Polar regions

The pan-Arctic loss of sea ice cover is a prominent indicator of climate change. Sea ice extent (the total area of the Arctic with at least 15% sea ice concentration) has declined since 1979 in each month of the year (very high confidence) (Barber et al., 2017⁴¹; Comiso et al., 2017b⁴²; Stroeve and Notz, 2018⁴³) (Figure 3.3). Changes are largest in summer and smallest in winter, with the strongest trends in September (1979–2018; summer month with the lowest sea ice cover) of –83,000 km² yr^–1 (–12.8% per decade ± 2.3% relative to 1981–2010 mean), and –41,000 km² yr^–1 (–2.7% per decade ± 0.5% relative to 1981–2010 mean) for March (1979–2019; winter month with the greatest sea ice cover) (Onarheim et al., 2018⁴⁴). Regionally, summer ice loss is dominated by reductions in the East Siberian Sea (explains 22% of the September trend), and large declines in the Beaufort, Chukchi, Laptev and Kara seas (Onarheim et al., 2018⁴⁵). Winter ice loss is dominated by reductions within the Barents Sea, responsible for 27% of the pan-Arctic March sea ice trends (Onarheim and Årthun, 2017⁴⁶). Summer Arctic sea ice loss since 1979 is unprecedented in 150 years based on historical reconstructions (Walsh et al., 2017⁴⁷) and more than 1000 years based on palaeoclimate evidence (Polyak et al., 2010⁴⁸; Kinnard et al., 2011⁴⁹; Halfar et al., 2013⁵⁰) (medium confidence).

Approximately half of the observed Arctic summer sea ice loss is driven by increased concentrations of atmospheric greenhouse gases, with the remainder attributed to internal climate variability (Kay et al., 2011⁵¹; Notz and Marotzke, 2012⁵²) (medium confidence). The sea ice albedo feedback (increased air temperature reduces sea ice cover, allowing more energy to be absorbed at the surface, fostering more melt) is a key driver of sea ice loss (Perovich and Polashenski, 2012⁵³; Stroeve et al., 2012b; Serreze et al., 2016⁵⁴) and is exacerbated by the transition from perennial to seasonal sea ice (Haine and Martin, 2017⁵⁵; see Section 3.2.1.1.2). Other drivers include increased warm, moist air intrusions into the Arctic during both winter (Box 3.1) and spring (Boisvert et al., 2016⁵⁶; Cullather et al., 2016⁵⁷; Kapsch et al., 2016⁵⁸; Mortin et al., 2016⁵⁹; Graham et al., 2017⁶⁰; Hegyi and Taylor, 2018⁶¹), radiative feedbacks associated with cloudiness and humidity (Kapsch et al., 2013⁶²; Pithan and Mauritsen, 2014⁶³; Hegyi and Deng, 2016⁶⁴; Morrison et al., 2018⁶⁵), and increased exchanges of sensible and latent heat flux from the ocean to the atmosphere (Serreze et al., 2012⁶⁶; Taylor et al., 2018⁶⁷). A lack of complete process understanding limits a more definitive differentiation between anthropogenic versus internal drivers of summer Arctic sea ice loss (Serreze et al., 2016⁶⁸; Ding et al., 2017⁶⁹; Meehl et al., 2018⁷⁰). The unabated reduction in Arctic summer sea ice since AR5 means contributions to additional global radiative forcing (Flanner et al., 2011⁷¹) have continued, with estimates of up to an additional 6.4 ± 0.9 W/m² of solar energy input to the Arctic Ocean region since 1979 (Pistone et al., 2014⁷²).

Although Arctic ice freeze-up is occurring later (Section 3.2.1.1.3), rapid thermodynamic ice growth occurs over thin ice areas after air temperatures drop below freezing in autumn. Later freeze-up also delays snowfall accumulation on sea ice, leading to a thinner and less insulating snowpack (Section 3.2.1.1.6) (Sturm and Massom, 2016⁷³). These two negative feedbacks help to mitigate sudden and irreversible loss of Arctic sea ice (Armour et al., 2011⁷⁴).

Total Antarctic sea ice cover exhibits no significant trend over the period of satellite observations (Figure 3.3; 1979–2018) (high confidence) (Ludescher et al., 2018⁷⁵). A significant positive trend in mean annual ice cover between 1979 and 2015 (Comiso et al., 2017a⁷⁶) has not persisted, due to three consecutive years of below average ice cover (2016–2018) driven by atmospheric and oceanic forcing (Turner et al., 2017b⁷⁷; Kusahara et al., 2018⁷⁸; Meehl et al., 2019⁷⁹; Wang et al., 2019⁸⁰). The overall Antarctic sea ice extent trend is composed of near-compensating regional changes, with rapid ice loss in the Amundsen and Bellingshausen seas counteracted by rapid ice gain in the Weddell and Ross seas (Holland, 2014⁸¹) (Figure 3.3). These regional trends are strongly seasonal in character (Holland, 2014⁸²); only the western Ross Sea has a trend that is statistically significant in all seasons, relative to the variance during the period of satellite observations.

Multiple factors contribute to the regionally variable nature of Antarctic sea ice extent trends (Matear et al., 2015⁸³; Hobbs et al., 2016b⁸⁴). Sea ice trends are closely related to meridional wind trends (high confidence) (Holland and Kwok, 2012⁸⁵; Haumann et al., 2014⁸⁶): poleward wind trends in the Bellingshausen Sea push sea ice closer to the coast (Holland and Kwok, 2012⁸⁷) and advect warm air to the sea ice zone (Kusahara et al., 2017⁸⁸), and the reverse is true over much of the Ross Sea. These meridional wind trends are linked to Pacific variability (Coggins and McDonald, 2015⁸⁹; Meehl et al., 2016⁹⁰; Purich et al., 2016b⁹¹). Ozone depletion may also affect meridional winds (Fogt and Zbacnik, 2014⁹²; England et al., 2016⁹³), but there is low confidence that this explains observed sea ice trends (Landrum et al., 2017⁹⁴).

Coupled climate models indicate that anthropogenic warming at the surface is delayed by the Southern Ocean circulation, which transports heat downwards into the deep ocean (Armour et al., 2016⁹⁵). This overturning circulation (Cross-Chapter Box 7 in Chapter 3), along with differing cloud and lapse rate feedbacks (Goosse et al., 2018⁹⁶), may explain the weak response of Antarctic sea ice cover to increased atmospheric greenhouse gas concentrations compared to the Arctic (medium confidence). Because Antarctic sea ice extent has remained below climatological values since 2016, there is still potential for longer-term changes to emerge in the Antarctic (Meehl et al., 2019⁹⁷), similar to the Arctic.

Historical surface observations (Murphy et al., 2014⁹⁸), reconstructions (Abram et al., 2013b⁹⁹), ship records (de la Mare, 2009; Edinburgh and Day, 2016¹⁰⁰), early satellite images (Gallaher et al., 2014¹⁰¹), and model simulations (Gagné et al., 2015¹⁰²) indicate a decrease in overall Antarctic sea ice cover since the early 1960s which is too modest to be separated from natural variability (Hobbs et al., 2016a¹⁰³) (high confidence).

The proportion of Arctic sea ice at least 5 years old declined from 30% to 2% between 1979 and 2018; over the same period first-year sea ice proportionally increased from approximately 40% to 60–70% (Stroeve and Notz, 2018¹⁰⁴) (very high confidence) (Sections 3.2.1.1.3 and 3.2.1.1.4). Arctic sea ice has thinned through volume reductions in satellite altimeter retrievals (Laxon et al., 2013¹⁰⁵; Kwok, 2018¹⁰⁶), ocean–sea ice reanalyses (Chevallier et al., 2017¹⁰⁷) and in situ measurements (Renner et al., 2014¹⁰⁸; Haas et al., 2017¹⁰⁹) (very high confidence). Data from multiple satellite altimeter missions show declines in Arctic Basin ice thickness from 2000 to 2012 of –0.58 ± 0.07 m per decade (Lindsay and Schweiger, 2015¹¹⁰). Integration of data from submarines, moorings, and earlier satellite radar altimeter missions shows ice thickness declined across the central Arctic by 65%, from 3.59 to 1.25 m between 1975 and 2012 (Lindsay and Schweiger, 2015¹¹¹). There is emerging evidence that this sea ice volume loss may be unprecedented over the past century (Schweiger et al., 2019¹¹²). New estimates of ice thickness are available for the marginal seas (up to a maximum thickness of ~1 metre) from low-frequency satellite passive microwave measurements (Kaleschke et al., 2016¹¹³; Ricker et al., 2017¹¹⁴) but data are only available since 2010. The shift to thinner seasonal sea ice contributes to further ice extent reductions through enhanced summer season melt via increased energy absorption (Nicolaus et al., 2012¹¹⁵), and it is vulnerable to fragmentation from the passage of intense Arctic cyclones in summer and increased ocean swell conditions (Zhang et al., 2013¹¹⁶; Thomson and Rogers, 2014¹¹⁷).

Surface observations of Antarctic sea ice thickness are extremely sparse (Worby et al., 2008¹¹⁸). There are no consistent long-term observations from which trends in ice volume may be derived. Calibrated model simulations suggest that ice thickness trends closely follow those of ice concentration (Massonnet et al., 2013¹¹⁹; Holland et al., 2014¹²⁰) (medium confidence). Satellite altimeter datasets of Antarctic sea ice thickness are emerging (Paul et al., 2018¹²¹) but definitive trends are not yet available.

