Editing IPCC:AR6/SROCC/Chapter-3 (section)

==== 3.2.3.1 Arctic ====

<div id="section-3-2-3-1-arctic-block-1"></div>

Climate change has, and is projected to continue to have, significant implications for Arctic marine ecosystems, with consequences at different trophic levels both in the pelagic, benthic, and sympagic (sea ice related) realms (Figure 3.5). Specifically, climate change is projected to alter the distribution and properties of Arctic marine habitats with associated implications for species composition, production and ecosystem structure and function (Frainer et al., 2017 <sup>[[#fn:r518|518]]</sup> ; Kaartvedt and Titelman, 2018 <sup>[[#fn:r519|519]]</sup> ; Moore et al., 2018 <sup>[[#fn:r520|520]]</sup> ). The rate and severity of ecosystem impacts will be spatially heterogeneous and dependent on future emission scenarios.

In the few Arctic regions where data is sufficient to assess trends in biodiversity, the ecosystem level responses appear to be products of multiple interacting physical, chemical and biological processes (Frederiksen, 2017 <sup>[[#fn:r521|521]]</sup> ) ( ''medium confidence'' ). Climate change impacts on vertical fluxes and stratification (Sections 3.2.1.2.3, 3.2.2.2) will contribute to changes in bentho-pelagic-sympagic coupling. For instance, projected climate driven changes in ocean properties and hydrography (Section 3.2.2.2) and the abundance of pelagic grazers (Box 3.4) could alter the export of organic matter to the sea floor with associated impacts on the benthos in some Arctic shelf ecosystems (Moore and Stabeno, 2015 <sup>[[#fn:r522|522]]</sup> ; Stasko et al., 2018 <sup>[[#fn:r523|523]]</sup> ) ( ''low confidence'' ). Projected future reductions in summer sea ice (Section 3.2.1.1), increased stratification in summer, shifting currents and fronts and increased ocean temperatures (Section 3.2.2.2) and ocean acidification (Section 3.2.2.3) are all expected to impact the future production and distribution of several marine fish and invertebrates ( ''high confidence'' ).

Ocean acidification (Section 3.2.2.3) will affect several key Arctic species ( ''medium confidence'' ). The effects of current and projected levels of acidification have been examined for a broad suite of species groups (bivalves, cephalopods, echinoderms, crustaceans, corals and fishes) and these studies reveal species-specific differences in sensitivity, as well as differences in the scope for, and energetic cost of, adaptation (Luckman et al., 2014 <sup>[[#fn:r524|524]]</sup> ; Howes et al., 2015 <sup>[[#fn:r525|525]]</sup> ; Falkenberg et al., 2018 <sup>[[#fn:r526|526]]</sup> ).

<div id="section-3-2-3-1-arctic-block-2"></div>

<span id="plankton-and-primary-production"></span>
===== 3.2.3.1.1 Plankton and primary production =====

There is evidence that the combination of loss of sea ice, freshening, and regional stratification (Sections 3.2.1.1 and 3.2.1.2) has affected the timing, distribution and production of primary producers (Moore et al., 2018 <sup>[[#fn:r527|527]]</sup> ) ( ''high confidence'' ). Satellite data show that the decline in ice cover has resulted in a &gt;30% increase in annual net primary production (NPP) in ice-free Arctic waters since 1998 (Arrigo and van Dijken, 2011 <sup>[[#fn:r528|528]]</sup> ; Bélanger et al., 2013 <sup>[[#fn:r529|529]]</sup> ; Arrigo and van Dijken, 2015 <sup>[[#fn:r530|530]]</sup> ; Kahru et al., 2016 <sup>[[#fn:r531|531]]</sup> ), a phenomenon corroborated by both ''in situ'' data (Stanley et al., 2015 <sup>[[#fn:r532|532]]</sup> ) and modelling studies (Vancoppenolle et al., 2013 <sup>[[#fn:r533|533]]</sup> ; Jin et al., 2016 <sup>[[#fn:r534|534]]</sup> ). Ice loss has also resulted in earlier phytoplankton blooms (Kahru et al., 2011 <sup>[[#fn:r535|535]]</sup> ) with blooms being dominated by larger-celled phytoplankton (Fujiwara et al., 2016 <sup>[[#fn:r536|536]]</sup> ). The longer open water season in the Arctic has also increased the incidence of autumn blooms, a phenomenon previously rarely observed in Arctic waters (Ardyna et al., 2017 <sup>[[#fn:r537|537]]</sup> ).

