) and could stabilize food security in rain-fed regions (e.g., southeastern Mexico) (Spring, 2014); water
 6 allocation must balance multiple needs and rights (medium confidence) (14.5.3) (Brown et al., 2015b; Levis
 7 et al., 2018; Gomez Diaz et al., 2019). Heritage livestock breeds, changing species, and precision ranching
 8 technology may promote ranch and rangelands resilience (Zhao et al., 2013). In loblolly pine plantations in
 9 the southern US, effective adaptation includes reducing tree density and, less effectively, shifting to slash
10 pine (Susaeta et al., 2014). Salvage logging following forest disturbances (e.g., insect outbreaks) can increase
11 timber harvest (Bogdanski et al., 2011; USDA Forst Service, 2011; Han et al., 2018; Morris et al., 2018a).

12

13 14.5.4.5 Food and Fibre Adaptation: Fisheries and Aquaculture

14

15 Proactive and ecosystem-based management increases climate resilience in fisheries (high confidence) but
16 effectiveness after 2050 may be limited without global carbon mitigation (medium confidence) (Gaichas et
17 al., 2017; Gaines et al., 2018; Kritzer et al., 2019; Barbeaux et al., 2020; Free et al., 2020; Holsman et al.,
18 2020). Flexibility (e.g., mobility, diverse incomes or harvest portfolios) underpins climate resilience across
19 regions, management policies, and fisheries, although small-scale fisheries have less scope for adaptation
20 (Aguilera et al., 2015; Young et al., 2019). Climate-informed and dynamic management (Hazen et al., 2018)
21 improves modeled fishery performance (medium confidence) (see section 14.5.2) (Froehlich et al., 2017;
22 Tommasi et al., 2017a; Tommasi et al., 2017b; Karp et al., 2019; Barbeaux et al., 2020), yet planning and
23 policies that directly incorporate climate change information remain limited (Skern-Mauritzen et al., 2015;
24 Marshall et al., 2019b). Expanding aquaculture across North America will likely address deficits in
25 nutritional and protein yields (Gentry et al., 2019; Costello et al., 2020), yet aquaculture initiatives have
26 largely progressed without explicitly considering climate impacts (FAO, 2018; Froehlich et al., 2019) and
27 critical elements for climate adaptation (e.g., climate-informed zoning, monitoring, insurance) are not widely
28 implemented (Li馻n-Cabello et al.; FAO, 2018 ; Stewart-Sinclair et al., 2020). Climate-informed and
29 standardized aquaculture governance, and increased coordination with fishery and coastal management, is
30 needed for climate resilience (high confidence) (Brug鑢e et al., 2019; Froehlich et al., 2019; Free et al., 2020;
31 Galparsoro et al., 2020).

32

33 14.5.5 Cities, Settlements and Infrastructure

34

35 Cities are complex social-ecological systems with large populations, concentrated wealth, ageing
36 infrastructure, reliance on extrinsic and increasingly stressed natural systems, social inequality, differential
37 institutional capacities, and impervious, heat-retaining surfaces (Maxwell et al., 2018a; Schell et al., 2020).
38 These factors interact with location (e.g., proximity to coast, in a flood plain) to create city-specific
39 vulnerabilities to climate change and requirements for resilience initiatives (Mercer Clarke et al., 2016).
40 Cities are home to diverse cultural and social communities, including large Indigenous populations, who can
41 be uniquely affected by climate change yet who bring valuable IK and leadership to urban adaptation efforts
42 (Statistics Canada, 2020; Brown et al., 2021). The rural and remote settlements of North America also
43 experience similar hazards and risks, but due to different factors such as geographic isolation, dependence on
44 local food resources, and socioeconomic conditions (Kearney and Bell, 2019; Vodden and Cunsolo, 2021).

45

46 14.5.5.1 Observed Impacts

47

48 14.5.5.1.1 Rising temperatures and extreme heat
49 Extreme heat events are affecting natural assets and built infrastructure as well as individuals in cities and
50 rural settlements across North America (high confidence) (Maria Raquel et al., 2016; Amec Foster Wheeler
51 Environment and Infrastructure, 2017; Howell and Brady, 2019; Martinich and Crimmins, 2019). Key urban
52 infrastructure systems (e.g., services in buildings, energy distribution) are interdependent and susceptible to
53 cascading impacts (e.g., electricity supply disruption during a heat wave compromising another system like
54 water delivery, high-rise cooling) (Brown et al., 2021). Urban social inequality and systemic racism has led
55 to disproportionately higher exposure to urban heat island effects in low-income and minority
56 neighbourhoods in US cities, due in part, to less green space and tree cover to offset heat retained in the built
57 environment (Hoffman et al., 2020; Schell et al., 2020; Hsu et al., 2021). In the rural context, extreme heat

Do Not Cite, Quote or Distribute  14-46       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 contributes to migration out of small communities (e.g., cases reported in Mexico (Nawrotzki et al., 2015a)).
 2 Extreme heat events pose a significant risk to residents of small towns across North America due to limited
 3 resources to address heat impacts and attendant increased morbidity and mortality (McDonald et al., 2016;
 4 Guo et al., 2018; D'ulisse, 2019) (See 14.5.6.1).

 5

 6 Hot and dry conditions increase risk of wildfires close to human settlements through collateral impacts on
 7 properties, economic activity and human health (Box 14.2, 14.5.6.3). These environmental conditions also
 8 stress natural assets (e.g., urban forests, wetlands, household gardens, green walls) and performance of green
 9 infrastructure leading to higher operation and maintenance costs (high confidence) (Kabisch et al., 2017;
10 Terton, 2017).

11

12 14.5.5.1.2 Storms and flooding
13 Short-duration, high-intensity rainfall and other extreme events (e.g., hurricanes, atmospheric river events)
14 create significant flooding risks and impacts for cities in North America and negatively affect the lives,
15 livelihoods, economic activities, infrastructure, and access to services (high confidence) (Amec Foster
16 Wheeler Environment and Infrastructure, 2017; Curry et al., 2019). In 2016, US flooding events caused 126
17 fatalities and US$11B (2016) in damages (NOAA, 2019). In Canada, flooding accounts for 40% of the costs
18 associated with weather-related disasters recorded since 1970 (Canadian Institute for Climate Choices,
19 2020); the most costly event was the 2013 Calgary flood (CA-PR) (CAD$1.8B in catastrophic insurance
20 losses and CAD$6B in direct costs such as uninsured losses) (Office of the Auditor General of Canada,
21 2016). Mexico City is seasonally impacted by high-intensity rainfall events that generate local flooding (de
22 Alba and Castillo, 2014). Rural and remote settlements are also threatened by floods; Indigenous lands in
23 Canada are disproportionately exposed to flooding, with almost 22 % of residential properties at risk of a
24 100-year flood (Thistlethwaite et al., 2020a; Yumagulova, 2020).

25

26 Wind storms and hurricanes are significant climate hazards for North American cities and settlements,
27 affecting urban forests, electricity distribution and service delivery, and damaging buildings and
28 transportation infrastructure (Amec Foster Wheeler Environment and Infrastructure, 2017; British Columbia
29 Hydro, 2019; Smith, 2020), with enduring impacts on small villages due to lost livelihoods and limited
30 recovery capacity (e.g., Rio Lagartos and Las Coloradas in Mexico (MX-SE) after Hurricane Isidore)
31 (Audefroy and Cabrera S醤chez, 2017). The Pacific coast of Mexico is also experiencing hurricanes such as
32 Patricia (Category IV) in 2015 and Newton (Category I) in 2016 (CONAGUA, 2015; CONAGUA, 2016);
33 hurricane Patricia affected 56 municipalities in the states of Colima, Nayarit and Jalisco (MX-CE, MX-NW)
34 (Calleja-Reina, 2016).

35

36 14.5.5.1.3 Sea level rise
37 SLR interacts with shoreline erosion, storm surge and wave action, saline intrusion, and coastal flooding to
38 directly threaten coastal cities and small communities in North America with impacts to public and private
39 buildings and infrastructure, port and transportation facilities, water resources (high confidence) (NOAA
40 National Weather Service, 2017; Boretti, 2019), and cultural heritage sites (Dawson et al., 2020) (Box 14.4).
41 SLR is creating conditions where considerable financial investments are needed and, in many cases, are
42 being raised to address adaptation needs (Fatori and Seekamp, 2017; Hinkel et al., 2018; Greenan et al.,
43 2019a) (see Box 14.4, CCP6). Across North America, high population density and concentrated development
44 along the coast generates exposure to SLR impacts.

45

46 14.5.5.2 Projected Impacts and Risks

47

48 Evidence since the AR5 highlights increased risk to quality of life in cities and rural communities as a result
49 of exposure to intensifying climate change hazards, and the compounding and interacting effects of climate
50 and non-climate factors (medium confidence).

51

52 14.5.5.2.1 Rising temperatures and extreme heat
53 Extreme heat events are projected to increase in frequency and intensity across North America in the coming
54 decades (14.2.2, Figure 14.2(F),(G)). Inland urban areas in southern and eastern US are susceptible to urban
55 heat island effects, particularly the Midwest/Great Lakes regions (Krayenhoff et al., 2018) and Mexico City
56 and many other cities in Mexico (Vargas and Maga馻, 2020). Climate change (RCP8.5) interacting with
57 urban form, development and systemic racism (Schell et al., 2020; Hsu et al., 2021), could worsen risks from

Do Not Cite, Quote or Distribute  14-47       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 extreme heat in North American cities, especially where there is limited adaptation (high confidence)
 2 (Krayenhoff et al., 2018). Impacts from extreme heat will be exacerbated when multiple hazards occur
 3 simultaneously (e.g., heat waves concurrent with droughts) (Mora et al., 2018; Zscheischler et al., 2018).
 4 Extreme heat events increase energy demand for space cooling in buildings, especially during peak demand
 5 periods and heat waves (IEA, 2018a). This can decrease cooling efficiency, increase emissions of GHG from
 6 electricity generation, increase refrigerant loads and associated emissions, and negatively affect air quality
 7 (IEA, 2018a). Major electrical grid failure (i.e., "blackouts") have increased across the US, and will continue
 8 to be particularly dangerous for human health when they coincide with extreme heat events (Stone et al.,
 9 2021). Efforts to increase resilience of the infrastructure that cities rely on are increasing (Climate-Safe
10 Infrastructure Working Group, 2018)

11

12 Warmer and/or drier conditions may reduce water supply reliability for cities and small communities that
13 rely on surface water sources fed by rain or snowmelt runoff (e.g., Victoria and Vancouver, Canada (CA-BC)
14 (Metro Vancouver, 2016; Vadeboncoeur, 2016; Islam et al., 2017); San Pedro, Hermosillo and Los Pargas,
15 Aguascalientes, M閤ico (MX-NW, MX-CE) (Vadeboncoeur, 2016; Soto-Montes-de-Oca and Alfie-Cohen,
16 2019); New York City, (US-NE) (N. Y. C. Department of Environmental Protection, 2014) and Washington
17 State (US-NW) (Fosu et al., 2017) (see 14.5.3.2).

18

19 14.5.5.2.2 Storms and flooding
20 Annual and winter precipitation is expected to increase for most of Canada (14.2, Figure 14.2(D), (E)) and
21 will increase flooding in cities and settlements (Bonsal et al., 2019) (high confidence). Although there is
22 more geographical variation across the continental US (e.g., between high-latitude and subtropical zones),
23 extreme precipitation events are projected to increase in frequency and intensity with impacts on flood
24 hazards (Easterling et al., 2017) (14.5.3.2). Winter (snow and ice) storms are expected to increase in northern
25 North America and decrease in southern North America under RCP 8.5 (Jeong and Sushama, 2018b).
26 Projected increases in wind-driven rain exposure is an emerging consideration for moisture-resilient design
27 and management of buildings, especially in western and northern Canada (Jeong and Cannon, 2020).

28

29 14.5.5.2.3 Sea level rise
30 In the US, many people are projected to be at risk of flooding from SLR (high confidence) (Box 14.4). A
31 projected SLR of 0.9m by 2100 could place 4.2 million people at risk of inundation in US coastal counties,
32 whereas a 1.8-m SLR exposes 13.1 million people (Hauer et al., 2016). In California, under an extreme 2-m
33 SLR by 2100, US$150B (2010) of property or more than 6% of the state's GDP and 600,000 people could be
34 affected by flooding (Barnard et al., 2019). A 1-m SLR would inundate 42% of the Albemarle-Pamlico
35 Peninsula in North Carolina and incur property losses of up to US$14B (2016) (Bhattachan et al., 2018). In
36 nine southeast US states, a 1-m SLR would result in the loss of more than13,000 recorded historical and
37 archaeological sites with over 1,000 eligible for inclusion in the National Register for Historic Places
38 (Anderson et al., 2017). SLR raises groundwater levels by impeding drainage and enhancing runoff during
39 rain events (Hoover et al., 2017); coastal flooding enhances saltwater intrusion affecting drinking water
40 supply in settlements (e.g., coast of Texas) (Anderson and Al-Thani, 2016).

41

42 In Canada, SLR is expected to increase the frequency and magnitude of extreme high water-level events
43 (Greenan et al., 2019a) and to create widespread impacts on natural and human systems (high confidence)
44 (Lemmen et al., 2016) (Box 14.4). Although coastal sensitivity is high in the Arctic, Canada's more
45 populated regions are also sensitive to the impacts of SLR (Manson et al., 2019). The Mi'kmaq community
46 of Lennox Island First Nation is exploring relocation options because of erosion from SLR (Savard et al.,
47 2016).

48

49 In Mexico, crucial coastal tourism cities such as Cancun, Isla Mujeres, Playa del Carmen, Puerto Morelos
50 and Cozumel (MX-SE) are at risk of SLR with an estimated economic impact of US$1.4 �2.3B (Ruiz-
51 Ram韗ez et al., 2019) (14.5.7.1.12). Negative effects of the "coastal squeeze" phenomena (generated by SLR,
52 land subsidence, sediment deficit, and current urbanization processes) have been documented on tourist
53 destinations along the coasts of Mexican Gulf of Mexico and Mexican Caribbean. Zoning, limiting
54 urbanization along the coastline, and using nature-base solutions (Box 14.7) are alternatives that could be
55 applied to improve the adaptation of these destinations (Mart韓ez et al., 2014; Salgado and Luisa Martinez,
56 2017; Lithgow et al., 2019).

57

Do Not Cite, Quote or Distribute  14-48       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Rural low-lying coastal areas are at risk from SLR where natural barriers or shoreline infrastructure are
 2 deteriorating and this interacts with remoteness, resource-dependent economies, and socioeconomic
 3 challenges to adaptive capacity (Bhattachan et al., 2018; Manson et al., 2019). The Northeast Atlantic region
 4 of North America (CA-AT, US-NE) is exposed to high risk by combined effects of land subsidence and
 5 climate-driven SLR (Lemmen et al., 2016; Sweet et al., 2017; Fleming et al., 2018; Greenan et al., 2018)
 6 (Box 14.4).

 7

 8 14.5.5.3 Adaptation

 9

10 In North American cities, present-day adaptation responses extend beyond the traditional focus on
11 infrastructure to include measures aimed to protect people, property, and ecosystems (medium confidence).
12 Barriers to adaptation include challenges related to the local physical and environmental setting, effects of
13 colonialism and racism, socioeconomic attributes of the population, institutional frameworks, and competing
14 interests of city stakeholders (medium confidence). Current scale of adaptation is generally not
15 commensurate with reducing risks from projected climatic hazards, although resources exist that provide
16 guidance and examples of effective adaptation (medium confidence). Some remote Canadian communities
17 have demonstrated strengths (e.g., strong social networks) that support resilience to climate change (Kipp et
18 al., 2020; Vodden and Cunsolo, 2021). In some US cities with political resistance to action on climate
19 change, adaptation measures focused on addressing extreme events (rather than climate change impacts)
20 were able to make progress (Hamin et al., 2014). Enhanced public awareness of the risks from extreme
21 events associated with climate change is important for motivating adaptation (Howe et al., 2019) (14.3) and
22 developing a climate change agenda (Arag髇-Durand, 2020).

23

24 Community-level planning tailors adaptation responses and disaster preparedness to the local context but
25 misalignment of policies within and between levels of government can prevent implementation (Oulahen et
26 al., 2018). Coordination, planning, and national support are needed to provide sufficient financial resources
27 to implement climate-resilient policies and infrastructure (USGCRP, 2018) (see 14.7.3).

28

29 Public health measures to address extreme heat events are more common across North America, with a focus
30 on vulnerable populations (e.g., City of Toronto, 2019) and innovative approaches for reaching at-risk
31 populations with an overarching intent of prevention (medium confidence) (Guilbault et al., 2016) (14.4.6.1).
32 The heatwave plan for Montreal includes visits to vulnerable populations, cooling shelters, monitoring of
33 heat-related illness, and extended hours for public pools (Lesnikowski et al., 2017); efforts have reduced heat
34 wave-related mortalities (Benmarhnia et al., 2016).

35

36 Other adaptation responses to reduce temperature effects include modifying structures (roofs, engineered
37 materials) and the urban landscape through green infrastructure (e.g., urban trees, wetlands, green roofs),
38 which increases climate resilience and quality of life by reducing urban heat effects, while additionally
39 improving air quality, capturing stormwater, and delivering other co-benefits to the community (e.g., access
40 to food, connection to nature, social connectivity) (Ballinas and Barradas, 2016; Emilsson and Sang, 2017;
41 Kabisch et al., 2017; Krayenhoff et al., 2018; Petrovic et al., 2019; Schell et al., 2020) (Box 14.7) (high
42 confidence). Green infrastructure can be flexible and cost-effective (Ballinas and Barradas, 2016; Emilsson
43 and Sang, 2017; Kabisch et al., 2017). Initiatives can be "bottom-up" community-led adaptation with support
44 from municipal governments, (e.g., East Harlem, New York City) (Petrovic et al., 2019). Valuing municipal
45 natural assets (e.g., assigning economic value to cooling from urban forests or stormwater retention by urban
46 wetlands) is becoming increasingly common in Canada and the US (Wamsler, 2015; Roberts et al., 2017a;
47 Municipal Natural Assets Initiative, 2018). Guidance assists municipalities to identify, value, and account for
48 natural assets in their financial planning and asset management programs (O'Neil and Cairns, 2017) and
49 consider future climate (Municipal Natural Assets Initiative, 2018).

50

51 Meeting increasing demand for indoor space cooling with equitable access, requires new approaches to
52 providing cooling (e.g., equipment efficiencies, refrigerants with lower global warming potential) and
53 electricity production and transmission innovation (Shah et al., 2015; IEA, 2018a). While energy efficiency
54 and building code standards are not directly established by local governments, they can encourage behaviour
55 change via incentives (e.g., rebates on efficient equipment) or disincentives (e.g., more onerous permit
56 approvals).

57

Do Not Cite, Quote or Distribute  14-49       Total pages: 157
    FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

1 Experiences with droughts, heat waves and other weather extremes has led many municipal water managers

2 to accept the importance of building resilience to the risks of future water shortages and costs posed by

3 climate change (Metro Vancouver, 2016; Misra et al., 2021; WUCA, 2021). In the US SW, water utilities

4 have introduced demand-management programs to encourage water conservation (e.g., tiered pricing,

5 incentives for water-efficient appliances and fixtures, and rewards for replacing water-guzzling lawns with

6 water-thrifty native vegetation) (Luthy et al., 2020; Baker, 2021) (14.5.3.3). Water providers also have

7 increased their adaptive capacity by diversifying water sources (Hanak et al., 2015).

8

9 Adaptation to the risks of wildland-urban interface fire is underway (Kovacs et al., 2020) (Box 14.2) but the

10 scope of adaptation required to sufficiently minimize wildfire risks for cities and settlements across North

11 America has not been assessed (medium confidence). Leadership at the local level is increasingly supported

12 by federal resources that provide guidance on hazard and exposure assessment, property protection,

13 community resilience and emergency planning (National Research Council of Canada, 2021).

14

15 Cities and settlements in North America can be susceptible to multiple flooding hazards (i.e., coastal SLR,

16 pluvial, fluvial); each presents unique adaptation challenges that can be addressed through structural (e.g.,

17 armouring coastlines, reservoirs, levees, floodgates; New York City commuter tunnels) and non-structural

18 approaches (e.g., land use planning and zoning, expanding green infrastructure; Chetumal, Mexico) (Hardoy

19 et al., 2014) (high confidence). Green infrastructure practices (Box 14.7) (e.g., open space preservation,

20 floodplain restoration, urban forestry, de-channelization of streams) can reduce urban flooding, erosion, and

21 harmful runoff (Kovacs et al., 2014; Angel et al., 2018b; Government of Canada, 2021c). Structural

22 approaches have limitations and require trade-offs that could be addressed with a focus on socio-ecological

23 solutions and stronger institutional coordination (e.g., flood risk management in Mexico City) (Arag髇-

24 Durand, 2020). In response to high intensity rainfall events, Mexico City invested in stormwater

25 infrastructure, although additional benefits could have been realized if water supply needs had been

26 incorporated (de Alba and Castillo, 2014). Some programs exist to facilitate stormwater and wastewater

27 infrastructure updating to accommodate increased precipitation across North America. The US federal Clean

28 Water State Revolving Fund, provides low-interest loans for states to upgrade infrastructure for climate

29 change, with US$42B provided since 1987 (ASCE, 2019). In Canada, local governments are important

30 leaders in managing engineered and green infrastructure decisions, incentivizing property-level flood

31 protection, and ensuring service delivery (Government of Canada, 2021c). The civil engineering profession

32 is playing an active role in facilitating an understanding of risks and prioritization of adaptation investments

33 in communities (Tye and Giovannettone, 2021).The high concentration of valuable assets in cities requires

34 mechanisms to facilitate replacement of assets including use of existing and proposed insurance mechanisms

35 (14.7) (medium confidence).

36

37 Adaptation planning and implementation to address SLR and coastal flooding has been initiated across cities

38 and settlements in North America but varies in preparedness (high confidence) (Box 14.4). Efforts are

39 supported by SLR design guidelines. In Canada, the Government of British Columbia provided SLR

40 projections for 2050 (i.e., +0.5m) and 2100 (i.e., +1m) in order to initiate community vulnerability and risk

41 assessment, and adaptation planning (The Arlington Group Planning + Architecture Inc et al., 2013). Based

42 on recent hurricane impacts in Yucatan, Mexico, recommendations to enhance the rules governing the

43 Mexican Recovery Program included incorporating local and Indigenous knowledge when rebuilding houses

44 and other structures on coasts (Audefroy and Cabrera S醤chez, 2017). Where adaptation in-place is

45 insufficient, planned retreat is being considered as an sustainable option for reducing future risks (Saunders-

46 Hastings et al., 2020).

47

48

49 [START BOX 14.4 HERE]

50

51 Box 14.4: Sea Level Rise Risks and Adaptation Responses for Selected North American Cities and

52  Settlements

53

54 Approximately 95 million Americans lived in coastal communities in 2017 (US Census Bureau, 2019) and in

55 2013, Canada had roughly 6.5 million coastal residents (Lemmen et al., 2016), while Mexico had 19 million

56 people living in coastal municipalities in 2015 (Azuz-Adeath et al., 2018). Sea level rise around North

57 American coastlines (Figure Box14.4.1) is projected to be greatest along the coasts of Atlantic Canada,

    Do Not Cite, Quote or Distribute  14-50                                                  Total pages: 157
    FINAL DRAFT                                         Chapter 14     IPCC WGII Sixth Assessment Report

1 northern Gulf of Mexico for the US, and the Pacific coast of Mexico (IPCC, In Press). Sections 14.5.2.1,
2 14.5.5.1.3, 14.5.5.2.3 describe SLR impacts. Status of adaptation to SLR by local governments is variable
3 (see Table Box 14.4.1, where progress is indicated by colour coding) and ranges from financed
4 implementation to preliminary/preparatory/scoping studies and workshops. Adaptation planning and
5 implementation to address SLR and coastal flooding have been initiated across many cities and settlements
6 in North America but preparedness varies (high confidence).