There is high confidence that the Arctic sea ice melt season has extended by 3 days per decade since 1979 due earlier melt onset, and 7 days per decade due to later freeze-up (Stroeve and Notz, 2018¹²²). This longer melt season is consistent with the observed loss of sea ice extent and thickness (Sections 3.2.1.1.1; 3.2.1.1.2). While the melt onset trends are smaller, they play a large role in the earlier development of open water (Stroeve et al., 2012b¹²³; Serreze et al., 2016¹²⁴) and melt pond development (Perovich and Polashenski, 2012¹²⁵) which enhance the sea ice albedo feedback (Stroeve et al., 2014b¹²⁶; Liu et al., 2015a¹²⁷). Observed reductions in the duration of seasonal sea ice cover are reflected in community-based observations of decreased length of time in which activities can safely take place on sea ice (Laidler et al., 2010¹²⁸; Eisner et al., 2013¹²⁹; Fall et al., 2013¹³⁰; Ignatowski and Rosales, 2013¹³¹).

Changes in the duration of Antarctic sea ice cover over 1979–2011 largely followed the spatial pattern of sea ice extent trends with reduced ice cover duration in the Amundsen/Bellingshausen Sea region in summer and autumn owing to earlier retreat and later advance, and increases in the Ross Sea due to later ice retreat and earlier advance (Stammerjohn et al., 2012¹³²).

Winds associated with the climatological Arctic sea level pressure pattern drive the Beaufort Gyre (Dewey et al., 2018¹³³; Meneghello et al., 2018¹³⁴) and the Transpolar Drift Stream (Vihma et al., 2012¹³⁵), which retains sea ice within the central Arctic Basin, and exports sea ice out of the Fram Strait, respectively. There is high confidence that sea ice drift speeds have increased since 1979, both within the Arctic Basin and through Fram Strait (Rampal et al., 2009¹³⁶; Krumpen et al., 2019¹³⁷), attributed to thinner ice (Spreen et al., 2011¹³⁸) and changes in wind forcing (Olason and Notz, 2014¹³⁹). Fram Strait sea ice area export estimates range between 600,000 to 1 million km2 of ice annually, which represents approximately 10% of the ice within the Arctic Basin (medium confidence) (Kwok et al., 2013¹⁴⁰; Krumpen et al., 2016¹⁴¹; Smedsrud et al., 2017¹⁴²; Zamani et al., 2019¹⁴³). Sea ice volume flux estimates through Fram Strait are now available from satellite altimeter datasets (Ricker et al., 2018¹⁴⁴), but they cover too short a time period for robust trend analysis. Observations of extreme Arctic sea ice deformation is attributed to the combination of decreased ice thickness and increased ice motion (Itkin et al., 2017¹⁴⁵).

Satellite estimates of sea ice drift velocity show significant trends in Antarctic ice drift (Holland and Kwok, 2012¹⁴⁶). Increased northward drift in the Ross Sea and decreased northward drift in the Bellingshausen and Weddell seas agree with the respective ice extent gains and losses in these regions, but there is only medium confidence in these trends due to a small number of ice drift data products derived from temporally inconsistent satellite records (Haumann et al., 2016¹⁴⁷).

Immobile sea ice anchored to land or ice shelves is referred to as ‘landfast’. The few long term surface (auger hole) records of Arctic landfast sea ice thickness all exhibit thinning trends in springtime maximum sea ice thickness since the mid-1960s (high confidence): declines of 11 cm per decade in the Barents Sea (Gerland et al., 2008¹⁴⁸), 3.3 cm per decade along the Siberian Coast (Polyakov et al., 2010¹⁴⁹), and 3.5 cm per decade in the Canadian Arctic Archipelago (Howell et al., 2016¹⁵⁰). Over a shorter 1976–2007 period, winter season landfast sea ice extent from measurements across the Arctic significantly decreased at a rate of 7% per decade, with the largest decreases in the regions of Svalbard (24% per decade) and the northern coast of the Canadian Arctic Archipelago (20% per decade) (Yu et al., 2013¹⁵¹). Svalbard and the Chukchi Sea regions are experiencing the largest declines in landfast sea ice duration (~1 week per decade) since the 1970s (Yu et al., 2013¹⁵²; Mahoney et al., 2014¹⁵³). While most Arctic landfast sea ice melts completely each summer, perennial landfast ice (also termed an ‘ice-plug’) occurs in Nansen Sound and the Sverdrup Channel in the Canadian Arctic Archipelago. These ice-plugs were in place continuously from the start of observations in the early 1960s, until they disappeared during the anomalously warm summer of 1998, and they have rarely re-formed since 2005 (Pope et al., 2017¹⁵⁴). The loss of this perennial sea ice is associated with reduced landfast ice duration in the northern Canadian Arctic Archipelago (Galley et al., 2012¹⁵⁵; Yu et al., 2013¹⁵⁶) and increased inflow of multi-year ice from the Arctic Ocean into the northern Canadian Arctic Archipelago (Howell et al., 2013¹⁵⁷).

Arctic landfast ice is important to northern residents as a platform for travel, hunting, and access to offshore regions (Sections 3.4.3.3, 3.5.2.2). Reports of thinning, less stable, and less predictable landfast ice have been documented by residents of coastal communities in Alaska (Eisner et al., 2013¹⁵⁸; Fall et al., 2013¹⁵⁹; Huntington et al., 2017¹⁶⁰), the Canadian Arctic (Laidler et al., 2010¹⁶¹), and Chukotka (Inuit Circumpolar Council, 2014). The impact of changing prevailing wind forcing on local ice conditions has been specifically noted (Rosales and Chapman, 2015¹⁶³) including impacts on the landfast ice edge and polynyas (Box 3.3) (Gearheard et al., 2013¹⁶⁴). Long-term records of Antarctic landfast ice are limited in space and time (Stammerjohn and Maksym, 2016¹⁶⁵), with a high degree of regional variability in trends (Fraser et al., 2011¹⁶⁶) (low confidence).

Snow accumulation on sea ice inhibits sea ice melt through a high albedo, but the insulating properties limit sea ice growth (Sturm and Massom, 2016¹⁶⁷) and inhibits photosynthetic light (important for in- and under-ice biota) from reaching the bottom of the ice (Mundy et al., 2007¹⁶⁸). If snow on first-year ice is sufficiently thick, it can depress the ice below the sea level surface, which forms snow-ice due to surface flooding. This process is widespread in the Antarctic (Maksym and Markus, 2008¹⁶⁹) and the Atlantic Sector of the Arctic (Merkouriadi et al., 2017¹⁷⁰), and may become more common across the Arctic (with implications for sea ice ecosystems) as the ice regime shifts to thinner seasonal ice (Olsen et al., 2017¹⁷¹; Granskog et al., 2018¹⁷²) (medium confidence).

Despite the importance of snow on sea ice (Webster et al., 2018¹⁷³), surface or satellite derived observations of snowfall over sea ice, and snow depth on sea ice are lacking (Webster et al., 2014¹⁷⁴). The primary source of snow depth on Arctic sea ice are based on observations collected decades ago (Warren et al., 1999¹⁷⁵) the utility of which are impacted by the rapid loss of multi-year ice across the central Arctic (Stroeve and Notz, 2018¹⁷⁶), and large interannual variability in snow depth on sea ice (Webster et al., 2014¹⁷⁷). Airborne radar retrievals of snow depth on sea ice provide more recent estimates, but spatial and temporal sampling is highly discontinuous (Kurtz and Farrell, 2011¹⁷⁸). Multi-source time series provide evidence of declining snow depth on Arctic sea ice (Webster et al., 2014¹⁷⁹) consistent with estimates of higher fractions of liquid precipitation since 2000 (Boisvert et al., 2018¹⁸⁰) but there is low confidence because surface measurements for validation are extremely limited and suggest a high degree of regional variability (Haas et al., 2017¹⁸¹; Rösel et al., 2018¹⁸²). 

Although there are regional estimates of snow depth on Antarctic sea ice from satellite (Kern and Ozsoy-Çiçek, 2016¹⁸³), airborne remote sensing (Kwok and Maksym, 2014¹⁸⁴), field measurements (Massom et al., 2001¹⁸⁵) and ship-based observations (Worby et al., 2008¹⁸⁶), data are not sufficient in time nor space to assess changes in snow accumulation on Antarctic sea ice.

Ocean temperatures and associated heat fluxes have a primary influence on sea ice (e.g., Carmack et al., 2015; Steele and Dickinson, 2016²¹⁸). WGI AR5 (their Section 3.2.2) reported that Canada Basin surface waters warmed from 1993 to 2007, and observations over 1950–2010 show the Arctic Ocean water of Atlantic origin (i.e., the Atlantic Water Layer) warming starting in the 1970s. Warming trends have continued: August trends for 1982–2017 reveal summer mixed layer temperatures increasing at about 0.5°C per decade over large sectors of the Arctic basin that are ice-free in summer (Timmermans et al., 2017²¹⁹) (Figure 3.3). This is primarily the result of increased absorption of solar radiation accompanying sea ice loss (Perovich, 2016²²⁰). Between 1979 and 2011, the decrease in Arctic Ocean albedo corresponded to more solar energy input to the ocean (virtually certain) of approximately 6.4 ± 0.9 Wm^-2 (Pistone et al., 2014²²¹), likely reducing the growth of sea ice by up to 25% in both Eurasian and Canadian basins (Timmermans, 2015²²²; Ivanov et al., 2016²²³) (Section 3.2.1.1).