Thinner Arctic sea ice cover has led to the appearance of intense phytoplankton blooms that develop beneath first-year sea ice ( ''medium confidence'' ). Blooms of this size (1000s of km 2 ) and intensity (peaks of approximately 30 mg Chla-m –3 ) were previously thought to be restricted to the marginal ice zone and the open ocean where ample light reaches the surface ocean for rapid phytoplankton growth (Arrigo et al., 2012 <sup>[[#fn:r538|538]]</sup> ). Evidence shows that these blooms can thrive beneath sea ice in areas of reduced thickness, increased coverage of melt ponds (Arrigo et al., 2014 <sup>[[#fn:r539|539]]</sup> ; Zhang et al., 2015 <sup>[[#fn:r540|540]]</sup> ; Jin et al., 2016 <sup>[[#fn:r541|541]]</sup> ; Horvat et al., 2017 <sup>[[#fn:r542|542]]</sup> ), first-year ridges at the snow-ice interface (Fernández-Méndez et al., 2018 <sup>[[#fn:r543|543]]</sup> ), and a large number of cracks (high lead fractions) in the ice (Assmy et al., 2017 <sup>[[#fn:r544|544]]</sup> ), although the latter has not changed significantly in the last three decades (Wang et al., 2016a <sup>[[#fn:r545|545]]</sup> ). Local features including snow-free or thin snow, hummocks and ridges commonly found on multi-year ice also provide habitat for ice algae (Lange et al., 2017 <sup>[[#fn:r546|546]]</sup> ).

The reduction in sea ice area and thickness in the Arctic Ocean appears to be indirectly impacting rates of NPP through increased exposure of the surface ocean to atmospheric forcing ( ''medium confidence'' ) and these indirect impacts will possibly increase in the future ( ''low confidence'' ). Greater wind stress has been shown to increase upwelling of nutrients at the shelf break both over ice-free waters (Williams and Carmack, 2015 <sup>[[#fn:r547|547]]</sup> ) and a partial ice cover (Schulze and Pickart, 2012 <sup>[[#fn:r548|548]]</sup> ), leading to more new production (Williams and Carmack, 2015 <sup>[[#fn:r549|549]]</sup> ). At the same time, enhanced vertical stratification (Section 3.2.1.2.2, SM3.2.2) and decreased upwelling of nutrients into surface waters (Capotondi et al., 2012 <sup>[[#fn:r550|550]]</sup> ; Nummelin et al., 2016 <sup>[[#fn:r551|551]]</sup> ) may reduce Arctic NPP in the future, especially in the central basin (Ardyna et al., 2017 <sup>[[#fn:r552|552]]</sup> ). It could also impact phytoplankton community composition and size structure, with small-celled phytoplankton, which require less nutrients, becoming more dominant as nutrient concentrations in surface waters decline (Yun et al., 2015 <sup>[[#fn:r553|553]]</sup> ).

In addition to its impact on phytoplankton bloom dynamics, the decline in the proportion of multi-year sea ice and proliferation of a thinner first year sea ice cover may favour growth of microalgae within the ice due to increased light availability ( ''medium confidence'' ). Recent studies suggest that the contribution of sea ice algae to total Arctic NPP is higher now than values measured previously (Song et al., 2016 <sup>[[#fn:r554|554]]</sup> ), accounting for nearly 10% of total NPP (ice plus water) and as much as 60% in places like the central Arctic (Fernández-Méndez et al., 2015 <sup>[[#fn:r555|555]]</sup> ).