7
8

9

10 Figure Box 14.4.1: Sea Level Rise (SLR) projections for 2050, 2100 and 2150 for selected North American cities.

11 Projections changes are relative to 2005, which is the central year for the 1994-2014 reference period. Horizontal lines
12 in the boxes represent the median projection, boxes represent 25th to 75th percentile and whiskers the 10th to 90th

13 percentile of SLR projections from all CMIP6 models as well as other lines of evidence (see Table 9.7 in WGI.9 for

14 more details). Two SLR scenarios are provided for lower (SSP 126) and higher emissions (SSP 585), and are consistent

15 with WGI AR6 Interactive Atlas. Numbers and colors (see Table Box14.4.1 for detailed readiness definitions) on the

16 map and in the projections represent sites and status of SLR adaptation progress. Information supporting SLR

17 adaptation status is summarized in Table Box14.4.1.

18

19

20 Table Box14.4.1. Status of adaptation actions associated with locations on SLR map, colour-coded according to level
21 of SLR preparedness through Adaptation (as discoverable on government websites): No climate change adaptation

22 action plan but processes underway such as workshops, studies, vulnerability assessments (red), General Climate

23 Change Adaptation Action Plan which mentions SLR as a risk/issue/impact but no concrete actions developed (orange),

24 Specific Plan for SLR but does not include specific actions (light blue), Specific Plan for SLR with concrete actions

25 identified but no evidence of actions taken to date (med blue), and Specific Plan for SLR with evidence of progress on

26 taking actions including allocating funding for projects (purple).

    Ocean Site Area/City Exposure (not exhaustive) Does the area/city have an Adaptation Plan for SLR? If

    Basin #                                              so, are they taking actions to implement it? (Status)

    Arctic 1     Tuktoyuktuk, Infrastructure, municipal  Tuktoyaktuk Coastal Erosion Study completed March
                                                         2019. Additional investments in both planning and
                 CA  services, transportation,           actual adaptation measures have occurred. Limited
                                                         financial resources remain a barrier. (Government of
                     homes, 900 people                   Canada, 2020)

    Do Not Cite, Quote or Distribute                     14-51         Total pages: 157
FINAL DRAFT                                           Chapter 14           IPCC WGII Sixth Assessment Report

Atlantic 2      Prince       PEI: residential, industrial Prince Edward Island government released a five year
            3   Edward
                Island       and commercial             climate change action plan in 2018 which includes both
            4   with Lennox
            5   Island, CA   infrastructure.            adaptation and mitigation (Prince Edward Island

            6                Lennox Island: 10 out of Government, 2018). Biennial progress reports were
            7
                             79 homes, causeway to the issued (Prince Edward Island Government, 2019). The
            8
Gulf of 9                    island, sacred grounds, Mi'kmaq community of Lennox Island First Nation has
Mexico
                             sewage treatment systems explored relocation options (Daigle et al., 2015).
            10
                Truro, CA    A regional centre of       Town of Truro, County of Colchester and Millbrook
                             12,000 residents, which    First Nations commissioned a flood risk study 2014�
                             has been vulnerable to     2017 (CBCL, 2017; Sherren et al., 2019) triggered by
                             repeated floods for        the 2012 flooding. Outcome was Truro-Onslow dyke
                             decades.                   project -- a voluntary retreat with realignment of dyke
                                                        infrastructure and habitat restoration by conversion of
                                                        agricultural land into salt marsh habitat (Saunders-
                                                        Hastings et al., 2020).

                Halifax, CA  Transportation causeways   HalifACT 2050 is a comprehensive plan adopted as of
                             and bridges, marine        2020 by the Halifax regional council which includes
                             facilities, municipal      reducing GHGs and adapting to climate change
                             infrastructure.            including a coastal preparedness section 5.2.9. (Halifax
                                                        Regional Council, 2020)

                New York,    20 million people at risk New York City has developed many adaptation plans
                US
                             by 2050; 40% of water for sustaining NYC in light of SLR and other climate

                             treatment plans will be hazards/impacts, especially since Hurricane Sandy

                             compromised by flooding, affected the city in 2012. It is unclear how much of the

                             60% of power plants will planning has moved forward into implementation

                             need to be relocated,      (NYC, 2013; New York City, 2015; NYC Mayor's

                             transportation systems will Office of Resiliency, 2020).

                             need to be upgraded to

                             avoid flooding

                Norfolk, US  Homes, massive US naval City of Norfolk published a very specific Coastal

                             base, shipyards, active Resilience Strategy in 2014. Capital improvement

                             waterfront, and deep water projects highlighted in this strategy have been funded

                             ports                      (City of Norfolk Virginia, 2014). Plan for protecting

                                                        Naval base and shipyard not evident.

                Miami, US    Homes, port,               Miami Dade County released a specific SLR Strategy
                             transportation             in 2021. Actions in the plan include elevating roads
                             infrastructure, tourism    and other infrastructure, designing ways to
                             (hotels, restaurants,      accommodate more water in and around buildings,
                             beaches)                   building on higher ground and expanding waterfront
                                                        parks and canals. The plan includes a map with current
                                                        and planned adaptation projects in the county (Miami-
                                                        Dade County, 2021).

                Cancun, MX   Tourism infrastructure     2013 Climate Change plan assigns adaptation in
                             (hotels, restaurants,      general to different government levels. No evidence of
                             beaches), homes, markets,  specific adaptation plan for SLR (Government of
                             service industry,          Quintana Roo, 2013)
                             transportation.

                New Orleans, Entire city, especially low- City of New Orleans adaptation is incorporated in the

                US           lying, low-income areas, is broader Louisiana coastal climate change adaptation

                             vulnerable as evidenced plan (CPRA, 2023). The process includes very specific

                             by Hurricane Katrina in projects with updates on risk based implementation.

                             2005.

                Ciudad del Freshwater access, 11,000 The Campeche State Climate Change plan was released

                Carmen, MX homes, aquaculture           in 2013 (Government of Campeche, 2013). The plan

                                                        does not include any specific recommended actions to

                                                        adapt to SLR in Cuidad del Carmen. Flood risk maps

                                                        for Ciudad del Carmen were created in 2011

                                                        (Audefroy, 2019).

Do Not Cite, Quote or Distribute                      14-52                                   Total pages: 157
FINAL DRAFT                                        Chapter 14  IPCC WGII Sixth Assessment Report

11 Veracruz,               Freshwater access, sewage State of Veracruz published a Climate Change plan in
        MX
                           treatment systems,          2008 (Government of Veracruz, 2008). Plan includes

                           electrical and              specific tables of actions needed to monitor and adapt

                           petrochemical industries to SLR. World Bank funded coastal adaptation in

                                                       Veracruz focused on mangroves to dissipate storm

                                                       surge but no investments in infrastructure to mitigate

                                                       SLR.

Pacific 12   Unalaska, US Loss of cultural resources, Climate Change Adaptation and Vulnerability
                               salinization of rivers/lakes, Assessment Workshops have been held with discussion
                                                                of coastal erosion. SLR not viewed as important as
                                                                impacts from sea ice and permafrost loss (Poe et al.,
                                                                2016).

13 Surrey                  Disruption in flow of       Surrey has a Coastal Flood Adaptation Strategy

             (Greater      goods in/out of Port of (CFAS) approved by Council (City of Surrey, 2019)

             Vancouver Vancouver,                      with 46 actions (policy and program, local area

             Area),CA communication facilities, infrastructure). Some local area infrastructure

                           road, rail and air          improvements received capital funding.

                           transportation

                           infrastructure, businesses

                           and agriculture.

14 Seattle, US Low-lying areas, near- Seattle released a Climate Change Response Plan in

                           shore habitats, stormwater 2017 which includes general approaches including

                           drains, roads, homes,       development of risk maps for SLR which are also

                           businesses, socially        available online (City of Seattle, 2017).

                           vulnerable communities.

15 Quinault 650 residents and                          Quinault Indian Reservation has a plan to move Tahola
                                                       to higher ground, one half mile from the existing
             Indian        buildings.                  village (EPA, 2021).

             Reservation

             (Tahola), US

16 San                     37,200 residents, 17,200 SF has an active, SLR planning process as well as an

             Francisco, businesses and 167,300 iterative Sea Level Rise Action Plan (City of San

             US            jobs are vulnerable to      Francisco, 2016), planning tools and iterative

                           inundation by 2100 at assessment (City and County of San Francisco,

                           upper bounds of SLR, 2020).The process specifically addresses wastewater,

                           mostly along the bay side water, transportation, power, public safety, open space,

                           of the city.                port, neighborhoods and changing shoreline.

17 Los Angeles, Power plants, wastewater Los Angeles has commissioned a projected SLR

             US            treatment plants, Port of impact report but not an action plan. The Port of Los

                           Los Angeles, beaches, Angeles is particularly vulnerable and, as of 2019, has

                           tourism                     a SLR Adaptation Plan (Newbold et al., 2019).

18 Acapulco,               Tourism infrastructure No climate change plan exists although the Mexican
        MX
                           (hotels, restaurants        Tourism Sector conducted a climate change

                           beaches), homes, markets, vulnerability assessment covering Acapulco (Guerrero,

                           service industry,           2017).

                           transportation.

 1

 2 [END BOX 14.4 HERE]

 3
 4

 5 14.5.6 Health and Wellbeing

 6

 7 Research examining climate change impacts on human health in North America has increased substantially
 8 since AR5 (Harper et al., 2021a). Using a systematic approach (Harper et al., 2021b), the assessment focused
 9 on advancements since AR5.

10

11 14.5.6.1 Heat-Related Mortality and Morbidity

12

13 High temperatures currently increase mortality and morbidity in North America (very high confidence), with
14 impacts that vary by age, gender, location, and socioeconomic factors (very high confidence). Observed

Do Not Cite, Quote or Distribute                       14-53                                      Total pages: 157
    FINAL DRAFT                                    Chapter 14  IPCC WGII Sixth Assessment Report

1 increases in heat-related mortality have been attributed to climate change in North America (Vicedo-Cabrera

2 et al., 2021). Temperature effects on health vary based on how unusual the temperature is for that time and

3 location (medium evidence, high agreement), highlighting the important role that temperature extremes and

4 variability play in mortality and morbidity (Li et al., 2013; Lee et al., 2014; Barreca et al., 2016; Allen and

5 Sheridan, 2018). Adaptation has played an important role in reducing observed heat-related deaths (Vicedo-

6 Cabrera et al., 2018b).

7

8 Rising temperatures are projected to increase heat-related mortality across emission scenarios this century in

9 North America (very high confidence), although the magnitude of increase varies geographically (Isaksen et

10 al., 2014; Petkova et al., 2014; Wu et al., 2014; Weinberger et al., 2017; Anderson et al., 2018a; Limaye et

11 al., 2018; Marsha et al., 2018; Morefield et al., 2018). The elderly (Isaksen et al., 2014; Limaye et al., 2018)

12 and urban areas (Limaye et al., 2018) are projected to experience the greatest increase in heat-related

13 mortality this century. Warming temperatures are also projected to increase heat-related morbidity (medium

14 confidence). For instance, the incidence and treatment costs of asthma attributed to warmer temperatures are

15 projected to increase in Texas by 2040-2050 (A1B) (McDonald et al., 2015).

16

17 While heat-related mortality is projected to increase across emissions scenarios and shared socio-economic

18 pathways, fewer deaths are projected under both lower emissions scenarios and higher adaptation scenarios

19 in North America (very high confidence). Heat-related mortality was projected to be 50% less under RCP4.5

20 compared to RCP8.5 in the US for SSP3 and SSP5 (Wu et al., 2014; Marsha et al., 2018) (Table 14.5).

21

22

23 Table 14.5: A summary of adaptation options for different health outcomes in North America.

    Health outcome             Adaptation Options

    Heat-related mortality Future temperature-related health impacts can be reduced by adaptation measures

    and morbidity              (Petkova et al., 2014; Wu et al., 2014; Mills et al., 2015b; Kingsley et al., 2016;

                               Anderson et al., 2018b; Marsha et al., 2018; Morefield et al., 2018), including more

                               effective warning and response systems and building designs, enhanced pollution

                               controls, urban planning strategies, and resilient health infrastructure (very high
                               confidence) (Figure Box 14.7.1).

    Wildfire-relate mortality  Air quality indices are correlated with many respiratory conditions (Yao et al., 2013;
                               Hutchinson et al., 2018), suggesting that providing air quality information to the public
                               could reduce smoke-related health impacts (Yao et al., 2013; Rappold et al., 2017).
                               Enhanced coordination between the health sector and fire suppression agencies can also
                               reduce the health impacts of wildfire smoke via improving communication, weather
                               forecasting, mapping, fire shelters, and coordinated decision making (Withen, 2015),
                               including transnational and cross-jurisdictional actions.

    Vectorborne disease        Prevention of vectorborne disease currently involves surveillance, reducing
                               environmental risks, and promoting individual behaviours to reduce human-vector
                               contact. Top ranked Canadian West Nile interventions include individual protection (i.e.,
                               window screens, wearing lightly coloured clothing), and regional management and
                               mosquito-targeting interventions (i.e., larvicides, vaccination of animal reservoirs,
                               modification of human-made larval sites) (Hongoh et al., 2016).

    Waterborne disease         Climate change is projected to increase waterborne disease risks (medium confidence),
                               particularly in areas with aging water and wastewater infrastructure in North America
                               (high confidence). In Wisconsin, US, precipitation changes are projected to increase
                               gastrointestinal illness in children this century (A1B, A2, B1) (Uejio et al., 2017). Slight
                               reductions in precipitation-associated gastrointestinal illness is projected if water
                               treatment infrastructure is upgraded slowly over time; however, if water treatment
                               infrastructure is installed more rapidly, large decreases in precipitation-associated
                               gastrointestinal illness incidence are projected (Uejio et al., 2017), highlighting the
                               benefits of rapidly implementing adaptation actions.

    Foodborne disease          Food safety programs play important roles in reducing the risk of climate-related
                               foodborne disease (high confidence). Integrated health surveillance, more stringent
                               refrigeration temperature controls to limit pathogen growth, targeted communication to
                               public and food sector, and enhanced coordination between health and food sectors can
                               reduce risk (Hueffer et al., 2013; Jones et al., 2013; Fillion et al., 2014; Doyle et al.,

    Do Not Cite, Quote or Distribute               14-54                                        Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report
 Mental health
                2015). In Mexico, the projected risk of Vibrio parahaemolyticus in oysters was 11 times
                higher in a high emissions scenario compared to a low emissions scenario by the end of
                the century; however, this risk could be substantially lowered with adaptation measures,
                including improving temperature control (Ortiz-Jim閚ez, 2018).

                Effectiveness of individual and/or group therapy, and place-specific mental health
                infrastructure, to treat mental health challenges is well-proven; yet, there is limited
                evidence evaluating these interventions within the context of climate change (e.g.
                Tschakert et al., 2017; Young et al., 2017b; Cunsolo and Ellis, 2018).

 1
 2

 3 14.5.6.2 Cold-Related Mortality

 4

 5 Winter season mortality rates are generally high in high income regions such as North America, with most of
 6 that mortality due to cardiovascular diseases (Ebi and Mills, 2013). It is important to differentiate between
 7 mortality related to cold temperatures and mortality due to other factors that vary with season (Ebi and Mills,
 8 2013; Ebi, 2015). Warmer temperatures do not always equate to lower winter mortality: many cold-related
 9 deaths do not occur during the coldest times of year or in the coldest places (high confidence) but occur
10 during the beginning or end of the winter season (Barnett et al., 2012; Lee et al., 2014; Schwartz et al., 2015;
11 Sarofim et al., 2016b; Smith and Sheridan, 2019). Warmer US cities generally experience more mortality
12 from extreme cold events and cold temperatures than colder cities in the US and Canada (Lee et al., 2014;
13 Gasparrini et al., 2015; Schwartz et al., 2015; Wang et al., 2016; Smith and Sheridan, 2019). While mortality
14 rates linked to direct cold-exposure (e.g. hypothermia, falls, and fractures) is generally low, the relatively
15 higher mortality during milder temperatures is thought to be largely due to respiratory infections and
16 cardiovascular impacts (Lee et al., 2014; Gasparrini et al., 2015), which, although correlate with temperature,
17 may not be caused by cold temperatures (Ebi and Mills, 2013; Ebi, 2015; Sarofim et al., 2016a). When
18 separating the effects of cold temperatures from the effects of the winter season, one study found cold
19 temperature did not drive mortality and suggested that winter season excess mortality was due to seasonal
20 factors other than temperature (e.g. influenza, seasonal gatherings) (Kinney et al., 2015).

21

22 Mortality attributed to cold temperatures has increased in the US and remained stable in Canada from 1985-
23 2012 despite increasing winter temperatures (Vicedo-Cabrera et al., 2018b). Some attenuation in cold-related
24 mortality in Mexico and warmer US states is projected under climate change, but less so in colder climates in
25 north-eastern US and Canada, with statistically insignificant trends in some regions and increasing cold-
26 related mortality in other regions (Li et al., 2013; Mills et al., 2015b; Schwartz et al., 2015; Sarofim et al.,
27 2016a; Wang et al., 2016; Gasparrini et al., 2017; Vicedo-Cabrera et al., 2018a; Lee et al., 2019). These
28 reductions in cold-mortality are generally considered relatively small.

29

30 Observed and projected trends in winter mortality highlight that non-climate factors may have a greater role
31 in driving winter mortality than cold temperature, and that these deaths are expected to occur with or without
32 climate change (Ebi and Mills, 2013; Ebi, 2015; Sarofim et al., 2016a). This challenges the assumption that
33 warmer winters due to climate change would dramatically lower winter season mortality (medium evidence,
34 medium agreement).

35

36 14.5.6.3 Wildfire-Related Morbidity

37

38 Smoke from intensified wildfire activity in North America is associated with respiratory distress (very high
39 confidence), and persists long distances from the wildfire and beyond the initial high-exposure time period
40 (Hutchinson et al., 2018)(Box 14.2). Exposure to wildfire smoke increases hospital admissions (McLean et
41 al., 2015; Alman et al., 2016; Reid et al., 2016; Yao et al., 2016; Rojas-Downing et al., 2017). Increased
42 wildfire smoke from climate change is projected to result in more respiratory hospital admissions in the
43 Western US by 2046-2051 (A1B) (Liu et al., 2016; Rojas-Downing et al., 2017).

44

45 The magnitude of health risks varies by age (Le et al., 2014; Reid et al., 2016; Liu et al., 2017a; Liu et al.,
46 2017b), gender (Delfino et al., 2009; Rojas-Downing et al., 2017), socio-economic conditions (Henderson et
47 al., 2011; Rappold et al., 2012; Reid et al., 2016), and underlying medical conditions (Liu et al., 2015). The
48 intersectionality of these subgroups plays an important role in health-related vulnerability to wildfire smoke.

Do Not Cite, Quote or Distribute  14-55       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Among the elderly in the western US, risks of respiratory admissions from wildfire smoke was significantly
 2 higher for African American women in lower-education counties (Liu et al., 2017b). For Indigenous Peoples,
 3 medical visits for respiratory distress, heart disease, and headaches increased during a wildfire in California
 4 (Lee et al., 2009). In Northern Canada, Indigenous livelihoods were disrupted during a wildfire, which
 5 negatively impacted mental, emotional, and physical health (Dodd et al., 2018a; Howard et al., 2021).

 6

 7 14.5.6.4 Vectorborne Disease

 8

 9 Climate change creates conditions that enable earlier seasonal activity and general northern expansion of
10 ticks (Ogden et al., 2014), increasing human exposure to tickborne diseases in North America (very high
11 confidence). Lyme disease incidence and geographic extent has already increased in Canada and the US
12 (Eisen et al., 2016), which has been associated with climate change (Ogden et al., 2014), including warmer
13 temperature (Cheng et al., 2017; Lin et al., 2019). Climate change is projected to increase disease spread into
14 new geographic regions, lengthen the season of disease transmission, and increase tickborne disease risk in
15 North America across emissions scenarios throughout this century (very high confidence), with regional
16 variability (Roy-Dufresne et al., 2013; Feria-Arroyo et al., 2014; Monaghan et al., 2015; Robinson et al.,
17 2015; McPherson et al., 2017). Chagas disease is transmitted by triatomines, and most of the Mexican
18 population (88.9%) already reside in areas with at least one infected vector species in both rural and urban
19 populations (Carmona-Castro et al., 2018). Chagas has already extended its range into the southern US, and
20 the triatomines' niche is projected to expand northward this century (Garza et al., 2014; Carmona-Castro et
21 al., 2018) in both rural and urban areas (Carmona-Castro et al., 2018).

22

23 Climate change is projected to impact the distribution, abundance, and infection rates of mosquitoes in North
24 America (high confidence), which will increase risk of mosquito-borne diseases including West Nile Virus,
25 chikungunya, and dengue (medium confidence). The geographic distribution of West Nile virus is projected
26 to expand in North America this century (A1B) (Harrigan et al., 2014). In the US and Canada, mosquitoes
27 are projected to emerge earlier in the year and remain active longer into the fall; however, mosquito
28 population dynamics vary by location with northern locations projected to have an increased vector
29 abundance, and currently hot areas may become too hot, thus negatively affecting mosquito survival (A2,
30 A1B, B1) (Chen et al., 2013; Morin and Comrie, 2013; Brown et al., 2015a).

31

32 Local transmission of chikungunya virus has emerged in Mexico and the US since AR5, and areas suitable
33 for transmission are projected to expand (RCP4.5, RCP8.5) (Tjaden et al., 2017). Although chikungunya
34 virus is not currently in Canada, climate change is projected to make southern British Columbia suitable for
35 virus transmission this century, particularly under RCP8.5 (Ng et al., 2017).

36

37 The dengue mosquito vector is well-established in Mexico and southeastern US. In northwestern Mexico,
38 incidence of dengue cases is associated with minimum monthly temperature (Diaz-Castro et al., 2017), and
39 the geographic range of the vector in the US is restricted, in part, by low temperatures. Thus, a northward
40 range expansion is projected; however, future dengue risk also depends on built environments and
41 competition with other mosquito species (Col髇-Gonz醠ez et al., 2013a; Eisen and Moore, 2013). Climate
42 change is projected to increase the geographic range and extend the seasonal activity of the dengue vector in
43 the southern US by 2045-2065 (A1B); however, transmission is projected to be limited by low winter
44 temperatures in the mainland US, potentially preventing its permanent establishment (Butterworth et al.,
45 2017). In Mexico, increased dengue cases are projected this century (A1B, A2, B1) (Col髇-Gonz醠ez et al.,
46 2013b).