While Atlantic Water Layer temperatures appear to show less variability since 2008, total heat content in this layer continues to increase (Polyakov et al., 2017²²⁴). Recent changes have been dubbed the ‘Atlantification’ of the Northern Barents Sea and Eurasian Basin (Arthun et al., 2012²²⁵; Lind et al., 2018²²⁶), characterised by weaker stratification and enhanced Atlantic Water Layer heat fluxes further northeast (medium confidence). Polyakov et al. (2017)²²⁷ estimate 2–4 times larger heat fluxes in 2014–2015 compared with 2007–2008. In the Canadian Basin, the maximum temperature of the Pacific Water Layer increased by ~0.5°C between 2009 and 2013 (Timmermans et al., 2014²²⁸), with a doubling in integrated heat content over 1987–2017 (Timmermans et al., 2018²²⁹). Over 2001–2014, heat transport associated with Bering Strait inflow increased by 60%, from around 10 TW in 2001 to 16 TW in 2014, due to increases in both volume flux and temperature (Woodgate et al., 2015²³⁰; Woodgate, 2018²³¹) (low confidence).

The Southern Ocean is important for the transfer of heat from the atmosphere to the global ocean, including heat from anthropogenic warming (Frölicher et al., 2015²³²; Shi et al., 2018²³³). The Southern Ocean accounted for ~75% of the global ocean uptake of excess heat during 1870–1995 (Figure SM3.2; Frölicher et al., 2015²³⁴), of which ~43% resided in the Southern Ocean with the remainder redistributed to lower latitudes. Over 1970–2017, observations show that the upper 2000 m of the ocean south of 30°S was responsible for 35–43% of the increase in global ocean heat content (Table 3.1). Both models and observations show that, relative to its size (Table SM3.1), the Southern Ocean is disproportionately important in the increase in global upper ocean heat content (high confidence). Multi-decadal warming of the Southern Ocean has been attributed to anthropogenic factors, especially the role of greenhouse gases but also ozone depletion (Armour et al., 2016²³⁵; Shi et al., 2018²³⁶; Swart et al., 2018²³⁷; Irving et al., 2019²³⁸) (medium confidence).

Salinity is the dominant determinant of polar ocean density, and exerts major controls on stratification, circulation and mixing. Salinity changes are induced by freshwater runoff to the ocean (rivers and land ice), net precipitation, sea ice, and advection of mid-latitude waters, with the potential to impact water mass formation and circulation (e.g., Thornalley et al., 2018; see also Section 6.7.1).

Updating WGI AR5 (their Section 3.3.3.3), recent Arctic-wide estimates yield a freshwater increase (relative to salinity of 34.8 on the Practical Salinity Scale, used throughout this chapter) of 600 ± 300 km3 yr–1 over 1992–2012, with about two-thirds concomitant with decreasing salinity, and the remainder with a thickening of the freshwater layer (medium confidence) (Rabe et al., 2014²⁵⁸; Haine et al., 2015²⁵⁹; Carmack et al., 2016²⁶⁰). The Beaufort Gyre region has increased its freshwater by ~40% (6600 km3) over 2003–2017; this, and the Gyre’s strengthening, have been attributed to dominance of clockwise wind patterns over the Canadian Basin over 1997–2016 and freshwater accumulation from sea ice-melt (Krishfield et al., 2014²⁶¹; Proshutinsky et al., 2015²⁶²). Freshwater decreases in the East Siberian, Laptev, Chukchi and Kara seas are estimated to be ~180 km3 over 2003–2014 (Armitage et al., 2016²⁶³). During the 2000s, freshwater content in the upper 100 m of the northern Barents Sea declined by about 32%, from a mean of ~2.5 m (relative to a salinity of 35) in 1970–1999, to 1.7 m in 2010–2016 (Lind et al., 2018²⁶⁴). An increasing trend of 30 ± 20 km3yr–1 in freshwater flux through Bering Strait, primarily due to increased volume flux, was measured from 1991 to 2015, with record maximum freshwater influx in 2014 of around 3500 km3 in that year (Woodgate, 2018²⁶⁵). Freshwater fluxes from rivers are also increasing (Section 3.4.1.2.2), and there have been observed increases in discharge of glacial ice from Greenland (Section 3.3.1.3).

Observed Southern Ocean freshening trends are consistent with WGI AR5; subsequent studies have increased confidence in their magnitude and sign, and also attributed them to anthropogenic influences (Swart et al., 2018²⁶⁶). Changes over 1950–2010 show persistent surface water freshening over the whole Southern Ocean, with subducted mode/intermediate waters carrying trends of 0.0002–0.0008 yr–1 to below 1500 m (Skliris et al., 2014²⁶⁷), whilst de Lavergne et al. (2014) observe a circumpolar freshening south of the ACC of 0.0011 ± 0.0004 yr–1 in the upper 100 m since the 1960s (medium confidence). This intensifies over the Antarctic continental shelves (except along the Western Antarctic Peninsula), where freshening of up to 0.01 yr–1 is observed (Schmidtko et al., 2014²⁶⁹). Freshening may be driven by increases in precipitation, but while models (Pauling et al., 2016²⁷⁰) and observations suggest an increase may have occurred over the last 60 years, uncertainty is presently too high to quantify its net impact (Skliris et al., 2014²⁷¹). Recently, there has been increased recognition of the importance of sea ice in driving Southern Ocean salinity changes, with Haumann et al. (2016)²⁷² demonstrating that wind driven sea ice export has increased by 20 ± 10% from 1982 to 2008, and that this may have driven freshening of 0.002 ± 0.001 yr–1 in the surface and intermediate waters. Separately, the central role of sea ice in driving water mass transformations in the Southern Ocean has been highlighted (Abernathey et al., 2016²⁷³; Pellichero et al., 2018²⁷⁴; Swart et al., 2018²⁷⁵), hence such changes have the potential to affect overturning circulation (Cross-Chapter Box 7 in Chapter 3). Freshwater input to the ocean from the Antarctic Ice Sheet also has the potential to affect the properties and circulation of Southern Ocean water masses; see Section 3.3.3.

See Supplementary Material (SM3.2.2).

Various elements of marine biogeochemistry and geochemistry in the polar regions are of global importance. Here we focus on aspects relevant to carbon and ocean acidification; others (e.g., changes in dissolved oxygen) are assessed in Section 5.2.2. Compiled datasets on observed trends in ocean acidification from different observational platforms can be found in Table SM5.3.

About a quarter of carbon dioxide (CO2) released by human activities is taken up by the ocean (WGI AR5, their Section 3.8). This dissolves in surface water to form carbonic acid, which, upon dissociation, causes a decrease in pH (acidification) and carbonate ion (CO32–) concentration. This can affect organisms that form shells and skeletons using calcium carbonate (CaCO3, aragonite and calcite as dominant mineral forms). Since AR5, new observations have demonstrated the spatial and temporal variability of ocean acidification and controlling mechanisms of carbon systems in different regions (Bellerby et al., 2018²⁷⁶). 

Robbins et al. (2013)²⁷⁷ showed aragonite undersaturation for about 20% of surface waters in the Canada and Makarov Basins, where substantial sea ice melt occurred. Qi et al. (2017)²⁷⁸ reported that aragonite undersaturation has expanded northward by at least 5° of latitude, and deepened by ~100 m between the 1990s and 2010 primarily due to increased Pacific Winter Water transport. In the East Siberian Arctic Shelf, extreme aragonite undersaturation was driven by the degradation of terrestrial organic matter and runoff of Arctic river water with elevated CO2 concentrations, reflecting pH changes in excess of those projected in this region for 2100 (Semiletov et al., 2016²⁷⁹) (high confidence); this was also observed along the continental margin and traced in the deep Makarov and Canada Basins (Anderson et al., 2017a²⁸⁰). The variable buffering capacities of rivers flowing through watersheds with different bedrock geology also influenced the state of ocean acidification in coastal regions (Tank et al., 2012²⁸¹; Azetsu-Scott et al., 2014²⁸²).

The dissolved inorganic carbon (DIC) concentration increased in subsurface waters (150–1400 m) in the central Arctic between 1991 and 2011 (Ericson et al., 2014²⁸³). The rate of increase was 0.6–0.9 µmol kg–1 yr–1 in the Arctic Atlantic Water and 0.4–0.6 µmol kg–1 yr–1 in the upper Polar Deep Water due to anthropogenic CO2, while no trend was observed in nutrient concentrations. In waters below 2000 m, no significant trend was observed for DIC and nutrient concentrations. Observation-based estimates (MacGilchrist et al., 2014²⁸⁴) revealed a net summertime pan-Arctic export of 231 ± 49 TgC yr–1 of DIC across the Arctic Ocean gateways to the North Atlantic; at least 166 ± 60 TgC yr–1 of this was sequestered from the atmosphere (medium confidence). Similar to other regions (Table SM5.3), observed changes in the carbonate chemistry of the Arctic are indicative of ongoing ocean acidification (high confidence).

Studies covering seasonal-to-decadal variability in the Arctic are limited, with most conducted in ice-free or low ice periods during summer to autumn. However, it has been demonstrated that biological processes, respiration and photosynthesis, control the CaCO3 saturation states in Chukchi Sea bottom water (Yamamoto-Kawai et al., 2016²⁸⁵). Sea ice formation and melt influence the dynamics of ikaite (CaCO3 precipitation trapped in sea ice during brine rejection), and therefore local carbonate chemistry (Rysgaard et al., 2013²⁸⁶; Bates et al., 2014²⁸⁷; Geilfus et al., 2016²⁸⁸; Fransson et al., 2017²⁸⁹). Although the increase of pH and saturation states by biological carbon fixation that consumes DIC in surface water is well documented (Azetsu-Scott et al., 2014²⁹⁰; Yamamoto-Kawai et al., 2016²⁹¹) (high confidence), it has been shown that long photoperiods in Arctic summers sustain high pH in kelp forests, slowing ocean acidification (Krause-Jensen et al., 2016²⁹²).

