FINAL DRAFT                       Cross-Chapter Paper 1            IPCC WGII Sixth Assessment Report


2                     Cross-Chapter Paper 1: Biodiversity Hotspots


4 Cross-Chapter Paper Leads: Mark John Costello (New Zealand/Ireland/Norway), Mariana M. Vale

5 (Brazil)


7 Cross-Chapter Paper Authors: Wolfgang Kiessling (Germany), Shobha Maharaj (Germany/Trinidad and

8 Tobago), Jeff Price (United Kingdom), Gautam Hirak Talukdar (India)


10 Cross-Chapter Paper Contributing Authors: Heath Beckett (South Africa), Will Darwall (United

11 Kingdom), Anindita Debnath (India), Eleanor Devenish-Nelson (United Kingdom), Yimnang Golbuu

12 (Palau), Kerry-Anne Grey (South Africa) Tamlin Jefferson (New Zealand/United Kingdom), Rhosanna

13 Jenkins (United Kingdom), Adam Kocsis (Germany/Hungary), Stella Manes (Brazil), Guy Midgley (South

14 Africa), David Schoeman (Australia), Pablo A. Tedesco (France)


16 Cross-Chapter Paper Review Editor: Andreas Fischlin (Switzerland)


18 Date of Draft: 1 October 2021


20 Notes: TSU Compiled Version



23 Table of Contents


25 Executive Summary..........................................................................................................................................2

26 CCP1.1 Point of Departure ..........................................................................................................................3

27 CCP1.2 Assessment.......................................................................................................................................7

28  CCP1.2.1 Global Perspective................................................................................................................7

29  CCP1.2.2 Terrestrial ...........................................................................................................................14

30  CCP1.2.3 Freshwater ..........................................................................................................................22

31  CCP1.2.4 Marine.................................................................................................................................25

32 CCP1.3 Adaptation and Solutions.............................................................................................................30

33 Box CCP1.1: Climate Change and Terrestrial Biodiversity Hotspots on Small Islands .........................32

34 FAQ CCP1.1: Why are biodiversity hotspots important?..........................................................................34

35 FAQ CCP1.2: How can society ensure conservation of biodiversity in climate policies?........................34

36 References ..................................................................................................................................................36



    Do Not Cite, Quote or Distribute  CCP1-1                           Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 Executive Summary
 3 Geographic areas that are exceptionally rich in species, ecologically distinct and/or show high endemism
 4 (species occurring in that defined geographic area and nowhere else), are broadly recognised as biodiversity
 5 hotspots and prioritised for conservation. Here, we assess the impacts and vulnerability to climate change of
 6 terrestrial, freshwater and marine regions considered to be biodiversity hotspots. {CCP1.1}
 8 Species in biodiversity hotspots already show changes in response to climate change (high confidence1).
 9 The animal and plant species assessed have been shifting their geographic ranges from low to high latitudes
10 in response to climate warming on land and in the ocean (very high confidence). On land, climate change-
11 induced shifts towards higher elevations are also common in biodiversity hotspots (high confidence); while
12 in the ocean climate-induced shifts to greater water depths are little studied. In the ocean, abrupt mortality of
13 habitat-forming species on coral reefs and kelp forests, especially following heatwaves, are increasing in
14 frequency in biodiversity hotspots (high confidence). {CCP1.2.1, 1.2.2, 1.2.4}
16 All biodiversity hotspots are impacted, to differing degrees, by human activities (very high confidence),
17 Climate change impacts are compounded by other anthropogenic impacts, including habitat loss and
18 fragmentation, hunting, fishing and its bycatch, over-exploitation, water abstraction, nutrient enrichment,
19 pollution, human introduction of invasive species, pests and diseases, all of which reduce climate resilience
20 (very high confidence), complicating the attribution of observed changes to climate change. {CCP1.2.1}
22 Observed climate velocities are approximately 20% lower inside than outside of terrestrial and
23 freshwater biodiversity hotspots, but 69% higher inside than outside marine hotspots (high
24 confidence). In spite of the lower climate velocities inside terrestrial hotspots, these areas are projected not
25 to serve as effective climate refugia from the effects of global warming, especially for endemic species
26 (unique to a hotspot) (medium confidence). The greater climate velocities inside marine hotspots exposes
27 their species to greater climate-induced pressures inside than outside hotspots (high confidence). The
28 differences between temperatures inside and outside of hotspots narrow with increasing warming (medium
29 confidence). {CCP1.2.2}
31 The risk of species extinction increases with warming in all climate change projections for native
32 species studied in hotspots (high confidence), being about ten-times greater for endemic species from
33 1.5°C to 3°C above pre-industrial levels (medium confidence). Of the 6,116 projections for more than
34 2,700 species assessed in biodiversity hotspots, ~44% were found to be at high extinction risk, and ~24% at
35 very high extinction risk due to climate change (medium confidence). Very high extinction risk in
36 biodiversity hotspots due to climate change is more common for endemic species than other native species
37 (high confidence). For these endemic species, considering all scenarios and time periods evaluated, ~100%
38 on islands, ~84% on mountains, ~12% on continents (high confidence) and ~54% in the ocean (notably the
39 Mediterranean) (low confidence) are projected to be threatened with extinction due to climate change. With
40 further warming, increasingly high risks of local and global extinctions are projected in biodiversity hotspots
41 from climate-related stressors (high confidence). {CCP1.2.1, Figure CCP1.7, Figure CCP1.6}
43 Adaptation options could enhance the persistence of biodiversity in hotspots (high confidence). Noting
44 that over 3 billion people live within biodiversity hotspots, reduction of existing (non-climatic) pressures due
45 to human activities is critical for building resilience within hotspots. Adaptation options for biodiversity
46 (e.g., expanding fully protected areas, restoration and sustainable use practices) are as applicable inside as
47 outside biodiversity hotspots (high confidence). Nevertheless, the protection of biodiversity hotspots is key
48 to prevent a substantial global biodiversity decline from climate change {CCP 1.3, Table CCP1.2, 2.6, 3.6,
49 Cross-Chapter Box NATURAL in Chapter 2}.

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  CCP1-2                 Total pages: 47
    FINAL DRAFT                         Cross-Chapter Paper 1               IPCC WGII Sixth Assessment Report

1 CCP1.1         Point of Departure


3 Biodiversity hotspots are geographic areas with an exceptionally high richness of species, including rare and

4 endemic species. Such hotspots have deep evolutionary roots and are concentrated in areas where past

5 climatic variability was moderate (Enquist et al., 2019; Brown et al., 2020; Trew and Maclean, 2021). An

6 important limitation of the biodiversity hotspot concept is that there may be species highly threatened with

7 extinction which do not occur within what has traditionally been classified as a hotspot (Grenyer et al.,

8 2006). Thus, biodiversity hotspot assessments need to be paralleled by assessments of highly endangered

9 species, and the threats they face.


11 Many studies have proposed biodiversity hotspots based on different criteria, taxa and geographic contexts

12 (e.g., Myers et al., 2000; Mittermeier et al., 2004; Mittermeier et al., 2011; Williams et al., 2011; Noss et al.,

13 2015; Asaad et al., 2017). A coherent comparative assessment of all such schemes is beyond the scope of

14 this chapter but is provided in recent reviews (Asaad et al., 2017; Jefferson and Costello, 2019). We base this

15 assessment on the "WWF Global 200" areas of conservation importance (Olson and Dinerstein, 2002). These

16 238 ecoregions have been used in a previous climate risk assessment (Warren et al., 2018b) and cover

17 terrestrial, freshwater and marine environments (Table CCP1.1, Figures CCP1.1, CCP1.2). In addition we

18 included the terrestrial "biodiversity hotspots" as defined by Myers et al. (2000) and extended later ( Table

19 CCP1.1; Mittermeier et al., 2011; Williams et al., 2011; Noss et al., 2015). This assessment thus covers the

20 "Global 200" (hereafter G200) and "Myers" biodiversity hotspots rather than particular species or ecological

21 systems such as rainforests, coral reefs or the deep sea. Such systems are assessed in Chapters 2 and 3.

22 Chapters 2 and 3 also cover observed and projected impacts, changes in ecosystem functioning, and species

23 extinction risks at a global level.


25 Biodiversity hotspots have not been explicitly covered in the Fifth Assessment Report (IPCC, 2014)

26 (hereafter AR5). Thus, the point of departure is AR4 (Parry et al., 2007), which assessed that climate change

27 exacerbates biodiversity risks in hotspots and that 15-40% of endemic species (species only occurring in one

28 region) were projected to become extinct at 3.5°C global warming (Fischlin et al., 2007). Risks of extinction

29 are assessed using the guidelines in Chapter 2, with species projected to lose 80% of its range or abundance

30 being classified as at very high risk of extinction and those projected to lose 50% being at high risk of

31 extinction (Chapter 2, Fig. 2.8 for definitions and a global overview).



    Do Not Cite, Quote or Distribute    CCP1-3                              Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.1: Recent human impacts on the terrestrial biodiversity hotspots (coloured, grey is non hotspot) (Table
3 SMCCP1.1). Impacts are scaled in five equal 20% categories. A, North and Central America, B, South America, ,C,
4 Europe and North Africa, ,D, Africa and Arabia, E, North Asia, ,F, Southeast Asia, G, Southeast Asian archipelagos,
5 Australia and New Zealand. See Table CCP1.1 for key to hotspot numbers.

Do Not Cite, Quote or Distribute  CCP1-4                 Total pages: 47
   FINAL DRAFT                       Cross-Chapter Paper 1             IPCC WGII Sixth Assessment Report

2 Figure CCP1.2: Recent human impacts from multiple factors on (A) freshwater hotspots since 2000 based on Janse et

3 al. (2015) and (B) marine hotspots based on Halpern et al. (2015). Human impacts in terrestrial and freshwater habitats

4 refer to the remaining wilderness. Marine impacts represent land-based, fishing, climate change and ocean-based

5 stressors. Impacts are scaled into five equal 20% categories. See Table CCP1.1 for the key to hotspot numbers.



8 Table CCP1.1: List of biodiversity hotspots names from (Olson and Dinerstein, 2002) as mapped in Figures CCP1.1

9 (terrestrial, black text) and CCP1.2 (freshwater and marine in green and blue text respectively). Hotspots containing

10 islands (> 100km2) are indicated with an asterisk.