Ongoing changes in NPP will impact the biogeochemistry and ecology of large parts of the Arctic Ocean ( ''high confidence'' ). In areas of enhanced nutrient availability and greater NPP, dominance by larger-celled microalgae increases vertical export efficiency from the surface downwards in both ice covered (Boetius et al., 2013 <sup>[[#fn:r556|556]]</sup> ; Lalande et al., 2014 <sup>[[#fn:r557|557]]</sup> ; Mäkelä et al., 2017 <sup>[[#fn:r558|558]]</sup> ) and open ocean (Le Moigne et al., 2015) areas. However, because exported biomass production may be increasing in some areas but declining in others, the net impact may be small (Randelhoff and Guthrie, 2016 <sup>[[#fn:r559|559]]</sup> ) (Sections 3.2.3.1.2, 5.3.6, SM3.2.6). Phytoplankton may have the capacity to compensate for ocean acidification under a range of temperatures and pH values (Hoppe et al., 2018 <sup>[[#fn:r560|560]]</sup> ).

Increased water temperatures (Section 3.2.1) and shifts in the spatial pattern and timing of the ice algal and phytoplankton blooms, have impacted the phenology, magnitude and duration of zooplankton production with associated changes in the zooplankton community composition ( ''medium confidence'' ). Negative effects of reductions in ice algae on zooplankton may be partially offset by predicted increases in water column phytoplankton production in the Bering Sea (Wang et al., 2015 <sup>[[#fn:r561|561]]</sup> ). Changes in sea ice coverage and thickness may alter the phenology, abundance and distribution of zooplankton in the future. Projected changes will initially have the most pronounced impact on sympagic amphipods, but will subsequently affect food web functioning and carbon dynamics of the pelagic system (Kohlbach et al., 2016 <sup>[[#fn:r562|562]]</sup> ).

At the more southern boundaries of the Arctic such as the southeastern Bering Sea, warm conditions have led to reduced production of large copepods and euphausiids ( ''medium confidence'' ) (Sigler et al., 2017 <sup>[[#fn:r563|563]]</sup> ; Kimmel et al., 2018 <sup>[[#fn:r564|564]]</sup> ). On more northern shelves, the increased open water period has led to increases in large copepods over a 60 year period within the Chukchi Sea (Ershova et al., 2015 <sup>[[#fn:r565|565]]</sup> ) and in recent years also the Beaufort Sea (Smoot and Hopcroft, 2017 <sup>[[#fn:r566|566]]</sup> ), while in the Central Basins zooplankton biomass in general has increased (Hunt et al., 2014 <sup>[[#fn:r567|567]]</sup> ; Rutzen and Hopcroft, 2018 <sup>[[#fn:r568|568]]</sup> ) ( ''medium confidence'' ).

There are inconsistent findings concerning the future development of copepods in the Arctic. Coupled biophysical model results suggest that sea ice loss will increase primary production and that will primarily be consumed pelagically by zooplankton grazers such as ''Calanus hyperboreus'' ; increasing their abundances in the central Arctic (Kvile et al., 2018 <sup>[[#fn:r569|569]]</sup> ). Feng et al. (2018) concluded that ''C. glacialis'' should continue to benefit from a warmer Arctic Ocean. On the other hand, in the transition zone between Arctic and Atlantic water masses, ''C. glacialis'' may face increasing competition from the more boreal ''C. finmarchicus'' (Dalpadado et al., 2016 <sup>[[#fn:r571|571]]</sup> ). Renaud et al. (2018) <sup>[[#fn:r572|572]]</sup> found the lipid content of ''Calanus'' spp. was related to size and not species. This suggests that climate driven shifts in dominant ''Calanus'' species may, because of overlap in size spectrum and contrary to earlier assumptions, not negatively impact their consumers in the Barents Sea.

The effects of ocean acidification on Arctic zooplankton and pteropods (small pelagic molluscs) have been examined for only a few species and these studies reveal that the severity of effects is dependent on emission scenarios and the species sensitivity and adaptive capacity. The copepod ''C. glacialis'' exhibits stage-specific sensitivities to ocean acidification with some stages being relatively insensitive to decreases in pH and other stages exhibiting substantial reductions in scope for growth (Bailey et al., 2017 <sup>[[#fn:r573|573]]</sup> ; Thor et al., 2018 <sup>[[#fn:r574|574]]</sup> ). Although there is strong evidence that pteropods are sensitive to the effects of ocean acidification (Manno et al., 2017 <sup>[[#fn:r575|575]]</sup> ) recent studies indicate they may exhibit some ability to adapt (Peck et al., 2016 <sup>[[#fn:r576|576]]</sup> ; Peck et al., 2018 <sup>[[#fn:r577|577]]</sup> ). However, the metabolic costs of adaptation may be constraining, especially during periods of low food availability (Lischka and Riebesell, 2016 <sup>[[#fn:r578|578]]</sup> ).