Since AR5, there are new constraints on the seasonal-to-decadal variability in the Southern Ocean CO2 flux (McNeil and Matear, 2013²⁹³; Landschützer et al., 2014²⁹⁴; Landschützer et al., 2015²⁹⁵; Gregor et al., 2017²⁹⁶; Ritter et al., 2017²⁹⁷; Keppler and Landschutzer, 2019²⁹⁸) (Figure SM3.4), with mean annual flux anomalies varying from 0.3 ± 0.1 Pg C yr–1 in 2001–2002 to –0.4 Pg C yr–1 in 2012 (Landschützer et al., 2015²⁹⁹); this can affect the magnitude of the global CO2 sink (Section 5.2.2). A weakening CO2 sink during the 1990s (Le Quéré et al., 2007) reversed in the 2000s as part of a decadal cycle (Landschützer et al., 2015³⁰⁰; Munro et al., 2015³⁺¹; Williams et al., 2017³⁰²) (SM3.2.3; Figure SM3.4), with a weakening again since 2011 (Keppler and Landschutzer, 2019³⁰³). While the weakening sink during the 1990s was explained as a response to changes in the circumpolar winds over the Southern Ocean enhancing the outgassing of natural CO2, the subsequent changes appear due to a combination of changes in regional winds, temperature and circulation (Landschützer et al., 2015³⁰⁴; Gregor et al., 2017³⁰⁵; Keppler and Landschutzer, 2019³⁰⁶). Data scarcity, especially in winter, remains a challenge (Ritter et al., 2017³⁰⁷; Fay et al., 2018³⁰⁸; Gruber et al., 2019b³⁰⁹); recent data from pH-enabled floats highlighted the potential role for winter outgassing south of the Polar Front (Williams et al., 2017³¹⁰; Gray et al., 2018³¹¹). Overall, there is medium confidence that the Southern Ocean CO2 sink has experienced significant decadal variations since the 1980s. 

Southern Ocean carbon storage is affected by changes in overturning circulation (Cross-Chapter Box 7 in Chapter 3), with the storage of anthropogenic and natural carbon being both variable and out of phase on decadal timescales (DeVries et al., 2017³¹²; Tanhua et al., 2017³¹³) (Table SM3.4). Mode and intermediate waters are strongly involved in changing storage, also showing high sensitivity to shifts in winds (Swart et al., 2014³¹⁴; Swart et al., 2015a³¹⁵; Tanhua et al., 2017³¹⁶; Gruber et al., 2019a³¹⁷). Zonal basin differences in the uptake and storage of anthropogenic carbon are not well resolved and there is weak agreement between reanalysis products and Coupled Model Intercomparison Project Phase 5 (CMIP5) models (Swart et al., 2014³¹⁸). The presence of subduction hotspots suggest that basin-wide studies may be underestimating the importance of mode water subduction as a principal storage mechanism (Langlais et al., 2017³¹⁹). 

Strengthening impacts of Southern Ocean acidification are illustrated by the 3.9 ± 1.3% decrease in derived calcification rates (1998–2014) (Freeman and Lovenduski, 2015³²⁰). These have strong regional character, with decreases in the Indian and Pacific sectors (7.5–11.6%) and increases in the Atlantic (14.3 ± 5.1%). There have also been changes in the seasonality of pCO2 linked to decreasing buffer capacity (McNeil and Sasse, 2016³²¹) (SM3.2.4) or adjustments to primary production (Conrad and Lovenduski, 2015³²²); seasonal changes are discussed further in Section 5.2.2.

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., 2018⁵²⁷) (high confidence). Satellite data show that the decline in ice cover has resulted in a >30% increase in annual net primary production (NPP) in ice-free Arctic waters since 1998 (Arrigo and van Dijken, 2011⁵²⁸; Bélanger et al., 2013⁵²⁹; Arrigo and van Dijken, 2015⁵³⁰; Kahru et al., 2016⁵³¹), a phenomenon corroborated by both in situ data (Stanley et al., 2015⁵³²) and modelling studies (Vancoppenolle et al., 2013⁵³³; Jin et al., 2016⁵³⁴). Ice loss has also resulted in earlier phytoplankton blooms (Kahru et al., 2011⁵³⁵) with blooms being dominated by larger-celled phytoplankton (Fujiwara et al., 2016⁵³⁶). 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., 2017⁵³⁷).

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 km2) 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., 2012⁵³⁸). Evidence shows that these blooms can thrive beneath sea ice in areas of reduced thickness, increased coverage of melt ponds (Arrigo et al., 2014⁵³⁹; Zhang et al., 2015⁵⁴⁰; Jin et al., 2016⁵⁴¹; Horvat et al., 2017⁵⁴²), first-year ridges at the snow-ice interface (Fernández-Méndez et al., 2018⁵⁴³), and a large number of cracks (high lead fractions) in the ice (Assmy et al., 2017⁵⁴⁴), although the latter has not changed significantly in the last three decades (Wang et al., 2016a⁵⁴⁵). 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., 2017⁵⁴⁶).

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, 2015⁵⁴⁷) and a partial ice cover (Schulze and Pickart, 2012⁵⁴⁸), leading to more new production (Williams and Carmack, 2015⁵⁴⁹). 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., 2012⁵⁵⁰; Nummelin et al., 2016⁵⁵¹) may reduce Arctic NPP in the future, especially in the central basin (Ardyna et al., 2017⁵⁵²). 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., 2015⁵⁵³). 

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., 2016⁵⁵⁴), 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., 2015⁵⁵⁵).

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., 2013⁵⁵⁶; Lalande et al., 2014⁵⁵⁷; Mäkelä et al., 2017⁵⁵⁸) 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, 2016⁵⁵⁹) (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., 2018⁵⁶⁰).

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., 2015⁵⁶¹). 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., 2016⁵⁶²).

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., 2017⁵⁶³; Kimmel et al., 2018⁵⁶⁴). 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., 2015⁵⁶⁵) and in recent years also the Beaufort Sea (Smoot and Hopcroft, 2017⁵⁶⁶), while in the Central Basins zooplankton biomass in general has increased (Hunt et al., 2014⁵⁶⁷; Rutzen and Hopcroft, 2018⁵⁶⁸) (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., 2018⁵⁶⁹). 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., 2016⁵⁷¹). Renaud et al. (2018)⁵⁷² 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., 2017⁵⁷³; Thor et al., 2018⁵⁷⁴). Although there is strong evidence that pteropods are sensitive to the effects of ocean acidification (Manno et al., 2017⁵⁷⁵) recent studies indicate they may exhibit some ability to adapt (Peck et al., 2016⁵⁷⁶; Peck et al., 2018⁵⁷⁷). However, the metabolic costs of adaptation may be constraining, especially during periods of low food availability (Lischka and Riebesell, 2016⁵⁷⁸).

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., 2013⁵⁷⁹) (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., 2015⁵⁸⁰). 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., 2017⁵⁸¹), 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., 2012⁵⁸²; Kortsch et al., 2015)⁵⁸³. Also, there are indications of declining benthic biomass in the northern Bering Sea (Grebmeier and Cooper, 2016⁵⁸⁴) and southern Chukchi Sea (Grebmeier et al., 2015⁵⁸⁵). It is unclear whether these rapid ecosystem changes will be tipping points for local ecosystems (Chapter 6, Table 6.1; Wassmann and Lenton, 2012⁵⁸⁶). 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., 2012⁵⁸⁷; Paar et al., 2016⁵⁸⁸) (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., 2013⁵⁸⁹; Long et al., 2015⁵⁹⁰; Ryer et al., 2016⁵⁹¹). In Newfoundland and Labrador waters and on the western Scotian Shelf, snow crab (Chionoecetes opilio) productivity has declined (Mullowney et al., 2014⁵⁹²; Zisserson and Cook, 2017⁵⁹³). Contrary to this, snow crabs have expanded their distribution in the Barents Sea and commercial harvesting increased (Hansen, 2016⁵⁹⁴; Lorentzen et al., 2018⁵⁹⁵) (high confidence). 

Bering sea crabs exhibit species-specific sensitivities to reduced pH (Long et al., 2017⁵⁹⁶; Swiney et al., 2017⁵⁹⁷; Long et al., 2019⁵⁹⁸). However, current pH levels do not appear to have negatively impacted crab production in the Bering or Barents Seas (Mathis et al., 2015⁵⁹⁹; Punt et al., 2016⁶⁰⁰).

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., 2015⁶⁰¹; Dalpadado et al., 2016⁶⁰²; Hunt et al., 2016⁶⁰³; 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., 2014⁶⁰⁴). 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, 2017⁶⁰⁵), and impaired growth and reproductive success (Nahrgang et al., 2014⁶⁰⁶). 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., 2015⁶⁰⁷; Frainer et al., 2017⁶⁰⁸).

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., 2016⁶⁰⁹). 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., 2018⁶¹⁰).