   1 Hawaii Moist Forest *           81 Drakensberg Montane Woodlands 161 Danube River Delta

                                     & Grasslands

   2 Hawaii Dry Forest *             82 Ural Mountains Taiga & Tundra  162 Balkan Rivers & Streams *

   3 Alaskan North Slope Coastal     83 Taimyr and Russian Coastal     163 Anatolian Freshwater

   Tundra *                          Tundra *

   4 Canadian Boreal Taiga *         84 Altai-Sayan Montane Forests    164 Upper Guinea Rivers & Streams *

   5 Canadian Low Arctic Tundra *    85 Central and Eastern Siberian Taiga 165 Niger River Delta *

   6 Muskwa/Slave Lake Boreal        86 Chukhote Coastal Tundra *      166 Cameroon Crater Lakes


   7 Pacific Temperate Rainforests*  87 Central Asian Deserts          167 Gulf of Guinea Rivers & Streams *

   Do Not Cite, Quote or Distribute                    CCP1-5                                       Total pages: 47
FINAL DRAFT                            Cross-Chapter Paper 1                   IPCC WGII Sixth Assessment Report

8 Northern Prairies                    88 Middle Asian Montane Woodlands       168 Congo River and Flooded Forests
                                       and Steppe
9 Klamath-Siskiyou Coniferous          89 Kamchatka Taiga & Grasslands *       169 Congo Basin Piedmont Rivers &
Forests                                                                        Streams
10 Sierra Nevada Coniferous Forests    90 Russian Far East Broadleaf &         170 Cape Rivers & Streams
                                       Mixed Forests *
11 California Chaparral & Woodlands *  91 Daurian/Mongolian Steppe             171 Rift Valley Lakes
12 Sonoran-Baja Deserts*               92 Tibetan Plateau Steppe               172 Madagascar Freshwater
                                                                               Ecosystem *
13 Chihuahuan-Tehuacan Deserts         93 Hengduan Shan Conifer Forests        173 Lena River Delta
14 Sierra Madre Oriental & Occidental  94 Southwest China Temperate            174 Lake Baikal
Pine-Oak                               Forests
15 Southern Mexican Dry Forests *      95 Western Himalayan Temperate          175 Russian Far East Rivers &
                                       Forests                                 Wetlands *
16 Mesoamerican Pine-Oak Forests       96 Rann of Kutch Flooded Grasslands     176 Lake Biwa *
17 Appalachian & Mixed Mesophytic      97 Terai-Duar Savannas & Grasslands     177 Indus River Delta *
18 Southeastern Conifer and Broadleaf  98 Eastern Himalayan Alpine             178 Western Ghats Rivers & Streams
Forests *                              Meadows
19 Everglades Flooded Grasslands       99 Eastern Himalayan Broadleaf and      179 Southwestern Sri Lanka Rivers *
                                       Conifer Forests
20 Greater Antillean Moist Forests *   100 Eastern Deccan Plateau Moist        180 Salween River
                                       Forests *
21 Greater Antillean Pine Forests *    101 Chhota-Nagpur Dry Forests           181 Lake Inle
22 Talamancan-Isthmian Pacific         102 Southwestern Ghats Moist Forest     182 Sundaland Rivers & Swamps *
23 Choco-Darien Moist Forests *        103 Sri Lankan Moist Forest *           183 Yangtze River & Lakes
24 South American Pacific Mangroves*   104 Sundarbans Mangroves                184 Xi Jiang Rivers & Streams
25 Galapagos Islands Scrub *           105 Naga-Manapuri-Chin Hills Moist      185 Yunnan Lakes & Streams
                                       Forests *
26 Northern Andean Páramo              106 Kayah-Karen/Tenasserim Moist        186 Mekong River *
27 Northern Andean Montane Forests     107 Indochina Dry Forests *             187 Philippines Freshwater *
28 Tumbesian-Andean Valleys Dry        108 Cardamom Mountains Moist            188 Central Sulawesi Lakes *
Forests *                              Forests *
29 Napo Moist Forests                  109 Peninsular Malaysia Lowland&        189 New Guinea Rivers & Streams *
                                       Montane Forests *
30 Southwestern Amazonian Moist        110 Sumatran Islands Lowland and        190 Lakes Kutubu & Sentani *
Forests                                Montane Forests *
31 Atacama-Sechura Deserts             111 Western Java Montane Forests *      191 Kimberley Rivers & Streams
32 Central Andean Dry Puna             112 Nusu Tenggara Dry Forests *         192 Southwest Australia Rivers &
33 Central Andean Yungas               113 Southeast China-Hainan Moist        193 New Caledonia Rivers & Streams *
34 Chilean Matorral                    114 North Indochina Subtropical Moist   194 Central Australian Freshwater *
35 Valdivian Temp. Rain Forests Juan   115 Annamite Range Moist Forests        195 Eastern Australia Rivers & Streams
Fernandez *
36 Patagonian Steppe*                  116 Kinabalu Montane Shrublands *       196 Bering Sea *
37 Amazon-Orinoco-Southern             117 Borneo Lowland and Montane          197 California Current *
Caribbean Mangroves *                  Forests *
38 Coastal Venezuela Montane Forests   118 Greater Sundas Mangroves *          198 Hawaiian Marine *
39 Llanos Savannas                     119 Nansei Shoto Archipelago Forests *  199 Gulf of California *
40 Guianan Highlands Moist Forests     120 Taiwan Montane Forests *            200 Mesoamerican Reef *
41 Rio Negro-Jurua Moist Forests       121 Philippines Moist Forests *         201 Panama Bight *
42 Guianan Moist Forests               122 Palawan Moist Forests *             202 Galapagos Marine *
43 Atlantic Dry Forests                123 Sulawesi Moist Forests *            203 Fiji Barrier Reef *
44 Cerrado Woodlands & Savannas        124 Moluccas Moist Forests *            204 Tahitian Marine *
45 Pantanal Flooded Savannas           125 New Guinea Mangroves *              205 Rapa Nui *
46 Chiquitano Dry Forests              126 Southern New Guinea Lowland         206 Humboldt Current *
                                       Forests *

Do Not Cite, Quote or Distribute       CCP1-6                                  Total pages: 47
   FINAL DRAFT                                 Cross-Chapter Paper 1        IPCC WGII Sixth Assessment Report

   47 Atlantic Forests*               127 Central Range Subalpine           207 Grand Banks *

                                      Grasslands *

   48 Fenno-Scandia Alpine Tundra &   128 New Guinea Montane Forests *      208 Chesapeake Bay

   Taiga *

   49 Caucasus-Anatolian-Hyrcanian    129 Solomons-Vanuatu-Bismarck Moist 209 Greater Antillean Marine *

   Temp. Forests                      Forests *

   50 European-Mediterranean Montane 130 Northern Australia & Trans-Fly     210 Southern Caribbean Sea *

   Forests                            Savannas *

   51 Mediterranean Forests, Woodlands, 131 Great Sandy-Tanami-Central      211 Northeast Brazil Shelf Marine *

   Scrub *                            Ranges Desert

   52 Sudd-Sahelian Flooded Grasslands 132 Carnavon Xeric Shrubs *          212 Patagonian Southwest Atlantic *

   & Savannas

   53 Guinean Moist Forests           133 Southwestern Australia Forests & 213 Antarctic Peninsula and Weddell

                                      Scrub *                               Sea *

   54 Gulf of Guinea Mangroves *      134 Southern Australia Mallee and     214 Barents-Kara Seas *

                                      Woodlands *

   55 Cameroon Highlands Forests *    135 Queensland Tropical Forests *     215 Northeast Atlantic Shelf Marine *

   56 Congolian Coastal Forests *     136 New Caledonia Moist Forests *     216 Mediterranean Sea *

   57 Sudanian Savannas               137 New Caledonia Dry Forests         217 Canary Current *

   58 Western Congo Basin Moist Forests 138 Lord Howe and Norfolk Island    218 Benguela Current

                                      Forests *

   59 Northeastern Congo Basin Moist 139 New Zealand Temperate Forests * 219 Arabian Sea *


   60 Central Congo Basin Moist Forests 140 Eastern Australia Temperate     220 Red Sea *

                                      Forests *

   61 Albertine Rift Montane Forests  141 Tasmanian Temperate Rainforests 221 West Madagascar Marine *


   62 Central and Eastern Miombo      142 Southern Pacific Islands Forests * 222 East African Marine *


   63 Zambezian Flooded Savannas      143 Gulf of Alaska Coastal Rivers*    223 Agulhas Current

   64 Namib-Karoo-Kaokoveld Deserts & 144 Pacific Northwest Coastal Rivers 224 Maldives-Chagos-Lakshadweep

   Shrublands *                                                             Atolls *

   65 Fynbos                          145 Colorado River                    225 Andaman Sea *

   66 Arabian Highlands Woodlands and 146 Chihuahuan Freshwater             226 Banda-Flores Sea *

   Shrublands *

   67 Socotra Island Desert *         147 Mexican Highland Lakes            227 Western Australia Marine *

   68 Ethiopian Highlands             148 Mississippi Piedmont Rivers &     228 Southern Australian Marine


   69 Horn of Africa Acacia Savannas  149 Lower Mississippi River           229 Okhotsk Sea *

   70 East African Coastal Forests *  150 Southeastern Rivers & Stream s* 230 Yellow Sea *

   71 East African Moorlands          151 Greater Antillean Freshwater *    231 Nansei Shoto *

   72 Eastern Arc Montane Forests     152 Orinoco River & Flooded Forests 232 Sulu-Sulawesi Seas

   73 East African Mangroves *        153 Upper Amazon Rivers & Streams 233 Palau Marine *

   74 East African Acacia Savannas    154 Brazilian Shield Amazonian Rivers 234 Bismarck-Solomon Seas

                                      & Streams

   75 Southern Rift Montane Woodlands 155 High Andean Lakes                 235 New Caledonia Barrier Reef *

   76 Madagascar Mangroves *          156 Guianan Freshwater                236 Great Barrier Reef *

   77 Madagascar Dry Forests *        157 Amazon River & Flooded Forests * 237 Lord Howe-Norfolk Islands Marine

   78 Seychelles and Mascarenes Moist 158 Upper Parana Rivers & Streams 238 New Zealand Marine *

   Forests *

   79 Madagascar Forests & Shrublands * 159 Volga River Delta

   80 Madagascar Spiny Thicket *      160 Mesopotamian Delta and Marshes *



3 CCP1.2          Assessment


5 Specific hotspot numbers (H) are indicated in this chapter text to aid their identification in Table CCP1.1 and

6 Figures CCP1.1 and CCP1.2.


8 CCP1.2.1 Global Perspective


   Do Not Cite, Quote or Distribute                  CCP1-7                                         Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 CCP1.2.1.1 Observed Impacts
 3 CCP1. Observed climatic hazards
 4 Terrestrial and freshwater hotspots have been warming less over the last 50 years than non-hotspot areas,
 5 whereas marine hotspots have been warming more (Kocsis et al., 2021). The warming inside terrestrial
 6 hotspots is 0.91°C (Myers) and 1.04°C (G200), respectively, while for freshwater hotspots it is 0.89°C,
 7 compared to a +1.08°C warming outside (Kocsis et al., 2021). In contrast, mean annual sea surface
 8 temperatures in the G200 marine biodiversity hotspots have warmed 41% more than the regions outside
 9 (0.53 vs 0.38°C) (Kocsis et al., 2021). Thus, terrestrial biodiversity hotspots have been warming slightly less,
10 and marine hotspots considerably more than non-hotspots (medium confidence).
12 Climate velocity, the direction and pace of movement in climate variables (typically temperature) in space, is
13 key to understanding the origin and fate of biodiversity hotspots under climate change (Loarie et al., 2009;
14 Burrows et al., 2011). Climate trajectories generally predict the direction and pace of past and future species
15 range shifts (Pinsky et al., 2013; Brito-Morales et al., 2018) although there are exceptions (Fuchs et al.,
16 2020). Spatial patterns of climate trajectories show regions where species are expected to leave, pass
17 through, and/or arrive under climate change (Burrows et al., 2014). Regions of high climate velocities are
18 those with low topographic relief on land, particularly flooded grasslands and deserts (Loarie et al., 2009),
19 and tropical as well as offshore and polar sea regions (Burrows et al., 2011; Burrows et al., 2014; García
20 Molinos et al., 2016; Brito-Morales et al., 2018; Brito-Morales et al., 2020).
22 On millennial time scales, some areas of low climate velocity have more endemic species and can be
23 considered climate refugia, at least on land (Sandel et al., 2011) and, for marine species, around Antarctica
24 (H213) (Costello et al., 2010). This suggests that if these areas are subject to increased velocities they may
25 lose species which may not be able to disperse fast enough to cope with the pace of climate change (medium
26 confidence) (Sandel et al., 2011; Brito-Morales et al., 2018).
28 Climate velocities are 47% (Myers), 29% (G200, terrestrial) and 10% (G200, freshwater) lower inside than
29 outside biodiversity hotspots, respectively (Kocsis et al., 2021), but are 69% higher inside marine hotspots
30 than outside (medium confidence). Climate velocities from 1970 to 2019 ranged from 3-4 km/decade
31 (terrestrial and freshwater) to ~11 km/decade in marine (Kocsis et al., 2021).
33 For terrestrial and freshwater hotspots, the highest climate velocities are in central South America, including
34 the Amazon (H153, 154) (Figure CCP1.3). Terrestrial hotspots also have high velocities in the Arctic (H196,
35 214) and east of the Caspian Sea, while freshwater hotspots have low velocities in eastern European
36 Mediterranean and eastern Australia.
38 Marine hotspots have a wider range of climate velocities than terrestrial and freshwater environments (Figure
39 CCP1.3), being faster in equatorial, Mediterranean (H216), Baltic (H215), North and Okhotsk (H229), and
40 Arctic hotspots (H196, 214), and slow in the Antarctic hotspot (H213). Marine species tend to follow climate
41 velocities more closely than terrestrial species (high confidence) (Sunday et al., 2012; Pinsky et al., 2019;
42 Lenoir et al., 2020). The reasons may be smaller thermal safety margins in the seas or greater human impacts
43 on land impeding species range shifts. Climate velocities are particularly fast in equatorial seas (Figure
44 CCP1.3; Burrows et al., 2011), which are therefore expected to be source areas for species shifting their
45 ranges towards the subtropics (Burrows et al., 2014). The subtropics are then source areas of species that
46 shift to temperate latitudes and so forth, such that observed impacts in marine biodiversity hotspots are
47 largely attributable to species range shifts (high confidence) (Pecl et al., 2017). Because marine climate
48 velocities are significantly greater within than outside hotspots, marine hotspots are especially prone to
49 species redistributions (medium confidence) (Figure CCP1.3; Kocsis et al., 2021).
51 While species from lower latitudes may shift their geographic ranges into higher latitudes to adapt to
52 changing climate, there are no species to replace low latitude species. Thus, as already observed in the
53 oceans around the equator, the loss of species in low latitudes will continue with future climate warming
54 (high confidence) (Yasuhara et al., 2020; Chaudhary et al., 2021). The issue also extends to altitudinal ranges
55 in terrestrial environments, with species moving to higher elevations where surface area generally declines
56 with increasing elevation, and mountaintop species may have nowhere to go (Flousek et al., 2015; Freeman
57 et al., 2018; Kidane et al., 2019).