47

48 14.5.6.5 Waterborne Disease

49

50 Heavy precipitation events are associated with contaminated drinking water and waterborne disease in North
51 America (high confidence). Acute gastrointestinal illnesses increase with many hydro-climatological
52 variables, including precipitation, streamflow, and snowmelt (Harper et al., 2011; Wade et al., 2014; Galway
53 et al., 2015). Extreme precipitation is associated with Campylobacter and Salmonella infections in the US,
54 particularly in counties characterized by farms and private well water (Soneja et al., 2016). In Canada,
55 human Giardia infections are associated with increased temperature, precipitation, pathogen presence in
56 livestock manure, and river water level and flow (Brunn et al., 2019). Land-use patterns and aquafer-types

Do Not Cite, Quote or Distribute  14-56       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 are associated with waterborne disease, and ecological zones with higher waterborne rates are projected to
 2 expand in range by 2080 in Canada (Brubacher et al., 2020).

 3

 4 In North America, stormwater and water treatment infrastructure play important roles in reducing waterborne
 5 disease risk during precipitation events (high confidence). In the US, heavy precipitation events are
 6 associated with higher rates of childhood gastrointestinal illness in municipalities with untreated drinking
 7 water, but not in municipalities with treated drinking water (Uejio et al., 2014). In Mexico, disparities in
 8 access to treated water are a key determinant of under age-5 morbidity (Jim閚ez-Mole髇 and G髆ez-
 9 Albores, 2011; Romero-Lankao et al., 2014a). In remote communities in Alaska and Northern Canada,
10 challenges in water service provision and maintenance can increase risk of waterborne disease during high
11 impact weather events (Harper et al., 2011; Bressler and Hennessy, 2018; Harper et al., 2020). In older
12 sections of many North American cities sewage treatment plant capacity is exceeded by overflow of
13 combined sanitary and storm sewer systems during heavy precipitation events, resulting in bypass of
14 untreated and microbiologically contaminated wastewater discharge into drinking water sources (Jagai et al.,
15 2017; Olds et al., 2018; Staley et al., 2018). These sewer overflow events are associated with increased
16 gastrointestinal illness across age groups (Jagai et al., 2017).

17

18 14.5.6.6 Foodborne Disease

19

20 Warmer air temperature, changes in precipitation, extreme weather events, and ocean warming can increase
21 microbial pathogen loads in food (very high confidence). Indeed, temperature and extreme weather are top
22 factors influencing food safety in Canada (Charlebois and Summan, 2015). Outbreaks of Vibrio
23 parahaemolyticus have been associated with the consumption of raw oysters harvested from higher-than-
24 usual ocean temperatures in Canada and Alaska (McLaughlin et al., 2005; Taylor et al., 2018). Warmer air
25 temperature increases Campylobacter, Salmonella, and E. coli prevalence in Canadian meat products (Smith
26 et al., 2019), higher microbial load in American produce (Ward et al., 2015), and increased Campylobacter
27 spp., pathogenic E. coli, and Salmonella spp. infections in humans (Akil et al., 2014; Valcour et al., 2016;
28 Uejio, 2017).

29

30 Climate change is projected to increase food safety risks (medium confidence); however, the actual burden of
31 foodborne disease will depend on the efficacy of public health interventions (high confidence). Increased
32 ciguatera fish poisoning is associated with increased SSTs and tropical storm frequency, and this risk is
33 projected to increase this century (Gingold et al., 2014). Campylobacter infection in humans due to food
34 contamination from flies is projected to increase this century in Canada (Cousins et al., 2019), and increased
35 housefly populations are projected this century in Mexico (Meraz Jimenez et al., 2019). Climate change may
36 also lead to new emerging foodborne disease risks. For instance, V. cholerae is a pathogen previously
37 restricted to tropical regions; however, due to warming ocean temperatures, its detection has significantly
38 increased along Canadian coasts (Banerjee et al., 2018).

39

40 Climate change is projected to increase human foodborne exposure to chemical contaminants (medium
41 confidence). Increases in SST have been associated with greater accumulation of mercury in seafood, marine
42 mammals, and fish (Ziska et al., 2016). This particularly increases food safety risks in the Arctic, with
43 methylmercury and polychlorinated biphenyl (PCB) concentrations in high trophic animals projected to
44 increase under high emission scenarios by 2100 (Alava et al., 2017; Alava et al., 2018).

45

46 Climate-related foodborne disease risks vary temporally, and are influenced, in part, by food availability,
47 accessibility, preparation, and preferences (medium confidence). For example, seafood risks are more
48 pronounced in coastal regions due to high seafood consumption (Radke et al., 2015). In Alaska and Northern
49 Canada, where locally harvested foods are critical to diet, climate change may introduce new pathogens to
50 local food sources through wildlife range changes, warming temperatures affecting safe fermentation and
51 drying preparation methods, and food temperature control in belowground cold storage in or near permafrost
52 (King and Furgal, 2014; Harper et al., 2015; Rapinski et al., 2018).

53

54 14.5.6.7 Nutrition

55

56 Agricultural productivity declines due to climate change (14.5.4) are projected to lower caloric availability
57 and increase the prevalence of underweight people and climate-related deaths in North America by 2050

Do Not Cite, Quote or Distribute  14-57       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 (IMPAACT) (Springmann et al., 2016a; Springmann et al., 2016b; Springmann et al., 2018); however, this
 2 lower caloric availability could also reduce obesity, which could result in deaths avoided (Springmann et al.,
 3 2016a; Springmann et al., 2016b). The climate-related deaths per capita due to reduced fruit and vegetable
 4 consumption is projected to exceed the mortality due to reduced caloric intake in North America by 2050,
 5 particularly in Canada and US (Springmann et al., 2016a; Springmann et al., 2016b). These climate change
 6 projections underscore the importance of focusing on nutritional security in North America, instead of only
 7 considering caloric intake.

 8

 9 Shifting to a more sustainable diet can have adaptation and mitigation co-benefits while simultaneously
10 improving health outcomes for North Americans. Transitioning to more plant-based diets is projected to
11 reduce climate-related deaths in Canada, US, and Mexico by 2050 (Springmann et al., 2016a; Springmann et
12 al., 2016b), while simultaneously reducing food-related GHG emissions per capita in North America by
13 2050 (Springmann et al., 2018).

14

15 Nutrition impacts will not be experienced uniformly within countries (Shannon et al., 2015; Zeuli et al.,
16 2018). In Alaska and Canada, Indigenous knowledge has documented how climate change has already
17 impacted locally harvested foods and challenged nutrition security (Lynn et al., 2013; Petrasek MacDonald
18 et al., 2013; Harper et al., 2015; Hupp et al., 2015; Bunce et al., 2016) (CCP6). For First Nations coastal
19 communities in western Canada, decreased access to traditionally harvested seafood is projected to reduce
20 nutritional status by 2050 (RCP2.5, RCP8.5), with higher nutritional impacts for men and older adults
21 (Marushka et al., 2019). Substitution of seafood with non-traditional foods (i.e., chicken, canned tuna) would
22 not replace the projected nutrients lost (Marushka et al., 2019), challenging assumptions that market food
23 substitutions could be effective adaptation strategies for Indigenous Peoples.

24

25 14.5.6.8 Mental Health and Wellness

26

27 Climate change has had, and will continue to have, negative impacts on mental health in North America
28 (high confidence) (Figure 14.8). Climate change impacts mental health through multiple direct and indirect
29 pathways stemming from extreme weather events, slower, cumulative events, and vicarious or anticipatory
30 events (Cunsolo Willox et al., 2013; Cunsolo Willox et al., 2014; Durkalec et al., 2015; Yusa et al., 2015;
31 Schwartz et al., 2017; Trombley et al., 2017; Burke et al., 2018b; Cunsolo and Ellis, 2018; Dodd et al.,
32 2018b; Hayes et al., 2018; Middleton et al., 2020b). Climate change disruptions to infrastructure, underlying
33 determinants of health, and changing place attachment are also stressors on mental health (Vida et al., 2012;
34 Cunsolo Willox et al., 2013; Burke et al., 2018b; Obradovich et al., 2018).

35

36 In North America, climate change has been linked to strong emotional reactions; depression and generalized
37 anxiety; ecological grief and loss; increased drug and alcohol usage, family stress, and domestic violence;
38 increased suicide and suicide ideation; and loss of cultural knowledge, and place-based identities and
39 connections (Cunsolo Willox et al., 2013; Durkalec et al., 2015; Harper et al., 2015; Fern醤dez-Arteaga et
40 al., 2016; Schwartz et al., 2017; Trombley et al., 2017; Burke et al., 2018b; Cunsolo and Ellis, 2018;
41 Clayton, 2020; Dumont et al., 2020).

42

43 Suicide is projected to increase in Mexico and the US by 2050 due to rising temperatures (RCP8.5) (Burke et
44 al., 2018b) (limited evidence). Literature on climate change and mental health in North America is
45 increasing; however, few population-level quantitative studies exist, although are increasing (e.g. Burke et
46 al., 2018b; Kim et al., 2019; Dumont et al., 2020; Middleton et al., 2021).

47
48

Do Not Cite, Quote or Distribute  14-58       Total pages: 157
      FINAL DRAFT                 Chapter 14  IPCC WGII Sixth Assessment Report

1

2

3 Figure 14.8: Pathways through which climate change impacts mental health risk in North America.

4
5

6 14.5.7 Tourism and Recreation

7

8 Tourism is one of the largest and fastest growing industries in North America, contributing USD$2.5 trillion
9 to North Americas' GDP in 2019 (WTTC, 2018; Duro and Turri髇-Prats, 2019). The US is the world's

Do Not Cite, Quote or Distribute  14-59       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 largest tourism economy (USD$1839 billion contribution to global GDP in 2019), Mexico is ranked 9th
 2 (USD$196 billion) and Canada 13th (USD$108 billion) (WTTC, 2018). The tourism industry is both
 3 impacted by climate change and significantly contributes to it through the emission of GHGs from travel and
 4 activities (Becken and Hay, 2007). By 2060 under RCP8.5 Canada and the US are projected to benefit from
 5 climate-induced changes in tourism expenditures of up to 92% and 21% respectively, whereas Mexico could
 6 experience a 25% decrease (OECD, 2015; Scott et al., 2019a).

 7

 8 14.5.7.1 Observed Impacts and Projected Risks of Climate Change

 9

10 14.5.7.1.1 Alpine and Nordic skiing, snowmobiling and other winter sports
11 Winter tourism activities with hard limits to adaptation, particularly those that occur at sea level where less
12 precipitation is expected to fall as snow (i.e., Nordic skiing, snowmobiling, snowshoeing), are at the highest
13 risk from climate change and may experience irreversible impacts well before 2癈 of warming above pre-
14 industrial levels (high confidence) (Figure 14.9). During record warm winters, alpine ski resorts in eastern
15 Canada experienced reductions in ski season lengths of between 11 and 17 days (Rutty et al., 2017) and
16 resorts in the US Northeast (US-NE) experienced decreased skier visits by 11.6% and reductions in
17 operational profits of 33% amounting to US$40-52 million (Dawson et al., 2009). Even with advanced
18 snowmaking as an adaptation to warmer temperatures, average ski season lengths are projected to decrease
19 8% (RCP2.6, 2050s) to 73% (RCP8.5, 2080s) in Ontario, Canada (CA-ON) (Scott et al., 2019b), 12%
20 (RCP4.5, 2050s) to 22% (RCP8.5, 2080s) in Quebec, Canada (CA-QC), and 13% (RCP 4.5, 2050s) to 45%
21 (RCP 8.5, 2080s) in the US Northeast (US-NE) (Wobus et al., 2017; Scott et al., 2020). Season length for
22 snowmobiling and cross-country skiing is projected to decrease more dramatically (high confidence) by from
23 80% (RCP4.5) to 100% (RCP 8.5) by mid-century (Wobus et al., 2017) (also see CCP5). The number of
24 outdoor skating-days may decrease by 34% in Toronto and Montreal and 19% in Calgary by 2090 under
25 RCP8.5 (Robertson et al., 2015). The skating season length for the Rideau Canal in Ottawa, Canada, a
26 UNESCO heritage site attracting 1.3 million visitors annually, may decrease by 3.8�2.0 days per decade
27 with later opening dates of 2.6�1.5 days per decade (Jahanandish and Alireza, 2019).

28

29 14.5.7.1.2 Beach, coral reef, and protected areas tourism
30 Sea level rise, increased storm surge, wave action, algae blooms, extreme air temperatures, and changes in
31 wind and precipitation patterns threaten coastal tourism infrastructure, submerge beaches, erode walking
32 paths on coasts, and impact destination attractiveness, tourism demand, and recreation economies (very high
33 confidence). Warm weather tourism activities, including beach tourism, snorkelling, and national park
34 visitation will have more time to implement adaptation strategies to reduce climate risks as significant and
35 widespread impacts are not expected until 3 to 4癈 of warming (Fig 14.9) (Rutty and Scott, 2015; Atzori et
36 al., 2018; Santos-Lacueva et al., 2018; Duro and Turri髇-Prats, 2019). Thirty percent of hotels along the Gulf
37 of Mexico and Caribbean Sea are exposed to flooding and 66% are located on eroding beaches (Lithgow et
38 al., 2019). Coral reef cover in Akumal Bay, Mexico decreased by 79% between 2011 and 2014 (Gil et al.,
39 2015; Manuel-navarrete and Pelling, 2015). The recreation value of coral reef tourism in Florida, Puerto
40 Rico, and Hawai'i is expected to decrease by 90% by mid-century under RCP8.5 (EPA, 2017) (14.4.2).
41 Wildfires and insect outbreaks have contributed to reduced desirability for tourism across forest and
42 mountain regions (Bawa, 2017; Hestetune et al., 2018; White et al., 2020). Visitors to Utah's National Parks
43 declined 0.5 to 1.5% during wildfire years between 1993 to 2015, resulting in US$2.7 to 4.5 million in lost
44 revenue (Kim and Jakus, 2019) (see Box 14.2). Trees damaged by insects have caused campground and
45 hiking trail closures in the western US and Alaska (Arnberger et al., 2018). SLR, flooding, coastal erosion,
46 changing air and sea temperatures, changing humidity, and extreme weather events are putting cultural
47 heritage sites at risks (Fatori and Seekamp, 2017; Hollesen et al., 2018; Tetu et al., 2019).

48

49 14.5.7.1.3 Arctic tourism
50 Cruise and yacht tourism in the North American Arctic increased rapidly over the past decade as changes in
51 sea ice has expanded open water areas and season length (Johnston et al., 2016; Pizzolato et al., 2016;
52 Dawson et al., 2018). The risk of a major accident or incident among Arctic-going yachts and some
53 expedition passenger vessels is very high relative to other ships (high confidence) due to the combined
54 increases in mobile ice, especially along the Northwest Passage (Barber et al., 2018a; Howell and Brady,
55 2019; Copland et al., 2021; Lemmen et al., 2021), limited regulation for private yachts (Dawson et al., 2014;
56 Dawson et al., 2017), the propensity for cruise ships to travel into newly ice-free and poorly charted areas,
57 and the increasing number of non-ice strengthened vessels operating in the region (Dawson et al., 2018;

Do Not Cite, Quote or Distribute  14-60       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Copland et al., 2019; Copland et al., 2021). Compounding risks include a lack of hydrographic charting and
 2 the lack of emergency response infrastructure (e.g., spill response, search and rescue, salvage) (Amap, 2017).
 3 Tourism demand for polar bear viewing in Churchill, Manitoba, Canada may change due to climate-related
 4 declines in polar bear health (Gil et al., 2015; Manuel-navarrete and Pelling, 2015), but may be offset by
 5 `Last Chance Tourism' (LCT), a niche tourism market of individuals who explicitly seek to visit vanishing
 6 landscapes and/or disappearing flora and fauna (Lemelin et al., 2010).The ethics of promoting LCT has been
 7 questioned considering that more visitation to sensitive sites increases local impacts as well as travel-related
 8 emissions (Groulx et al., 2016; Groulx et al., 2019).

 9

10 14.5.7.2 Emerging Responses and Adaptation

11

12 Compared to other economic sectors (see section 14.5.8), the tourism industry has high adaptive capacity
13 (high confidence) (Figure 14.9). Investments in climate-resilient infrastructure within Canadian National
14 Parks have increased visitation rates during the shoulder seasons (Fisichelli et al., 2015; Lemieux et al.,
15 2017; Wilkins et al., 2018), regional collaboration among US and Canadian park agencies has enhanced
16 adaptive capacity through integrated planning and management (Lemieux et al., 2015), and technological
17 advancements have reduced the vulnerability of alpine winter sports from warming temperatures (e.g.,
18 snowmaking, refrigerated surfaces, chemical additives) (Rutty and Scott, 2015; Scott et al., 2019b; Scott et
19 al., 2020). Snowmaking as an adaptation strategy affects mitigation efforts by increasing the need for energy
20 and fuel (Scott et al., 2019b).

21

22 Tourists are also highly adaptable and, depending on their levels of place attachment, location loyalty, and
23 socio-demographics, are very likely to substitute the timing or location of their travel activity based on
24 climate and climatic-driven environmental changes (Rutty and Scott, 2015; Atzori et al., 2018). Lemieux
25 (2017) found that if the state of the Athabasca Glacier (CA-PR, Figure14.1) were to change negatively as a
26 result of climate change, 83% would travel elsewhere, and if large infrastructure was built as an adaptive
27 measure for viewing receding glaciers at Jasper National Park, 40% of tourists would no longer visit.

28

29 Hard and soft limits to adaptation exist in the tourism sector (Manuel-navarrete and Pelling, 2015). For
30 example, machine-made snow without the use of environmentally harmful chemical additives that are
31 banned in most jurisdictions, can only be made efficiently in temperatures below -2 癈, but projections
32 indicate warming temperatures above this threshold (Wobus et al., 2017; Scott et al., 2019a). Multi-
33 jurisdictional adaptation planning for parks and protected areas in the US has been hindered by a lack of
34 funding, communication, and funding trade-offs that could be remedied through coordination (Lemieux et
35 al., 2015). Social inequalities generated by the tourism development process must also be considered by
36 climate-related interventions to avoid the perpetuation of inequalities that may exist, particularly in less
37 developed regions and rapidly developing regions. For example, New developments in Hawai'i, Florida,
38 Quebec, and popular resort areas in Mexico have led to social inequalities through increased property taxes
39 leading to the marginalization of local residents away from these areas in favour of wealthy tourists (Manuel-
40 navarrete and Pelling, 2015) (also see 14.5.9).

41
42

Do Not Cite, Quote or Distribute  14-61       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Figure 14.9: Burning ember of the relative risks to select tourism activities in North America with and without
 3 adaptation as a function of global mean surface temperature increase since pre-industrial times. Risks to tourism
 4 activities include: 1) season length reductions from warming temperatures for Nordic skiing and snowmobiling, 2)
 5 season length reductions from warming temperatures and precipitation changes for alpine skiing, 3) visitor experience
 6 changes as a result of warming surface and ocean temperatures for beach tourism and degrading coral reef systems for
 7 snorkelling, 4) visitor experience changes related to warming temperatures and changing landscape aesthetic for Parks
 8 and Protected Areas. Risks assessed cover all of North America (3, 4), or are specific to certain regions (1, 2). The
 9 supporting literature and methods are provided in Supplementary Materials (SM14.4).

10

11

12 14.5.8 Economic Activities and Sectors in North America

13

14 Economic sectors highly reliant on climate, such as agriculture, tourism, fisheries, and forestry, have higher
15 levels of exposure and sensitivity (high confidence) and greater overall risk to climate change compared to
16 other economic sectors such as mining, construction, and manufacturing (medium confidence). However, the
17 cascading nature of climate impacts related to trade (Box 14.5), labour productivity (14.5.8.1.5), and
18 infrastructure (14.5.8.1.2) means that there is no economic sector in North America that will be unaffected
19 by climate change (very high confidence) (Figure 14.10). For Canada, this assessment is further supported by
20 the Canadian Climate Assessment (Lemmen et al., 2021). The combined economies of Canada, Mexico and
21 the US represented ~28% of the global GDP in 2019, with the US accounting for almost 90% of the total
22 activity for North America (World Bank, 2020a). The risks posed at different GWLs for any given economic
23 activity or sector are presented in Figure 14.10. By combining expert judgement with a systematic review of
24 the literature for each sector, the information in this Figure represents a broader synthesis, especially for

Do Not Cite, Quote or Distribute  14-62       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 sectors with a smaller literature base and at higher GWLs. The assessment of the risks of climate change on
 2 tourism (14.5.7) and the interactions between sectors through trade (Box 14.5) are discussed separately.

 3

 4 14.5.8.1 Observed Impacts and Projected Risks of Climate Change

 5

 6 14.5.8.1.1 Agriculture, fisheries, and forestry
 7 The wide range of observed and projected impacts of climate hazards on food and fibre in North America are
 8 documented in 14.5.4 (also see Chapter 5). Agriculture (USNW - corn and soybeans), fisheries (cod and
 9 pollock), and forestry (Boreal Forest timber yield) are expected to experience substantial and widespread
10 risks by 2癈 of global warming above pre-industrial (medium to high confidence) (Figure 14.10). Economic
11 models generally show economic losses in the agricultural sector across North America, especially at higher
12 GWL (14.5.4) (EPA, 2017; Boyd and Markandya, 2021), although the effects in local economies, especially
13 rural areas of the US that are highly dependent on agriculture, will be substantial even at lower GWLs
14 (Gowda et al., 2018). Full evaluations of climate risks for forestry and fisheries are presented in 14.5.1,
15 14.5.4 (also see 14.6), respectively.

16

17 14.5.8.1.2 Transportation
18 Transportation infrastructure, including roads, bridges, rail, air, sea, and pipelines, are highly vulnerable to
19 rising temperatures, SLR, weather extremes, changing ice conditions, permafrost degradation, and flooding
20 (high confidence), resulting in damage, disruption to operations, unsafe conditions, and supply-chain impacts
21 (Board and Council, 2008; Natural Resources Conservation Service; Andrey and Palko, 2017; Jacobs et al.,
22 2018; Lemmen et al., 2021) (Box 14.5). In the Mexican states of Veracruz, Tabasco, San Luis Potos�,
23 Chiapas and Oaxaca, 105,000 infrastructure sites, mostly major connecting roads, were found to be at risk of
24 flooding from tropical storms (De la Pe馻 et al. 2018). Low water levels in the Great Lakes has severely
25 impacted US grain transport (Attavanich et al., 2013). High intensity rain events destroyed 1,000km of roads
26 and washed out hundreds of bridges and culverts in 2013 resulting in an estimated CAD$6 billion (2013
27 dollars) in damages and recovery costs in Alberta, Canada (CA-PR) (Palko and Lemmen, 2017). In 2019, the
28 rail line from Winnipeg to Churchill Manitoba, which is the only ground transportation to the community
29 and to Canada's only deep-water Arctic port, was reopened after being closed for over two years due to the
30 cumulative effects of flooding, permafrost degradation, and political challenges (Lin et al., 2020). In the US,
31 the number of heat-related train delays has increased (Bruzek et al., 2013; Chinowsky et al., 2019) and by the
32 end of the century may cause economic losses of US$25 to 45 billion (RCP4.5) or US$35 to 60 billion
33 (RCP8.5) (Chinowsky et al., 2019). Sea ice reduction in the North American Arctic has led to a rapid
34 increase in ship traffic (Huntington et al., 2015; Phillips, 2016; Pizzolato et al., 2016; Huntington et al.,
35 2021b; Li et al., 2021) with cascading risks related to invasive species introduction, accident rates, black
36 carbon emissions, underwater noise pollution for marine mammals, and risks to subsistence harvesting
37 activities in Indigenous communities. (Ware et al., 2014; Council of Canadian Academies, 2016, Huntington,
38 2021; Verna et al., 2016; Chan et al., 2019)

39

40 14.5.8.1.3 Energy, oil and gas, and mining
41 Climate change is increasing the demand for electric power for cooling and threatens existing power supply
42 (high confidence) (see 14.5.5). Increased energy demand often occurs during peak energy usage and
43 especially during heat waves (Cruz and Krausmann, 2013; Leong and Donner, 2015). Cooling represented
44 74% of peak electricity demand in Philadelphia on a particularly hot day in July 2011 (Waite et al., 2017;
45 IEA, 2018b). In Canada, warming temperatures are expected to reduce demand for heating by 18 - 33% and
46 increase demand for cooling by 14 - 126% by 2070 compared to 1959-89 and 1998-2014 baseline periods,
47 respectively (Berardi and Jafarpur, 2020). The effects on hydropower are uneven across the region with the
48 potential for increases in capacity in Canada but declines of over 20% in Mexico (RCP4.5 and RCP8.5)
49 (Turner et al., 2017). Electricity demand in the US is projected to increase by 5.3 % per degree C rise in
50 temperature (Hsiang et al., 2017). Energy infrastructure, such as drilling platforms, refineries and pipelines
51 and evacuation routes are also increasingly vulnerable to higher sea levels, hurricanes, storm surges, mobile
52 multi-year sea ice, erosion, inland flooding, wildfires, and other climate-related changes (Zamuda et al.,
53 2018).