<div id="section-3-2-3-1-arctic-block-3"></div>

<span id="benthic-communities"></span>
===== 3.2.3.1.2 Benthic communities =====

There is evidence that earlier spring sea ice retreat and later autumn sea ice formation (Section 3.2.1.1) are changing the phenology of primary production with cascading effects on Arctic benthic community biodiversity and production (Link et al., 2013 <sup>[[#fn:r579|579]]</sup> ) ( ''medium confidence'' ). In the Barents Sea, evidence suggests that factors directly related to climate change (sea ice dynamics, ocean mixing, bottom-water temperature change, ocean acidification, river/glacier freshwater discharge; Sections 3.2.1.1, 3.2.1.2) are impacting the benthic species composition (Birchenough et al., 2015 <sup>[[#fn:r580|580]]</sup> ). Other human influenced activities, such as commercial bottom trawling and the introduction of non-native species are also regarded as major drivers of observed and expected changes in benthic community structure (Johannesen et al., 2017 <sup>[[#fn:r581|581]]</sup> ), and may interact with climate impacts.

Rapid and extensive structural changes in the rocky-bottom communities of two Arctic fjords in the Svalbard Archipelago during 1980–2010 have been documented and linked to gradually increasing seawater temperature and decreasing sea ice cover (Kortsch et al., 2012 <sup>[[#fn:r582|582]]</sup> ; Kortsch et al., 2015) <sup>[[#fn:r583|583]]</sup> . Also, there are indications of declining benthic biomass in the northern Bering Sea (Grebmeier and Cooper, 2016 <sup>[[#fn:r584|584]]</sup> ) and southern Chukchi Sea (Grebmeier et al., 2015 <sup>[[#fn:r585|585]]</sup> ). It is unclear whether these rapid ecosystem changes will be tipping points for local ecosystems (Chapter 6, Table 6.1; Wassmann and Lenton, 2012 <sup>[[#fn:r586|586]]</sup> ). However, biomass of kelps have increased considerably in the intertidal to shallow subtidal in Arctic regions over the last two decades, connected to reduced physical impact by ice scouring and increased light availability as a consequence of warming and concomitant fast-ice retreat (Kortsch et al., 2012 <sup>[[#fn:r587|587]]</sup> ; Paar et al., 2016 <sup>[[#fn:r588|588]]</sup> ) ( ''medium confidence'' ) (see Section 5.3.3 and SM3.2.6 for further information on kelp).

The growth, early survival and production of commercially important crab stocks in the Bering Sea are influenced by time-varying exposure to multiple interacting drivers including bottom temperature, larval advection, predation, competition and fishing (Burgos et al., 2013 <sup>[[#fn:r589|589]]</sup> ; Long et al., 2015 <sup>[[#fn:r590|590]]</sup> ; Ryer et al., 2016 <sup>[[#fn:r591|591]]</sup> ). In Newfoundland and Labrador waters and on the western Scotian Shelf, snow crab ( ''Chionoecetes opilio'' ) productivity has declined (Mullowney et al., 2014 <sup>[[#fn:r592|592]]</sup> ; Zisserson and Cook, 2017 <sup>[[#fn:r593|593]]</sup> ). Contrary to this, snow crabs have expanded their distribution in the Barents Sea and commercial harvesting increased (Hansen, 2016 <sup>[[#fn:r594|594]]</sup> ; Lorentzen et al., 2018 <sup>[[#fn:r595|595]]</sup> ) ( ''high confidence'' ).