Do Not Cite, Quote or Distribute  CCP1-8                 Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 2 Figure CCP1.3: Climate velocities in terrestrial (A), freshwater (B) and marine (C) hotspots between 1970-2019.
 3 Values are presented in km/decade and derived using VoCC (García Molinos et al., 2019) from gridded temperature
 4 data, sea-surface temperatures for marine (Rayner et al., 2003) and near surface air temperatures on land and freshwater
 5 (Harris et al., 2020). Positive and negative velocities indicate warming and cooling, respectively.
 8 CCP1. Observed impacts on biodiversity
 9 Although the conservation status has only been assessed globally for about 6% of all species (Costello, 2019)

10 and most confirmed extinctions and threatened species are terrestrial, a higher proportion of freshwater

11 species are threatened, which is reflected in the higher proportion of freshwater hotspots impacted by
12 humans (Collen et al., 2014; Costello, 2015; Harrison et al., 2018). The rate of species endemicity is

13 exceptionally high in freshwater biogeographic realms (i.e., large regions of distinct species composition and

Do Not Cite, Quote or Distribute  CCP1-9                 Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 endemicity), at 89-96% for fish in all but one realm, compared to 11-98% for terrestrial vertebrate groups
 2 (Leroy et al., 2019) and 17-84% for marine realms (Costello et al., 2017). Already, one third of wetlands
 3 have been lost and 9,000 freshwater species threatened with extinction without considering the effects of
 4 climate change (Darwall et al., 2018), and only 13% of world rivers were recently classified as least
 5 impacted (Su et al., 2021).
 7 Globally, observed climate-driven changes in biodiversity are typically of species distributions shifting to
 8 higher latitudes (virtually certain) (Lenoir et al., 2020, Ch.2, Ch. 3.4). Since the 1950s, marine species
 9 richness has shifted poleward in the northern hemisphere, increased in mid-latitudes, and declined at the
10 equator in concert with ocean warming (medium confidence) (Chaudhary et al., 2021). Climate-driven
11 altitudinal shifts are common on land (high confidence) (Lenoir and Svenning, 2015; Steinbauer et al., 2018),
12 and depth shifts in the ocean may occur but are little studied (low confidence) (Burrows et al., 2019; Jorda et
13 al., 2020). While climate-induced range expansions can be viewed as opportunities for increasing regional
14 biodiversity, range contractions adversely affect biodiversity through regional extirpations (high confidence)
15 (Cahill et al., 2013; Chaudhary et al., 2021).
17 Both of the two climate change associated global species extinctions to date support the predictions that
18 endemic species on mountains and islands are at the greatest risk of extinction (Manes et al., 2021). The
19 golden toad (Bufo periglenes) became extinct after some years of decline associated with changes in climate
20 warming and precipitation in the Talamancan-Isthmian Pacific Forests biodiversity hotspot (H22) (Section
21; Pounds et al., 1999; Cahill et al., 2013). The Bramble Cays Melomys, a rodent endemic to an island
22 between Australia and Papua New Guinea closely related to a mainland Australian species, became extinct
23 due to habitat loss arising from climate change related sea-level rise and cyclone activity (Chapter11; Fulton,
24 2017; Roycroft et al., 2021).
26 CCP1.2.1.2 Projected Impacts
28 CCP1. Projected climatic hazards
29 Comparison of climate warming projected for air and sea temperature shows biodiversity hotspots will
30 continue to experience the greatest net increases in temperature at higher northern hemisphere latitudes;
31 particularly in tundra regions (Figures CCP1.4, CCP1.5, Table CCP1.1). Generally, terrestrial and freshwater
32 hotspots are projected to continue to warm more than marine (Figure CCP1.3). Predicted actual temperatures
33 are projected to continue to be the highest in the tropics, indicating where there are more thermally stressful
34 conditions for more species (high confidence) (Stuart-Smith et al., 2015; Stuart-Smith et al., 2017; Foster et
35 al., 2018; Waldock et al., 2019). By the end of this century, all terrestrial biodiversity hotspots in central and
36 south America, Africa, India and southern and eastern Asia (including the Indo-west Pacific islands) are
37 projected to experience climates unprecedented in their species' evolutionary history (medium confidence)
38 (Williams et al., 2007).
40 Based on WGI Atlas data (Gutiérrez et al., 2021), global warming is projected to affect terrestrial hotspots
41 less than non-hotspot areas: 80% less for Myers and 95-96% less for G200 terrestrial and freshwater
42 hotspots at global warming of 1.5-3°C (medium confidence) (Kocsis et al., 2021). In contrast, warming is
43 projected to be 12-13% greater inside than outside marine hotspots (medium confidence) (Kocsis et al.,
44 2021). Precipitation is generally projected to increase more in terrestrial and freshwater biodiversity hotspots
45 compared to outside them (low confidence) (Kocsis et al., 2021). The exception is Myers hotspots, which are
46 projected to have, on average, ~28% less precipitation at 1.5°C warming, but ~33% more at 2°C, and ~65%
47 more at 3°C (low confidence). However, precipitation changes are often difficult to assess as many hotspots
48 cover large areas, with some areas projected to be wetter and some drier with wide differences between
49 different climate models.

Do Not Cite, Quote or Distribute  CCP1-10                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.4: Projected loss of climatically suitable area in terrestrial biodiversity hotspots for a global average of
3 1.5°C (upper row, A-B), 2°C (middle, C-D) and 3°C (lower E-F). Left-hand column displays the projected human
4 impact using the five equal 20% categories of present-day impact (Figure CCP1.1). The right-hand column indicates the
5 changes of impact categories compared to present day-impact. See Table SMCCP1.1 for more details.

Do Not Cite, Quote or Distribute  CCP1-11                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 2 Figure CCP1.5: Projected future warming in °C for freshwater (left column, near-surface air temperature, panels A, C
 3 and E) and marine (right column as sea surface temperature, panels B, D and F) hotspots for a global average warming
 4 of +1.5°C (A, B), +2°C (C, D) and +3°C (E, F) compared to pre-industrial conditions. Values in text boxes in the
 5 Figures indicate temperature increase from present-day (2005-2014) settings. Projected temperatures were calculated
 6 with averages of multi-model, yearly means across Shared Socioeconomic Pathways (SSP) 1.26 (only for +1.5°C),
 7 SSP2-45, SSP3-70 and SSP5-85.
10 CCP1. Projected impacts on biodiversity
11 Biodiversity hotspots are expected to be especially vulnerable to climate change because their endemic
12 species have smaller geographic ranges (high confidence) (Sandel et al., 2011; Brown et al., 2020; Manes et
13 al., 2021). Manes et al. (2021) reviewed over 8,000 projections of climate change impacts on biodiversity in
14 232 studies, including 6,116 projections on endemic, native and introduced species in terrestrial (200
15 studies), freshwater (14 studies) and marine (34 studies) environments in biodiversity hotspots. Only half of
16 the hotspots had studies on climate change impacts. All measures of biodiversity were found to be negatively
17 impacted by projected climate change, i.e., species abundance, diversity, area, physiology, and fisheries
18 catch potential (medium confidence). However, introduced species' responses were neutral to positive
19 (medium confidence). Land areas were projected to be more negatively affected by climate warming than
20 marine. Land plants, insects, birds, reptiles, and mammals were all projected to be negatively affected
21 (medium confidence), as well as fish, coral reef, benthic, planktonic and other marine species (medium
22 confidence).
24 Of the 6,116 projections for more than 2,700 species assessed in biodiversity hotspots, ~44% were found to
25 be at high extinction risk, and ~24% at very high extinction risk due to climate change (Manes et al., 2021)
26 (medium confidence). Risks of extinction were estimated based on the projections for all warming levels
27 combined, showing that endemic species were about 2.7 times more at very high risk of extinction compared
28 to non-endemic native species (Manes et al., 2021). Extinction risks were highest for endemic species of
29 both land and ocean (medium confidence), and were higher for those living on islands (~100%, medium
30 confidence) and mountains (~84%, medium confidence) than in the ocean (~54%, low evidence, medium

Do Not Cite, Quote or Distribute  CCP1-12                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 agreement; low confidence) and on continents (~12%, robust evidence, medium agreement; medium
 2 confidence) (Figure CCP1.6). Extinction risks for non-endemic natives were ~20% for both terrestrial and
 3 marine species, with introduced species projected to become more rather than less invasive. At 1.5°C
 4 warming, ~2% of both terrestrial and marine species and at 3°C, ~20% and ~32% respectively, were
 5 projected to be at very high risk of extinction in the hotspots (Figure CCP1.6). Thus, a doubling of warming
 6 results in a roughly tenfold increase in species at very high extinction risk.
 8 Manes et al. (2021) found that any benefits to species (e.g., range or abundance increase) were projected to
 9 be localised and transient (e.g., Arctic, H196, 214). This and previous assessments indicate that while
10 climate change varies spatially and taxa may respond differently, a loss of biodiversity is projected across all
11 terrestrial hotspots (high confidence) (Foden et al., 2013; Warren et al., 2018a; Manes et al., 2021). Abrupt
12 changes across species assemblages may occur under all scenarios: in 9% of assemblages at 1.75°C and 35%
13 at 4.4°C on both land and sea (Trisos et al., 2020). However, species losses may be reduced if species have
14 thermal microclimate refugia and behavioural thermoregulation, or greater due to extreme events, such as
15 heatwaves.

19 Figure CCP1.6: A summary of the projected risks of species extinction at global warming levels of < 1.5 oC, 1.5 ­ 2.0
20 oC, and > 3°C in terrestrial and marine biodiversity hotspots. Data from Manes et al. (2021).