54

55 Operational efficiency and human safety at mining and energy production sites is expected to be adversely
56 affected by increases in extreme events (Section 14.2), including storms, heavy rains, riverine flooding, and
57 wildfires (high confidence). General remoteness of many mining sites (especially in the North American

Do Not Cite, Quote or Distribute  14-63       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Arctic) exacerbates risks related to emergency responses to extreme events such as wildfire (medium
 2 confidence). The 2016 Fort McMurray wildfire in Alberta Canada forced the evacuation of 88,000 people
 3 and the shutdown of mine operations. Damages were minimal because companies had undertaken proactive
 4 FireSmart interventions specifically developed for the industry (Council of Canadian Academies, 2019) (see
 5 Box 14.1). Onshore oil field production in Tabasco, Mexico, which accounts for 16% of the country's daily
 6 output, was interrupted by extensive flooding (Cruz and Krausmann, 2013). Two-thirds of mine operators
 7 globally, including major operators in North America, have experienced production challenges related to
 8 water shortages and flooding (Carbon Disclosure Project, 2013).Water availability stress due to climate
 9 change is lower in Canada than in the US and Mexico and mines in Canada may be less exposed to this risk
10 (World Resourcs Institute, 2012), with some exceptions, i.e., water-intensive oil sands mining in the
11 Athabasca River basin in Canada (Leong and Donner, 2016) (also see 14.5.3).Warming temperatures also
12 has the potential to alter the nature, characteristics and quality of mineral resources such as kaolin or
13 limestone (Phillips, 2016).

14

15 14.5.8.1.4 Construction
16 In the US, construction workers comprise 6% of the total workforce but accounted for 36% of all
17 occupational heat-related deaths from 1992-2016 (Dong et al., 2019). It is expected that total labour hours
18 among outdoor construction workers will decrease by 0.53 (+/- 0.01)% per 癈 based on existing warming
19 trends (Hsiang et al., 2017) also see (EPA, 2017). Risks are expected to be exacerbated as SLR and storm
20 surge expands the risk zone for coastal flooding exposing more property to inundation and enhancing
21 construction demand (EPA, 2017) (Box 14.4, section 14.5.5.1.3). Meeting existing and projected demand for
22 water in affected regions could also require building new desalination plants. Texas has constructed over 44
23 desalination plants across the state because of a lack of freshwater to meet potable water demand and due to
24 climate driven droughts (Kloesel et al., 2018b). Other infrastructure damaged by floods and SLR will need to
25 be reassessed and perhaps relocated away from the coast. Relocation requires availability of land that
26 frequently does not exist within urban areas (Lithogow, 2019). Some US tribes and Indigenous groups in
27 Canada lack the financial resources to build climate-resilient infrastructure such as housing and sewage
28 treatment facilities to assure clean drinking water (Mart韓ez et al., 2014; Salgado and Luisa Martinez, 2017;
29 Lithgow et al., 2019).

30

31 Permafrost thaw in northern North America will result in increased construction and reconstruction needs
32 (medium confidence) related to direct damage to buildings, roads, airport runways and other critical
33 infrastructure including decreased bearing capacities of building and pipeline foundations, damage to road
34 surfaces, and deterioration of reservoirs and impoundments used for wastewater and mine tailings
35 containment (Pendakur, 2017; Meredith et al., 2019). Ice roads have become less safe due to warming,
36 pavement damage has increased related to seasonal thaw/freeze cycles, and there have been interruptions in
37 airport operations, water and sewage service, and school operations in the Canadian territories of Yukon and
38 Nunavut (Canadian Western and Eastern Arctic � (CA-WA and CA-EA), fig 14.1) (Council of Canadian
39 Academies, 2019). By the end of the century, the economic impact of projected reconstruction of Alaska's
40 public infrastructure due to climate change (mainly from permafrost thaw) is estimated to range from
41 USD$4.2B (RCP4.5) to USD$5.5B (RCP8.5) (Melvin et al., 2017; Markon et al., 2018).

42

43 14.5.8.1.5 Manufacturing
44 Twelve million Americans (Bureau of Labor Statistics, 2015), 1.5 million Canadians (Statistics Canada,
45 2020) and 9 million Mexicans (Statistics Mexico, 2021) are employed in manufacturing. The southeast US
46 and Texas have the highest manufacturing output, with 34% of total US output ($700 billion per year). The
47 impact of climate change on manufacturing varies greatly by region. Vulnerability of the sector to climate
48 change stems from exposure of workers to increasing temperatures and humidity, exposure of facilities to
49 SLR and flooding, and changes in water supply and quality required in many manufacturing processes (Lall
50 et al., 2018).

51

52 14.5.8.1.6 Labour Productivity
53 Climate change is negatively affecting working conditions and labour productivity in North America
54 (medium confidence) (Section 14.5.6.1 and Box 14.5). Working conditions in temperatures above a Heat
55 Index of 85篎 (29.4篊) are correlated with potentially hazardous health conditions (Tustin et al., 2018) and
56 for every 篊 increase in temperature, labour productivity is estimated to be reduced by 0.11% for low risk
57 workers and 0.53% for high risk workers (i.e., construction, mining, agriculture and manufacturing) (Hsiang

Do Not Cite, Quote or Distribute  14-64       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 et al., 2017). By mid-century (RCP8.5), temperature increase, changing water availability and SLR, are
 2 projected to result in a 0.6% drop in labour productivity in auto, timber, textile and chemical manufacturing
 3 in the Southeast and Texas regions (Kinniburgh et al., 2015; Hsiang et al.). Labour productivity in the US
 4 automobile industry decreases by 8% for every six or more days of consecutive unusually hot weather
 5 (above 90篎/32.2篊) (Cachon et al., 2012). Thirty percent of California workers are employed in high-risk
 6 industries, such as agriculture, with exposure to high temperature leading to loss in productivity (Rogers et
 7 al., 2015). Under RCP8.5 increases in extreme temperatures, labour productivity in the US is projected to
 8 decrease, costing US$190 billion in lost wages by 2090 (EPA, 2017; Kjellstrom et al., 2019)(also see
 9 (Gubernot et al., 2014; Kiefer et al., 2016; Carter et al., 2018).

10

11 14.5.8.2 Current and Potential Adaptation

12

13 Adaptation options are highly diverse and sector-specific (EPA, 2017). Regardless of economic sector,
14 companies that implement effective and rapid response options that address climate change stressors will
15 have a competitive advantage (Gasbarro et al., 2016, Lemmen, 2021). Most companies focus on short-term
16 risk management and consequently short-term adaptation is often favoured over long-term approaches
17 particularly in the private sector, which will be ineffective for climate change risk reduction over the long
18 term (Gasbarro et al., 2016).

19

20 Investment and coordination of climate services (forecasting) can support many economic sectors across
21 North America. In 2017, 15% of S&P 500 companies publicly disclosed an effect on earnings from weather
22 events, reflecting a growing trend (Williams et al., 2018).Existing US federal-sponsored planning tools
23 provide guidance to states and to plan for SLR and flooding with large threats to commercial sectors (US
24 Department of Transportation, 2015). The NOAA Coastal Services Center SLR and coastal inundation
25 viewer (https://coast.noaa.gov/digitalcoast/tools/slr.html), the Army Corps of Engineers Sea Level Change
26 Curve simulator, and Climate Central's interactive portal (Ocean at the Door) all provide access to
27 visualizations of future sea level rise that are available to US coastal cities and towns for commercial
28 planning purposes. Similar resources are being developed and are available for Canada including Canada's
29 Climate Atlas (https://climateatlas.ca/).

30

31 Adaptation options for transportation and related infrastructure include engineering and technological
32 solutions, as well as innovative policy, planning, management, and maintenance approaches (Natural
33 Resources Conservation Service, 2008; Jacobs et al., 2018). For northern transportation, new technologies
34 and infrastructure adaptations can be employed to facilitate heat extraction (e.g. air convection
35 embankments, heat drains, thermosyphons, high albedo surfacing, gentle embankment slopes) (McGregor et
36 al., 2010b; United Nations, 2020) Adaptation options for roads include changing pavement mixes to be more
37 tolerant to heat or frost heaving, expanding drainage capacity, reducing flood risks, enhancing travel
38 advisories and alerts, elevating or relocating new infrastructure where feasible and changing infrastructure
39 design requirements to include climate change considerations or to introduce new flood event thresholds
40 (Natural Resources Conservation Service, 2008; EPA, 2017; Pendakur, 2017). Railroads are testing
41 temperature sensors on rail tracks to provide early warning of buckling. Sensors that signal when tracks are
42 approaching dangerous temperatures may help to avoid accidents (Hodge et al., 2014; Chinowsky et al.,
43 2019).

44

45 Adapting building codes more uniformly to changing climate conditions such as SLR, storms, winds and
46 wildfires reduces risk (Olsen, 2015; Maxwell et al., 2018b). North America has not, on the whole, adapted its
47 building code regulations to consider the dynamic challenges of climate change, although some specific
48 efforts have been made, including the addition of requirements for wildfire within California's building
49 codes and Canada's Climate-resilient building and core public infrastructure initiative, which involves
50 updating building codes and standards to improve climate resiliency (Lacasse et al., 2020) (Box 14.4). To
51 enhance safety, some outdoor workers have been fitted with heat sensors to analyse/assess how warming
52 may affect productivity and well-being (Runkle et al., 2019). Other options include raising public roads and
53 seawalls, initiating buy-outs of property owners in flood-risk areas, and improving storm water drainage.
54 Adopting approaches like the International Future Living Institute's Living Building Challenge (LBC) may
55 inform future regulatory processes (Eisenberg, 2016). The LBC (https://living-future.org/basics/) has seven
56 thematic areas that inform building design, although only a subset of those are relevant for climate change
57 including water, energy, and materials considerations.

Do Not Cite, Quote or Distribute  14-65       Total pages: 157
      FINAL DRAFT                 Chapter 14  IPCC WGII Sixth Assessment Report

1
2

 3

 4 Figure 14.10: Burning Ember of the relative risks to economic sectors in North America as a function of
 5 projected global mean surface temperature increase since pre-industrial times. Impacts on economic sectors
 6 include: 1) changing crop yield leading to economic loss for agriculture, 2) changes in the quality and quantity of timber
 7 yields, 3) reductions in season length and economic viability for tourism activities, 4) increased maintenance and
 8 reconstruction costs to transportation infrastructure, 5) changes in fisheries catch, 6) reduced productivity in mining and
 9 energy operations, 6) reduced labour productivity in outdoor construction, and 7) increased maintenance and
10 reconstruction costs to transportation systems. Risks to economic sectors and activities were sometimes assessed across
11 all of North America (3, 4), within specific regions (1, 2), and for specific crops or species (1 - corn and soybean, 2 �
12 cod and pollock). The supporting literature and methods are provided in Supplementary Material (SM14.4).

13

14

15 [START BOX 14.5 HERE]

16

17 Box 14.5: Climate Change Impacts on Trade Affecting North America

18

19 In North America, trade - defined as the sum of export and import of goods and services - is valued at $1.3
20 trillion USD annually (2019 dollars) and represents 30% of North American GDP. Variations within the
21 region are notable; Mexico relies on trade for 80% of its GDP and Canada for 66% (World Bank, 2020a).
22 Canada and the US traded over USD$55.2 billion worth of products related to the agriculture industry
23 between 2015 and 2018 (Government of Canada, 2019). Canada, the US and Mexico have the longest
24 running trade pacts globally and these agreements have played a major role in supporting economic and
25 social development in the region (see (Frankel and Rose, 2005; Eaton et al., 2016; World Bank, 2020b).
26 However, recent changes to the North American Free Trade agreement do not clearly address climate change
27 (Lucatello, 2019).

28

29 Climate risks may create shocks to the trade system by damaging infrastructure and disrupting
30 supply-chains in North America (medium confidence). Sea level rise, flooding, permafrost thaw,
31 landslides, and increased frequency and magnitude of extreme weather events are projected to impact
32 transportation infrastructure which will pose challenges to the movement of goods, especially in coastal
33 areas (Lantuit et al., 2012; Dor� et al., 2016; Hjort et al., 2018; Koks et al., 2019; Lemmen et al., 2021).
34 Maritime ports are at the greatest risk from climate hazards (Messner et al., 2013; Slack and Comtois, 2016),

Do Not Cite, Quote or Distribute  14-66       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 followed by roads, rail, and airports (Anarde et al., 2017). Due to the trans-national nature of trade, extreme
 2 weather disruptions in one region are likely to lead to cascading effects in other regions (high confidence)
 3 (Lemmen et al., 2021). For example, climate change will have negative impacts for global food and energy
 4 trade where reductions in crop production and fish stocks in some regions could cause food and fish price
 5 spikes elsewhere (Beaugrand et al., 2015; Lam et al., 2016; Shukla, 2019) (also see 14.5.4, Figure 14.10, and
 6 5.11.8).

 7

 8 Climate change impacts may alter current trade practices and patterns with implications for regional
 9 economic development in North America, especially in the Arctic (medium confidence). Climate change
10 is causing modal-shifts in cargo shipping. For example, lower water levels in lakes and rivers (e.g.,
11 Mackenzie River, Mississippi River) impact freight transport and may cause a shift from marine transport to
12 more GHG-intensive rail, road, or air transport (Koetse and Rietveld, 2009; Du et al., 2017; Pendakur, 2017).
13 Sea ice change is creating new Arctic marine trade corridors (Melia et al., 2016; Pizzolato et al., 2016; Ng et
14 al., 2018; Bennett et al., 2020; Mudryk et al.), including shorter and potentially more economical routes such
15 as the Northwest Passages (see CCP6, Box 6.1). Warming temperatures have also reduced the season length
16 for ice roads, which are heavily relied upon to service remote communities and remote industries including
17 forestry and mining (Pendakur, 2017) (see 14.5.8.1.2).

18

19 Effective and equitable trade policies can act as important adaptation strategies (medium confidence).
20 Higher temperatures have had no direct effect on developed countries' exports, but have significantly
21 reduced growth in exports among developing countries, which in turn can increase the price of goods that
22 developed countries then import (Costinot et al., 2016; Constant and Davin, 2019). Schenker (2013)
23 estimated that the climate impacts on trade from developing to developed countries could be responsible for
24 16.4% of the total expected cost of climate change in the US in 2100 and thus, North America would benefit
25 from increased investment in effective and equitable trade policies and adaptation in developing regions.
26 Under an RCP8.5 scenario (~2.6 to 4.8 degree C warming) and within current trade integration, climate
27 change could lead to up to 55 million undernourished people by 2050; these projections decrease by 64% (20
28 million people) with the introduction of reduced trade tariffs and the lessening of institutional and
29 infrastructure barriers (Janssens et al., 2020). Although most studies focus on global food security
30 (agriculture), it is likely that the same challenges exist for other commodities and manufactured goods.

31

32 [END BOX 14.5 HERE]

33
34

35 [START BOX 14.6 HERE]

36

37 Box 14.6: The Costs and Economic Consequences of Climate Change in North America

38

39 Observed Impacts

40

41 Extreme weather events, including hurricanes, droughts, and flooding, and wildfires, have been partly
42 attributed to anthropogenic climate change (e.g., Rupp et al., 2015; Emanuel, 2017); attribution table in
43 Chapter 16) (high confidence). Direct, indirect and non-market economic damages from extreme events
44 have increased in some parts of North America (high confidence). The number of extreme events with
45 inflation-adjusted damages totalling more than US$1B has risen in the US over the past decades (NOAA,
46 2020; Smith, 2020), and similar increases have been observed in Canada (Boyd and Markandya, 2021).
47 Factors other than climate change, including increases in exposure and the value of the assets at risk, also
48 explain increasing damage amounts (Freeman and Ashley, 2017; Vano et al., 2018). Climate change explains
49 a portion of long-term increases in economic damages of hurricanes (limited evidence, low agreement).
50 Studies of US hurricanes since 1900 have found increasing economic losses that are consistent with an
51 influence from climate change (Estrada et al., 2015; Grinsted et al., 2019), although another study finds no
52 increase (Weinkle et al., 2018).

53

54 Formal attribution of economic damages from individual extreme events to anthropogenic climate change
55 has been limited, but climate change could account for a substantial fraction of the damages (limited
56 evidence, medium agreement). Two recent studies have shown approaches for how damages may be
57 attributed for individual events in the US. Assuming a direct proportionality between attributable risk of the

Do Not Cite, Quote or Distribute  14-67       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 event to the attributable economic damages, one study suggested that 30-75% of the direct damages from
 2 Hurricane Harvey was caused by climate change, with a best estimate of US$67B out of an estimated
 3 US$90B total of attributable damages (Frame et al., 2020). Another study modeled the component of the
 4 flooding from Hurricane Sandy due to rising SLR and mapped that to coastal damages. That study estimated
 5 that US$8.1 billion (13% of the total) was attributable to the climate influence on SLR (Strauss et al., 2021).

 6

 7 The effect of climate change has been identified in aggregate measures of economic performance, such as
 8 GDP, in North America and globally (medium confidence), although the magnitude of these changes is
 9 difficult to constrain (medium confidence). Climate change has been observed to affect national GDP level
10 and economic growth (low confidence). The extent to which climate has affected GDP may be challenging to
11 identify statistically (Cross-Working Group Box ECONOMIC in Chapter 16). Observed GDP effects are
12 generally slightly negative in the US, higher and negative for Mexico, and the directionality of the effects in
13 Canada varies by study and modeling approach (Burke et al., 2015; Colacito et al., 2018; Kahn et al., 2019).

14

15 Projected Risks

16

17 Projections of market and non-market economic damages demonstrate the substantial economic risks of
18 climate impacts associated with high temperature pathways (RCP8.5) (high confidence). Since AR5, a wide
19 range of estimates of the costs of climate change have been developed for the US (EPA, 2015a; Houser et al.,
20 2015; EPA, 2017; Hsiang et al., 2017; Martinich and Crimmins, 2019), with ongoing processes to update
21 national estimates for Canada and Mexico (Semarnat, 2009; NRTEE, 2011; Estrada et al., 2013; Sawyer et
22 al., 2020). While the magnitude of the estimates depend on approach and assumptions in the methods and
23 expectations of future socioeconomic conditions, these studies show substantial projected economic damages
24 across North America by the end of the century, especially for warming greater than 4篊 (high evidence,
25 high agreement). Whether these damages translate into GDP effects is not clear for Canada. Some modeling
26 approaches show modest GDP increases in 2050 and 2100, while others suggest modest decreases although
27 it is anticipated that the economic effects for Canada will be large and negative (Boyd and Markandya,
28 2021). Large costs and risks, such as those associated with extreme events such as wildfires (Hope et al.,
29 2016) and the increased need for infrastructure replacement (Neumann et al., 2015; Maxwell et al., 2018a)
30 will have compounding effects in the markets by disrupting economic activities (Box 14.5).

31

32 Market and non-market risks and costs will not be experienced equally across countries, sectors and regions
33 in North America (high confidence). For the US, reductions in mortality, energy expenditures and
34 improvements in agricultural yields are projected to result in net gains in the North and Pacific Northwest
35 whereas in the South, higher heat-related mortality, increases in energy expenditures, SLR and storm surge
36 are projected to result in economic losses by the end of century (Hsiang et al., 2017). No region in the US is
37 expected to avoid some level of adverse effects (EPA, 2017; Martinich and Crimmins, 2019) (medium
38 evidence, high agreement). Economic models generally show losses in the agricultural sector across North
39 America, especially at higher GWL (Boyd and Markandya, 2021, EPA 2017). Some models show large
40 gains in parts of Canada, although these models do not capture the full range of climate hazards including
41 change in precipitation or extreme events (Boyd and Markandya, 2021).

42

43 Economics of Adaptation Opportunities

44

45 Economic analysis can help reveal where the avoided economic damages are greater than the costs of
46 adaptation, improving decision-making for adaptation planning and efforts in North America (high
47 confidence). Detailed assessment of total needs and costs of climate adaptation are limited (Sussman et al.,
48 2014), but estimates suggest that the costs are large (low evidence, high agreement). Cost-benefit and other
49 economic analyses that incorporate damage estimates are expanding for adaptation decision-making (Li et
50 al., 2014), especially for technical options in areas with high exposure such as coastal areas in Mexico (Haer
51 et al., 2018) and Alaskan infrastructure (Melvin et al., 2017). Cost-benefit analysis has also been applied to
52 coordinating planning across jurisdictions in North America for SLR and flood control (Adeel et al., 2020).
53 Adaptation costs in the US are lower on RCP4.5 compared to RCP8.5 emission pathways (Martinich and
54 Crimmins, 2019). Adaptation, however, cannot be based solely on the cost benefit analysis due to the high
55 level of uncertainty related to climate risks (Cross-Chapter Box DEEP in Chapter 17).

56

Do Not Cite, Quote or Distribute  14-68       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Improving projections of future economic risk and damages facilitates the development of tools that can be
 2 used for economic analysis of climate policies (high confidence). Monetized estimates of the damages from
 3 climate change have been developed and refined since AR5, motivated in part by efforts to estimate the
 4 Social Cost of Carbon (SCC) (National Academies of Sciences, 2017). Support for these efforts and the use
 5 of SCC in regulatory analysis of mitigation and adaptation efforts have been pledged across the national and
 6 sub-national governments of Canada, the US and Mexico. Harmonizing SCC and consistent use can further
 7 enhance coordination of mitigation and adaptation decision-making (Auffhammer, 2018; Aldy et al.,
 8 2021).Using these damages estimates can also inform other policy and tools that improve the consideration
 9 of climate impacts in markets and decision-making (Report of the Climate-Related Market Risk
10 Subcommittee, 2020).