Bering sea crabs exhibit species-specific sensitivities to reduced pH (Long et al., 2017 <sup>[[#fn:r596|596]]</sup> ; Swiney et al., 2017 <sup>[[#fn:r597|597]]</sup> ; Long et al., 2019 <sup>[[#fn:r598|598]]</sup> ). However, current pH levels do not appear to have negatively impacted crab production in the Bering or Barents Seas (Mathis et al., 2015 <sup>[[#fn:r599|599]]</sup> ; Punt et al., 2016 <sup>[[#fn:r600|600]]</sup> ).

<div id="section-3-2-3-1-arctic-block-4"></div>

<span id="fish"></span>
===== 3.2.3.1.3 Fish =====

Since AR5, additional evidence shows climate-induced physical and biogeochemical changes are impacting, and will continue to impact, the distribution and production of marine fish ( ''medium confidence'' ). Changes in the spatial distribution and production of Arctic fish are best documented for ecologically and commercially important stocks in the Bering and Barents Seas (Box 3.4; Figure 3.5), while data is severely limited in other Arctic shelf regions and the Central Arctic Ocean (CAO).

Higher temperature and changes in the quality and distribution of prey is already affecting marine fish (Wassmann et al., 2015 <sup>[[#fn:r601|601]]</sup> ; Dalpadado et al., 2016 <sup>[[#fn:r602|602]]</sup> ; Hunt et al., 2016 <sup>[[#fn:r603|603]]</sup> ; Section 3.2.3.1) ( ''high confidence'' for detection '', medium confidence'' for attribution). In the northern Barents Sea, Atlantic Sector, higher temperatures (Section 3.2.1.2) have expanded suitable feeding areas for boreal/subarctic species (Box 3.4) and has contributed to increased Atlantic cod ( ''Gadus morhua'' ) production (Kjesbu et al., 2014 <sup>[[#fn:r604|604]]</sup> ). In contrast, Arctic species like polar cod ( ''Boreogadus saida'' ) are expected to be affected negatively by a shortened ice covered season and reduced sea ice extent through loss of spawning habitat and shelter, increased predatory pressure, reduced prey availability (Christiansen, 2017 <sup>[[#fn:r605|605]]</sup> ), and impaired growth and reproductive success (Nahrgang et al., 2014 <sup>[[#fn:r606|606]]</sup> ). These changes may cause structural changes in food webs, with large piscivorous and semipelagic boreal fish species replacing small bodied Arctic benthivores (Box 3.4; Fossheim et al., 2015 <sup>[[#fn:r607|607]]</sup> ; Frainer et al., 2017 <sup>[[#fn:r608|608]]</sup> ).

Time series on responses of anadromous fish (including salmon) in the high Arctic are limited, although these stocks will also be exposed to a wide range of future stressors (Reist et al., 2016 <sup>[[#fn:r609|609]]</sup> ). There is some evidence that environmental variability influences the production of anadromous species such as Arctic char ( ''Salvelinus alpinus'' ), brown trout ( ''Salmo trutta'' ), and Atlantic salmon ( ''Salmo salar'' ) through its influence on growth and winter survival (Jensen et al., 2018 <sup>[[#fn:r610|610]]</sup> ).

<span id="figure-3.5"></span>
<!-- START IMG -->
<!-- IMG TITLE -->
'''Figure 3.5'''

<span id="schematic-summary-of-key-drivers-that-are-causing-or-are-projected-to-cause-direct-effects-on-arctic-marine-ecosystems-section-3.2.1.2.-effects-presented-here-are-described-in-the-main-text-sections-3.2.3.1-3.2.4.1.1-3.2.4.2-3.2.4.3-with-associated-confidence-levels-and-citations.-for-mixed-effects-no-confidence-level-is-given-see-main-text-for-details-on"></span>
<!-- IMG CAPTION -->
'''Schematic summary of key drivers that are causing, or are projected to cause, direct effects on Arctic marine ecosystems (Section 3.2.1.2). Effects presented here are described in the main text (Sections 3.2.3.1; 3.2.4.1.1; 3.2.4.2; 3.2.4.3) with associated confidence levels and citations. For mixed effects, no confidence level is given (see main text for details on […]'''
<!-- IMG FILE -->
[[File:b9febfe6dc3285f5ce48ec8097e23c96 IPCC-SROCC-CH_3_5.jpg]]