Do Not Cite, Quote or Distribute  CCP1-13                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 CCP1.2.1.3 Compounding and Cascading Effects
 3 All biodiversity hotspots are already impacted, to differing degrees, by human activities (high confidence)
 4 (Table CCP1.1, Figures CCP1.1, CCP1.2, Myers et al., 2000; Le Roux et al., 2019). At present, over three
 5 billion people live within the terrestrial and (catchments of) freshwater biodiversity hotspots, many of which
 6 border marine hotspots (WGII Atlas; Figures CCP1.1, CCP1.2; Table SMCCP1.1). Thus, climate change
 7 impacts on biodiversity hotspots are compounded by other anthropogenic impacts, increasing the
 8 vulnerability and reducing the resilience of biodiversity to climate change (very high confidence). Projections
 9 of changing climate alone may overestimate or underestimate the impacts on biodiversity (medium evidence,
10 high agreement). The additional risk of the combined effects of climate change and other impacts (e.g., land-
11 use change, overhunting, pollution and invasive species) on species has been raised since the Third
12 Assessment Report. The terrestrial hotspots projected to be most affected by global warming are, in general,
13 those already being impacted by loss of habitat due to land use change (Figure CCP1.4; Table SMCCP1.1)
14 (Warren et al., 2018a). This remains a trend in the recent literature, although most studies still address only
15 one stressor (Titeux et al., 2016). For example, Mantyka-Pringle et al. (2015) show that when the interaction
16 between projected climate change and habitat loss is taken into account, the extinction risk of birds and
17 mammals in 15­32% of terrestrial biodiversity hotspots changes. Similarly, Bellard et al. (2014b) found
18 different results when examining the impact of climate change, invasive species and land use change
19 independently, as opposed to synergistically. When combining those three impacts they identified the
20 Atlantic Forest (H47), Cape Floristic Region (H65) and Polynesia-Micronesia (H1, 2, 138, 139, 142) as
21 particularly vulnerable.
23 In a global assessment of the threat of climate change to river fish biodiversity, Tedesco et al. (2013)
24 projected that current extinction rates of species may be 7% greater due to climate change (low confidence).
25 The main threat is due to the effects of drought and reduced river flows, which would be 18 times greater
26 than without climate change. However, just 20 of the 110 river basins studied would experience sufficient
27 climate-driven water loss to cause fish extinctions by 2090. Moreover, the present rates of species loss due to
28 present human activities are 130 times greater than those projected by likely by future climate change
29 (medium confidence) (Tedesco et al., 2013).
31 Marine systems are also vulnerable to cumulative human impacts, which can be direct (e.g., pollution,
32 overfishing) and indirect (altered food webs) (very high confidence) (Halpern et al., 2008; Halpern et al.,
33 2015). The marine hotspots most currently threatened by non-climate related human impacts are all situated
34 in the northern hemisphere, specifically along the northern European, Mediterranean, and Asian coasts,
35 where the overlap of overfishing and pollution is especially large (Halpern et al., 2008; Halpern et al., 2015;
36 Ramírez et al., 2018), part B in Figure CCP1.2). Although there is a strong overlap of non-climatic and
37 climatic impacts in marine ecosystems (Blowes et al., 2019; Bowler et al., 2020), the effects suggest that
38 climate change impacts are most severe in tropical and northern high latitude seas (high confidence) (Doney
39 et al., 2012; Gattuso et al., 2015; Cheung et al., 2018; IPCC, 2019b). Temperature-driven range-shifts and
40 range-expansions are projected to also lead to cascading effects on marine biodiversity through ecological
41 interactions (high confidence) (Pecl et al., 2017; Vergés et al., 2019). Cascading effects may be especially
42 pronounced in temperate reefs, where tropicalisation may lead to arrival of herbivorous fish and predators
43 previously absent (Vergés et al., 2019). However, how these indirect effects of climate change on species
44 may change food webs and ecosystem function, including carbon sequestration, is unknown. Direct and
45 indirect human impacts due to fisheries and pollution can also lead to cascading effects that may be additive
46 with climate impacts on biodiversity. Destruction of marine biogenic habitats due to trawling and dredging,
47 and loss of large proportions of marine megafauna, particularly fish, mammals, birds and reptiles, alters food
48 webs and reduces resilience to additional disturbances, such as due to climate change (medium evidence,
49 high agreement) (Brander, 2007; Wernberg et al., 2011; Ramírez et al., 2017; Cheung et al., 2018; Bates et
50 al., 2019; Costello, in press).
52 The following sections report observed and projected climate change impacts on terrestrial, freshwater and
53 marine environments.
55 CCP1.2.2 Terrestrial

Do Not Cite, Quote or Distribute  CCP1-14                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 The 177 terrestrial hotspots assessed here (including 142 G200) cover ca. 61,000,000 km2 (41% of global
 2 land area), with a 37% overlap with freshwater hotspots (Table CCP1.1, Figure CCP1.2). They include wet
 3 and dry forests, woodland and scrub, highlands, mangroves, deserts, steppe, savanna, grasslands, moorlands
 4 and tundra. Over 77% of publications on climate change impacts on hotspots since AR5 have been on
 5 terrestrial ecosystems, most of it on projected (as opposed to observed) impacts (Manes et al., 2021).
 7 CCP1.2.2.1 Observed Impacts
 9 There is high confidence that climate change has already had impacts in North American hotspots.
10 Phenological and range shifts have been reported for bird and mammal species within the boreal forest
11 hotspot (Davidson et al., 2020), and earlier egg laying in birds in tundra hotspots (H3, 5) owing to changes in
12 snowmelt (Grabowski et al., 2013). Woody vegetation is already shifting north into the tundra (Larsen et al.,
13 2014).
15 In Central and South America, observed impacts within Mesoamerica (H15, 16) and the Tropical Andes
16 hotspots (H26, 27, 28, 32, 33) comprise upward altitudinal range shifts of birds, frogs, beetles and butterflies
17 (Narins and Meenderink, 2014; Molina-Martínez et al., 2016; Moret et al., 2016; Freeman et al., 2018)
18 (medium confidence). A shift of the Guianan-Amazon mangroves (H37) to higher grounds inland was
19 attributed to the effects of observed sea-level rise (low confidence) (Cohen et al., 2018).
21 In Europe, the Mediterranean hotspot (H216) has seen increases in wildfires and droughts attributed to
22 anthropogenic climate change (Gudmundsson et al., 2017; Barbero et al., 2020), and range shifts in birds
23 have been observed at higher elevations (medium confidence) (Tellería, 2020).
25 In Africa, multiple lines of evidence suggest woody plants are increasing in area, density and cover in
26 previously lightly wooded savanna and grassland hotspots (H65, H82) (Poulsen and Hoffman, 2015; Stevens
27 et al., 2017). Significant vulture and cheetah range reductions in these hotspots are at least partially
28 attributable to bush encroachment (Nghikembua et al., 2016; Wolter et al., 2016; Santangeli et al., 2018).
29 Thus, climate-driven bush encroachment has adversely affected unique mammal and bird diversity (robust
30 evidence, medium agreement, medium confidence). Warming and drying trends have historically been shown
31 to reduce the range of the Ethiopian Wolf, and they interact with land-use pressures in the Ethiopian hotspot
32 (H68) (Sintayehu, 2018) and plant species richness in the Cape Fynbos (H65) of southern Africa by reducing
33 post-wildfire recruitment (low confidence) (Slingsby et al., 2017).
35 Observed impacts in Asia were mostly restricted to the Himalaya (H95, 98, 99), Sundaland (H109, 110, 111,
36 112, 117, 118) and Indo-Burma (H105, 106, 107, 114, 115) hotspots, showing negative impacts through
37 increased invasion by exotic plants, decreased suitable area for endemic species and significant changes in
38 phenology (medium confidence) (Telwala et al., 2013; Braby et al., 2014; Padalia et al., 2015; Lamsal et al.,
39 2017). In the Central Asian mountain landscape (H87), studies have shown increased aridity induced by
40 climate change impacts on several shrub species (Seim et al., 2016). Some positive effects were observed for
41 native species in terms of an increase of suitable habitat (limited evidence, low agreement) (Priti et al., 2016;
42 Tang et al., 2017; Rathore et al., 2019).
44 In Australia, climate change has been implicated in: drought-induced canopy dieback across a range of forest
45 and woodland types due to decades of declining rainfall in the southwestern hotspot (H133); fires in the
46 palaeo-endemic pencil pine forests (Tasmania H142); declines in vertebrates in the Australian Wet Tropics
47 World Heritage Area, which overlaps with the eastern part of the Northern Australia hotspot (H131), related
48 to warming and increased length of the dry season; and declines in grass and increases in shrubs in the
49 Bogong High Plains (high confidence) (Hoffmann et al., 2019). The Australian Alps have seen increased
50 species diversity following retreat of the snow line (Slatyer, 2010), replacement of long-lived trees by short-
51 lived shrubs following multiple wildfires (Zylstra, 2018), and changing ecological interactions due to
52 climate-related snow loss, drought and fires (high confidence) (Hoffmann et al., 2019). While warming is
53 allowing mangroves to expand their range in coastal hotspots of Asia and Australia (Ward et al., 2016;
54 Hughes et al., 2019a), drought and associated salinity stress has killed mangroves in northern Australia
55 hotspots (Chapter 11, Babcock et al., 2019).

Do Not Cite, Quote or Distribute  CCP1-15                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 Approximately 76% of biodiversity hotspots within this assessment either contain, or, are comprised of
 2 islands >100 km2 (Table CCP1.1). However, just 0.08% of these hotspots were represented in post-AR5
 3 literature examining climate change impacts on terrestrial biodiversity. Most observed impacts were assessed
 4 with low evidence, but high agreement, and focus on plants and insects. Impacts described included
 5 abundance changes and extirpations (Jenouvrier et al., 2014), altitudinal range shifts (Koide et al., 2017),
 6 increased invasive alien species' abundance and extent in Madagascar (H76, 77), Balearic (H51) and Pacific
 7 Islands (Ghulam, 2014; Silva-Rocha et al., 2015; Goulding et al., 2016; Dawson et al., 2017), increased
 8 temperature affecting physiology, body size and behaviour of frogs in the Caribbean (H20) (Narins and
 9 Meenderink, 2014) and phenological alterations (Fontúrbel et al., 2018). One positive observation was the
10 high resilience to recovery of intact forest ecosystems to tropical cyclones within Caribbean (H20) and
11 Pacific islands (medium confidence) (Keppel et al., 2014; Marler, 2014; Shiels et al., 2014).
13 CCP1.2.2.2 Projected Impacts
15 Most terrestrial species in biodiversity hotspots in North America have been projected to be negatively
16 impacted by climate change (medium evidence, medium agreement, medium confidence). About ~80% of
17 projections for assessed species showed a negative impact of climate change climate change, with ~25% at
18 very high risk of extinction (Figure CCP1.7; Manes et al., 2021). Alterations to vegetation that would have
19 ecosystem-wide impacts, such as a shift from oak-dominated forests to predominantly hickory and maple
20 species in the Appalachian Forests (H17) (Ma et al., 2016) or the continued shrinking of tundra ecosystems,
21 have also been projected. Range shifts have been projected for a variety of plants (Beltrán et al., 2014;
22 Riordan and Rundel, 2014) and vertebrate taxa (Warren et al., 2014; Stralberg et al., 2015; McKelvy and
23 Burbrink, 2017). Sizeable range loss, which particularly affects endemic species, is projected with higher
24 levels of climate change. Adaptation in the agricultural sector poses an additional risk to remaining wildlife
25 habitat (e.g., wine in California: Roehrdanz and Hannah, 2016; Figure CCP1.9, CCP1.8).
27 In Central and South America, risks have been assessed in at least twenty-four terrestrial hotspots, especially
28 within the Atlantic Forest, Cerrado, Mesoamerica and the Caribbean, the most studied hotspots in the world
29 in terms of climate change impacts (H47, 44, 15, 16, 20, respectively) (Manes et al., 2021). About 85% of
30 projections for assessed species showed a negative impact of climate change (high confidence), with ~26%
31 of projections predicting species extinctions (Figure CCP1.7; Manes et al., 2021). Projected impacts include
32 contraction or loss of species' geographic range, loss of diversity and high species turnover (high
33 confidence). Most studies had focused on vertebrates and plants in the Atlantic Forest (H47) and Cerrado
34 (H44) (Loyola et al., 2014; de Oliveira et al., 2015; Vale et al., 2018; Vasconcelos et al., 2018; Hidasi-Neto
35 et al., 2019; Lima et al., 2019; Lourenço-de-Moraes et al., 2019; Vasconcelos and Prado, 2019; Velazco et
36 al., 2019). Several insect species are projected to lose suitable climatic conditions, including Cerrado (H44)
37 moths (Khormi and Kumar, 2014). There were projected negative impacts on vegetation such as rupestrian
38 grasslands in Cerrado (H44) (Fernandes et al., 2018) and tropical and temperate forests in hotspots in
39 Mesoamerica (H15, H16) (Mendoza-Ponce et al., 2018; Mendoza-Ponce et al., 2019). Endemic species face
40 consistent risks of decrease in suitable habitat in the Atlantic Forest (H47) (Vale et al., 2018), Cerrado (H44)
41 (Vasconcelos, 2014), Tumbes-Chocó-Magdalena (H28, H23) (Hermes et al., 2018), and Mesoamerica (H15,
42 H16) (Garcia et al., 2014; Ramírez-Amezcua et al., 2016). Climate change may also benefit invasive plant
43 species in terms of range expansion (Wang et al., 2017) and physiology (de Faria et al., 2018) in the region
44 (Figure CCP1.8).
46 In European biodiversity hotspots, about 75% of projections for assessed species showed a negative impact
47 of climate change, with ~30% at very high risk of extinction (medium confidence) (Figure CCP1.7; Manes
48 et al., 2021). These threats are projected to be worse under higher levels of warming. Increased wildfire size
49 and frequency is projected to have a strong effect on, the Mediterranean basin (H216) ecosystems (medium
50 confidence) (Lozano et al., 2017). Range reductions have been projected for endemic plants (Pérez-García et
51 al., 2013; Casazza et al., 2014), reptiles (Ahmadi et al., 2019), birds (Abolafya et al., 2013) and insects
52 (Sánchez-Guillén et al., 2013) (medium confidence).
54 In African biodiversity hotspots, about 80% of projections for assessed species showed a negative impact of
55 climate change, with ~10% at very high risk of extinction, especially of endemic species including birds,
56 plants, bees across several taxa and hotspots if warming exceeds 2°C (high confidence) (Figure CCP1.7;