11

12 [END BOX 14.6 HERE]

13
14

15 14.5.9 Livelihoods

16

17 Exposure and vulnerability to climate hazards have varied across North America by region and by
18 population (high confidence). These differences have been often underpinned by social and economic
19 inequalities and have been observed between households, social groups, rural and urban communities, and
20 Indigenous Peoples (high confidence). These vulnerabilities have also been observed to contribute to
21 maladaptation (medium confidence) (14.5.9.1). Social and economic trends and development will determine
22 near-term impacts on livelihoods from projected climate hazards; livelihoods will also adapt to the risks and
23 opportunities (high confidence) (14.5.9.2). Actions to enhance the livelihoods of the most vulnerable social
24 groups in North America will lessen the impacts of climate hazards on them (high confidence) (14.5.9.3).

25

26 14.5.9.1 Observed Impacts

27

28 Livelihoods are `the resources used and the activities undertaken in order to live. Livelihoods are usually
29 determined by the entitlements and assets to which people have access' (IPCC, 2018) (8.1.1). While often
30 understood as subsistence or traditional ways of life (Oswal, 1991), livelihoods are often conceptualized
31 more broadly as encompassing the economic, cultural, and social capitals or assets, capabilities, and
32 activities that individuals, households, and social groups use as the means to make a living (DFID, 1999;
33 Obrist et al., 2010).

34

35 Past and current patterns of development in North America have propagated and perpetuated vulnerabilities
36 that have created differential impacts on livelihoods from climate hazards (high confidence). Predatory and
37 extractive economies have underpinned economic activity in North America historically and currently.
38 While generating substantial wealth, these patterns have also driven social and economic inequality (medium
39 evidence, high agreement) (Jasanof, 2010; Shove, 2010; Klinsky et al., 2016; Robinson and Shine, 2018).
40 Patterns of development that reinforce these structures remain a large contributor to current social-
41 environmental risks and have affected all kinds of contemporary livelihoods (Cannon and M黮ler-Mahn,
42 2010; Koch et al., 2019) (also, see Chapter 18).

43

44 Climate impacts have damaged livelihoods across North America, especially those of marginalized people
45 (high confidence) and deepened inequalities for these groups (medium confidence). Across North America,
46 climate change has affected livelihoods with larger effects on individuals, households and communities that
47 are already more vulnerable due to a range of pre-existing social and environmental stressors (Olsson et al.,
48 2014; Hickel, 2017; Koch et al., 2019) such as Indigenous Peoples, urban ethnic minorities, and immigrants
49 (Guyot et al., 2006; Gronlund, 2014; Klinenberg, 2015). These impacts have also contributed to a deepening
50 of inequalities for marginalized groups (Audefroy and Cabrera S醤chez, 2017; Garc韆 et al., 2018) (medium
51 evidence, high agreement). As climate hazards further degrade their livelihoods, these groups have faced
52 additional challenges to avoiding or escaping poverty (Ruiz Meza, 2014). Furthermore, these groups have
53 needed to use their more limited resources to manage present challenges, restricting their future capacities to
54 adapt (Tolentino-Ar関alo et al., 2019). Climate impacts have also affected the livelihoods of the middle
55 classes (Dom韓guez et al., 2020) who have become more vulnerable due to changes in their social and
56 economic security (Garza-Lopez et al., 2018). Gender has also been recognized as a determinant of

Do Not Cite, Quote or Distribute  14-69       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 differential vulnerability with implications for women's livelihoods (Cross-Chapter Box GENDER in
 2 Chapter 18).

 3

 4 Migration and mobility have been an important part of livelihoods in North America (high confidence).
 5 Movement across North America has been reinforced by social, cultural and economic ties (Box 14.5). For
 6 example, middle class retirees from Canada and the US engage from temporary, seasonal to permanent
 7 migration to the warmer climates of the Southern US and Mexico, often benefiting from the lower cost of
 8 living (Dom韓guez et al., 2018). Temporary or semi-permanent labor migration, generally followed by
 9 remittances, has been an important part of livelihoods for rural areas in Mexico (high confidence) and has
10 been employed as a response to climate hazards (low evidence). Drought in rural areas which are highly
11 dependent on subsistence agriculture have observed migration to urban areas in Mexico (Nawrotzki et al.,
12 2017). Evidence of international migration in response to climate hazards is sparse with difficulties in
13 identifying a climate signal due to the multi-causal nature of migration decision-making (Cross-Chapter Box
14 MIGRATE in Chapter 7). There is limited evidence of extreme weather events or climate hazards on
15 migration from Mexico to the United States (Nawrotzki et al., 2015b; Nawrotzki et al., 2015c; Nawrotzki et
16 al., 2016; Murray-Tortarolo and Salgado, 2021).

17

18 Pre-existing social vulnerabilities have also led to forced displacement from extreme weather events (low
19 confidence). In the US, compounding effects of SLR and storm surge interacted with pre-existing social
20 vulnerabilities of local communities to generate large-scale displacement after the effects of Hurricane
21 Katrina on New Orleans in 2005 (Jessoe et al., 2018). The processes of relocation and recovery in New
22 Orleans was further shaped by vulnerability where out-migration was more likely to be minorities and
23 economically disadvantaged while the recovery was predominantly in neighborhoods that were wealthier
24 prior to the disaster (Fussell et al., 2014; Fussell, 2015). Newer evidence from Hurricane Maria in Puerto
25 Rico in 2017 has shown an initial spike in displacement with slower recovery with more vulnerable
26 communities returning at higher rates (DeWaard et al., 2020); however, overall out-migration trends have
27 been consistent with long-term economic migration (Santos-Lozada et al., 2020). Interactions of slower onset
28 climate hazards with displacement, such as observed in Shishmaref, Alaska, have revealed the challenges in
29 attribution of migration to climate as it intersects with socioeconomic conditions and lived experiences
30 (Marino and Lazrus, 2015).

31

32 Maladaptation has also been occurring in livelihoods, especially as it relates to agricultural practices that are
33 less resilient to climate hazards and competition for land use (limited evidence, high agreement). Focusing
34 on examples in Mexico (see 14.5.4.3 for US and Canada examples), for some Mexican Indigenous Peoples,
35 the replacement of ancestral farming practices with technological adaptations like transgenic crops has
36 reduced their resilience by making them more dependent on external inputs and more expensive supplies
37 while increasing putting their health at risk with herbicide and insecticide use (Mercer et al., 2012). Existing
38 power structures have also interacted with climate hazards to generate maladaptive outcomes (Quintana,
39 2013). Mennonite communities in the northern state of Chihuahua, Mexico have pursued commercial
40 agricultural markets that lead them to shift to transgenic crops and to overexploit local groundwater
41 resources in a region experiencing multi-year droughts. These actions have led to conflict with other local
42 farming groups with less economic capital to access groundwater (Quintana, 2013). Climate mitigation
43 measures may also have adverse effects on local livelihoods with implications for adaptive capacity. The
44 Reducing Emissions from Deforestation and Forest Degradation in Developing Countries (REDD+)
45 mitigation program has been highlighted as a trade-off between an international/national carbon mitigation
46 strategy and the ability of some Mexican rural communities to improve their food security (Barbier, 2014)
47 (5.6.3.3).

48

49 14.5.9.2 Projected Risks

50

51 Livelihoods will evolve as a result of both challenges presented directly or indirectly from climate impacts as
52 well as socioeconomic changes and technological developments (high confidence). Livelihoods, however,
53 can be undermined by many of the projected climate risks with the impacts depending on adaptive capacity
54 and adaptation limits (high confidence) (8.4.5.1). Real areas in Mexico and the southern US with agriculture-
55 based livelihoods and projected reduction in precipitation will be adversely affected (Esperon-Rodriguez et
56 al., 2016) (14.5.4). Outdoor workers in rural and urban areas will be exposed to higher health risks from

Do Not Cite, Quote or Distribute  14-70       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 higher temperatures and heatwaves (14.5.8). Reduced livelihoods will also be associated with adverse mental
 2 health effects (14.5.6.8).

 3

 4 Future climate hazards will deepen patterns of social inequality as vulnerable groups may also experience
 5 intersecting impacts that adversely affect their livelihoods (medium confidence). Health, in particular, will be
 6 a key intersection as marginalized and disadvantaged groups often have poorer health status and hold
 7 occupations that may involve higher exposure to climate hazards. African Americans are expected to
 8 experience the largest impacts on their health status due to differential exposure and vulnerability to climate
 9 hazards (Marsha et al., 2016) (Section 14.5.6).

10

11 Displacement, migration and resettlement will increase along higher emission pathways (medium
12 confidence). Combining projections of SLR and population scenarios for the US, Haer et al. (2013) and
13 Hauer et al. (2016) have estimated the magnitude of the population at risk in coastal communities,
14 numbering in the millions. In the near-term, where climate hazards influence out-migration, it will mostly
15 augment existing patterns as migration is strongly influenced by existing social networks (7.3.2). Planned
16 relocation and resettlements will reduce the exposure to climate hazards for the involved populations but
17 could adversely affect their livelihoods in the absence of supportive programs (Jantarasami et al., 2018a)
18 (7.3.2), since livelihood outcomes strongly depend on socioeconomic conditions.

19

20 14.5.9.3 Adaptation

21

22 Climate hazards undermine adaptation by damaging livelihoods (high confidence). Many actions that
23 enhance and promote resilient livelihoods can have substantial benefit for adaptation to climate hazards
24 (medium confidence). Livelihoods in the context of climate change are characterized by adjustments that
25 then feedback into the assets that comprise a livelihood. Social capital - in the form of household and
26 community cohesion - facilitates the development of adaptation strategies to the impacts of climate change in
27 rural and urban communities at the household level and for small groups (Barbier, 2014; Nawrotzki et al.,
28 2015b; Nawrotzki et al., 2015c). Cultural capital, especially in the form of local knowledge and Indigenous
29 knowledge, can guide adaptation practices in North America (Akpinar Ferrand and Cecunjanin, 2014),
30 preserving Indigenous cultures and enhancing future adaptation and resilience (Pearce et al., 2012 2015;
31 Audefroy and Cabrera S醤chez, 2017) (Box 14.1). In Mexico, rain-water harvesting (practiced by some
32 Mayan communities) and the use of local-traditional varieties of maize have assisted in the adaptation to
33 climate impacts and promoted food security (Akpinar Ferrand and Cecunjanin, 2014; Hellin et al., 2014).
34 Funding and support for these social adaptation strategies have been uneven (Barbier, 2014; Romeo-Lankao
35 et al., 2014). The legacy of colonialism and historical patterns of development will continue to shape the
36 adaptation responses and resiliency of Indigenous Peoples (Todd, 2015; Davis and Todd, 2017; Whyte,
37 2017; Cameron et al., 2019).

38

39 Migration is a common adaptation strategy to maintain and diversify people's livelihoods and will continue
40 to play an important role when households manage climate and social risks (high confidence) (7.4.3). In the
41 near-term, actions that enhance in-situ adaptive capacities as well as fostering safe and orderly migration can
42 result in synergies for both adaptation and development (Cross-Chapter Box MIGRATE in Chapter 7).
43 Populations that experience less mobility or cannot engage in voluntary migration as an adaptation may need
44 additional support to adapt to climate hazards, for example northern communities that are at risk of climatic
45 events (Hamilton et al., 2016). Policies associated with the transition from high GHG intensive extractive
46 industries, sometimes referred to as "just transitions", may also support in-situ livelihoods if they also aim to
47 address and redress existing inequalities to reduce vulnerabilities (McCauley, 2018); however, these policies
48 could result in maladaptation if they create new inequalities or generate other environmental damages.

49

50 14.5.10 Violence, Crime, and Security

51

52 Elevated rates of various types of crime have been associated with higher temperatures in the US and
53 Mexico (medium confidence based on limited evidence and high agreement) (14.5.10.1). If social
54 relationships prevailing now and in the recent past continue, projections show future crime rates in the US
55 and Mexico increasing with increasing temperatures (low confidence) (14.5.10.2). Degradation of human
56 security and conflicts exacerbated by climate change--even outside of North America--will increase the
57 demand for humanitarian assistance, foreign aid and resettlement (medium confidence) (14.5.10.2).

Do Not Cite, Quote or Distribute  14-71       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 14.5.10.1 Observed Impacts

 3

 4 14.5.10.1.1 Violence and crime in the past and present
 5 Crime, including violent crime, has been associated with higher temperatures in the US (medium
 6 confidence). Studies of crime statistics in the US have revealed a relationship between temperature and a
 7 range of violent crimes including aggravated assaults, rapes, and homicides; effects for property crimes are
 8 weaker (Ranson, 2014; Houser et al., 2015; Heilmann and Kahn, 2019; Mares and Moffett, 2019) (limited
 9 evidence, medium agreement). These effects have been observed in US urban centres (Hsiang et al., 2013;
10 Mares, 2013; Ranson, 2014; Schinasi and Hamra, 2017; Heilmann and Kahn, 2019) and more generally
11 across the US (Mares and Moffett, 2019). Differential effects have also been observed within urban areas.
12 Observed higher rates of domestic and intimate partner violence during periods of high heat in less affluent
13 neighbours in Los Angeles have been associated with disparities in access to air conditioning and greenery
14 (Heilmann et al., 2021). By contrast, (Lynch et al., 2020a) found no significant correlation between annual
15 homicide rate and annual temperature for New York City (Lynch et al., 2020b). For Mexico, (Burke et al.,
16 2018a) found temperature linkages with intergroup killings by drug-trafficking organizations, homicides, and
17 suicides. No linkages between temperature and crime have been reported for Canada. Differences in spatial
18 and temporal aggregation of the crime statistics as well as in the measure of climate change or variability
19 explain some of the differences between studies. Several causal pathways can explain these relationships
20 (Miles-Novelo and Anderson, 2019; Lynch et al., 2020b). The dominant theory is that weather changes result
21 in changes in behavioural patterns that lead to more opportunities for crimes. For example, studies that
22 disaggregate by month often report significant positive associations between temperature anomalies and
23 violent crime (especially aggravated assaults, rapes, and homicides), particularly in the cold season (Harp
24 and Karnauskas, 2018; Mares and Moffett, 2019)). Smaller increases in crime during positive warm-season
25 temperature anomalies may be due to people seeking shelter in cooler indoor spaces, decreasing crimes of
26 opportunity (Gamble and Hess, 2012) (7.2.7).

27

28 The archaeological record has been used to infer linkages between climatic variability and social
29 process, including violence (inferred with medium confidence). Past North American societies have been
30 exposed to greater climatic variability than is documented in the instrumental record. Because future climatic
31 conditions are likely to exceed those known for the recent past (Cross-Chapter Box PALEO in Chapter 1),
32 the North American archaeological record can illuminate possible relationships between climate variability
33 and violence that cannot be observed in the present record. In the upland US Southwest between A.D. 600
34 and 1280, one study found that violence significantly increased as climatically-controlled maize production
35 decreased and interannual variability increased (Kohler et al., 2014) (low evidence, high agreement); massive
36 emigration from the northern Southwest in the last half of the AD 1200s is connected with though not
37 completely explained by climatic variability (Scheffer et al., 2021). In the central and southern Maya
38 lowlands, following centuries of increasing populations and attempts to produce more maize (Roman et al.,
39 2018), episodes of drought and/or increased summer temperatures in the 9th and 10th centuries AD
40 (Dunning et al., 2012; Kennett et al., 2012) accompanied increased conflicts and social disintegration
41 including collapse of long-lived dynasties, cessation of monumental inscriptions (Carleton et al., 2017) and
42 emigration (medium evidence, medium agreement). Such findings reinforce research on contemporary
43 societies that climate-induced farming shortfalls in regions dependent on agriculture may induce or
44 exacerbate conflict, especially in interaction with unfavourable demographic, political, and socioeconomic
45 factors (e.g. (Koubi, 2019))(medium evidence, medium agreement) (7.2.7.).

46

47 14.5.10.1.2 Security
48 Climate change poses risks to peace (16.5.2.3.8) that could affect North America (medium confidence).
49 Military and security communities are adapting their planning, operations and infrastructure to current
50 impacts of climate change in North America and globally (medium agreement, medium evidence). Arctic
51 nations are renewing their military capacity and expanding their constabulary presence around their existing
52 boundaries (Choi, 2020). There is increasing awareness that climate change causes weather patterns and
53 extreme events that directly harm military installations and readiness through infrastructure damage, loss of
54 utilities, and loss of operational capability (Duffy-Anderson et al., 2019). Transboundary disputes and
55 competition over resources such as fish (豷thagen, 2020) are a concern in the changing Arctic and increases
56 in military and constabulary operations are being observed (J鰊sson et al., 2012; Smith et al., 2018;
57 Eyzaguirre et al., 2021).

Do Not Cite, Quote or Distribute  14-72       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 14.5.10.2 Projected Risks

 3

 4 14.5.10.2.1 Violence and Crime
 5 Projections of future crime derived from the empirical relationships between temperature and crime
 6 in the US show the potential for increased criminality under RCP8.5 compared to RCP4.5 (low
 7 confidence). For RCP8.5, holding all socioeconomic conditions at 2015 levels, violent crime could increase
 8 0.6 �2.1% by mid-century and 1.9 �4.5% by late-century (Houser et al., 2015). The rise in property crime is
 9 projected to be smaller as property crime flattens at higher temperatures (Hsiang et al., 2013). Using
10 relationships between crime and monthly temperatures established for five US regions by Harp and
11 Karnauskas (2018), Harp and Karnauskas (2020) project 18,800 additional violent crimes annually beyond
12 2014 levels by the end of the 21st century under 1.5篊 warming, rising to 48,200 under 4篊 warming.
13 Aggregating data by states weighted by population density, (Mares and Moffett, 2019) project an average
14 annual increase of 0.94% across seven categories of violent and property crime for each anomalous 篊
15 warming (an average annual increase of about 100,000 crimes). Changing socioeconomic conditions in the
16 future may either reduce or exacerbate the projected contemporaneous relationship between temperature
17 anomalies and crime (Agnew, 2011; Lynch et al., 2020b) whereas adaptation could weaken these
18 relationships.

19

20 14.5.10.2.2 Defense and Security
21 Climate change will affect ecosystems (16.5.2.3), living standards (16.5.2.3.4), health (16.5.2.3.5), and
22 food security (16.5.2.3.6) globally and these changes may exacerbate violence and political instability
23 (medium confidence) with implications for national security in North America (medium confidence).
24 Climate variability, hazards, and trends to date have played a role in exacerbating conflict, but the influence
25 of climate appears to be minor and more uncertain than the roles of low socioeconomic development, low
26 state capability and high intergroup inequality (Mach et al., 2019). More profound impacts from climate
27 change on weather and seasons as well as changing socioeconomic conditions could lead to patterns of
28 violence that cannot be predicted by projecting relationships between current climate and violence into the
29 future (14.6.3) (Mach et al., 2019). If global levels of violence increase, there will be increased demand for
30 international efforts, including disaster aid and humanitarian efforts (Eyzaguirre et al., 2021). Climate
31 change and geopolitical goals interact in the Arctic (Smith et al., 2018). New transportation corridors and the
32 potential access to natural resources could lead to competition for access to and control over the region
33 (Estrada, 2021) (CCP6.2.6; Box CCP6.1; FAQ CCP6.2). Governance structures exist to manage geopolitical
34 manoeuvring and to protect the human security of Arctic populations (14.5.10.3; 7.2.7.1).

35

36 14.5.10.3 Adaptation Options

37

38 14.5.10.3.1 Violence and Crime
39 Co-benefits from adaptation options include improving the liveability of and quality of life in cities,
40 reducing socioeconomic vulnerability and exposure to locally higher temperatures (medium
41 confidence). Urban settings in the US have disproportionately higher exposure to urban heat island effects in
42 low-income and minority neighbourhoods in US cities (14.5.5.1). Co-benefits from adaptation responses in
43 the urban landscape can reduce socioeconomic vulnerabilities and exposure to higher temperatures
44 (14.5.5.3). Evaluation of adaptation efforts to reduce crime rates that have been associated with temperature
45 are limited. In LA, a link has been inferred between violence and older buildings that may lack air
46 conditioning (Heilmann et al., 2021). By contrast, access to air conditioning did not appear to lessen crime
47 rates in Mexico (Baysan et al., 2019).

48

49 14.5.10.3.2 Defence and Security
50 Existing environmental and international agreements that consider climate risks can contribute to
51 cooperation (medium confidence). Strengthening and empowering existing environmental and diplomatic
52 avenues (e.g., the Arctic Council and international agreements such as the United Nations Convention on the
53 Law of the Sea, and various subnational actors and agreements (CCP6.3.2)) to incorporate risks from climate
54 impacts could enhance cooperative avenues for defusing conflict (Huebert et al., 2012). Improving the
55 consideration of climate risks in efforts to expand economies and trade (Box 14.5), and improvements in
56 peace-keeping (7.4.4) (Barnett, 2018) could also reduce future conflict risks.

57

Do Not Cite, Quote or Distribute  14-73       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

1

2 14.6 Key Risks

3

4 Ten key risks from climate change were identified for North American based on definitions and assessment
5 approaches outlined in Chapter 16, which were extended to include the development of a risk database and
6 analysis that included expert evaluation of interactions between climate hazards and sectors (Figure 14.11,
7 SM14.3).

8
9

10

11 Figure 14.11: Rapid assessment of relative risk by sector (y axis) and climate hazard (x-axis) for North America based
12 on an assessment of asset-specific vulnerability and exposure across climate hazards (see SM14.3 for methodological
13 details). For each unique combination, the hazard by sector risk was ranked as very high (very high risk and high
14 confidence), high (significant impacts and risk, high to medium confidence), medium (impacts are detectable and
15 attributable to climate change, medium confidence), low or not detected (risk is low or not detectable). Blank cells are
16 those where the assessment was not applicable or not conducted. Risks identified through the rapid assessment were
17 further evaluated in the chapter assessments (see corresponding sector text for full assessment of risk and impacts).

18

19

20 14.6.1 Key Risks of Climate Change for North America

21

22 In North America, divergent perceptions regarding the attribution and implications of climate change pose a
23 key risk to adaptation mainstreaming (KR1). This lack of adequate adaptation in turn amplifies threats to
24 human life and safety from intensifying extreme events, fires, and storms (KR2). Climate change hazards
25 pose risks to economic and social well-being (KR3), marine social-ecological systems (KR4), unique
26 terrestrial ecosystems and their services (KR5), freshwater services (KR6), physical and mental health
27 (KR7), food and nutritional security (KR8), and commerce and trade (KR9). Cumulatively, these risks
28 interact to imperil the quality of life for North American communities, cities, and towns (KR10).

Do Not Cite, Quote or Distribute  14-74       Total pages: 157
    FINAL DRAFT                               Chapter 14                     IPCC WGII Sixth Assessment Report

1

2 14.6.2 Key Risks across Sectors in North America

3

4 KR1: In the public and policy domains, divergent perceptions of anthropogenic climate change pose a risk of

5   inaction on adaptation efforts to reduce exposure and socioeconomic vulnerability

6

7 Complex factors including individual beliefs, ideology, worldview, partisan identity, as well as societal

8 context influence how the public, as well as professional groups, communities, and policy makers, perceive

9 and understand climate change (14.3.3; 14.3.4) (high confidence). While there is expert scientific consensus

10 on anthropogenic climate change, rhetoric, misinformation and politicization of science have contributed to

11 misperceptions (high confidence), polarization on the severity of impacts and risks to society, indecision, and

12 delayed action (14.3.1) (high confidence). In North America, this impedes adaptation efforts (14.3.4) and

13 inflates climate risks (high confidence).