Schematic summary of key drivers that are causing, or are projected to cause, direct effects on Arctic marine ecosystems (Section 3.2.1.2). Effects presented here are described in the main text (Sections 3.2.3.1; 3.2.4.1.1; 3.2.4.2; 3.2.4.3) with associated confidence levels and citations. For mixed effects, no confidence level is given (see main text for details on how multiple drivers cause interacting positive and negative effects). Projected effects are conceptual representations based on high emission scenarios (Section 3.2.1.2). The cross-sectional view of the Arctic ecosystem shows the association of key functional groups (marine mammals, birds, fish, zooplankton, phytoplankton and benthic assemblages) with Arctic marine habitats. Species depicted in the fishing net are not a comprehensive depiction of all target species.

The scope for adaptation of marine fish to a changing ocean conditions is uncertain, but knowledge is informed by previous biogeographic studies (Chernova, 2011 <sup>[[#fn:r611|611]]</sup> ; Lynghammar et al., 2013 <sup>[[#fn:r612|612]]</sup> ). The present niche partitioning between subarctic and Arctic pelagic fish species is expected to become more diffuse with potential negative impacts on cold adapted species such as polar cod (Laurel et al., 2017 <sup>[[#fn:r613|613]]</sup> ; Alabia et al., 2018 <sup>[[#fn:r614|614]]</sup> ; Logerwell et al., 2018 <sup>[[#fn:r615|615]]</sup> ) ( ''low confidence'' ). Winter ocean conditions in the high Arctic are projected to remain cold in most regions (Section 3.2.3.1), limiting the immigration of subarctic species that spawn in positive temperatures onto the high Arctic shelves (Landa et al., 2014 <sup>[[#fn:r616|616]]</sup> ). Projected increases in summer temperature may open gateways to subarctic pelagic foragers in summer, particularly in the inflow regions of the Kara and Chukchi Seas, and the shelf regions of east and west Greenland (Mueter et al., 2017 <sup>[[#fn:r617|617]]</sup> ; Joli et al., 2018 <sup>[[#fn:r618|618]]</sup> ). For example, t he pelagic capelin ( ''Mallotus villosus'' ) are capable of entering the CAO, but may be restricted in winter by availability of suitable spawning areas and lack of antifreeze proteins (Hop and Gjøsæter, 2013 <sup>[[#fn:r619|619]]</sup> ; Christiansen, 2017 <sup>[[#fn:r620|620]]</sup> ).

Regional climate scenarios, derived from down-scaled global climate scenarios, have been used to drive environmentally linked fish population models (Hermann et al., 2016 <sup>[[#fn:r621|621]]</sup> ; Holsman et al., 2016 <sup>[[#fn:r622|622]]</sup> ; Ianelli et al., 2016 <sup>[[#fn:r623|623]]</sup> ; Hermann et al., 2019 <sup>[[#fn:r624|624]]</sup> ). Hermann et al. (2019) <sup>[[#fn:r625|625]]</sup> contrasted future production of copepods and euphausiids in the eastern Bering Sea under scenarios derived from projected downscaled high spatial and temporal resolution ocean habitats under RCP4.5 and RCP8.5. Consistent with AR5, these updated scenarios project future declines in the abundance of large copepods under RCP8.5, a result that has been shown to negatively impact production of walleye pollock, Pacific cod ( ''Gadus microcephalus'' ) and arrowtooth flounder ( ''Atheresthes stomias'' ) (Sigler et al., 2017 <sup>[[#fn:r626|626]]</sup> ; Kimmel et al., 2018 <sup>[[#fn:r627|627]]</sup> ) ( ''medium confidence'' ). Hedger et al. (2013) predicts increases in Atlantic salmon abundance in northern Norway (river Alta around 70°N) with future warming ( ''low confidence'' ). Under end of century RCP8.5 projections, ocean acidification and higher ocean temperatures are expected to reduce production of Barents Sea cod (Stiasny et al., 2016 <sup>[[#fn:r629|629]]</sup> ; Koenigstein et al., 2018 <sup>[[#fn:r630|630]]</sup> ) ( ''low confidence'' ).