Do Not Cite, Quote or Distribute  CCP1-16                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 Figure CCP1.10; Huntley and Barnard, 2012; Kuhlmann et al., 2012; Baker et al., 2015; Lee and Barnard,
 2 2016; Young et al., 2016; Hannah et al., 2020; Manes et al., 2021).
 4 In Asia, there is a bias in studies towards Indo-Burma (H105, 106, 107, 114, 115), followed by Himalaya
 5 (H95, 98, 99) and South-East Asian montane tropical and temperate forests. About ~70% of projections for
 6 assessed species showed a negative impact of climate change, with ~30% at very high risk of extinction
 7 (medium confidence) (Figure CCP1.7; Manes et al., 2021) (). Impacts include species' range changes, habitat
 8 loss for endemic plants, expansion of invasive species, decreased connectivity and overall species richness
 9 decline (high confidence) (DasGupta and Shaw, 2013; Telwala et al., 2013; Sridhar et al., 2014; Zomer et al.,
10 2014; Ali and Begum, 2015; Aryal et al., 2016). A projected decrease in habitat suitability for large species
11 like the Asiatic Black Bear is of concern as alternative habitats are outside protected areas, and may lead to
12 human-wildlife conflicts (Farashi and Erfani, 2018). The few positive impacts of climate change were
13 projected as increases in suitable habitat and distribution range for a few endangered plants and mammals
14 (medium confidence) (Banag et al., 2015; Shrestha et al., 2018). Animals benefiting from increased fruit and
15 seed production in Southeast Asian forests during warm El Nino cycles were also projected to increase with
16 climate warming ( Figure CCP1.8; Corlett, 2011).
18 All projections for assessed species in Australia and New Zealand terrestrial biodiversity hotspots showed a
19 negative impact of climate change, with half at very high risk of extinction (low confidence) (Figure CCP1.7;
20 Manes et al., 2021) (). Observed impacts in the Australian Alps were projected to continue under future
21 climate change (Zylstra, 2018). The Northern Australia savanna (H131) may experience increased rainfall
22 and carbon dioxide due to climate change (Scheiter et al., 2015), and exotic grasses are projected to reduce
23 their range under climate warming (Gallagher et al., 2009). In Australian tropical wet forests, ground-living
24 vertebrates may be more sensitive than arboreal species to unstable climates (Scheffers et al., 2017). Bellard
25 et al. (2016) projected losses of land due to sea-level rise in the East Australian Forest hotspot (H140), and
26 González-Orozco et al. (2016) projected the contraction of eucalyptus species towards the coast of the
27 Southwest Australia hotspot (H134), exposing them to sea-level rise. In New Zealand forests (H139), native
28 plants may be replaced by more fire-resistant introduced species following climate-change related fires (low
29 confidence) (Perry et al., 2014). While forest growth is projected to potentially increase due to carbon
30 dioxide fertilization, this may be compromised by drought (low confidence) (Ausseil et al., 2013). Seed
31 production in native New Zealand beech forests is projected to increase due to climate warming, fuelling the
32 abundance of invasive rats and stoats, which then predate native species and lead to loss of endemic fauna
33 and flora (medium confidence) (Figure CCP1.8; Tompkins et al., 2013, Ch. 11).
35 About 80% of projections for assessed terrestrial species within insular biodiversity hotspots showed a
36 negative impact of climate change, with ~50% at very high risk of extinction, including 100% of endemic
37 species (medium confidence) (Figure CCP1.7; Manes et al., 2021). In addition to habitat loss and species
38 range reductions, changes in precipitation may be a major driver impacting tropical and subtropical island
39 species (medium confidence) (Maharaj and New, 2013; Harter et al., 2015; Struebig et al., 2015; Vogiatzakis
40 et al., 2016; Maharaj et al., 2018). Compared to continents, island species are projected to undergo greater
41 impacts from spatial change, especially birds and amphibians (high confidence) (Box CCP1.1; Fortini et al.,
42 2015; Holmes et al., 2015; Manes et al., 2021,). Of all biodiversity hotspots, island species face the highest
43 proportion of extirpation risk at high elevations due to decreasing habitat area (e.g., Brown et al., 2015) and
44 at low elevations from sea-level rise, habitat loss and introduced species (medium confidence) ( Figure
45 CCP1.11; Bellard et al., 2014a).

Do Not Cite, Quote or Distribute  CCP1-17                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.7: The projected impacts of climate change on species in 232 studies (a) terrestrial and (b) marine
3 hotspots (adapted from Manes et al., 2021), showing the number and percentage of species showing positive and
4 negative responses to climate change, and threatened with extinction. Note Oceania includes Australia, New Zealand,
5 Wallacea, New Guinea, New Caledonia, Polynesia and Micronesia and overlaps with the global Small Islands category,
6 which excludes Australia. The "small Islands" category represents oceanic and continent associated small islands, and
7 thus overlaps with Oceania and continental data.

Do Not Cite, Quote or Distribute  CCP1-18                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.8: Terrestrial biodiversity hotspots in the Americas, Asia and New Zealand. Photos by Denis Costello
3 (top four), Mariana M. Vale (Brazil), and Mark Costello (other three).

Do Not Cite, Quote or Distribute  CCP1-19                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.9: Polar and boreal biodiversity hotspots in the Arctic (Norway) taiga. Photos by Galice Hormeau (top
3 three) and Mark Costello (bottom two).

Do Not Cite, Quote or Distribute  CCP1-20                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.10: African biodiversity hotspots. Photos by Denis Costello (top two), Mark Costello (with elephant),
3 Frank Zachos (lower four).

Do Not Cite, Quote or Distribute  CCP1-21                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.11: Island biodiversity hotspots. Photos by Galice Hormeau (top two) and Mark Costello (other four).
5 CCP1.2.3 Freshwater


Do Not Cite, Quote or Distribute  CCP1-22                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 The 53 hotspots in freshwater ecosystems assessed here cover ca. 32,830,000 km2 (17% of global freshwater
 2 habitats and 22% of the global land area), with a 68% overlap with terrestrial ecosystems (Table CCP1.1,
 3 Figure CCP.2). They include lakes, rivers, and streams (Figure CCP1.12).
 5 CCP1.2.3.1 Observed Impacts
 7 An analysis of trends in 190 river basins in Australia found that stream-flows have been declining, including
 8 in the Central Australian (H194) and Kimberley (H191) hotspots, due to greater terrestrial plant uptake of
 9 water in response to climate-related increases in carbon dioxide (low confidence) (Ukkola et al., 2016). We
10 did not find any other publications providing evidence of impacts of climate change on freshwater
11 biodiversity within the hotspots. Whether this is because freshwater temperatures tend to be cooler due to
12 inputs from groundwater and/or mountain streams (Knouft and Ficklin, 2017), resilience of freshwater
13 species, or lack of research is unclear.
15 CCP1.2.3.2 Projected Impacts
17 Cold-water species are projected to lose habitat in Canada and this may apply in the Alaskan river (H143)
18 and Russian Far East Lake Inle (H181) hotspots (medium confidence) (Comte et al., 2013). Water abstraction
19 is significant in the Colorado river hotspot (H145) and reduces its resilience to climate change effects on
20 flow rates (Grafton et al., 2013).
22 In South America, in the Brazilian Amazon hotspot (H153, 154, 157) half the assessed fish species were
23 considered sensitive to increased temperatures and reduced oxygen due to climate change (low confidence)
24 (Frederico et al., 2016), and the use of protected areas was recommended to reduce the impacts of
25 deforestation and water pollution (Jézéquel et al., 2020). El Nino related floods have led to declines in
26 numbers of caiman, a top predator in the Brazilian Paraná river hotspot (H158), which indicates that
27 increased floods due to climate change may reduce its population and alter food webs (low confidence)
28 (Herrera et al., 2015).
30 In Europe, including the Mediterranean freshwater hotspots, climate change is projected to result in reduced
31 river flow, low oxygen in summer, salinity incursions, further eutrophication and spread of invasive species,
32 compromising the survival of native biodiversity (medium confidence) (Moss et al., 2009). The longer
33 growth season in the boreal and Arctic latitudes is projected to aid the invasion of exotic species, and
34 increase lake stratification resulting in lower oxygen below the hypolimnion (medium confidence). In
35 addition, strict cold-water species are projected to lose suitable habitat (Moss et al., 2009). An analysis of
36 1,648 species of freshwater fish, amphibians, turtles, plants, molluscs, crayfish and dragonflies, projected
37 ~6% of common and ~77% of rare species would lose 90% of their geographic range (low confidence)
38 (Markovic et al., 2014). Even if some species may be able to spread to other areas and follow the climate,
39 Markovic et al. (2014) projected a loss of species, especially molluscs, from the south-eastern
40 Mediterranean, including the Balkan biodiversity hotspot (H162) (medium confidence). Similarly, within
41 Europe, Mediterranean fish (Jari et al., 2019) and insects (Conti et al., 2014) are the most threatened by
42 climate warming, droughts and floods. The fish species of the Danube river delta hotspot (H161) are less
43 susceptible to climate change than in the Balkans (H162) and Anatolian (H163) hotspots. The rest of Europe,
44 from the Iberian Peninsula to Scandinavia is not classified as a biodiversity hotspot. Thus, the areas where
45 freshwater biodiversity is most threatened by climate change in Europe are in two of the three hotspots (high
46 confidence).
48 The African Rift Valley Lakes (H171), including Lakes Tanganyika and Turkana, are suffering from climate
49 change influenced drought, potentially impacting freshwater biodiversity (medium confidence) (Dudgeon et
50 al., 2006). Africa and Madagascar (H172) are projected to see a climate-driven 10% reduction in freshwater
51 flow that is projected to threaten the survival of ~9% of freshwater dependent fish and birds (low confidence)
52 (Thieme et al., 2010). Climate change is projected to increase the extinction vulnerability of most freshwater
53 fish in the western South Africa Cape hotspot (H170) (low confidence) (Shelton et al., 2018).
55 In Asia, although climate change impacts on the Yangtze (H183) and Mekong river (H186) biodiversity
56 hotspots have not been reported, they are subject to the range of human impacts of over-exploitation,
57 pollution, water abstraction, altered flow regimes, habitat loss, and spread of invasive species which makes

Do Not Cite, Quote or Distribute  CCP1-23                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 them more vulnerable to climate effects (medium confidence) (Dudgeon et al., 2006). The release of water
 2 from shrinking glaciers in Asia to some extent protects downstream freshwaters against drought, but half of
 3 these glaciers are projected to disappear by 2100 (medium confidence) (Pritchard, 2019).
 5 In Australia, the Murray-Darling river basin occupies much of the Eastern Rivers hotspot (H195) and climate
 6 related drought exacerbated by water abstraction is projected to drive declines in freshwater birds, fish and
 7 invertebrates (high confidence) (Grafton et al., 2013). However, a national scale analysis projected that
 8 climate change would cause freshwater species range shifts, but no losses of species in this hotspot (low
 9 confidence) (James et al., 2017, Chapter 11).