The scope for adaptation of marine fish to a changing ocean conditions is uncertain, but knowledge is informed by previous biogeographic studies (Chernova, 2011⁶¹¹; Lynghammar et al., 2013⁶¹²). 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., 2017⁶¹³; Alabia et al., 2018⁶¹⁴; Logerwell et al., 2018⁶¹⁵) (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., 2014⁶¹⁶). 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., 2017⁶¹⁷; Joli et al., 2018⁶¹⁸). For example, the 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, 2013⁶¹⁹; Christiansen, 2017⁶²⁰). 

Regional climate scenarios, derived from down-scaled global climate scenarios, have been used to drive environmentally linked fish population models (Hermann et al., 2016⁶²¹; Holsman et al., 2016⁶²²; Ianelli et al., 2016⁶²³; Hermann et al., 2019⁶²⁴). Hermann et al. (2019)⁶²⁵ 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., 2017⁶²⁶; Kimmel et al., 2018⁶²⁷) (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., 2016⁶²⁹; Koenigstein et al., 2018⁶³⁰) (low confidence).

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., 2012⁶³¹; Laidre et al., 2015⁶³²; Gall et al., 2017⁶³³) (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., 2017⁶³⁴).

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., 2013⁶³⁵; Hamilton et al., 2019⁶³⁶) (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., 2012⁶³⁷; Hamilton et al., 2015⁶³⁸). 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, 2015⁶³⁹; Hamilton et al., 2017⁶⁴⁰; Hunt et al., 2018⁶⁴¹).

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., 2016⁶⁴²). Shifts in the temporal and spatial distribution and availability of suitable areas of sea ice for ice-breeding seals have occurred (Bajzak et al., 2011⁶⁴³; Øigård et al., 2013⁶44) with increases in strandings and pup mortality in years with little ice (Johnston et al., 2012c⁶⁴⁵; Soulen et al., 2013⁶⁴⁶; Stenson and Hammill, 2014⁶⁴⁷). 

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., 2015⁶⁵¹), also causing diet shifts (Lowther et al., 2017⁶⁵²). 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., 2017⁶⁵³), 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., 2017⁶⁵⁴; Udevitz et al., 2017⁶⁵⁵; Szpak et al., 2018⁶⁵⁶). 

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., 2012⁶⁵⁷; Hamilton et al., 2017⁶⁵⁸; Escajeda et al., 2018⁶⁵⁹) (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., 2017⁶⁶⁰; Pilfold et al., 2017⁶⁶¹; Lone et al., 2018⁶⁶²). Cumulatively, changes in sea ice patterns are driving demographic changes in polar bears, including declines in some populations (Lunn et al., 2016⁶⁶³; McCall et al., 2016⁶⁶⁴), while others are stable or increasing (Voorhees et al., 2014⁶⁶⁵; Aars et al., 2017⁶⁶⁶). 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., 2016⁶⁶⁷; Stapleton et al., 2016⁶⁶⁸). 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., 2014⁶⁶⁹; Breed et al., 2017⁶⁷⁰; Smith et al., 2017a⁶⁷¹). 

Several studies from different parts of the Arctic show evidence that changing temperatures impact seabird diets (Dorresteijn et al., 2012⁶⁷²; Divoky et al., 2015⁶⁷³; Vihtakari et al., 2018⁶⁷⁴), reproductive success and body condition (Gaston et al., 2012⁶⁷⁵; Provencher et al., 2012⁶⁷⁶; Gaston and Elliott, 2014⁶⁷⁷) (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., 2017⁶⁷⁸), nutritional stress and decreased reproductive output (Dorresteijn et al., 2012⁶⁷⁹; Divoky et al.; Kokubun et al., 2018⁶⁸⁰) and survival (Renner et al., 2016⁶⁸¹; Hunt et al., 2018⁶⁸²).

Changes in column-integrated phytoplankton biomass for the Southern Ocean are coupled with changes in the spatial extent of ice-free waters, suggesting little overall change in biomass per area at the circumpolar scale (Behrenfeld et al., 2016⁶⁹⁸). Arrigo et al. (2008)⁶⁹⁹ also report no overall trend in remotely-sensed column-integrated primary production south of 50°S from 1998 to 2006. At a regional scale, local-scale forcings (e.g., retreating glaciers, topographically steered circulation and sea ice duration) and associated changes in stratification are key determinants of phytoplankton bloom dynamics at coastal stations on the West Antarctic Peninsula (Venables et al., 2013⁷⁰⁰; Schofield et al., 2017⁷⁰¹; Kim et al., 2018⁷⁰²; Schofield et al., 2018⁷⁰³) (medium confidence). For example, a shallowing trend in mixed layer depth in the southern part of the Peninsula (as opposed to no trend in the north) associated with changes in sea ice duration over a 24-year period (from 1993 to 2017) has been linked to enhanced phytoplankton productivity (Schofield et al., 2018⁷⁰⁴). The phenology of Southern Ocean phytoplankton blooms in this region may also be shifting to earlier in the growth season (Arrigo et al., 2017a⁷⁰⁵). However, the effect of climate change on Southern Ocean pelagic primary production is difficult to determine given that the length of time series data is insufficient (less than 30 years) to enable the climate change signature to be detected and attributed; and that, even when records are of sufficient length, data trends are often reported as being driven by climate change when they are due to a combination of climate change and variability.

Recent studies on the ecological effects of acidification in coastal waters near the Antarctic continent indicate a detrimental effect of acidification on primary production and changes to the structure and function of microbial communities (Hancock et al., 2017⁷⁰⁶; Deppeler et al., 2018⁷⁰⁷; Westwood et al., 2018⁷⁰⁸) (medium confidence). Trimborn et al. (2017) report that Southern Ocean diatoms are more sensitive to ocean acidification and changes in irradiance than the prymnesiophyte Phaeocystis antarctica, which may have implications for biogeochemical cycling because diatoms and prymnesiophytes are generally considered key drivers of these cycles. Both laboratory manipulations and in situ experiments indicate that sea ice algae are tolerant to acidification (McMinn, 2017⁷⁰⁹) (medium confidence). Model projections of trends in primary production in the Southern Ocean due to climate change from Leung et al. (2015)⁷¹⁰ are summarised in Table 3.2.

Carbon uptake and storage by Antarctic benthic communities is predicted to increase with sea ice losses, because across-shelf growth gains from longer algal blooms outweigh ice scour mortality in the shallows (Barnes, 2017⁷¹⁶). Bentho-pelagic coupling and vertical energy flux will also influence Southern Ocean ecosystem responses to climate change (Jansen et al., 2017⁷¹⁷). Benthic communities in shallow water habitats mostly consist of dark-adapted invertebrates and rely on sea ice to create low-light marine environments. Increases in the amount of light reaching the shallow seabed under climate change may result in ecological regime shifts, in which invertebrate-dominated communities are replaced by macroalgal beds (Clark et al., 2015⁷¹⁸; Clark et al., 2017⁷¹⁹) (low confidence) (Table 6.1). Griffiths et al. (2017a)⁷²⁰ modelled distribution changes for 963 benthic invertebrate species in the Southern Ocean under RCP8.5 for 2099. Their results suggest that 79% of Antarctica’s endemic species will face a reduction in suitable temperature habitat (an average 12% reduction) over the current century. Predicted reductions in the number of species are most pronounced for the West Antarctic Peninsula and the Scotia Sea region (Griffiths et al., 2017a⁷²⁰).

Many Antarctic fish have a narrow thermal tolerance as a result of physiological adaptations to cold water (Pörtner et al., 2014⁷²¹; Mintenbeck, 2017⁷²²), which makes them vulnerable to the effects of increasing temperatures (Mueller et al., 2012⁷²³; Beers and Jayasundara, 2015⁷²⁴). Increasing water temperatures may displace icefish (family Channichthyidae) in marginal habitats (e.g., shallow regions around subantarctic islands) as they lack haemoglobin and are unable to adjust blood parameters to an increasing oxygen demand (Mintenbeck et al., 201250‹) (low confidence). Future warming may also reduce the planktonic duration and increase egg and larval mortality for fish species, which is predicted to affect dispersal patterns, with implications for population connectivity and the ability of fish species to adapt to ongoing environmental change (Young et al., 2018⁷⁵¹). The Antarctic silverfish (Pleuragramma antarctica) is an important prey species in some regions of the Southern Ocean, and has an ice-dependent life cycle (Mintenbeck et al., 2012⁷⁵²; Vacchi et al., 2012⁷⁵³). Documented declines in the abundance of this species in some parts of the West Antarctic Peninsula may have consequences for associated food webs (Parker et al., 2015⁷⁵⁴; Mintenbeck and Torres, 2017⁷⁵⁵) (low confidence).

Myctophids and toothfish are important fish groups from both a food web (myctophids) and fishery (toothfish) perspective. Species distribution models for Electrona antarctica, a dominant myctophid species in the Southern Ocean, project habitat loss for this species under RCP4.5 (6.2 ± 6.0% loss) and RCP8.5 (13.1 ± 10.2% loss) by 2090, associated with increased sea surface temperature (Freer et al., 2018⁷⁵⁶). There have been no observed effects of climate change on the two species of toothfish that are found in the Southern Ocean: Patagonian and Antarctic toothfish (Dissostichus eleginoides and D. mawsoni), but recruitment is inversely correlated with sea surface temperature for Patagonian toothfish at South Georgia (Belchier and Collins, 2008⁷⁵⁷). Given differences in temperature tolerances for Patagonian toothfish (with a wide temperature tolerance) and Antarctic toothfish (limited by a low tolerance for water temperatures above 2°C), the latter may be faced with reduced habitat and potential competition with southward-moving Patagonian toothfish under climate change (Mintenbeck, 2017⁷⁵⁸) (very low confidence).