14

15 KR2: Risk to life, safety, and property from intensifying extreme events

16

17 Human life and safety across North America and especially along the coasts of Mexico, the Hawaiian

18 Islands, Gulf of Mexico, Atlantic Canada and southeastern US will be placed at risk from SLR and severe
19 storms and hurricanes, even at 1.5oC GWL (very high confidence) (14.5.2,14.5.5, Box 14.4). Warming,

20 heatwaves, and increases in wildfire activity in many regions of North America pose risks to air quality,

21 health, lives, and property (Box 14.2). More extreme precipitation and flooding pose a risk to human

22 morbidity, mortality, and safety in fluvial flood zones and areas downstream of levees, dams, and flood

23 culverts. Increasing intensity of storm events poses a risk of landslides, erosion, and flooding in shoreline

24 and urban communities, especially high bank areas along exposed coasts, in Arctic and temperate areas

25 where winter sea ice has diminished, and in low-lying coastal areas where SLR and storm surge often

26 overwhelm existing natural coastal features and engineered structures (14.5.5, Box 14.4).

27

28 KR3: Cumulative damages from climate hazards pose a substantial risk to economic well-being and shared

29  prosperity

30

31 Climate change impacts are projected to cause large market and non-market damages (high confidence). By

32 end-of-century under higher GWL scenarios (>4癈), these damages are expected to reach several tens of

33 billions of dollars/annually in Canada and hundreds of billions/annually in the United States. Losses in

34 labour productivity and wages, and damages to coastal properties will be especially large; however, all

35 sectors in the US and most sectors in Canada are projected to see substantial relative damages on high

36 emission pathways by mid to end-of-century compared to lower emission pathways. Economic sectors with

37 hard limits to adaptation (i.e., winter tourism) or that are highly affected by climate variability (i.e.,

38 agriculture and fisheries) will be at more risk at lower temperatures than other economic sectors (14.5.7;

39 14.5.8). Strategic implementation of adaptation strategies coupled with lower emissions scenarios result in

40 multi-billion-dollar reductions in economic damages (14.5.8, Box 14.6).

41

42 KR4: Risk of degradation of marine and coastal ecosystems, including loss of biodiversity, function, and

43  related services with cascading effects for communities and livelihoods

44

45 Ocean warming will increase the frequency and intensity of marine heatwaves (MHWs, Box 14.3),

46 accelerate unprecedented rates of sea ice loss, and alter ocean circulation, chemistry, and nutrient cycling in

47 ways that profoundly impact marine productivity, biodiversity, and foodwebs (very high confidence)

48 (Section 14.5.2). Collectively these impacts pose a risk to nearshore ecological and human systems (high

49 confidence), increasing the probability of phenological mismatches, large-scale redistribution of species, and

50 species population declines (14.5.4) with cascading impacts that strain cultural and economic systems reliant

51 on marine productivity across North America (high confidence). Nearshore areas of Chesapeake Bay (US)

52 and Akimiski Island, mid-western James Bay and the coasts in the Pacific ranging from of the Gulf of

53 Alaska through Baja Peninsula have a high proportion of species near their upper thermal limit, and are areas

54 of particularly climate change risk.

55

56 KR5: Risk to major terrestrial ecosystems leading to disruptions of species, ecosystems and their services

57

    Do Not Cite, Quote or Distribute          14-75                                           Total pages: 157
    FINAL DRAFT                          Chapter 14                   IPCC WGII Sixth Assessment Report

1 Major risks to terrestrial ecosystems across North America, such as semi-arid landscapes, rangelands, boreal

2 and temperate forests, and Arctic tundra, include significant ecosystem transformations and shifts in species

3 abundances and ranges and major vegetation types (e.g., transitions from forests to grasslands), with

4 cascading implications for regional biodiversity (very high confidence). Warming increases the risk of

5 permafrost thaw with propagating impacts on species and communities in the Canadian and US Arctic (high

6 confidence; Ch. 6). Forest disturbances, including wildfire, drought, insects, and pathogens are expected to

7 increase with warming, acting synergistically to raise the prevalence of tree mortality and ecosystem

8 transformation (medium confidence; 14.5.1). These changes will reduce services provided by terrestrial

9 ecosystems, including timber yields and carbon sequestration (medium confidence).

10

11 KR6: Risk to freshwater resources with consequences for ecosystems, reduced surface water availability for

12  irrigated agriculture and other human uses.

13

14 Droughts and earlier snowmelt runoff will increase water scarcity during the summer peak water demand

15 period especially in regions with extensive irrigated agriculture, leading to economic losses and increased

16 pressures on groundwater as a substitute for diminished surface water supplies (medium to high confidence;

17 14.5.3). Streams in North America are expected to continue to warm, with important ramifications for

18 aquatic ecosystems (high confidence), reducing habitat for salmon and trout species that are economically

19 and culturally important (14.5.1). Warming and drying coupled with other stressors (e.g., pollutants,

20 nutrients, and invasive species) pose a risk to ecosystem structure and function in lakes, streams and

21 reservoirs across many parts of North America (high confidence) (14.5.1, 14.5.3). Warming, increases in

22 heavy rainfall and nutrient loading pose risks for water quality and harmful algal blooms (medium to high

23 confidence; 14.5.3).

24

25 KR7: Risk to human health and wellbeing, including mental health.

26

27 Heat-related human mortality is projected to increase in North America as a result of climate change and

28 aging populations, poverty, chronic diseases and inadequate public health systems (very high confidence)

29 (14.5.6.1). Gradual changes to temperature and precipitation are impacting urban ecosystems and creating

30 ecosystem regime changes resulting in the poleward expansion among insects that bring risks related to

31 vector-borne diseases such as West Nile virus and Lyme disease (high confidence) (14.5.6). Climate change

32 is expected to lead to wide-ranging mental health challenges related to an increase in the psychological

33 burdens of climate change (high confidence), particularly for individuals with existing mental health

34 conditions, live in severely impacted areas, or who are reliant on climate for livelihoods and cultural well-

35 being (e.g., Indigenous Peoples and farmers) (14.5.6.8).

36

37 KR8: Risk to food and nutritional security through changes in agriculture, livestock, hunting, fisheries, and

38  aquaculture productivity and access

39

40 Cascading and interacting impacts of climate change threatens food systems and food and nutritional security

41 for many North Americans, especially those already experiencing food and nutritional scarcity, women and

42 children with high nutritional needs, and Indigenous Peoples reliant on subsistence resources (high

43 confidence) (14.5.6). In agricultural regions experiencing aridification and where water scarcity precludes

44 substantial expansion of irrigation, warming and extreme heat pose a risk to food and forage crop and

45 livestock production (14.5.4) (high confidence). Ocean warming and marine heatwaves will continue to

46 disrupt commercial capture fisheries through species redistribution and changes to yield (high confidence)

47 and warming waters and OA will increasingly impact aquaculture production (high confidence) (14.5.4).

48 Interactions between competing aspects of human security (e.g., food, energy, and water) will be exacerbated

49 by climate change (high confidence) (Sections 14.5.3, 43).

50

51 KR9: Risks to major infrastructure supporting commerce and trade with implications for sustainable

52  economic development, regional connections, and livelihoods

53

54 Climate change and extreme events are expected to increase risks to the North America economy via

55 infrastructure damage and deterioration (high confidence), disruption to operations, unsafe conditions for

56 workers (medium confidence), and interruptions to international and interregional supply chains (medium
57 confidence) (14.5.8, Box 14.5). These climatic impacts will have cascading implications for local livelihoods,

    Do Not Cite, Quote or Distribute     14-76                                       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 sustainable economic development pathways, regional connectivity and will reinforce pre-existing social
 2 inequities (medium confidence). Infrastructure damage will also disrupt economic activities, including
 3 manufacturing, tourism, fisheries, natural resource extraction, and energy production (high confidence) (14.5.8).

 4

 5 KR10: Risk to the quality of life in North American communities, cities, and towns

 6

 7 In major North American cities and settlements, vulnerability to climate change has increased and is
 8 projected to continue to rise (medium confidence) (14.5.5). Concentrated populations with unequal adaptive
 9 capacities, exposure of valuable assets, ageing infrastructure, differing degrees of institutional capacity and
10 effectiveness will underpin climate hazards (14.5.5). Coastal, riverine, and urban flooding displacing
11 communities and coastal ecosystems (14.5.5.2) will become a dominant risk to urban centres (high
12 confidence), will cause disruptions to transportation and trade infrastructure (14.5.8); large wildfires
13 endangering lives, livelihoods, property and key infrastructure, and economic activities and contributing to
14 compromised air quality and municipal water contamination (14.5.6, Box 14.2).

15

16 14.6.3 Cumulative risk, tipping points, thresholds and limits

17

18 Across North America, climate change poses a risk to social-ecological systems increasingly destabilized by
19 compounding climate impacts and non-climate pressures (high confidence) (14.5.1-3) that erode the
20 connectivity and redundancy underpinning system resilience (14.5.1-5) (Xiao et al., 2017a; Koven et al.,
21 2020; Malhi et al., 2020; Turner et al., 2020). Accelerating climate change and increasingly severe hazards
22 and shocks may induce abrupt changes or push systems, people, and species to critical points--i.e., tipping
23 points, where a small additional change causes a disproportionately large response, triggering feedbacks that
24 lock systems into novel regimes (Scheffer et al., 2001; Scheffer, 2010; Anderies et al., 2013; Lenton, 2013;
25 Iglesias and Whitlock, 2020; Lenton, 2020a). Climate change tipping points can compound and amplify
26 climate impacts and risk, induce disparate climate burdens and benefits across human and ecological
27 systems, and irreversibly restructure ecosystems and livelihoods (e.g., species extinctions, fisheries collapse,
28 community managed relocation) (Lynham et al., 2017). Examples of systems with potential tipping points in
29 North America include permafrost and sea-ice loss triggering transformation of ecological and human
30 systems (including substantial shipping opportunities) in the Arctic that are permanent and irreversible
31 except on geological timescales, and which are potentially underway (high agreement, low evidence) (14.6.2,
32 Box 14.3, CCP6), mid-latitude forest ecosystems at low to middle elevations in western North America
33 where wildfire and cumulative climate and non-climate pressures may restructure forests and succession in
34 ways that promote transition to new vegetation types (Section 14.5.1), and agricultural communities in
35 northern Mexico and the SW United States where aridification and drought may interact with water resource
36 policies, economic opportunities and pressures, and farm practices to induce either adaptation (via changes in
37 irrigation practices), or farm abandonment, land-use transformation, and livelihood changes (due to heat
38 stress, soil deterioration, or reduced economic viability) (14.5.3, 14.5.4, CCP 6) (Yumashev et al., 2019;
39 Turner et al., 2020; Heinze et al., 2021).

40

41 Identification of critical thresholds, elements, and connections within a system may also help identify
42 potential positive tipping points, i.e., focal components or processes in a system where a relatively small
43 investment or intervention can induce a large benefit and enable self-reinforcing transformative adaptation
44 (14.7, Ch 17) (T郻ara et al., 2018; Lenton, 2020b; Otto et al., 2020). Under low mitigation scenarios,
45 compounding risks and higher carbon emission scenarios increase the potential that amplifying feedback
46 loops and fatal synergies across sectors could lead to existential threats to the socio-ecological systems of
47 North America (medium confidence). Societal collapse has been linked to shifts in climate regimes,
48 especially when societies have lost resilience due to slowly mounting socio-ecological challenges; while
49 other studies reveal that social continuity and flexibility enable historical climate resilience and prosperity
50 under changing environments (Lenton et al., 2019; Otto et al., 2020; Degroot et al., 2021; Richards et al.,
51 2021) (FAQ 14.2).

52

53 Accounting for tipping points, interactions, and reinforcing dynamics among ecological, social, and climate
54 processes is necessary for comprehensive analyses of climate change risk, cost, and urgency, as well as
55 effective adaptation design and implementation (14.7) (Cai et al., 2015; Steffen and et al., 2018; Lenton et
56 al.; Narita et al., 2020; Dietz et al., 2021). Multiple lines of evidence across sectors assessed in this chapter
57 suggest that after mid-century and without carbon mitigation, climate-driven changes to ecological and social

Do Not Cite, Quote or Distribute  14-77       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 boundary conditions may rapidly push many systems into disequilibrium (medium confidence), emphasizing
 2 the importance of prioritizing adaptation actions with co-benefits for mitigation (14.5.4, Box 14.3). Reducing
 3 climate hazards through mitigation and removing catalysts of system instability through adaptation measures
 4 that increase system resilience (e.g., ecosystem restoration) will help reduce the risk that systems move
 5 across a tipping point from a desirable to an alternate or undesirable state (14.5.4, 14.7, Box 14.3) (Narita et
 6 al., 2020; Turner et al., 2020; Heinze et al., 2021).

 7
 8

 9 [START FAQ 14.2 HERE]

10

11 FAQ 14.2: What can we learn from the North American past about adapting to climate change?

12

13 The archaeology and history of Indigenous Peoples and Euroamerican farmers show that climate variability
14 can have severe impacts on livelihoods, food security, and personal safety. Traditional societies developed
15 numerous methods to cope with variability, but have always expanded to the limits of what those adaptations
16 permit. Current knowledge and technology can buffer societies from many negative effects of climate change
17 already experienced but will be severely challenged by the novel conditions we are now creating.

18

19 People came into North America more than 15,000 years ago and have experienced both massive and minor
20 shifts in climate ever since. At the end of the last very cold phase of the most recent Ice Age, about 11,500
21 years ago, temperatures rose extremely rapidly--as much as 10篊 (18篎) in a decade in some regions. This
22 undoubtedly contributed to the extinction of large mammals like mammoths and mastodons that people
23 hunted alongside many other resources (see Cross-Chapter Box PALEO in Chapter 1). There were so few
24 people on the land though, and other resources were so abundant, that the long-standing human means of
25 coping with climate variability--switching foods and moving on--were sufficient.

26

27 Following the end of the Ice Age, populations across North America grew for the next few thousand years, at
28 a rate that increased once people began to domesticate corn (maize), beans, and squash (the "Three Sisters")
29 as well as other crops. However, more people meant less mobility, and farmers are also more invested in
30 their fields and remaining in place than foragers are to hunting grounds. Other means of coping with
31 vulnerability to food shortage caused by climate variability included some continued hunting and gathering
32 of wild resources, planting fields in multiple locations and with different crops, storage in good years, and
33 exchange with neighbours and neighbouring groups.

34

35 According to archaeological evidence, however, these adaptation strategies were not always sufficient during
36 times of climate-induced stress. Human remains showing the effects of malnutrition are fairly common, and
37 conflict caused in part by climate-induced shortfalls in farming has left traces that include fortified sites, sites
38 placed in defensible locations, and trauma to human bone. Larger and more hierarchical groups emerged,
39 first in Mesoamerica and then in the US Southwest, Midwest, and Southeast. These groups offered the
40 possibility of buffering poor production in one area with surplus from another, but they also tended to
41 increase inequality within their borders and often attempted to expand at the expense of their neighbours,
42 introducing new sources of potential conflict. Dense hierarchical societies also arose in other areas such as
43 the Northwest coast where agriculture was not practiced but resources such as salmon and roots were
44 abundant and either relatively constant or storable.

45

46 These societies were not immune to climate hazards despite their greater population and more formal
47 organization. Archaeological evidence strongly suggests that drought, or growing conditions that were too
48 hot or cold, contributed to the decline of groups ranging from Classic period Maya states in Mesoamerica, to
49 the somewhat less hierarchical societies of Chaco in the US Southwest and Cahokia in the US Midwest
50 (Figure FAQ14.2.1). The usual pattern seems to be that climatic variability compounded social and
51 environmental problems that were already challenging these societies.

52

53 If societies in North America prior to the Euroamerican colonization were vulnerable to climate variability,
54 surely the more recent and technologically advanced societies of North America were at lower risk? The
55 20th Century Dust Bowl created in the US and Canadian prairies suggests otherwise. Severe drought
56 conditions throughout the 1930s--which to make matters worse peaked during the Great Depression--did
57 not cause either the US or Canada to collapse. But both countries suffered massive economic losses, regional

Do Not Cite, Quote or Distribute  14-78       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 loss of topsoil, and regional human strife (including loss of crops, income, and farms) leading to migration.
 2 Yet anthropogenic global climate change was of little or no consequence in the 1930s. While farming
 3 practices made climate stress worse, the climate variability itself was either completely, or mostly within the
 4 envelope of historical climate variability that earlier human societies had experienced.

 5

 6 Indigenous Peoples and Euroamerican farmers and ranchers have a long history of mostly successful
 7 adaptation to changing weather patterns. The wisdom held by Indigenous families and communities includes
 8 deep knowledge of how plants, animals, and atmospheric conditions provide early-warning signals of
 9 approaching weather shifts, and stories about how past communities have tried to cope with climate-related
10 resource shortfalls. Long-standing community-level management of resources also helps prevent shortfalls,
11 and institutions such as kin groups, church groups, clubs, and local governments (which exist in communities
12 of both Euroamericans and Indigenous Peoples, in different forms) can be powerful aids in ameliorating
13 shortfalls and resolving conflict.

14

15 Still, Indigenous knowledge, and traditional knowledge among Euroamerican farming communities, provide
16 guidelines for how to cope with traditional problems. Contemporary governmental restrictions (such as legal
17 water rights allocations, international borders and tribal lands boundaries) have limited the adaptive capacity
18 that Indigenous societies developed over the centuries. Now human-caused climate forcing, if not mitigated
19 by reducing heat-trapping greenhouse gases, is expected to produce climates in North America that have no
20 local analogs in human history even as it destroys heritage sites that are sources of knowledge about
21 paleoclimates and the diverse ways of coping with them that past people have discovered. Just as past
22 peoples often avoided local climate change by moving on, in a world where mobility options are severely
23 limited a lesson from archaeology and history is that we should use our hard-won knowledge of the causes of
24 climate change to avoid creating futures with no past analogs to provide useful guidance.

25
26

Do Not Cite, Quote or Distribute  14-79       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 Figure FAQ 14.2.1: Examples of areas where past climate variability has contributed to crises. Climatic variability is
 3 mostly likely to lead to crisis when it is accompanied by social, demographic, and political conditions or environmental
 4 mismanagement that compound climatic impacts on societies.

 5

 6 [END FAQ 14.2 HERE]

 7

 8

 9 [START FAQ 14.3 HERE]

10

11 FAQ 14.3: What impacts do changes in the North American Arctic have within and outside the
12 region?

13

14 The North American Arctic is warming at nearly three times the global average, creating a cascading web of
15 local, regional, and global impacts within and beyond polar regions. Changes in the Arctic not only effect
16 global ocean circulation and climate regulation, but also facilitate new Arctic transportation routes and
17 support transboundary resources with geopolitical, environmental, and cultural implications as conditions
18 change.

19

20 Rapid warming and extreme temperatures in the Arctic is leading to unprecedented seasonal sea ice loss,
21 permafrost thaw, and increasing ocean temperatures. Cascading from these biophysical changes are cultural,
22 socio-economic and political consequences that are widespread and largely unprecedented in human history.

Do Not Cite, Quote or Distribute  14-80       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Changes in sea ice create safety hazards for Indigenous Peoples and northerners who rely on frozen seas and
 2 rivers for transportation between remote communities and to subsistence hunting areas. Thawing permafrost,
 3 especially that of ice rich permafrost, creates challenges and costs for a region with low population density
 4 and a small tax base to support major infrastructure investments. Warmer ocean temperatures induce large-
 5 scale distributional shifts and reduced productivity and access to the largest North America fisheries. Ice-
 6 associated marine mammals, such as Polar bears, seals, and walrus, have declined precipitously with
 7 decreasing sea ice in the Bering Sea, and widespread ecosystem changes from fish through birds and marine
 8 mammal species have altered the system with uncertain outcomes for these productive ice-driven
 9 ecosystems. Newly ice-free shipping routes are increasing regional and geopolitical tensions and may
10 facilitate novel threats like the spread of invasive species and safety hazards to local hunters and fishers. The
11 local and regional impacts of climate change in the North American Arctic are profound and span social,
12 cultural, health, economic, and political imperatives.

13

14 Although the region is remote, changes in the Arctic impact the rest of the world. The Arctic serves as a
15 regulator of global climate and other ecological processes through large-scale patterns related to air and
16 ocean circulation. These vitally important processes are nearing points beyond which rapid and irreversible
17 (on the scale of multiple human generations) changes are possible. The magnitude of cascading changes over
18 the next two centuries includes regional warming and temperature extremes, permafrost declines, and sea ice
19 loss beyond that experienced in human existence. This includes macro-scale risks related to sea level rise
20 from the melting of glaciers and thermal expansion of oceans. Changes in the Arctic are more pronounced
21 than elsewhere and portend climate change impacts in other areas of the globe.

22

23 Adaptation in the Arctic is underway and lessons learned on what works and what is effective and feasible to
24 implement can provide global insights. Successful adaptation in the North American Arctic region has been
25 attributed, in part, to the explicit and meaningful inclusion of Indigenous Knowledge and Indigenous self-
26 determination, and diverse perspectives in decision making processes, strong local leadership, co-
27 management approaches, technological investment in integrated climate modeling and projections, and
28 multilateral cooperation.

29

30 [END FAQ 14.3 HERE]

31
32

33 14.7 Adaptation in North America

34

35 14.7.1 Overview of Observed Adaptation in North America

36

37 Climate adaptation efforts have increased across all North American regions and sectors (high confidence).
38 Support for and implementation of adaptation policies, plans and measures have not been equal across the
39 public and private sectors, regions or varying levels of governance (high confidence) (Table 14.7). To date,
40 reactive (coping-based) and incremental adaptations have helped North Americans avoid greater damages
41 from observed climate impacts (medium confidence). There is increasing agreement that worsening impacts
42 and expanding risk conditions may exceed current adaptation capacities by mid-century under high
43 emissions scenarios (RCP8.5) (medium confidence).

44

45 14.7.1.1 Individuals and Households

46

47 Across North America, individuals and households have taken action to reduce climate-influenced risks
48 (high confidence). These autonomous adaptations comprise the majority of the observed responses in the
49 peer-reviewed literature (Berrang-Ford and et al., Accepted). The increased use of cooling systems (which
50 could be maladaptive unless there are innovations (14.5.5.3)) (Barreca et al., 2016), creating defensible space
51 around homes in wildfire-prone areas (Box 14.2), and the modification or redesign of housing structures
52 along coasts (Koerth et al., 2017) are important household responses to existing risks. Although these actions
53 have played a role in reducing risks, the capacity to undertake such actions is not uniform across individuals
54 in North America and has exacerbated existing social inequities, especially in coastal areas (Keenan et al.,
55 2018; de Koning and Filatova, 2020). Additionally, these adaptation activities often are taken without
56 consideration of the impact on mitigation efforts (Kates et al., 2012; Fedele et al., 2019; Shi and Moser,
57 2021).

Do Not Cite, Quote or Distribute  14-81       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1

 2 14.7.1.2 Local and Sub-National Governments

 3

 4 A majority of local jurisdictions in North America have undertaken some level of adaptation; these efforts
 5 largely focused on planning and less on implementation (high confidence). Some sub-national governments,
 6 namely States and Provinces, have engaged in advanced adaptation planning efforts (high confidence).
 7 Indigenous Peoples in North America have undertaken substantial activities (14.4, Box 14.1).