<!-- END IMG -->
<div id="section-3-2-3-1-arctic-block-5"></div>

<span id="seabirds-and-marine-mammals"></span>
===== 3.2.3.1.4 Seabirds and marine mammals =====

Environmental alterations caused by global warming are resulting in phenological, behavioural, physiological, and distributional changes in Arctic marine mammal and seabird populations (Gilg et al., 2012 <sup>[[#fn:r631|631]]</sup> ; Laidre et al., 2015 <sup>[[#fn:r632|632]]</sup> ; Gall et al., 2017 <sup>[[#fn:r633|633]]</sup> ) ( ''high confidence'' ). These changes include altered ecological interactions as well as direct responses to habitat degradation induced especially via loss of sea ice. Population responses to warming have not all been linear, some have been particularly strong and abrupt due to environmental regime shifts, as seen in black-legged kittiwakes ( ''Rissa tridactyla'' ). A steep population decline in kittiwake colonies distributed throughout their breeding range coincided with an abrupt warming of sea surface temperature in the 1990s, while their population dynamics did not seem to be affected during periods of more gradual warming (Descamps et al., 2017 <sup>[[#fn:r634|634]]</sup> ).

Seabirds and marine mammals are mobile animals that respond to changes in the distribution of their preferred habitats and prey, by shifting their range, altering the timing or pathways for migration or prey shifting when this is feasible (Post et al., 2013 <sup>[[#fn:r635|635]]</sup> ; Hamilton et al., 2019 <sup>[[#fn:r636|636]]</sup> ) ( ''very high confidence'' ). However, some species display strong site fidelity that can be maladaptive in a changing climate and Arctic endemic marine mammals (all of which are ice-affiliated for breeding) in general have little scope to move northward in response to warming (Kovacs et al., 2012 <sup>[[#fn:r637|637]]</sup> ; Hamilton et al., 2015 <sup>[[#fn:r638|638]]</sup> ). Changes in the location or availability of polar fronts, polynyas, tidal glacier fronts or ice edges have impacted where Arctic sea birds and marine mammals concentrate because of the influence these physical features have on productivity; traditionally these areas have been key foraging sites for top predators in the Arctic (deHart and Picco, 2015 <sup>[[#fn:r639|639]]</sup> ; Hamilton et al., 2017 <sup>[[#fn:r640|640]]</sup> ; Hunt et al., 2018 <sup>[[#fn:r641|641]]</sup> ).

In some species, shifts in distribution in response to changes in suitable habitat have been associated with increased mortality. Increased mortality rates of walrus ( ''Odobenus rosmarus)'' calves have been observed during on-shore stampedes of unusually large herds, because Pacific walrus females are no longer able to haul out on ice over the shelf in summer due to the retraction of the southern ice edge into the deep Arctic Ocean (Kovacs et al., 2016 <sup>[[#fn:r642|642]]</sup> ). Shifts in the temporal and spatial distribution and availability of suitable areas of sea ice for ice-breeding seals have occurred (Bajzak et al., 2011 <sup>[[#fn:r643|643]]</sup> ; Øigård et al., 2013 <sup>[[#fn:r6|6]]</sup> 44) with increases in strandings and pup mortality in years with little ice (Johnston et al., 2012c <sup>[[#fn:r645|645]]</sup> ; Soulen et al., 2013 <sup>[[#fn:r646|646]]</sup> ; Stenson and Hammill, 2014 <sup>[[#fn:r647|647]]</sup> ).