Since AR5, there has been an increasing body of evidence of climate-induced changes in populations of some Antarctic higher predators such as seabirds and marine mammals. These changes vary between different regions of the Southern Ocean and reflect differences in key drivers (Bost et al., 2009⁷⁵⁹; Gutt et al., 2015⁷⁶⁰; Constable et al., 2016⁷⁶¹; Hunt et al., 2016⁷⁶²; Gutt et al., 2018⁷⁶³), particularly sea ice extent and food availability (high confidence) across regions (Sections 3.2.1.1.1, 5.2.3.1, 5.2.3.2, 5.2.4). The predictability of foraging grounds and ice cover are associated with variations in climate (Dugger et al., 2014⁷⁶⁴; Youngflesh et al., 2017⁷⁶⁵; Abrahms et al., 2018⁷⁶⁶) (Section 3.2.1.1) and are the main drivers of observed population changes of Southern Ocean higher predators (high confidence) (Descamps et al., 2015⁷⁶⁷; Jenouvrier et al., 2015⁷⁶⁸; Sydeman et al., 2015⁷⁶⁹; Abadi et al., 2017⁷⁷⁰; Bjorndal et al., 2017⁷⁷¹; Fluhr et al., 2017⁷⁷²; Hinke et al., 2017a⁷⁷³; Hinke et al., 2017b⁷⁷⁴; Pardo et al., 2017⁷⁷⁵). The suitability of breeding habitats and the location of environmental features that facilitate the aggregation of prey are also influenced by climate change, and in turn influence the distribution in space and time of marine mammals and birds (Bost et al., 2015⁷⁷⁶; Kavanaugh et al., 2015⁷⁷⁷; Hindell et al., 2016⁷⁷⁸; Santora et al., 2017⁷⁷⁹) (medium confidence). Finally, biological parameters (reproductive success, mortality, fecundity and body condition), life history traits, morphological, physiological and behavioural characteristics of top predators in the Southern Ocean, as well as their patterns of activity (migration, distribution, foraging and reproduction) are also changing as a result of climate change (Braithwaite et al., 2015a⁷⁸⁰; Whitehead et al., 2015⁷⁸¹; Seyboth et al., 2016⁷⁸²; Hinke et al., 2017b⁷⁸³) (high confidence).

Trends of populations of Antarctic penguins affected by climate change include both increases for gentoo penguins, (Pygoscelis papua) (Lynch et al., 2013⁷⁸⁴; Dunn et al., 2016⁷⁸⁵; Hinke et al., 2017a⁷⁸⁶), and decreases for Adélie (P. adeliae), chinstrap (P. antarctica), king (Aptenodytes patagonicus) and Emperor (A. forsteri) penguins (Trivelpiece et al., 2011⁷⁸⁷; LaRue et al., 2013⁷⁸⁸; Jenouvrier et al., 2014⁷⁸⁹; Bost et al., 2015⁷⁹⁰; Southwell et al., 2015⁷⁹¹; Younger et al., 2015⁷⁹²; Cimino et al., 2016⁷⁹³) (high confidence). Yet population shifts in Adélie penguins (Youngflesh et al., 2017⁷⁹⁴) may have resulted from strong interannual environmental variability in good and bad years for prey and breeding habitat rather than climate change (low confidence). New evidence suggests that present Emperor penguin population estimates should be evaluated with caution based on the existence of breeding colonies yet to be discovered/confirmed (Ancel et al., 2017⁷⁹⁵) as well as studies that draw conclusions based on trend estimates from single colonies (Kooyman and Ponganis, 2017⁷⁹⁶).

Evidence for climate change impacts on Antarctic flying birds indicates that contraction of sea ice (seasonally and in specific regions), increases in sea surface temperatures, extreme events (snowstorms) and wind regime shifts can reduce breeding success and population growth rates in some species: southern fulmars (Fulmarus glacialoides), Antarctic petrels (Thalassoica antarctica) and black-browed albatrosses (Thalassarche melanophris) (Descamps et al., 2015⁷⁹⁷; Jenouvrier et al., 2015⁷⁹⁸; Pardo et al., 2017⁷⁹⁹) (low confidence). Poleward population shifts with increased intensity and frequency of westerly winds affect functional traits, demographic rates, foraging range, rates of travel and flight speeds of flying birds (Weimerskirch et al., 2012⁸⁰⁰; Jenouvrier et al., 2018⁸⁰¹) but also increase overlap with fisheries activities thus increasing the risk of bycatch and the need for mitigation measures (Krüger et al., 2018⁸⁰²) (medium confidence).

Changes in local- and regional-scale oceanographic features (Section 3.2.1.2) together with bathymetry control prey aggregation and distribution, and affect the ecological responses and biological traits of higher predators (particularly marine mammals) in the Southern Ocean (Lyver et al., 2014⁸⁰³; Bost et al., 2015⁸⁰⁴; Jenouvrier et al., 2015⁸⁰⁵; Whitehead et al., 2015⁸⁰⁶; Cimino et al., 2016⁸⁰⁷; Hinke et al., 2017a⁸⁰⁸; Pardo et al., 2017⁸⁰⁹) (medium confidence) and likely explain most of the observed population shifts (Kavanaugh et al., 2015⁸¹⁰; Hindell et al., 2016⁸¹¹; Gurarie et al., 2017⁸¹²; Santora et al., 2017⁸¹³). Decadal climate cycles affect access to mesopelagic prey by southern elephant seals (Mirounga leonina) in the Indian Sector of the Southern Ocean and breeding females are excluded from highly productive continental shelf waters in years of increased sea ice extent and duration (Hindell et al., 2016⁸¹⁴) (medium confidence). To date there is no unified global estimate of the abundance of Antarctic pack ice seal species (Ross seals (Ommatophoca rossi), crabeater seals (Lobodon carcinophaga), leopard seals (Hydrurga leptonyx) and Weddell seals (Leptonychotes weddellii)) as a reference point for understanding climate change impacts on these species (Southwell et al., 2012⁸¹⁵; Bester et al., 2017⁸¹⁶), although some regional population estimates for pack ice seals are available (Gurarie et al., 2017⁸¹⁷ and references therein). Analysis of long-term data suggests a genetic component to adaptation to climate change (low confidence) in Antarctic fur seals (Arctocephalus gazella, Forcada and Hoffman (2014)) and pigmy blue whales (Balaenoptera musculus brevicauda, Attard et al. (2015)).

Population trends of migratory baleen whales have been associated with krill abundance in the Atlantic and Pacific sectors of the Southern Ocean which is reflected in increased reproductive success, body condition and energy allocation (milk availability and transfer) to calves (Braithwaite et al., 2015a⁸¹⁸; Braithwaite et al., 2015b⁸¹⁹; Seyboth et al., 2016⁸²⁰) (high confidence). There have been predictions of negative future impacts of climate change on krill and all whale species, although the magnitude of impacts differs among populations (Tulloch et al., 2019⁸²¹) as for other higher predators (Section 5.2.3). Pacific blue (Tulloch et al., 2019⁸²²) (Balaenoptera musculus), fin (B. physalus) and southern right whales (Eubalaena australis) are the most at risk but humpback whales (Megaptera novaeangliae) are also at risk, as consequence of reduced prey and increasing interspecific competition. Importantly, climate-related risks for whale populations are a product of environmental conditions and connectivity between whale foraging grounds (Southern Ocean) and breeding grounds (lower latitudes) (Section 5.2.3.1).

This section assesses the impacts of ocean and sea ice changes on pelagic foodwebs and ecosystem structure. The ecological impacts of loss of ice shelves and retreat of coastal glaciers around Antarctica are assessed in Section 3.3.3.4. Recent syntheses of Southern Ocean ecosystem structure and function recognise the importance of at least two dominant energy pathways in pelagic foodwebs—a short trophic pathway transferring primary production to top predators via krill, and at least one other pathway that moves energy from smaller phytoplankton to top predators via copepods and small mesopelagic fishes—and indicate that the relative importance of these pathways will change under climate change (Murphy et al., 2013⁸²³; Constable et al., 2016⁸²⁴; Constable et al., 2017⁸²⁵; McCormack et al., 2017⁸²⁶) (medium confidence). Using an ecosystem model, Klein et al. (2018)⁸²⁷ found that the effects of warming on krill growth off the AP and in the Scotia Sea translated to increased risks of declines in krill predator populations, particularly penguins, under both RCP2.6 and RCP8.5. The relative importance of different energy pathways in Southern Ocean foodwebs has important implications for resource management, in particular the management of krill and toothfish fisheries by the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) (Constable et al., 2016⁸²⁸; Constable et al., 2017⁸²⁹) (Sections 3.2.4.1.2, 3.5.3.2.2). 