 8

 9 Many cities across North America have undertaken adaptation planning (Hughes, 2015; Reich et al., 2016;
10 Moser et al., 2017; Auditors General, 2018; McMillan et al., 2019) (14.5) with some financing adaptation
11 implementation, for example, in the case of SLR (Box 14.4). Adaptation actions commonly implemented in
12 cities include climate-informed building codes, enacting energy conservation measures, modifying zoning,
13 and increasing green infrastructure (see 14.5.5.3; Box 14.7) (Binder et al., 2015; Maxwell et al., 2018a; Moss
14 et al., 2019; Brown et al., 2021). The majority of cities have formed practitioner networks to share
15 information (ICLEI Canada, 2016; Vogel et al., 2016; C40 Cities, 2018) and supporting learning and
16 collaboration through regional collaborations that include utility managers and the private sector (F黱fgeld,
17 2015; Moser et al., 2017).

18

19 In Canada, the Map of Adaptation Actions Canada (https://changingclimate.ca/case-studies/) presents over
20 200 adaptation case studies addressing a variety of climate-related impacts (Warren and Lulham, 2021). The
21 City of Saskatoon, in developing its Climate Action Plan (which includes a Corporate Climate Adaptation
22 Strategy), engaged with local businesses, NGOs, residents and experts to identify potential risks (and
23 benefits) requiring action (City of Saskatoon, 2019). Similarly, the City of Surrey specifically used
24 community outreach programs to develop its Coastal Flood Adaptation Strategy (CFAS) through a value-
25 based planning approach (City of Surrey, 2019). Municipal asset management, local services and
26 community well-being were key considerations for the City of Selkirk, Manitoba when developing an
27 adaptation strategy as well as ensuring a budgeting process that supports implementation (City of Selkirk,
28 2019). As of 2019, eight of thirteen Canadian provinces and territories have high-level climate adaptation
29 strategies. The scope of these efforts vary by jurisdiction as a review conducted by federal and provincial
30 auditors in Canada identified several deficiencies related to a lack of detailed implementation plans,
31 obligated funding, and specific timelines (Auditors General, 2018).

32

33 Progress in Mexico on adaptation implementation at the local level has been extensive (INECC and
34 Semarnat, 2018). Activities include executing programs for relocating infrastructure in high-risk zones in
35 priority tourist sites, incorporating adaptation criteria in public investment projects that involve construction
36 and infrastructure management, water management, application of climate adaptation norms for the
37 construction of tourist buildings in coastal zones, and improving the security of key water, communication,
38 and transportation infrastructure (14.5.5, 14.5.7, 14.5.8). Additionally, local capacity and protocol to respond
39 to extreme weather events as a function of climate change have been integrated more regularly into
40 community-based hazard mitigation plans. States and municipalities in Mexico must have climate policies
41 that are consistent with the guidelines of national strategies (see 14.7.1.5) and state-level programs on
42 climate change, in addition to other state and municipal laws. As a result, these entities have developed and
43 implemented early warning systems designed to protect the population from climate-related risks, such as
44 strong storms and hurricanes (INECC and Semarnat, 2018).

45

46 Implementation of adaptation initiatives and specific actions in US cities has increased in the approximately
47 five years between the 3rd US National Climate Assessment (NCA3) (Melillo et al., 2014) and the 4th
48 Assessment (NCA4), and adaptation responses have been observed widely (Lempert et al., 2018). ICLEI-
49 USA provides numerous resources for adaptation planning and implementation for cities, Indigenous
50 Peoples, and Regional Governments (https://icleiusa.org/). The Georgetown Center for Climate maintains a
51 comprehensive resource for tracking adaptation progress for States
52 (https://www.georgetownclimate.org/adaptation/plans.html). As of 2021, 18 US states have completed
53 climate adaptation plans, and six states have plans underway as of the time of this report (Georgetown
54 Climate Center, 2021). California, in particular, has adopted sustained climate assessment to allow for more
55 rapid iterations on adaptation planning (Bedsworth et al., 2018; Miao, 2019). Across all US states, however,
56 adaptation activities do not have readily accessible budgets, such that levels of funding cannot be assessed
57 directly (Gilmore and St. Clair, 2018).

Do Not Cite, Quote or Distribute  14-82       Total pages: 157
    FINAL DRAFT                                Chapter 14  IPCC WGII Sixth Assessment Report

1

2 14.7.1.3 National/Multi-National Governance

3

4 The federal government of each North American country has developed policies and actions that promote

5 climate adaptation (Figure 14.12). Recognizing the cultural, economic and social networks that span North

6 America, the federal governments have also committed to engagement on adaptation and resilience across

7 borders and through cooperation on domestic adaptation efforts (The White House, 2016). Each country also

8 outlines their respective adaptation efforts through submissions under the UN Framework Convention on

9 Climate Change (UNFCCC), including their Nationally Determined Contributions (NDCs) under the Paris

10 Agreement. The federal governments also support adaptation efforts in other countries through international

11 climate negotiations as well as related agreements, such as the Sendai Framework for Disaster Risk

12 Reduction and efforts to support the achievement of the Sustainable Development Goals (SDGs).

13

14 Mexico's 2020 update to its first NDC communicated extensive adaptation efforts (Government of Mexico,

15 2020). The measures outlined in this document highlight the importance of co-benefits for adaptation efforts

16 as they relate to the SDGs and to support mitigation commitments. Ecosystem- and nature-based solutions

17 (Box 14.7) are the basis for much of the synergies between adaptation and mitigation efforts. These plans are

18 supported by domestic legislation through the General Law on Climate Change, which includes the Climate

19 Change Adaptation Process (CCAP). CCAP provides a holistic systems-approach for identifying instruments

20 and institutional arrangements for adaptation implementation (Semarnat and INECC, 2015; INECC and

21 Semarnat, 2018). This approach includes guidance for planning (e.g., the Climate Change Mid-Century

22 Strategy, the Special Climate Change Program 2014�2018 (PECC)) and formalizes its adaptation

23 commitments to the Paris Agreement.

24

25 In Canada, the Federal Adaptation Policy Framework (Government of Canada, 2011) guides domestic action

26 to develop adaptation knowledge, build adaptive capacity, and mainstream adaptation into federal policy, in

27 support of the Pan-Canadian Framework on Clean Growth and Climate Change (Government of Canada,

28 2016), which included specific adaptation measures and investments to build resilience. In August 2021, the

29 government initiated a National Adaptation Strategy with development anticipated through 2022.

30 Additionally, the government facilitates efforts and funds research, capacity building, and information

31 sharing across sectors and amongst government departments (Government of Canada, 2021a). The Canadian

32 Centre for Climate Services provides access to climate data, tools, and information

33 (https://www.canada.ca/en/environment-climate-change/services/climate-change/canadian-centre-climate-

34 services.html). In Canada's revised NDC, near-term commitments to protecting land and oceans and efforts

35 related to sustainable and resilient energy systems are highlighted as examples of co-benefits between

36 climate change adaptation and mitigation (Government of Canada, 2021b).

37

38 The US has experienced substantial revisions to its climate policy and its international engagement since

39 AR5 with implications still unclear (Bomberg, 2021). Since AR5 and until early 2020, many congressionally

40 mandated federal efforts (Beavers et al., 2016; Parris et al., 2016; Rockman et al., 2016; Caffrey and

41 Hoffman, 2018) faced programmatic challenges, but most continued to provide research and capacity

42 development to support adaptation implementation across the US. Importantly, the US government sustained

43 the national climate assessments (Lempert et al., 2018). Recently, the administration has re-engaged with the

44 Paris Agreement and the US has submitted an NDC (Government of the United State of America, 2021);

45 however, adaptation was not directly addressed. Subsequent Executive orders mandate adaptation planning

46 at the federal level (e.g., USEO 13754; USEO 14008). As of the time of this report, the US climate policy

47 landscape is rapidly evolving, including major legislative initiatives (e.g., Green New Deal (Boyle et al.,

48 2021).

49

50 14.7.1.4 Private Sector, Including Companies, NGOs, Professional Organisations, Academic Institutions,

51         and Communities of Practice

52

53 The private sector comprises a diverse set of actors who influence, interact with and support adaptation

54 efforts, generally through shared governance with the public sector. The weight of evidence points to the

55 benefits of these collaborations and the importance of voluntary code-making and self-regulation

56 (17.4.2.1.6). In North America, non-governmental organisations (NGO) and professional organisations have

57 been important agents of change in the adaptation field (Bennett and Grannis, 2017; Stults and Meerow,

    Do Not Cite, Quote or Distribute           14-83                                    Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 2017). Efforts included supporting community-based resilience efforts, network-building, web-based
 2 guidance and resources, case studies, workshops and other services to support adaptation action (e.g.,
 3 vulnerability assessments, scenario-based planning).

 4

 5 Market and financial mechanisms have provided important buffering capacity against climate shocks in
 6 North America. Insurance products are being developed to meet emerging climate risks, especially related to
 7 availability and pricing of flood insurance in Canada (Thistlethwaite, 2017; Davies, 2020) and the United
 8 States (Kousky et al., 2021). Some existing US flood insurance products provided through joint public and
 9 private arrangements has led to rebuilding in flood-prone locations (Zellmer and Klein, 2016). The price of
10 these products may limit their uptake in low income neighbourhoods (Cannon et al., 2020).

11

12 Professional organisations have participated in the development and adoption of measures to integrate
13 climate resilience into the built environment. This includes new designs, guidelines, codes, standards, and
14 specifications, in addition to infrastructure inventories that incorporate evaluation of vulnerabilities and
15 identification of priority at-risk areas (Amec Foster Wheeler Environment and Infrastructure, 2017; ASCE,
16 2018a). These efforts are supported by provincial/state and federal initiatives (e.g., Canada's Climate Lens
17 (Infrastructure Canada, 2018), and California's Climate-Safe Infrastructure Working Group (Climate-Safe
18 Infrastructure Working Group, 2018)). Infrastructure Canada has undertaken Canada-wide initiatives to
19 improve infrastructure resilience to climate change (https://www.infrastructure.gc.ca/plan/crbcpi-irccipb-
20 eng.html). The Standards Council of Canada (SCC) established the Northern Infrastructure Standardization
21 Initiative (NISI) (https://www.scc.ca/en/nisi) engaging stakeholders including Indigenous Peoples to develop
22 standards specific for addressing climate change impacts on northern infrastructure design, planning and
23 management, and community development (Standards Council of Canada, 2020).

24

25 Professional organisations in the US (e.g., National Medical Association, American Institute of Architects,
26 Association of Metropolitan Water Agencies, Water Utility Climate Alliance, American Society of
27 Adaptation Professionals, etc.) have engaged with their members particularly through training about urban
28 adaptation (Stults and Meerow, 2017). The private sector and citizens (Klein et al., 2018) have been involved
29 in the management of increasing flood risk, such as the adoption of property-level flood protection
30 (Thistlethwaite and Henstra, 2018; Valois et al., 2019), implementing FireSmart Canada and Firewise USA
31 guidance (Box 14.2). In Canada, Engineers Canada developed the PIEVC Protocol to provide guidance for
32 professionals in engineering and geoscience (https://www.pievc.ca/).

33

34 Research-based institutions have accelerated the development of web-based tools for visualizing and
35 exploring climate information, in addition to furthering the scholarship on adaptation. In the US, joint
36 university, foundation, and government programs have contributed to advancing the field with products such
37 as oceanographic and fishery climate forecasting tools (14.5.2), in addition to methods for evaluating water
38 resource plans under uncertainty about future mean and extreme conditions (ASCE, 2018a; Ray et al., 2020).
39 Some regional research centres focus on stakeholder engagement in addition to research; these include the
40 National and Regional Climate Adaptation Science Center Network of the US Geological Survey
41 (https://www.usgs.gov/ecosystems/climate-adaptation-science-centers), the US Department of Agriculture's
42 Climate Hub Network (https://www.climatehubs.usda.gov/), and the Climate Program Office of NOAA
43 (https://cpo.noaa.gov/) includes the Regional Integrated Science Assessment Network
44 (https://cpo.noaa.gov/Meet-the-Divisions/Climate-and-Societal-Interactions/RISA/About-RISA) to support
45 delivery of climate services. So-called "networks of networks," consisting of NGOs, state and city
46 government programs, have provided an alternative to federal support. For example, the Science for
47 Adaptation Network (SCAN) formed subsequent to dismantling the Federal Advisory group to the US
48 National Climate Assessment (Moss et al., 2019).

49

50 14.7.2 The Solution Space

51

52 14.7.2.1 Incremental Adaptation, Barriers and Limits

53 Adaptation actions to moderate the effects of climate impacts are well-documented in North America and have
54 buffered much of the past and currently observed climate impacts (e.g. Lempert et al., 2018; Lemmen et al.,
55 2021). While it is challenging to catalogue adaptation activities as many are not published or are not necessarily
56 undertaken with climate adaptation as the primary rationale (1.3.2.2), most of the activities identified by sector

Do Not Cite, Quote or Distribute  14-84       Total pages: 157
FINAL DRAFT                                         Chapter 14             IPCC WGII Sixth Assessment Report

1 in this Chapter have been primarily incremental adaptation measures (medium evidence, high agreement).
2 Many actions are extensions of existing practices for managing climate variability and there is broad agreement
3 that worsening future conditions will exceed the capacity of many of these efforts (Kates et al.; Termeer et al.,
4 2017; Fazey et al., 2018; Fedele et al.; Shi and Moser, 2021).

 5 Progress in adaptation planning and implementation between regions in North America is uneven (Bierbaum
 6 et al., 2013; Moser et al., 2017; Auditors General, 2018; INECC and Semarnat, 2018; Shi and Moser, 2021)
 7 (Table 14.6, Box 14.7). At the local level (cities) in the US, commitment of elected officials, financial
 8 resources and awareness of climate change hazards and risks have been identified as driving the variation in
 9 climate adaptation (Shi et al., 2015). Adaptation programs have come under budgetary and political
10 pressures that limit continuity of efforts (Moss et al., 2019). Implementation of adaptation has also faced
11 challenges due to institutional arrangements, constraints, and gaps that prevent different levels of
12 government, social organizations and academia to act in an integrated and timely way to consider
13 biodiversity, agriculture and water systems (i.e., Box 14.7) (Bourne et al., 2016; Nalau et al., 2018).

14
15

16 Table 14.6: Adaptation trends and progress across sectors. Adaptation progress consists of assessment (A), planning
17 (P), implementation of strategies (I), and evaluation of efficacy (E). L=low, M=moderate, H=high.

                                                          Adaptation                           Limits
                                                           progress

Sector       Strategies           Cases                   A PI E           Soft                        Hard
Terrestrial
Eco-         Broad use of Planning for climate H                 H L- L    Management agency           Some species
Systems                                                               M    internal policies may       may face local
(14.5.1.1)   tools such as refugia in the Sierra                           prevent the flexibility     extirpation or
                                                                 H MM      required for                even
Oceans       scenario             Nevada of                                implementation of           extinction if
(14.5.2)                                                         L- L- L-  adaptation strategies       adaptive
             planning,            California, USA                M M M                                 capacity is
Freshwater                                                                                             overwhelmed
Resources    structured           (Morelli et al., 2016)
(14.5.3)
             decision making,

             and adaptation

             planning

             frameworks

             Proactive and Dynamic closure H                               Lack of coordination and Marine

             rapid                areas to reduce                          planning at multiple        species

             management loggerhead turtle                                  scales as species           mortality

             approaches to bycatch in Hawaiian                             redistribute across fishery events

             minimize impacts shallow-set longline                         areas, marine protected

             of increasingly fisheries (Howell et                          zones, and international

             frequent             al., 2015; Lewison et                    and jurisdictional

             entanglements of al., 2015), blue                             boundaries

             protected species, whale ship-strike

             caused by            risk in near-real time

             climate-driven (Hazen et al., 2017;

             changes in prey Abrahms et al.,

             and fishery          2019a), and bycatch

             activities           of multiple top

                                  predator species in a

                                  West Coast drift

                                  gillnet fishery

                                  (Hazen et al., 2018).

             Forecasting and Reduced human M                               Financial resources         Severe human

             warning of           exposure to the                          required to enhance water health effects;

             harmful algal increased risk of                               treatment facilities to deal mortality of

             blooms (HABs) toxins from HABs in                             with HABs; technological aquatic

             that affect water the Great Lakes                             innovation to improve species

             quality                                                       treatment and removal of

                                                                           HABs; closure of

                                                                           recreational water use

Do Not Cite, Quote or Distribute                          14-85                                        Total pages: 157
FINAL DRAFT                                            Chapter 14             IPCC WGII Sixth Assessment Report

Water         Water allocation US Colorado River H               H M L-       Complex legal and          Depletion of
Availability                                                               M
(14.5.3)      policies            interstate shortage                         administrative challenges, finite
                                                                 H M- M
              reassessed to sharing agreement                         H       heightening lengthy        groundwater

              enhance equity,                                    H M L-       disputes and costly        resources and
                                                                           M
              sustainability and                                              interstate legal battles reduced flow
                                                                 H L- L-
              flexibility in                                          M M                                in

              times of shortage                                  H ML                                    hydrologically

              through sharing                                    H ML                                    connected

              agreements,                                                                                rivers

              improved

              groundwater

              regulation and

              voluntary water

              transfers

Food & Fibre Improved climate Fishing communities H                           Lack of high resolution    Collapse of
                                                                              and locally tailored       fisheries and
(14.5.4)      resilience through in the US-SW and                             climate change             loss of crops
                                                                              information                due to
              increasing          US-NE through                                                          excessive
                                                                                                         warming and
              income and          nature-based                                                           extreme
                                                                                                         events
              harvest/crop        aquaculture solutions

              portfolio           (Messier et al., 2019;

              diversification Rogers et al., 2019;

                                  Young et al., 2019;

                                  Fisher et al., 2021)

Cities &      Consideration of Municipal Natural H                            Organizations'             Rate and
                                                                              willingness to take on     magnitude of
Infrastructure the value of green Assets Initiative                           solutions that are         climate
                                                                              emergent and less tested;  changes
(14.5.5)      infrastructure and (MNAI) assists                               capacity for               exceed
                                                                              municipalities to          capacity of
              natural assets to Canadian                                      undertake the              natural/green
                                                                              development and            infrastructure
              meet a range of municipalities to                               assessment this new        to cope
                                                                              infrastructure
              adaptation needs integrate natural

              related to          assets in financial

              flooding, extreme planning and asset

              urban heat, SLR, management

              drought             programs and

                                  consider projected

                                  climate changes

                                  (Municipal Natural

                                  Assets Initiative,

                                  2018)

Health &      Access to green The heatwave plan H                             Lack of effective warning Extreme

Communities spaces, cooler for Montreal                                       and response systems, increase heat-

(14.5.5, 14.5.6) infrastructure, includes visits to                           ability to reach at-risk related

              and cooling         vulnerable                                  populations, building mortality and

              stations            populations, cooling                        designs, enhanced          morbidity

                                  shelters, monitoring                        pollution controls, urban

                                  of heat-related                             planning strategies, and

                                  illness, and extended                       affordable, resilient

                                  hours for public                            health infrastructure

                                  pools (Lesnikowski

                                  et al., 2017)

Tourism &     Diversification of Investments in           H                   Social inequalities        Lack of
recreation                                                                    generated by the tourism   precipitation
(14.5.7)      winter-focused climate-resilient                                development process not    that falls as
                                                                              considered, such as        snow
              recreation and infrastructure within                            increased property taxes   particularly in
                                                                              leading to the             lower
              tourism             Canadian National                           marginalization of local   elevation
                                                                              residents in favour of     areas
              opportunities Parks have increased                              wealthy tourists

                                  visitation rates

                                  during the shoulder

                                  seasons (Fisichelli et

                                  al., 2015; Lemieux et

                                  al., 2017; Wilkins et

                                  al., 2018)

Commerce & Improved               For roads, changing H                       Lack of financial          Extreme
                                                                              resources to build         events may
transportation engineering and pavement mixes to                              climate-resilient          cause

(14.5.8)      technological be more tolerant to

Do Not Cite, Quote or Distribute                          14-86                                          Total pages: 157
    FINAL DRAFT                                          Chapter 14    IPCC WGII Sixth Assessment Report

                  solutions, in       heat or frost heaving,           infrastructure,                 significant

                  addition to         expanding drainage               particularly in                 and

                  innovative policy, capacity, reducing                marginalized                    irreversible

                  planning,           flood risks,                     communities                     impacts on the

                  management, and enhancing travel                                                     transportation

                  maintenance advisories and alerts,                                                   sector with

                  approaches          elevating or                                                     major

                  enhance climate relocating new                                                       implications

                  resilience for infrastructure where                                                  for supply-

                  transportation & feasible and                                                        chains and

                  related commerce changing                                                            global trade

                                      infrastructure design

                                      requirements

                                      (Natural Resources

                                      Conservation

                                      Service, 2008; EPA,

                                      2017; Pendakur,

                                      2017).

1

2

3 Adaptive capacity in the face of climate risks and impacts has not been equal across North American

4 communities (Sarkodie and Strezov, 2019). Lack of representation, health inequities, and economic

5 constraints adversely affect the capacity to respond to change and further exacerbate marginalization. For

6 example, within many water basins in Canada and the US, planning processes are often hampered by

7 conflicting interests, asymmetric information and differential power (ICLEI Canada, 2016; Nordgren et al.,

8 2016; Woodruff and Stults, 2016).

9

10 The absence of evidence about the current effectiveness of proposed adaptation actions to guide future

11 actions and investments presents a serious risk to North America, especially at higher global warming levels

12 (GWLs) (medium confidence). Evaluating the limits to adaptation and the effectiveness of adaptation actions

13 is hindered by a lack of monitoring and evaluation (Auditors General, 2018; Dilling et al., 2019; Berrang-

14 Ford and et al., Accepted). Incremental, passive adaptations are often characterized by soft limits due to

15 differing access to resources and by perceptions and tolerance of risk (Moser, 2010; Dow et al., 2013). At

16 current warming levels, socio-ecological systems have been reaching limits to adaptation in regions with

17 high exposure and high sensitivity (medium confidence). However, the implications for adaptation are

18 unclear as soft adaptation limits are mutable and change with evolving knowledge, values, interests, and

19 perspectives involved in decision making (Adger et al., 2009; Moser et al., 2017). Hard limits have been

20 identified for some natural systems, such as species extinctions (14.5.1.3, Table 14.2, 14.5.2.1).

21

22 Adaptation actions in one place or sector can have adverse side effects elsewhere (medium confidence). For

23 example, increased use of groundwater for irrigation in response to aridification can reduce baseflows into

24 rivers with adverse impacts on stream ecology and water availability for communities far downstream

25 (14.5.3). Additionally, across multiple sectors in North America, adaptation actions have tended to be sector-

26 specific rather than integrating across systems (Gao and Bryan, 2017; Fulton et al., 2019), despite the

27 increasing awareness of cascading impacts and interdependencies (Zimmerman and Faris, 2010; C40 Cities

28 and AECOM, 2017) and risks from possible ecological and social thresholds that have been identified under

29 higher GWL (14.6.3). For example, the water, energy and food nexus in North America has highlighted that

30 food, water, and energy security depend on transportation infrastructure (Romero-Lankao et al., 2018)

31 (14.5.8.1.2).