Do Not Cite, Quote or Distribute  CCP1-24                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

2 Figure CCP1.12: Photographs of freshwater biodiversity hotspots. Photos by Will Darwall, Pablo Tedesco, and Mark
3 Costello (bottom row).
6 CCP1.2.4 Marine


Do Not Cite, Quote or Distribute  CCP1-25                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 The 43 hotspots in marine ecosystems cover 46,600,000 km2, representing 9% of the ocean area (Table
 2 CCP1.1, Figure CCP1.2). They include coral reef ecosystems, kelp forests, seagrass meadows, polar, and
 3 upwelling zones (Figures CCP1.13, CCP1.14).
 5 CCP1.2.4.1 Observed Impacts
 7 Observed impacts attributable to climate change are strongly biased geographically, with most data from the
 8 temperate Northern Hemisphere, followed by subtropical to temperate Australia and few long-term data in
 9 the tropics (Poloczanska et al., 2013; Poloczanska et al., 2016). Marine heatwaves have increased over the
10 past century, causing mass mortalities in the hotspots of the Mediterranean (H216), Great Barrier Reef
11 (H236), western and southern Australia (H227, 228), north-west Atlantic (H207), and north-east Pacific
12 (H197) (high confidence) (Hobday et al., 2018; Oliver et al., 2018). The shift of thousands of species from
13 equatorial latitudes since the 1950s has been attributed to climate warming (medium confidence) (Chaudhary
14 et al., 2021).
16 Climate-change related hazards, particularly marine heat events, have caused widespread coral bleaching and
17 mass mortalities as the time between consecutive bleaching events decreases (high confidence) (IPCC, 2018;
18 Bindoff et al., 2019; IPCC, 2019b). Coral reefs in some Indian Ocean hotspots (H230, 234) already exhibit
19 net loss of coral reefs (low confidence) (Perry et al., 2018). While coral bleaching is a visible symptom of
20 heat stress, warming has also induced restructuring of associated fish and invertebrate communities in the
21 Great Barrier Reef (H236) (medium confidence) (Stuart-Smith et al., 2018).
23 Although the number of coral species that are both exposed and vulnerable to climate hazards is greatest in
24 the central Indo-Pacific, the proportion of corals at risk is greater in the lower diversity Caribbean hotspots
25 (H209) (medium confidence) (Foden et al., 2013). Some reef corals may acclimate to heatwaves (low
26 confidence) (DeCarlo et al., 2019), and some have expanded their latitudinal ranges polewards (high
27 confidence), up to 14 km/yr in the northwest Pacific (Yamano et al., 2011). Although future latitudinal
28 expansions may be limited by winter light availability (Muir et al., 2015) new coral reefs are already
29 emerging in Japan (Kumagai et al., 2018).
31 The Mediterranean Sea hotspot (H216) is negatively affected by climate change (high confidence) (Cross-
32 Chapter Paper 4). Species entering via the Suez Canal from the Red Sea (H220) are facilitated by warming
33 and lead to profound community changes (high confidence) (Yeruham et al., 2015; Rilov, 2016;
34 Vasilakopoulos et al., 2017; Givan et al., 2018; Bianchi et al., 2019). In contrast, the more open coastal seas
35 of both coasts of North America have had increasing species richness since the 1970s (Batt et al., 2017).
37 Kelp forests are in decline in mid-latitudes due to warming and associated increased herbivory (medium
38 confidence) (Section, Chapter 11). South and south-eastern (H228), and south-western (H227)
39 Australia have experienced a climate-related decline of kelp forests (Wernberg et al., 2011; Vergés et al.,
40 2016; Wernberg et al., 2016). West Australia (H227) has been affected by extreme climate events leading to
41 tropicalisation characterised by the replacement of kelp and sessile invertebrates by algal turfs and warm-
42 water fish species (Wernberg et al., 2013; Wernberg et al., 2016). Australia's Great Barrier Reef (H236),
43 kelp forests, seagrass meadows, and mangroves (due to drought), have suffered mortalities due to climate
44 change (medium confidence) (Babcock et al., 2019). Climate warming driven changes in seaweed
45 assemblages have been reported not only in Australia, but in the marine biodiversity hotspots of Atlantic
46 Canada, Japan, Mediterranean, New Zealand (Laffoley and Baxter, 2016; Thomsen et al., 2019; Thomsen
47 and South, 2019), and California (H207, 231, 216, 238, 199) (Arafeh-Dalmau et al., 2019; McPherson et al.,
48 2021). However, while climate change is having measurable effects on kelp, the dominant effects on kelp
49 projected to 2025 are fishing, through its effects on herbivores and predators (medium confidence) (Steneck
50 et al., 2002). Although fishing affected Atlantic cod in the Barents Sea (H214) and Gulf of Maine (H207)
51 biodiversity hotspots, so did climate change, but negatively and positively, respectively (Kjesbu et al., 2014;
52 Pershing et al., 2015).
54 Range expansions out of the Nansei Shoto (H231) hotspot south of Japan led to the replacement of temperate
55 kelp forests by tropical coral and herbivorous fishes on Japanese coasts (Kumagai et al., 2018). The Yellow
56 Sea (H230) is one of the most exploited marine hotspots, with decreasing ecosystem services compounded

Do Not Cite, Quote or Distribute  CCP1-26                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 by climate change but low confidence for climate change contributing substantially to ecological degradation
 2 (Wang et al., 2016; Song and Duan, 2019).
 4 Upwelling systems are best known for bringing nutrients to the surface that stimulate phytoplankton blooms,
 5 which in turn support important fisheries (Section However, this deep water also tends to be low in
 6 oxygen, which can be further depleted by respiration and surface warming. Prolonged marine heatwaves in
 7 this Californian hotspot (H197) drove major shifts in the geographic range of birds, mammals, fish,
 8 crustaceans, molluscs and other species, and toxic algal blooms (Sanford et al., 2019).
10 In both the Antarctic (H213) and Arctic (H196, 214), the loss of ice impacts on the behaviour and foraging
11 ability of marine mammals and birds (Doney et al., 2012). The retreat of sea-ice in the Bering Sea (H196)
12 hotspot has been followed by a reorganisation of the seabed and fish communities, a northward shift in
13 species, and greater species' biomass and richness (Mueter and Litzow, 2008; Grebmeier et al., 2018). In the
14 Eurasian Arctic (H214), species richness has similarly been increasing (Wslawski et al., 2011; Kortsch et
15 al., 2012; Certain and Planque, 2015; Fossheim et al., 2015; Wslawski et al., 2018) as has phytoplankton
16 productivity (Arrigo et al., 2008). The distribution of krill has already contracted with ocean warming in the
17 Southern Ocean (medium confidence) (Cox et al., 2018; Atkinson et al., 2019).
20 CCP1.2.4.2 Projected Impacts
22 Tropical extirpations, as already underway (CCP1.2.4.1), are projected to reduce hotspot diversity especially
23 in the Coral Triangle (H226, 232, 234), Maldives (H224), and to a lesser extent in the Caribbean (H200,
24 H210) (Jones and Cheung, 2015; García Molinos et al., 2016), and Persian Gulf (H219) (Wabnitz et al.,
25 2018). Paleo evidence supports projections of tropical biodiversity loss under high global warming (high
26 confidence) (Kiessling et al., 2012; Yasuhara et al., 2020).
28 Warm-water coral reefs are expected to decline with 1.5ºC warming (very high confidence) (King et al.,
29 2017; Bindoff et al., 2019) leading to systems with reduced biodiversity and structural complexity (high
30 confidence) (Chapters 3 and 11, Box 11.2). In the Coral Triangle, marine heatwaves are projected to have the
31 same effect as an added mean annual 0.5°C SST increase (McManus et al., 2020). While some corals may
32 survive in deep `mesophotic' reefs (Laverick and Rogers, 2019), the shallow coral reefs of today appear
33 unlikely to last the century if climate warming continues without mitigation (high confidence) (Hughes et al.,
34 2018a; Hughes et al., 2018b; IPCC, 2018; Bindoff et al., 2019; Hughes et al., 2019b).
36 In temperate shelf seas, ocean acidification may lead to increases of fleshy algae at the expense of calcifying
37 algae (low confidence) (Zunino et al., 2017). However, seagrass has been projected to decline (Chefaoui et
38 al., 2018) and increase (Zunino et al., 2017) in the Mediterranean Sea hotspot (H216). Kelp forests are
39 expected to decline in the northwest Atlantic (Grand Banks, H207), whereas gains and losses are
40 approximately balanced in the Northeast Atlantic Shelf (H215) under RCP 8.5 (Assis et al., 2018; Wilson et
41 al., 2019), but leading to impoverished benthic assemblages (medium confidence) (Teagle and Smale, 2018).
43 Projected climate caused changes in biodiversity in coastal upwelling regions are uncertain. While
44 productivity in the California Current (H197) system is projected to increase with future climate change,
45 non-linear plankton responses and uncertain interactions with food-web dynamics hinder predictions of
46 ecosystem responses (Xiu et al., 2018). In addition, this hotspot is projected to suffer from ocean
47 acidification by 2050 (Gruber et al., 2012).
49 Around Antarctica (H213), almost half of all species are endemic (Costello et al., 2010), and warming during
50 this century is projected to cause a reduction in suitable thermal environment for 79% of its species given
51 continued warming (RCP8.5) (low confidence) (Basher and Costello, 2016; Griffiths et al., 2017). The
52 previously mentioned declines in Southern Ocean krill due to climate change contribute to projected declines
53 in baleen whales there (Tulloch et al., 2019).
55 Species richness in the northern polar hotspots is expected to increase substantially (high confidence)
56 (Cheung et al., 2015). However, population sizes of presently occurring native species are expected to
57 decline, especially in the Barents Sea (H214) (Koenigstein et al., 2018). Ocean acidification is projected to

Do Not Cite, Quote or Distribute  CCP1-27                Total pages: 47
FINAL DRAFT                               Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

1 continue globally, and while its impact is uncertain and projected to be less than the effect of warming, it
2 may lead to changes in marine food webs due to varying effects on marine species (Terhaar et al., 2020).
3 Hotspots in temperate latitudes are projected to have assemblages modified by immigration from the tropics
4 and emigration to polar waters. Where land barriers and other geographical limits to range shifts occur,
5 limited dispersal and habitat fragmentation may also limit the capacity of species to track climate velocities
6 such as in the Baltic Sea (H215) (Jonsson et al., 2018), Mediterranean Sea (H216) (Burrows et al., 2014;
7 Arafeh-Dalmau et al., 2021), and Antarctica (H213) (medium confidence) (Cristofari et al., 2018).