In summary, advances in knowledge regarding the impacts of climate change on Antarctic marine ecosystems since AR5 are consistent with the impacts described in Larsen et al. (2014)⁸³⁰ (also summarised in Figure 3.6). These advances include further descriptions of local-scale, climate-related influences (sea ice and stratification) on primary productivity, particularly in the West Antarctic Peninsula region (Section 3.2.3.2.1) (medium confidence). At the circumpolar scale, primary production is projected to increase in regions south of 65°S over the period from now to 2100 under RCP8.5 (Leung et al., 2015⁸³¹) (low confidence). However, ocean acidification may have a detrimental effect on coastal phytoplankton communities around the Antarctic continent (Section 3.2.3.2.1) (medium confidence). Increased information is also available regarding climate-driven changes in Antarctic krill populations in the south Atlantic, including the observed southward shift in the spatial distribution of krill in this region (Atkinson et al., 2019⁸³²) (medium confidence) but evidence of a long-term trend in overall abundance in the region is equivocal (Section 3.2.3.2.1). Further habitat contraction for Antarctic krill is predicted in the future (medium confidence) (references detailed in Section 3.2.3.2.1). Under high emissions scenarios the majority of Antarctic seafloor species are projected to be negatively impacted by the end of the century (Griffiths et al., 2017a⁸³³) (low confidence). Observed changes in the geography of ice-associated habitats (sea ice, ice shelves and polynyas) have both positive and negative effects on seabirds and marine mammals, and will interact with ice dependent changes in Antarctic krill populations to compound the impacts on krill dependent predators (Klein et al., 2018⁸³⁴) (Sections 3.2.3.2.1, 3.2.3.2.4) (medium confidence).

Arctic fisheries are important economically and societally. Large commercial fisheries exist off the coasts of Greenland and in the Barents and Bering Seas (Holsman et al., 2018⁸³⁵; Peck and Pinnegar, 2018⁸³⁶). First-wholesale value for commercial harvest of all species in 2017 in the Eastern Bering Sea was 2.68 billion USD, and for the Barents Sea around 1 billion USD to Norwegian fishers alone. The target species for these commercial fisheries include gadoids, flatfish, herring, red fish (Sebastes sp.), salmonids, and capelin. Fisheries in other Arctic regions are relatively small-scale, locally operated, and target a limited number of species (Reist, 2018⁸³⁷). Still, these fisheries are of considerable cultural, economic and subsistence importance to local communities (Section 3.5.2.1).

Climate change will affect the spatial distribution and productivity of some commercially important marine fish and shellfish under most RCPs (Section 3.2.3.1) with associated impacts on the distribution and economic viability of commercial fisheries (high confidence). Past performance suggests that high latitude fisheries have been resilient to changing environmental and market drivers. For example, the Norwegian cod fishery has exported dried cod over an unbroken period of more than a thousand years (Barrett et al., 2011⁸³⁸), reflecting the resilience of the northern Norwegian cod fisheries to historic climate variability (Eide, 2017⁸³⁹). Also, model projections indicate that expansions in suitable habitat for subarctic species and increased production of planktonic prey due to increasing temperatures and ice retreat, will continue to support commercially important fisheries (Lam et al., 2016⁸⁴⁰; Eide, 2017⁸⁴¹; Haug et al., 2017⁸⁴²; Peck and Pinnegar, 2018⁸⁴³) (Section 3.2.3.1.3, Box 3.4) (medium confidence).

However, recent studies in the Bering Sea suggest that future fish production will also depend on how climate change and ocean acidification will alter the quality, quantity and availability of suitable prey; the thermal stress and metabolic demands of resident fish; and species interactions (Section 3.2.3.1.3), suggesting that the future of commercial fisheries in Arctic regions is uncertain (Holsman et al., 2018⁸⁴⁴). It is also uncertain whether future autumn and winter ocean conditions will be conducive to the establishment of resident overwintering spawning populations that are large enough to support sustainable commercial fishing operations at higher latitude Arctic shelf regions (Section 3.2.3.1) (medium confidence).

Projecting the impacts of climate change on marine fisheries is inextricably intertwined with response scenarios regarding risk tolerance in future management of marine resources, advancements in fish capture technology, and markets drivers (e.g., local and global demand, emerging product lines, competition, processing efficiencies and energy costs) (Groeneveld et al., 2018⁸⁴⁵). Seasonal and interannual variability in ocean conditions influences product quality and costs of fish capture (Haynie and Pfeiffer, 2012⁸⁴⁶) (Table 3.4). Further, past experience suggests that barriers to diversification may limit the portfolio of viable target fisheries available to small-scale fisheries (Ward et al., 2017⁸⁴⁷) (low confidence).

This section examines climate change impacts on Southern Ocean fisheries for Antarctic krill and finfish. Management of these fisheries by CCAMLR and responses to climate change are discussed in Section 3.5.2.1. The main Antarctic fisheries are for Antarctic krill, and for Antarctic and Patagonian toothfish; in 2016 the reported catches for these species were approximately 260 thousand tons for krill (CCAMLR, 2017b⁸⁴⁸) and 11 thousand tons for Antarctic and Patagonian toothfish combined (CCAMLR, 2017a⁸⁵⁰). The mean annual wholesale value of the Antarctic krill fishery was 69.5 USD million yr-1 for the period from 2011 to 2015, and 206.7 million USD yr-1 for toothfish fisheries (combined) over the same period (CCAMLR, 2016b). The fishery for Antarctic krill in the southern Atlantic Sector and the northern West Antarctic Peninsula (together the current area of focus for the fishery) has become increasingly concentrated in space over recent decades, which has raised concern regarding localised impacts on krill predators (Hinke et al., 2017a⁸⁵¹). The krill fishery has also changed its peak season of operation. In the early years of the fishery, most krill were taken in summer and autumn, with lowest catches being taken in spring. In recent years the lowest catches have occurred over summer, catches have peaked in late autumn, and very little fishing activity has occurred in spring (Nicol and Foster, 2016⁸⁵²). Some of these temporal and spatial shifts in the fishery over time have been attributed to reductions in winter sea ice extent in the region (Kawaguchi et al., 2009⁸⁵³) (low confidence). Recent increases in the use of krill catch to produce krill oil (as a human health supplement) has also led to vessels concentrating on fishing in autumn and winter when krill are richest in lipids (Nicol and Foster, 2016⁸⁵⁴). Available evidence regarding future changes to Antarctic krill populations (Section 3.2.3.2.1) indicates that the impacts of climate change will be most pronounced in the areas that are currently most important for the Antarctic krill fishery: the Scotia Sea and the northern tip of the AP. Major future changes in the krill fishery itself are expected to be driven by global issues external to the Southern Ocean, including conservation decision making and socioeconomic drivers.

There is limited understanding of the consequences of climate change for Southern Ocean finfish fisheries. Lack of recovery of mackerel icefish (Champsocephalus gunnari) after cessation of fishing in 1995 has been related to anomalous water temperatures (~2°C increase related to a strong El Niño) in the subantarctic Indian Ocean and to availability of krill prey in the Atlantic region (Mintenbeck, 2017⁸⁵⁵) (low confidence). Differences in temperature tolerance of Patagonian and Antarctic toothfish described in Section 3.2.3.2.3 may have implications for future fisheries of these two species.

The reduction of ice shelf buttressing that has dominated AIS mass loss (Section 3.3.1.2) has been driven primarily by increases in sub-ice shelf melting (Khazendar et al., 2013¹⁰⁷¹; Pollard et al., 2015¹⁰⁷²; Cook et al., 2016¹⁰⁷³; Rintoul et al., 2016¹⁰⁷⁴; Walker and Gardner, 2017¹⁰⁷⁵; Adusumilli et al., 2018¹⁰⁷⁶; Dow et al., 2018a¹⁰⁷⁷; Minchew et al., 2018¹⁰⁷⁸) (high confidence). Shoaling of relatively warm Circumpolar Deep Water has controlled recent variability in melting in the Amundsen and Bellingshausen seas, Wilkes Land (Roberts et al., 2018¹⁰⁷⁹) and the AP (medium confidence) (Jacobs et al., 2011¹⁰⁸⁰; Pritchard et al., 2012¹⁰⁸¹; Depoorter et al., 2013¹⁰⁸²; Rignot et al., 2013¹⁰⁸³; Dutrieux et al., 2014¹⁰⁸⁴; Paolo et al., 2015¹⁰⁸⁵; Wouters et al., 2015¹⁰⁸⁶; Christianson et al., 2016¹⁰⁸⁷; Cook et al., 2016¹⁰⁸⁸; Jenkins et al., 2018¹⁰⁸⁹; Roberts et al., 2018¹⁰⁹⁰). Changes in winds have driven this shoaling by affecting continental shelf edge undercurrents (Walker et al., 2013¹⁰⁹¹; Dutrieux et al., 2014¹⁰⁹²; Kimura et al., 2017¹⁰⁹³) and overturning in coastal polynyas (St‐Laurent et al., 2015¹⁰⁹⁴; Webber et al., 2017¹⁰⁹⁵) (medium confidence). Winds over the Amundsen Sea are highly variable, however, with complex interactions between SAM, El Niño/Southern Oscillation (ENSO), Atlantic Multidecadal Oscillation, and the Amundsen Sea Low (Uotila et al., 2013¹⁰⁹⁵; Li et al., 2014¹⁰⁹⁶; Turner et al., 2016¹⁰⁹⁷) (SM3.1.3). 

Through their effects on Antarctic coastal ocean circulation, ENSO or other tropical-ocean variability may have triggered changes to Pine Island Glacier in the 1940s (Smith et al., 2017c¹⁰⁹⁸) and again in the 1970s and 1990s (Jenkins et al., 2018¹⁰⁹⁹), and recent ENSO variability is correlated with recent changes in ice shelf thickness (Paolo et al., 2018¹¹⁰⁰) (medium confidence). Such coupling between wind variability, ocean upwelling, ice shelf melt, buttressing and glacier flow rate has also been observed in EAIS, at Totten Glacier, Wilkes Land (Greene et al., 2017¹¹⁰¹).