Climate impacts that reduce the availability of prey resources can negatively impact marine mammals (Asselin et al., 2011; Øigård et al., 2014; Choy et al., 2017) ( ''very high confidence'' ). Sea ice changes have increased the foraging effort of ringed seals ( ''Pusa hispida'' ) in the marginal ice zone north of Svalbard (Hamilton et al., 2015 <sup>[[#fn:r651|651]]</sup> ), also causing diet shifts (Lowther et al., 2017 <sup>[[#fn:r652|652]]</sup> ). Ringed seals in Svalbard are using terrestrial haul out sites during summer for the first time in observed history, following major declines in sea ice (Lydersen et al., 2017 <sup>[[#fn:r653|653]]</sup> ), an example of an adaptive behavioural response to extreme habitat changes. Sea ice related changes in the export of production to the benthos (Section 3.3.3.1) and associated changes in the benthic community (Section 3.4.1.1.2) may impact marine mammals dependent on benthic prey (e.g., walruses and gray whales, ''Eschrichtius robustus'' ) (Brower et al., 2017 <sup>[[#fn:r654|654]]</sup> ; Udevitz et al., 2017 <sup>[[#fn:r655|655]]</sup> ; Szpak et al., 2018 <sup>[[#fn:r656|656]]</sup> ).

Changes in the timing, distribution and thickness of sea ice and snow (Sections 3.2.1.1, 3.4.1.1) have been linked to phenological shifts, and changes in distribution, denning, foraging behaviour and survival rates of polar bears ( ''Ursus maritimus'' ) (Andersen et al., 2012 <sup>[[#fn:r657|657]]</sup> ; Hamilton et al., 2017 <sup>[[#fn:r658|658]]</sup> ; Escajeda et al., 2018 <sup>[[#fn:r659|659]]</sup> ) ( ''high confidence'' ). Less ice is also driving polar bears to travel over greater distances and swim more than previously both in offshore and in coastal areas, which can be particularly dangerous for young cubs (Durner et al., 2017 <sup>[[#fn:r660|660]]</sup> ; Pilfold et al., 2017 <sup>[[#fn:r661|661]]</sup> ; Lone et al., 2018 <sup>[[#fn:r662|662]]</sup> ). Cumulatively, changes in sea ice patterns are driving demographic changes in polar bears, including declines in some populations (Lunn et al., 2016 <sup>[[#fn:r663|663]]</sup> ; McCall et al., 2016 <sup>[[#fn:r664|664]]</sup> ), while others are stable or increasing (Voorhees et al., 2014 <sup>[[#fn:r665|665]]</sup> ; Aars et al., 2017 <sup>[[#fn:r666|666]]</sup> ). This is because protective management measures have been successful in allowing severely depleted populations to recover or because new food sources, such as carrion, are becoming available to polar bears in some regions (Galicia et al., 2016 <sup>[[#fn:r667|667]]</sup> ; Stapleton et al., 2016 <sup>[[#fn:r668|668]]</sup> ). Changes in the spatial distribution of polar bears and killer whales can have top-down effects on other marine mammal prey populations (Øigård et al., 2014 <sup>[[#fn:r669|669]]</sup> ; Breed et al., 2017 <sup>[[#fn:r670|670]]</sup> ; Smith et al., 2017a <sup>[[#fn:r671|671]]</sup> ).

Several studies from different parts of the Arctic show evidence that changing temperatures impact seabird diets (Dorresteijn et al., 2012 <sup>[[#fn:r672|672]]</sup> ; Divoky et al., 2015 <sup>[[#fn:r673|673]]</sup> ; Vihtakari et al., 2018 <sup>[[#fn:r674|674]]</sup> ), reproductive success and body condition (Gaston et al., 2012 <sup>[[#fn:r675|675]]</sup> ; Provencher et al., 2012 <sup>[[#fn:r676|676]]</sup> ; Gaston and Elliott, 2014 <sup>[[#fn:r677|677]]</sup> ) ( ''high confidence'' ). Recent studies also show that changes in sea surface temperature and sea ice dynamics have impacts on the distribution and abundance of seabird prey with cascading impacts on seabird community composition (Gall et al., 2017 <sup>[[#fn:r678|678]]</sup> ), nutritional stress and decreased reproductive output (Dorresteijn et al., 2012 <sup>[[#fn:r679|679]]</sup> ; Divoky et al.; Kokubun et al., 2018 <sup>[[#fn:r680|680]]</sup> ) and survival (Renner et al., 2016 <sup>[[#fn:r681|681]]</sup> ; Hunt et al., 2018 <sup>[[#fn:r682|682]]</sup> ).

<div id="section-3-2-3-2-southern-ocean"></div>

<span id="southern-ocean"></span>