10                                        CCP1-28                Total pages: 47
        Do Not Cite, Quote or Distribute
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

1 Figure CCP1.13: High latitude marine biodiversity hotspots. North-east Atlantic temperate seagrass beds, soft-corals,
2 and kelp forests in Norway (photos by Galice Hormeau). South African fynbos and Agulhas current and Antarctic
3 Peninsula and Weddell Sea (photos by Denis Costello). In the Americas, the Humboldt Current Chile and Chesapeake
4 Bay (photos by Mark Costello).

8 Figure CCP1.14: Species in island coral and rocky reef biodiversity richspots. Photos by Galice Hormeau (top four
9 Sulawesi), and Mark Costello (other nine).

Do Not Cite, Quote or Distribute  CCP1-29                Total pages: 47
    FINAL DRAFT                            Cross-Chapter Paper 1               IPCC WGII Sixth Assessment Report



3 CCP1.3         Adaptation and Solutions


5 Terrestrial, freshwater and coastal marine ecosystems are impacted by the three billion people that currently

6 live in in biodiversity hotspots (WGII Atlas). At the same time, biodiversity in hotspots supports the

7 livelihoods of the local communities. The suite of adaptation options for biodiversity are as applicable inside

8 as outside hotspots (Table CCP1.2). However, autonomous adaptation options are more limited in hotspots,

9 due to, for example, high endemism within hotspots, the fact that hotspots have evolved in isolation, their

10 species often have narrow thermal tolerances, or lack connectivity (e.g., mountaintops and islands) (high

11 confidence). In addition, many of these hotspots are now faced with widespread fragmentation and habitat

12 degradation (high confidence) (Table SMCCP1.1).


14 Because projected changes in biodiversity increase disproportionately with warming, climate change

15 mitigation is the primary action to conserve biodiversity within hotspots. If global warming is kept within the

16 1.5°C limit of the Paris Agreement, just ~4% of endemic species in biodiversity hotspots would be

17 threatened with extinction from climate change. However, at the current commitments there is projected to

18 be ~3°C warming by 2100 and ~20% and ~32% for terrestrial and marine species, respectively, fall into the

19 category of very high extinction risk (Figure CCP1.6; Manes et al., 2021).


21 Although mitigation can sharply reduce extinction risk associated with climate change (high confidence), it

22 cannot reduce all of the risk, nor the risk associated with other drivers that can have a compound effect with

23 climate change. Thus, in addition to mitigation, the literature consistently calls for reducing current non-

24 climate impacts (e.g., habitat conversion, over-exploitation, hunting, fishing, wildfire, pollution, human

25 introduced invasive species) in order to increase biodiversity resilience to climate change (very high

26 confidence) (e.g., Mantyka-Pringle et al., 2015; Warren et al., 2018a; Costello, in press). The main strategies

27 to increase resilience rely on the combination of well-planned protected areas, restoration of degraded areas,

28 and the sustainable use of biodiversity (high confidence) (Chapter 2; Section 3.6; Table CCP1.2; IPCC,

29 2019a; Pörtner et al., 2021) . On land, creating corridors for species is key for facilitating species movements

30 (high confidence) (McGuire et al., 2016; Heikkinen et al., 2020; Pörtner et al., 2021). Habitat protection has

31 numerous co-benefits, including potential climate mitigation through carbon storage and sequestration, in

32 addition to climatic regulation (Alkama and Cescatti, 2016; Mackey et al., 2020), and pandemic prevention

33 (Cross-Chapter Box COVID in Chapter 7; Allen et al., 2017; Dobson et al., 2020,).


35 Active relocation of endangered species to areas where they may be safer from predation and human

36 impacts, as already practised for a few charismatic fauna, is expensive and fraught with complex regulations

37 and concerns over impacts on native species (Brodie et al., 2021). Therefore, managed relocation of species

38 threatened by climate change is questionable for most species.


40 Healthier marine ecosystems are more resilient to additional stressors, such as storms and climate change

41 (high confidence) (Isbell et al., 2015; Duffy et al., 2016; Roberts et al., 2017; Bates et al., 2019; Mariani et

42 al., 2020; Donovan et al., 2021; Costello, in press). Extinction risk is lower when populations are larger and

43 more genetically diverse, individuals are larger and older, and seabed habitats (e.g., coral, kelp, seagrass) are

44 flourishing as occurs in marine reserves (high confidence) (Costello, 2014; Roberts et al., 2017; Bates et al.,

45 2019; Costello, in press). Similarly, global fish biomass may be less affected from climate change if

46 biodiversity is greater (Duffy et al., 2016). Thus, a network of reserves representative of global biodiversity,

47 helps attenuate the effects of climate change (medium confidence), e.g. by having more abundant fish and top

48 predator populations (Roberts et al., 2017; Beyer et al., 2018; Carter et al., 2020; Sala et al., 2021). However,

49 the impacts of marine heatwaves on corals across marine reserves illustrates that enhanced resilience is not

50 enough to protect against extreme and future climate change conditions (high confidence) (Bruno et al.,

51 2018; Hughes et al., 2018a; Kleypas et al., 2021).

52 Mangroves occupy the interface of terrestrial, freshwater and marine environments, dominate in eight

53 hotspots (Table CCP1.1) and are connected to or integral habitats within one-third of all terrestrial and

54 freshwater, and two-thirds of marine, hotspots (Figures CCP1.1, 2). A global analysis of sediment cores from

55 mangroves indicated that mangroves seem to be able to accrete sediment at levels of sea-level rise projected

56 under low emission scenarios, but may decline at their seaward edge under high emission scenarios (high

57 confidence) (Saintilan et al., 2020, Section 3.4.2, Cross-Chapter Box SLR in Chapter 3). However, even if

    Do Not Cite, Quote or Distribute                   CCP1-30                        Total pages: 47
    FINAL DRAFT                               Cross-Chapter Paper 1            IPCC WGII Sixth Assessment Report

1 this seaward erosion occurs, the expansion of mangroves inland due to sea-level rise will increase carbon

2 sequestration, because they capture carbon from seawater and freshwater runoff, in addition to

3 photosynthesis, into their underlying sediments. If coastal management permits the expansion of mangroves

4 inland with rising sea level, this will increase carbon sequestration because mangroves capture and preserve

5 more carbon in their sediments than other terrestrial and marine forests and biomes (high confidence) (Table

6 CCP1.2; Alongi, 2020; Goldstein et al., 2020; Lovelock and Reef, 2020; Saintilan et al., 2020).

7 On land, fragmentation and habitat degradation are particularly pervasive, imposing hard limits to adaptation

8 of terrestrial and freshwater ecosystems (Ibisch et al., 2016; Lenoir et al., 2020; Mechler et al., 2020). Thus,

9 the protection of existing natural habitats coupled with the restoration of the surrounding non-protected

10 habitat can increase the effectiveness of adaptation strategies in terrestrial and freshwater hotspots (very high

11 confidence) (Table CCP1.2; IPCC, 2019a; Jung et al., 2021) . Additionally, strategic allocation of new

12 protected areas within gaps across elevational and climatic gradients could enhance biodiversity conservation

13 across hotspots. This would align with target 3 of the Convention on Biological Diversity's post-2020 global

14 biodiversity framework, and could include underrepresented climate and elevation spaces as well as potential

15 climate refugia currently not under protection (Pörtner et al., 2021). In terrestrial ecosystems, restoration

16 initiatives can help sustain biodiversity, improve resilience in a changing climate, and avoid maladaptation

17 by selecting appropriate native species to be planted (Cross-Chapter Box BIOECONOMY in Chapter 5;

18 Gann et al., 2019). In freshwater ecosystems, conservation needs catchment level management of human

19 activities (Saunders et al., 2002; Dudgeon et al., 2006) (Chapter 11, Box 11.6) especially as 37% of the

20 terrestrial biodiversity hotspots overlap with freshwater (Figure CCP1.2), and 23% border marine hotspots

21 (Olson and Dinerstein, 2002).


23 Protecting biodiversity hotspots is a pragmatic way to conserve biodiversity that is representative of a

24 substantive fraction of genetic and species diversity on Earth (Mittermeier et al., 2011) while achieving co-

25 benefits (Bonan, 2016; Sala et al., 2021). Protecting hotspots also helps protect important ecosystem

26 services. In a global ranking of areas that combine biodiversity conservation while maximizing carbon

27 retention and water quality regulation, for example, the terrestrial and freshwater hotspots assessed here

28 ranked high (41st and 34th on average, respectively, on a scale of 100) (Jung et al., 2021). The solutions

29 needed to reverse biodiversity decline are well known and articulated in numerous international agreements

30 and goals, such as the Convention of Biological Diversity Aichi Targets, the United Nations' Sustainable

31 Development Goals, the Nationally Determined Contributions under the Paris Agreement, the International

32 Union for Nature Conservation's Bonn Restoration Challenge, and the Ramsar Convention on Wetland

33 Conservation. Thus, expanding and enhancing protection of a worldwide network of fully protected areas

34 and protection and restoration of non-protected areas representative of the biodiversity hotspots -- inclusive

35 of marine, freshwater and terrestrial environments, is a highly recommended adaptation strategy to increase

36 resilience of biodiversity to climate change (Brito-Morales et al., 2018). However, adaptation strategies

37 alone cannot protect biodiversity from climate impacts without complementary and concomitant reduction of

38 greenhouse gas emissions.



41 Table CCP1.2: Examples of adaptation actions that benefit the conservation of biodiversity and climate

42 change mitigation.

    Actions                   Terrestrial         Freshwater                   Marine

                                                  Stop pollution and

    Protect            Protect native forests, bush, sedimentation into        Ban seabed trawling and dredging
                              and grasslands      streams, rivers, ponds,

    hotspots                                      lakes

                                      Control introduction and spread of invasive species and pests

    Increase           Use riverbank and hedgerow corridors to connect         Already connected
    connectivity                      protected native habitats

                       Reduce habitat and species loss outside protected areas to add species dispersal (corridors)

    Outside            Environmentally sustainable agriculture, tourism, and    Environmentally sustainable
    biodiversity                     other land and freshwater uses            aquaculture, fisheries, tourism

    Restoration and    Actively rehabilitate old   Stabilise riverbanks.          Ban removal of marine life and
    recovery             mines, quarries and         Remove weirs and          habitat, and fishing in selected areas
                            industrial lands      artificial barriers to fish  to allow passive recovery of habitats,


    Do Not Cite, Quote or Distribute              CCP1-31                                            Total pages: 47
    FINAL DRAFT                              Cross-Chapter Paper 1        IPCC WGII Sixth Assessment Report

                     Reintroduce extirpated native species                natural population structure, and food

                     Preserve, reduce degradation, and restore habitats to enable uplands to absorb rainfall and

    Reduce erosion,                                             reduce flash floods.
    soil loss, and         Protect sand-dune systems from erosion due to human and farm animal trampling.
    flooding         Set aside land for salt marshes and mangroves to buffer against river and seawater flooding.
                     Link estuarine and upriver protected areas to provide more wildlife habitat and absorb storm

                                                   surges and floods.

    Urban            Concentrate development to         Limit upland      Avoid construction in areas at risk of
    development      more cost efficiently manage  development where it    sea level rise and associated storm
                                                   may affect freshwater                    surges.
                          transport and waste
                      management infrastructure            quality

                     Prevent deforestation;                               Limit seabed disturbance by trawling

    Greenhouse gas    Reforestation (especially    Expand wetlands to      and dredging that releases CO2 and
    mitigation       mangroves); Revegetation;     capture and deposit                       CH4.

                       Fewer farm mammals.           carbon in soils.        Eliminate fishery subsidies, and
                         Minimise release of                              remove tax breaks on fuel for fishing

                     greenhouse gases from soils.                                     boats.

                     Allow biodiversity to flourish and capture CO2 from the air and sequester it in biomass, soils

    Carbon                                         and sediments.

    sequestration    Manage forestry to

    and preservation maximise in situ food web     Manage fisheries to maximise in situ food web biomass.