32

33 14.7.2.2 Adaptation Through Participatory and Robust Decision-Making, Indicators and Sustained

34  Assessments

35

36 In response to some of the challenges presented in 14.7.2.1, substantial progress has been made in the North

37 American context on the development of climate services, indicators, sustained assessments, and

38 participatory and stakeholder-driven robust decision-making (medium confidence) (Fazey et al.; Fedele et al.,

39 2019; Moss et al., 2019; Boon et al., 2021; Werners et al., 2021).

40

    Do Not Cite, Quote or Distribute                          14-87                     Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Decision-making related to adaptation policies, plans and projects has become more formalized, emphasizing
 2 participatory governance and co-production of knowledge. Canada has improved capacity with its Canadian
 3 Expert Panel on Climate Change Adaptation and Resilience Results (EPCCARR) and the recent National
 4 Adaptation Plan (14.7.1.5), with the development of a series of indicators to measure progress on adaptation
 5 (EPCCAR, 2018; Government of Canada, 2021a). In the US, indicators have been developed to
 6 communicate climate risks and guide adaptation efforts from Federal (Kenney et al., 2020) to more regional
 7 initiatives (Kenney and Gerst, 2021). These climate indicators have been used to support user-driven
 8 assessments and to articulate adaptation goals (Moss et al., 2019; Kenney et al., 2020). However, these
 9 frameworks have not sufficiently incorporated monitoring and evaluation into adaptation plans (Lempert et
10 al., 2018; Kenney et al., 2020). Tools and services to facilitate risk assessment and action planning have been
11 made available through federal government climate service efforts and guidance for their use has been
12 developed (Vano et al., 2018). However, these products have been characterized as insufficiently developed
13 to allow all adaptation practitioners to use these services (Meerow and Mitchell, 2017).

14

15 Throughout North America, co-development (or co-production) of adaptation efforts among stakeholders
16 who share common climate vulnerabilities or risk levels (e.g., individuals, groups, communities, businesses
17 or institutions) has been a core attribute of adaptation planning (Mees et al., 2016) and ranges across many
18 sectors (e.g., Box 14.1, 14.5.2.2, 14.5.3.3, 14.5.4.3). Participatory efforts and robust decision-making have
19 also been observed; some integrated watershed planning processes have high degrees of sustained
20 stakeholder involvement (14.5.3.3) (FAQ 14.4; (Harris-Lovett et al., 2015; Cant�, 2016)).

21

22 14.7.2.3 Transformational Adaptation and Climate Resilience

23

24 Climate change and its projected impacts pose a substantial risk to North America as a region as well as to
25 sectors, communities, and individuals (14.6.2). Incorporating different values and knowledge systems,
26 consideration of equity and justice as core objectives, and addressing underlying vulnerabilities are
27 principles that can guide transformational adaptation and resilience (medium confidence).

28

29 Approaches that advance adaptation within the existing contexts (finances, institutions, processes) have been
30 increasingly promoted by governments to mainstream climate risk into all considerations (Rosenzweig and
31 Solecki, 2014; Van der Brugge and Roosjen, 2015; Boon et al.; Shi and Moser, 2021). Policies and programs
32 that build upon existing approaches that have inherent climate resilience including Indigenous knowledge-
33 based land and resource management (14.5.4), co-management of agriculture and freshwater resources
34 (Section 14.5.3), nature-based solutions (Box 14.7), links between health and equity, and ecosystem-based
35 management (Section 14.5.2, 14.5.3, 14.5.4) have advanced sustainable and equitable climate resilience.
36 Implementing the recommendations in the ASCE committee's report on adaptation to a changing climate
37 (2018a) and Canada's Infrastructure and Buildings Working Group report has been identified as an
38 opportunity to improve social equity by ensuring the resilience of infrastructure and the services it provides,
39 through adoption of standards and good asset management practices (Amec Foster Wheeler Environment
40 and Infrastructure, 2017; ASCE, 2018a).

41

42 Long-term policy signals to incentivize ongoing, scalable adaptation action that is coordinated with
43 mitigation efforts will increase actions and avoid potential maladaptive investment (Moser, 2018;
44 Shi & Moser 2021). Using SDG goals and the NDCs as a framework for inclusive and coordinated
45 partnership and vertical integration across sub-national, national and regional planning can promote
46 climate resilient development (CRD) (18.1.3). Coordination of policies and responses have been
47 identified as supporting longer-term, transformational adaptation and minimizing risk (Termeer et
48 al., 2017; Fazey et al., 2018). New approaches for enabling and incentivizing transformative
49 adaptation in North America are rapidly emerging (Colloff et al. 2017, Fedel et al. 2019, Werners et
50 al. 2021). Evaluation of the feasibility of evolving adaptation strategies is only in the early stages,
51 but recent work has provided the foundation for assessing these considerations (Chapter 16, Table
52 14.7).

53
54

55 Table 14.7: Simplified example for transitioning from incremental to transformative adaptation approaches to support
56 future climate-resilient sustainable development. Modified from IPCC SR1.5 adaptation feasibility assessment for Land

Do Not Cite, Quote or Distribute  14-88       Total pages: 157
FINAL DRAFT                                  Chapter 14           IPCC WGII Sixth Assessment Report

1 and Ecosystem Transitions (IPCC, 2018). Feasibility Dimensions (can be barriers and/or enablers): Economic (EC),
2 Technological (TEC), Institutional (INST), Socio-cultural (SOC), Environmental/Ecological (ENV), Geophysical
3 (GEO) (Chapter 16).

                      Adaptation Approaches                       Mitigation Feasibility Dimensions

Hazard Response       Incremental Transformational Evidence/ Co-benefits Barriers             Enablers
                                                                   Agreement

Extreme   Integrated  Restoration of  Restoration of      Medium  Conservation   Sectors      Develop
storms    Ecosystem   stream          streambanks and             of soil and    working      coordinated
causing   and         corridors to    beds to stabilize           increased      in silos,    suite of
severe    Watershed   incorporate     and slow flows;             opportunity    inadequate   adaptation
flooding  management  environmental   use drought-                for carbon     financing,   efforts, co-
and                   flows;          tolerant plantings          sequestration  inability    produced
erosion               continuing to   and shade trees to                         to identify  among
                      build           reduce                                     shared       stakeholders
                      hardened        evaporation rates;                         goals (EC,   and across
                      surfaces and    incorporate                                INST,        sectors
                      stream          impervious                                 SOC,         (INST,
                      diversions in   surfaces in urban                          GEO)         SOC, ENV)
                      urban areas to  settings in
                      accommodate     combination with
                      infrequent yet  designating wide
                      extreme storm   buffer area within
                      events          floodplains to
                                      accommodate
                                      increased
                                      frequency of
                                      extreme events;
                                      integrate equity
                                      & justice
                                      considerations

 4
 5

 6 Differing values, perspectives, interests, and needs of relevant actors (Dittrich et al., 2016) through
 7 participatory processes, such as co-production of knowledge (Meadow et al., 2015; Wall et al., 2017), have
 8 been incorporated through the Resilience Dialogues (http://www.resiliencedialogues.org/), and the
 9 development of guidance on climate scenarios (Chaumont, 2014). Framing of adaptation goals strongly
10 determines beneficiaries of resultant policies and underscores the importance of a plurality of perspectives in
11 adaptation governance (Cochran et al., 2013; Plummer, 2013; Allison and Bassett, 2015; Raymond-
12 Yakoubian and Daniel, 2018). Sustained engagement through iterative knowledge development, learning,
13 and negotiation has been identified as core for addressing climate risks (Kates et al.; Seijger et al., 2014).
14 Interdisciplinary and inclusive adaptation programs that embrace and plan for conflict and resolution, and
15 address inequalities have been part of broadening the opportunities for engagement (Cant�, 2016; Termeer et
16 al., 2017; Parlee and Wiber, 2018; Sterner et al., 2019; Haasnoot et al., 2020).

17

18 Equity and justice in climate adaptation have been identified as providing a foundation for resilience in
19 natural, social, and built systems (Cochran et al., 2013; Reckien et al., 2017; Schell et al., 2020). This
20 approach recognizes that social vulnerability undermines efforts to increase adaptive capacity and that
21 adaptation may also entrench existing social inequities, such as marginalization of communities of colour,
22 gender discrimination, legacy effects of colonisation, and gentrification of coastal communities (Schell et al.,
23 2020; Thomas, 2020). Thus, identifying systemic racism and effects colonialism within and across
24 institutions has also been identified as part of achieving more just and equitable adaptation (Shi & Moser
25 2021). Acknowledgment and incorporation of Indigenous knowledge in adaptation planning and
26 implementation also recognizes Indigenous sovereignty issues and the importance of the equitable role of
27 Indigenous self-determination in governance and planning (Raymond-Yakoubian and Daniel, 2018) (Box
28 14.1; 14.4).

29

Do Not Cite, Quote or Distribute             14-89                                 Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Strategies have been emerging to facilitate progress by including specific guidance on tools for financing and
 2 funding climate change adaptation infrastructure (Berry and Danielson, 2015; Chen et al., 2016; Zerbe,
 3 2019). This includes facilitating transitions between incremental and transformational efforts to facilitate
 4 CRD (Chapter 18, the Five Transitions) (Fig. 14.12).

 5

 6 The extent to which resilient infrastructure contributes to social justice and equity has also been taken into
 7 consideration (Climate-Safe Infrastructure Working Group, 2018; Doorn, 2019). Proactive actions focused
 8 on small towns and rural areas--including the interdependencies between cities and surrounding areas--
 9 increases the potential that small and medium cities can build adaptive capacity at a pace that is
10 commensurate with present and future risks (Moss et al., 2019; Vodden and Cunsolo, 2021). This
11 coordination also creates greater opportunity for translation of knowledge into practice and assessing
12 knowledge in the context that it is to be applied to improve decision-making across scales (Enquist et al.,
13 2017; Moss et al., 2019).

14
15

16

17 Figure 14.12: Conceptual diagram of the key elements for expanding the adaptation solution space and implementing
18 climate-resilient development (Chapter 18). Figure adapted from Shi & Moser (2021).

19

20

21 [START BOX 14.7 HERE]

22

23 Box 14.7: Nature-based Solutions to Support Adaptation to Climate Change

24

25 Nature-based Solutions (NbS) are "actions to protect, sustainably manage, and restore natural or modified
26 ecosystems, that address societal challenges effectively and adaptively, simultaneously providing human
27 well-being and biodiversity benefits" (IUCN, 2016). NbS in the context of climate change, or Nature-based
28 Adaptation (NbA; Box 1.3), can jointly address multiple social-ecological issues related to climate change
29 hazards, impacts, adaptation and mitigation (Figure Box14.7.1, Cross-Chapter Box NATURAL in Chapter
30 2). Successful NbA draws from existing adaptation approaches (Borsje et al., 2011; Temmerman et al.,
31 2013; Law et al., 2018; Reguero et al., 2018; Buotte et al., 2019) and is applied across ecological and human
32 systems (Table Box 14.7.1; Figure Box14.7.1; high confidence).

33

34 Through a capacity to evolve to keep pace with climate change, these approaches can impart self-sustaining
35 and cost-efficient long-term protection in addition to serving as biodiverse, carbon sinks (Scyphers et al.,
36 2011; Cheong et al., 2013; Temmerman et al., 2013; Rodriguez et al., 2014; Herr and Landis, 2016; Sasmito
37 et al., 2016; Reguero et al., 2018). NbA is generally less expensive and strengthens over time, as compared
38 with built infrastructure which erodes with time (medium confidence) (Narayan et al., 2016; Smith et al.,

Do Not Cite, Quote or Distribute  14-90       Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 2017; Sutton-Grier et al., 2018). Analysis of the impacts of Hurricane Sandy determined that communities
 2 located behind wetlands experienced 20% less damage (Narayan et al., 2016). Coral reefs are providing
 3 $544M per year (Beck et al., 2018a) and mangroves $22USDB in property protection for coastal
 4 communities in the US and Mexico (Beck et al., 2018b). By 2030, flooding from changes in storms, SLR
 5 (based on RCP8.5) and increases in built infrastructure in the US Gulf Coast may result in net economic
 6 losses of up to US$176 billion, of which US$50 billion could be avoided through implementation of nature-
 7 based measures including wetland and oyster reef restoration and other green infrastructure (Box 14.4,
 8 14.5.2) (EPA, 2015b; Reguero et al., 2018).

 9

10 Innovative approaches in Canada (Borsje et al., 2011; Spalding et al., 2014; Soto-Navarro et al., 2020) and
11 the US (Law et al., 2018; Buotte et al., 2019; Soto-Navarro et al., 2020) have led to social and environmental
12 co-benefits and could address both future climate risk and long-standing social injustices (Hobbie and
13 Grimm, 2020; Schell et al., 2020; Cousins, 2021). Effective NbA requires a well-coordinated suite of
14 adaptation efforts (e.g., assessment, planning, funding, implementation, and evaluation) that is co-produced
15 among stakeholders and across sectors (high confidence) (Millar and Stephenson, 2015; Kabisch et al., 2016;
16 Dilling et al., 2019; Morecroft et al., 2019; Lavorel et al., 2020). Evaluating the efficacy of NbA may
17 become more tractable with more uniform guidelines for implementation (Scarano, 2017; Malhi et al., 2020;
18 Seddon et al., 2020), and coordination in scaling-up local-level NbA measures is likely to facilitate long-term
19 success (Gao and Bryan, 2017).

20
21

22

23 Figure Box14.7.1: Climate hazards protection services provided by nature-based solutions.

24 Table Box 14.7.1: Nature-based adaptation in North America.

Sector       NbS Actions          Benefits    References

Do Not Cite, Quote or Distribute  14-91                                                       Total pages: 157
FINAL DRAFT                            Chapter 14  IPCC WGII Sixth Assessment Report
 Coasts
              Conservation and              Wave attenuation; erosion      (Borsje et al., 2011;
 Aquaculture  restoration of barrier        and flood reduction from       Scyphers et al., 2011;
 Agriculture  habitats, salt marshes,       storm events exacerbated       Cheong et al., 2013; Pinsky
 Urban Areas  mangroves, coral and          by SLR; novel, created         et al., 2013a; Temmerman
              oyster reefs, sand dunes,     habitats, connectivity;        et al., 2013; Ferrario et al.,
              and river deltas; combined    recreation, quality of life    2014; M鰈ler et al., 2014;
              natural and built                                            Rodriguez et al., 2014;
              infrastructure, e.g., oyster                                 Spalding et al., 2014; Yates
              reef in front of breakwall                                   et al.; EPA, 2015b; Grenier
                                                                           et al., 2015; Brandon et al.,
                                                                           2016; Herr and Landis,
                                                                           2016; Narayan et al., 2016;
                                                                           Sasmito et al., 2016; Ward
                                                                           et al., 2016; Aerts et al.,
                                                                           2018; Beck et al., 2018a;
                                                                           Morris et al., 2018b;
                                                                           Moudrak et al.; Reguero et
                                                                           al., 2018; Sutton-Grier et
                                                                           al., 2018)

              Watershed approaches          Create a less                  (Deutsch et al., 2015b)
              such as protecting and        flashy/variable hydrology;     Boesch 2019,CENR 2010
              restoring forests and         reduce sediment, nutrient,
              wetlands in coastal           hazardous chemical input
              watersheds, adopting          to coastal waters and
              stream buffers in             reduce eutrophication and
              agricultural areas (see       other water quality
              agriculture below)            impairments, notably in in
                                            deep waters where fish
                                            seek refuge from rising sea
                                            surface temperatures

              Controlled culture of fish,   Enhance, restore and           (Froehlich et al., 2017; Reid
              bivalves, corals and other    reduce pressure on wild        et al., 2019; Theuerkauf et
              marine species                species and ecosystems;        al., 2019)
                                            Restore threatened species
                                            such as coral reef species.
                                            Store carbon.

              Re-vegetate stream buffer     Self-sustaining and cost-      (CENR, 2010; Boesch;
              zones; plant winter cover     efficient long-term            Seddon et al., 2020)
              crops; wetland protection     protection from soil
              and restoration;              erosion; maintain and
              agroforestry                  enhance crop yields;
                                            enhance carbon sinks;
                                            enhance biodiversity;
                                            reduce nutrient input to
                                            coasts

              Replace impervious            Reduce urban heat-island       (Hobbie and Grimm, 2020;
              surfaces with permeable       effects, air pollution; self-  Brown et al., 2021)
              pavement, green space,        sustaining and cost-
              parks, wetlands and green     efficient long-term
              infrastructure, e.g.,         protection from flooding,
              stormwater ponds,             erosion, SLR; enhance
              bioswales, rain gardens,      carbon sequestration
              green roofs; community        biodiversity, habitat and
              gardens, urban forests;       connectivity; improved
              restore natural habitats;     quality of life, human
                                            health benefits

Do Not Cite, Quote or Distribute            14-92                          Total pages: 157
    FINAL DRAFT                                          Chapter 14  IPCC WGII Sixth Assessment Report

    Terrestrial                Forest conservation based    Increase carbon storage     (Law et al., 2018; Buotte et
                               on productivity and          and biodiversity            al., 2020; Soto-Navarro et
                               vulnerability to drought                                 al., 2020; Mori et al., 2021)
                               and fire; longer harvest
                               rotations

                               Forest thinning; prescribed  Reduce wildfire risk and    (Box 14.2 and citations
                               burning; cultural burning    severity; increase forest   therein)
                                                            resilience to fire; reduce
                                                            forest drought stress;
                                                            increase carbon storage

                               Protecting and restoring     Regulate stream flow;       (Lawler et al., 2020;
                               natural forests
                                                            reduce soil erosion; protect Seddon et al., 2020)

                                                            and enhance biodiversity

                               Beaver (Castor               Regulate seasonal stream (McKelvey and Buotte,

                               canadensis) reintroduction flow                          2018; Vose et al., 2018)

    Freshwater                 Forests to Faucets and       Improve water quality;      (Gartner et al., 2017;
                               other watershed restoration  reduced drinking water      Claggett and Morgan, 2018;
                               projects for stream &        treatment costs; increase   Price and Heberling, 2018)
                               drinking water protection    and regulate streamflow

1

2 [END BOX 14.7 HERE]

3

4

5 [START FAQ 14.4 HERE]

6

7 FAQ 14.4: What are some effective strategies for adapting to climate change that have been

8   implemented across North America, and are there limits to our ability to adapt successfully to

9   future change?

10

11 Climate adaptation is happening across North America. These efforts are differential across sectors, scale

12 and scope. Without more integrative and equitable approaches across broad scales, known as

13 transformational adaptation, the continent may face limits to the future effectiveness of adaptation actions.

14

15 Across North America, progress in introducing climate adaptation is steady, but incremental. Adaptation is

16 typically limited to planning, while implementation is often hindered by "soft" limits, such as access to

17 financial resources, disparate access to information and decision-making tools, the existence of antiquated

18 policies and management frameworks, lack of incentives, and highly variable political perceptions of the

19 urgency of climate change.

20

21 Cities and other state and local entities are taking the lead in adaptation efforts, particularly in terms of

22 mainstreaming the use of many approaches to adaptation. These approaches include a suite of efforts ranging

23 from assessment of impacts and vulnerability (relative to individuals, communities, jurisdictions, economic

24 sectors, natural resources, etc.), planning processes, implementation of identified strategies, and evaluation

25 of the effectiveness of these strategies. Other institutions (e.g., non-governmental organizations, professional

26 societies, private engineering and architecture businesses) also are making significant progress in the

27 adaptation arena, particularly at local to regional levels.

28

29 The water management and utilities sectors have made significant progress toward implementation of

30 adaptation strategies using broad-based participatory planning approaches. Consideration of climate change

31 is now folded into some ongoing watershed-wide planning efforts. An example is provided by the One-

32 Water-One-Watershed (OWOW) approach followed by the Santa Ana Watershed Project Authority

33 (SAWPA) in southern California. SAWPA is a Joint Powers Authority comprising five regional water

34 districts that provide drinking water to more than 6 million people as well as industrial and irrigation water

35 across the 2,400-square-mile watershed. The OWOW perspective focuses on integrated planning for multi-

36 benefit projects and explicit consideration of the impacts of any planning option across the entire watershed.

    Do Not Cite, Quote or Distribute                      14-93                                Total pages: 157
FINAL DRAFT                       Chapter 14  IPCC WGII Sixth Assessment Report

 1 Planning is supported by stakeholder-driven advisory bodies organized along themes that consider a full
 2 suite of technical, political, environmental and social considerations. SAWPA provides member agencies
 3 with decision-support tools and assistance to implement water conservation policies and pricing regimes, and
 4 one member agency is an industry leader on potable water recycling.

 5

 6 The marine and coastal fisheries sector also has shown considerable progress in climate adaptation planning,
 7 particularly in terms of assessing impacts and vulnerability of fisheries. Along the Pacific Northwest coast of
 8 the US and Alaska, seasonal and sub-seasonal forecasts of ocean conditions exacerbated by warming (e.g.,
 9 O2, pH, temperature, sea ice extent) already have informed fisheries and aquaculture management. Similarly,
10 forecasts and warnings have reduced human exposure to the increased risk of toxins from harmful algal
11 blooms in the Gulf of Mexico, the Great Lakes, California, Florida, Texas and the Gulf of Maine.

12

13 Professional organizations and insurance play an important part in mainstreaming climate adaptation.
14 Government and private sector initiatives can help address adaptation effort through building design
15 guidelines and engineering standards, as well as insurance tools that reflects the damages from climate
16 impacts. Through the identification of climate risks and proactive adaptation planning, the private sector can
17 contribute to reducing risks throughout North America by securing operations, supply chains, and markets.

18

19 Indigenous Peoples and rural community efforts across the continent show great potential for enhancing and
20 accelerating adaptation efforts particularly when integrated with western-based natural resource management
21 practices, such as cultural burning, traditional forest "tending" that reduces build-up of fuels (in addition to
22 prescribed fire and mechanical thinning). In the agricultural sector, examples include planting and cultivation
23 of culturally significant plants, as a traditional practice of soil conservation, in addition to food crops or in
24 lieu of synthetic or mechanical soil treatments.

25

26 Future changes in climate (e.g., more intense heat waves, catastrophic wildfire and post-fire erosion, sea
27 level rise and forced relocations) could exceed the current capacity of human and natural systems to
28 successfully adapt (or "hard limits"). The inclusion and equitable contribution of Indigenous Peoples and
29 rural communities in decision-making and governance processes--including recognition of the
30 interdependencies between cities and surrounding areas--increases the likelihood of building adaptive
31 capacity at a pace that is commensurate with present and future climate change risks.

32

33 Large-scale, equitable transformational adaptation likely will be required to respond to the growing rate and
34 magnitude of changes before crossing tipping points where hard limits exist, beyond which adaptation may
35 no longer be possible. Increasingly, there are calls for accelerating and scaling up adaptation efforts, in
36 addition to aligning policies and regulatory legislation at multiple levels of government. Improved processes
37 for adaptation decision-making, governance, and coordination, across sectors and jurisdictions, could
38 enhance North America's capacity to adapt to rapid climatic change. These actions include a focused societal
39 shift, across governments, institutions, and trans-national boundaries, from primarily technological
40 approaches to nature-based solutions that help foster changes in perception of risk and, ultimately, human
41 behaviour.

42

43 [END FAQ 14.4 HERE]

44

Do Not Cite, Quote or Distribute  14-94       Total pages: 157