    Social           Communicate information on the benefits of adaptation measures to the public

    Political and    Provide leadership and governance of mitigation and adaptation measures, including through
    economic            regulations and economic incentives that guide the transition to a low carbon emissions

                     Address data gaps and make monitoring data and its meaning rapidly available to society so that

                     the public and policy makers are informed of trends in biodiversity and related factors,

                     including climate variables, extreme weather related events, threatened and invasive species,

    Scientific                                 natural habitats, and their relationships.
                     Conduct research to improve understanding of cause-effect relationships regarding

                     environmental factors and biodiversity trends, including in nature conservation, forestry,

                     agriculture, fisheries and food production sectors, and improve projections of consequences of

                                                   management action and inaction





5 Box CCP1.1: Climate Change and Terrestrial Biodiversity Hotspots on Small Islands


7 Despite covering approximately 2% of the Earth's land area, islands harbour more than 20% of extant

8 terrestrial species (Wetzel et al., 2013). Islands have disproportionately higher rates of endemism and threat

9 when compared to continents, with 80% of historical extinctions (since 1500 AD) having occurred on islands

10 (high confidence) (Taylor and Kumar, 2016; Spatz et al., 2017; Dueñas et al., 2021). Current climate change

11 projections suggest that insular species are particularly sensitive, and even at mild warming levels,

12 substantial losses are expected (high confidence) (Pouteau and Birnbaum, 2016; Taylor and Kumar, 2016;

13 Dawson et al., 2017; Manes et al., 2021). Given islands' characteristic high endemicity, current high threat

14 levels and the fact that islands host almost half of all species currently considered to be at risk of extinction,

15 at especially higher warming levels (high confidence) (Taylor and Kumar, 2016; Spatz et al., 2017) --

16 further losses could contribute disproportionately to global biodiversity decline (medium evidence, high

17 agreement) (Harter et al., 2015; Pouteau and Birnbaum, 2016; Manes et al., 2021).


19 The high vulnerability of terrestrial biodiversity on islands to global change can be explained by a number of

20 limitations, characteristic of both islands and insular species. Older, isolated islands tend to have fewer

21 species and lower functional redundancy but higher proportion of endemism (Pouteau and Birnbaum, 2016;

22 Médail, 2017). Many of these islands contain species with inherently high sensitivity to environmental

23 change (narrow habitat ranges, small population sizes, low genetic diversity and poor adaptive, dispersal and

24 defensive capabilities) (Harter et al., 2015). Unlike continental environments, insular species often have

25 limited opportunities for autonomous adaptation from not having enough geographic space to shift their

    Do Not Cite, Quote or Distribute               CCP1-32                                    Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 ranges to track suitable climatic conditions (high confidence) (Fortini et al., 2015; Manes et al., 2021). Local
 2 extinction risks are amplified by even small losses of habitat due to global change including human-induced
 3 disturbances, extreme events, sea-level rise (Chapter 15; Cross-Chapter Box SLR in Chapter 3) and invasive
 4 species.
 6 However, some insular species may be resilient to climate change (low confidence). Intact island forests, for
 7 example, have shown rapid recovery rates after tropical cyclones, despite high levels of initial damage,
 8 especially in the Caribbean (medium confidence) (Luke et al., 2016; Richardson et al., 2018). Additionally,
 9 many Mediterranean islands are `disturbance adapted', with continued persistence of some single-island
10 endemic plants, despite exposure to multiple threats (Vogiatzakis et al., 2016). This continued persistence
11 has been attributed, at least partially, to climate refugia, oceanic buffering, and high habitat heterogeneity
12 within topographically complex mountainous regions (Chapter 15, Table 15.1; Pouteau and Birnbaum, 2016;
13 Médail, 2017). However, this climate resilience may not be sustained under climate change, especially when
14 coupled with habitat degradation (high confidence) (Wiens, 2016).
16 Adaptation strategies depend on the ability to project future impacts from climate change, but this is
17 hampered by lack of fine-scale climate data especially for developing small island nations. There is a paucity
18 of robust impacts-based modelling output for terrestrial biodiversity from these islands due to the wide,
19 chronic unavailability of Regional Climate Model (RCM) data premised on the most recent suite of scenarios
20 (RCPs and especially SSPs) (medium evidence, high agreement) (Gutiérrez et al., 2021, Ch.15.8; Pörtner et
21 al., 2021; WMO, 2021). Additionally, realistic assessments of changing climate on such small ecosystems
22 require further RCM downscaling and verification to sub-island resolutions of < 5 km. Furthermore, widely
23 used statistically (bias-corrected) downscaled data at sub-5 km resolutions such as WorldClim are often
24 unsuitable due to limited spatial and temporal resolutions of observation station data from small islands
25 (Maharaj and New, 2013; Gutiérrez et al., 2021), and higher errors associated with statistical downscaling
26 and locations with complex topography and coastlines (Fick and Hijmans, 2017; Lanzante et al., 2018).
27 Widespread unavailability of such data constrains accurate simulations of climatic variation within the small-
28 scale mountainous and coastal regions of islands, associated with climate refugia and high habitat
29 heterogeneity (high confidence) (Balzan et al., 2018). This is a key element contributing to the continued
30 delay in development of robust adaptation strategies towards not only biodiversity conservation but other
31 important cross-sectoral issues (medium confidence) (Robinson, 2020b).
33 Due to islands' limited size and isolation, conventional conservation measures focused on expanding
34 protected areas, dispersal corridors and buffer zones are of limited effectiveness on islands (high confidence)
35 (Vogiatzakis et al., 2016). Instead, multifaceted, locally-driven holistic climate-smart strategies across
36 mosaics of human-impacted, often heavily degraded and fragmented landscapes are required. These should
37 ideally be long-term, flexible and sustainable solutions that incorporate social and biocultural knowledge as
38 well as economic co-benefits to island communities in order to `buy time' (Betzold, 2015; Robinson, 2020a).
39 Examples include ecosystem-based approaches such as ridge-to-reef management (Table CCP1.2.4; Struebig
40 et al., 2015; Ferreira et al., 2019), which incorporates conservation partnerships among lands inside and
41 outside protected areas to increase connectivity and reduce land-use impacts, while building on the
42 interconnections among terrestrial, freshwater, coastal and marine ecosystems. Such strategies require raising
43 awareness of biodiversity values among local communities, and cross-sectoral planning and policy at both
44 island, regional and trans-boundary scales. These lend to private-public partnerships, increasing the potential
45 of solutions reaching beyond protected areas boundaries and affecting socio-political change (high
46 confidence) (Scobie, 2016).
48 Limited terrain, natural, economic and data resources across small developing-nation islands mean that
49 unconstrained habitat destruction and degradation cannot be sustained, as this harms both people and the
50 biodiversity upon which they depend. This limitation of resources compromises climate adaptation -- which
51 is often further complicated by varying governance and states of economic development (Petzold and
52 Magnan, 2019). With changing climate conditions, there is an increased urgency to re-think how progress
53 can be measured, and to create opportunities building on synergies between disaster risk reduction, food
54 security and social justice ­ so that islands can most benefit from their natural resources and biodiversity in a
55 sustained manner (Box 15.2; Section ).

Do Not Cite, Quote or Distribute  CCP1-33                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 5 FAQ CCP1.1: Why are biodiversity hotspots important?
 7 Biodiversity hotspots are regions that are exceptionally rich in species, ecologically unique and which may
 8 contain geographically restricted species. They are thus priority targets for nature conservation.
10 Recognizing that the Convention on Biological Diversity definition of biodiversity includes the variation
11 within and between species and of ecosystems, different schemes have been applied to define hotspots,
12 leading to hundreds of different areas being proposed as hotspots. However, all identify a set of priority areas
13 that cover a small portion of the Earth, but house an exceptionally high proportion of its biodiversity.
14 Because biodiversity underpins all life on Earth, these hotspots have significant global value as they contain
15 species and habitats that are found nowhere else. Their loss would mean loss of species and habitats that
16 provide wild and farmed food, medicine and other materials, and services such as climate regulation,
17 pollination and water purification, all of which maintain the health of the ecosystems we depend upon.
19 Healthy ecosystems, with flourishing biodiversity in natural conditions, are more resilient to disturbances,
20 whether natural or human in origin. Environmentally sustainable development inside and outside hotspots
21 could help reverse human impacts on biodiversity. The hotspots also capture and store carbon, thereby
22 helping to mitigate climate change. Prioritization of protecting biodiversity in hotspots thus benefits nature
23 conservation and helps mitigate climate change. A global network of protected areas and restoration
24 initiatives inside biodiversity hotspots can also help increase resilience to the effects of climate change on
25 biodiversity.
32 FAQ CCP1.2: How can society ensure conservation of biodiversity in climate policies?
34 To reduce the effects of climate change on biodiversity it is first essential to address direct human impacts
35 that are already leading to a loss of biodiversity. This can be achieved by protecting biodiversity in
36 conservation areas, restoring biodiversity everywhere possible and promoting sustainable development.
37 Climate policies should thus integrate with policies to protect and restore nature.
39 Avoiding further loss of biodiversity is implicit in sustainable development. This needs to happen on land,
40 rivers, lakes and in the oceans. It is especially important in "biodiversity hotspots" (FAQ-CCP1.1) and
41 protected areas. Hence calls by the International Union for the Conservation of Nature (IUCN), Convention
42 on Biological Diversity, United Nations Sustainable Development Goals and scientific community to
43 increase the size and connectivity of fully protected areas (which aim to have biodiversity in a near natural
44 condition) and include in them the biodiversity hotspots, need to be immediately implemented.
46 Five of the Sustainable Development Goals (SDGs) are life on land, life below water, good health and well-
47 being, food security, and climate action. They underpin and interact with many other SDGs. Healthy
48 ecosystems play a role in mitigating greenhouse gas emissions, not only protecting areas to prevent the
49 release of carbon through land conversion activities but also restoring otherwise degraded land. The United
50 Nations has declared 2021- 2030 as the Decade of Restoration, and the Decade of the Oceans. Restoration
51 means actively or passively allowing habitat to return to its natural state (e.g., grassland, forest, peatland,
52 oyster beds), including replanting native vegetation. This can benefit the recovery of biodiversity, help
53 remove carbon dioxide from the atmosphere, and improve the delivery of nature's contributions to people
54 such as climate regulation, water purification, pollination and pest and disease control. Thus, protecting
55 biodiversity helps to meet two SDGs directly, and three indirectly.

Do Not Cite, Quote or Distribute  CCP1-34                Total pages: 47
FINAL DRAFT                       Cross-Chapter Paper 1  IPCC WGII Sixth Assessment Report

 1 On land, the loss of natural forests and grasslands not only means a loss of carbon and many of their
 2 associated species, but exposes soils to erosion, affecting food production, and can affect the climate by
 3 altering the water-cycle. Sustainable development, even within hotspots, involves active restoration of
 4 natural biodiversity, reducing poaching and trafficking of wildlife (UN SDG 15), and needs to include
 5 agriculture. This includes working to ensure biodiverse soils, and supporting healthy pollinator populations.
 6 Biodiversity includes not only wild species but also the genetic diversity, including crops and wild crop
 7 relatives. These wild relatives may contain important genes that could help farmed crops survive better in a
 8 changed climate. At least some of these wild relatives come from areas designated as hotspots. In the ocean,
 9 sustainable development means reducing pollution, carefully managed aquaculture development, increased
10 protected areas (from the present 2.5% of the ocean area), enforcement of fisheries regulations, and removal
11 of fishery subsidies that perpetuate over-fishing within Exclusive Economic Zones and on the High Seas
12 (UN SDG 14). Generally, the use of freshwaters, both rivers, lakes and groundwaters, has not been
13 sustainable and needs to restore biodiversity and water quality by eliminating pollution, and better
14 management of abstraction, river flows, fishing and invasive species. Thus, as is the case with land and
15 oceans, climate policies must prioritize the restoration of freshwater biodiversity, and reduction of the
16 current negative impacts of human activities.

Do Not Cite, Quote or Distribute  CCP1-35                Total pages: 47