Around Greenland, an anomalous inflow of subtropical water driven by wind changes, multi-decadal natural ocean variability (Andresen et al., 2012¹¹⁰²), and a long-term increase in the North Atlantic’s upper ocean heat content since the 1950s (Cheng et al., 2017¹¹⁰³), all contributed to a warming of the subpolar North Atlantic (Häkkinen et al., 2013¹¹⁰⁴) (medium confidence). Water temperatures near the grounding zone of GIS outlet glaciers are critically important to their calving rate (O’Leary and Christoffersen, 2013) (medium confidence), and warm waters have been observed interacting with major GIS outlet glaciers (high confidence) (e.g., Holland et al., 2008; Straneo et al., 2017¹¹⁰⁵). 

The processes behind warm-water incursions in coastal Greenland that force glacier retreat remain unclear, however (Straneo et al., 2013¹¹⁰⁶; Xu et al., 2013b¹¹⁰⁷; Bendtsen et al., 2015¹¹⁰⁸; Murray et al., 2015¹¹⁰⁹; Cowton et al., 2016¹¹¹⁰; Miles et al., 2016¹¹¹¹), and there is low confidence in understanding coastal GIS glacier response to ocean forcing because submarine melt rates, calving rates (Rignot et al., 2010¹¹¹²; Todd and Christoffersen, 2014¹¹¹³; Benn et al., 2017¹¹¹⁴), bed and fjord geometry and the roles of ice melange and subglacial discharge (Enderlin et al., 2013¹¹¹⁵; Gladish et al., 2015¹¹¹⁶; Slater et al., 2015¹¹¹⁷; Morlighem et al., 2016¹¹¹⁸; Rathmann et al., 2017¹¹¹⁹) are poorly understood, and extrapolation from a small sample of glaciers is impractical (Moon et al., 2012¹¹²⁰; Carr et al., 2013¹¹²¹; Straneo et al., 2016¹¹²²; Cowton et al., 2018¹¹²³).

Snow accumulation and surface melt in Antarctica are influenced by the Southern Hemisphere extratropical circulation (SM3.1.3), which has intensified and shifted poleward in austral summer from 1950 to 2012 (Arblaster et al., 2014¹¹²⁴; Swart et al., 2015a¹¹²⁵) (medium confidence). The austral summer SAM has been in its most positive extended state for the past 600 years (Abram et al., 2014¹¹²⁶; Dätwyler et al., 2017¹¹²⁷), and from 1979 to 2013 has contributed to intensified atmospheric circulation and increasing and decreasing snowfall in the western and eastern AP respectively (Marshall et al., 2017¹¹²⁸) (medium confidence). WAIS accumulation trends (Section 3.3.1.2) resulted from a deepening of the Amundsen Sea Low over recent decades (Raphael et al., 2016¹¹²⁹) (high confidence). 

During the 1990s, WAIS experienced record surface warmth relative to the past 200 years, though similar conditions occurred for 1% of the preceding 2000 years (Steig et al., 2013¹¹³⁰), and WAIS surface melting remains limited. In contrast, AP surface melting has intensified since the mid-20th century and the last three decades were unprecedented over 1000 years (Abram et al., 2013a¹¹³¹). The northeast AP began warming 600 years ago and past-century rates were unusual over 2000 years (Mulvaney et al., 2012¹¹³²; Stenni et al., 2017¹¹³³). Increased föhn winds due to the more positive SAM (Cape et al., 2015¹¹³⁴) caused increased surface melting on the Larsen ice shelves (Grosvenor et al., 2014¹¹³⁵; Luckman et al., 2014¹¹³⁶; Elvidge et al., 2015¹¹³⁷) and after 11,000 years intact, the 2002 melt-driven collapse of the Larsen B ice shelf followed strong warming between the mid–1950s and the late 1990s (Domack et al., 2005¹¹³⁸) (medium confidence). 

In Greenland, associations between atmospheric pressure indices such as the North Atlantic Oscillation (NAO) and temperature, insolation and snowfall indicate with high confidence that, as in Antarctica, variability of large-scale atmospheric circulation is an important driver of SMB changes (Fettweis et al., 2013¹¹³⁹; Tedesco et al., 2013¹¹⁴⁰; Ding et al., 2014¹¹⁴¹; Tedesco et al., 2016b¹¹⁴²; Ding et al., 2017¹¹⁴³; Hofer et al., 2017¹¹⁴⁴). A post-1990s decrease in summer NAO reflects increased anticyclonic weather (e.g., Tedesco et al., 2013; Hanna et al., 2015¹¹⁴⁵) that advected warm air over the GIS, explaining ~70% of summer surface warming from 2003 to 2013 (Fettweis et al., 2013¹¹⁴⁶; Tedesco et al., 2013¹¹⁴⁷; Mioduszewski et al., 2016¹¹⁴⁸), and reduced the cloud cover, increasing shortwave insolation (Tedesco et al., 2013¹¹⁴⁹) that, combined with albedo feedbacks (Box et al., 2012¹¹⁵⁰; Charalampidis et al., 2015¹¹⁵¹; Tedesco et al., 2016a¹¹⁵²; Stibal et al., 2017¹¹⁵³; Ryan et al., 2018¹¹⁵⁴) (high confidence), explains most of the post-1990s melt increase (Hofer et al., 2017¹¹⁵⁵). These drivers culminated in July 2012 in exceptional warmth and surface melt up to the ice sheet summit (Nghiem et al., 2012¹¹⁵⁶; Tedesco et al., 2013¹¹⁵⁷; Hanna et al., 2014¹¹⁵⁸; Hanna et al., 2016¹¹⁵⁹; McLeod and Mote, 2016¹¹⁶⁰). 

Dramatic reductions in Arctic (land areas north of 60°N) spring snow cover extent have occurred since satellite charting began in 1967 (Estilow et al., 2015). Declines in May and June of –3.5% (± 1.9%) and –13.4% respectively per decade (± 5.4%) between 1967 and 2018 (relative to the 1981–2010 mean) were determined from multiple datasets based on the methodology of (Mudryk et al., 2017¹³⁵¹) (Figure 3.10) (high confidence). The loss of spring snow extent is reflected in shorter snow cover duration estimated from surface observations (Bulygina et al., 2011¹³⁵²; Brown et al., 2017¹³⁵³), satellite data (Wang et al., 2013¹³⁵⁴; Estilow et al., 2015¹³⁵⁵; Anttila et al., 2018¹³⁵⁶), and model-based analyses (Liston and Hiemstra, 2011¹³⁵⁷) (high confidence). These trends range between –0.7 and –3.9 days per decade depending on region and time period, but all spring snow cover duration trends from all datasets are negative (Brown et al., 2017¹³⁵⁸). These same multi-source datasets also identify reductions in autumn snow extent and duration (-0.6 to -1.4 days per decade; summarized in Brown et al., 2017) (high confidence). There is low confidence in positive October and November snow cover extent trends apparent in a single dataset (Hernández-Henríquez et al., 2015¹³⁵⁹) because they are not replicated in other surface, satellite and model datasets (Brown and Derksen, 2013¹³⁶⁰; Mudryk et al., 2017¹³⁶¹).

Weather station observations across the Russian Arctic identify negative trends in the maximum snow depth between 1966 and 2014 (Bulygina et al., 201¹³⁶²; Osokin and Sosnovsky, 2014¹³⁶³). There is medium confidence in this trend because the pointwise nature of these measurements does not capture prevailing conditions across the landscape. Seasonal maximum snow depth trends over the North American Arctic are mixed and largely statistically insignificant (Vincent et al., 2015¹³⁶⁴; Brown et al., 2017¹³⁶⁵). The timing of maximum snow depth has shifted earlier by 2.7 days per decade for the North American Arctic (Brown et al., 2017¹³⁶⁶); comparable analysis is not available for Eurasia. Gridded products from remote sensing and land surface models identify negative trends in snow water equivalent between 1981 and 2016 for both the Eurasian and North American sectors of the Arctic (Brown et al., 2017¹³⁶⁷). While the snow water equivalent anomaly time series show reasonable consistency between products when averaged at the continental scale, considerable inter-dataset variability in the spatial patterns of change (Liston and Hiemstra, 2011¹³⁶⁸; Park et al., 2012¹³⁶⁹; Brown et al., 2017¹³⁷⁰) mean there is only medium confidence in these trends.

Despite uncertainties due to sparse observations (Cowtan and Way, 2014¹³⁷¹), surface temperature has increased across Arctic land areas in recent decades (Hawkins and Sutton, 2012¹³⁷²; Fyfe et al., 2013¹³⁷³), driving reductions in Arctic snow extent and duration (high confidence). Changes in Arctic snow extent can be directly related to extratropical temperature increases (Brutel-Vuilmet et al., 2013¹³⁷⁴; Thackeray et al., 2016¹³⁷⁵; Mudryk et al., 2017¹³⁷⁶). Based on multiple historical datasets, there is a consistent temperature sensitivity for Arctic snow extent, with approximately 800,000 km2 of snow cover lost per degrees Celsius warming in spring (Brown and Derksen, 2013¹³⁷⁷; Brown et al., 2017), and 700,000–800,000 km2 lost in autumn (Derksen and Brown, 2012¹³⁷⁹; Brown and Derksen, 2013¹³⁸⁰) (high confidence).

There is high confidence that darkening of snow through the deposition of black carbon and other light absorbing particles enhances snow melt (Bullard et al., 2016¹³⁸¹; Skiles et al., 2018¹³⁸²; Boy et al., 2019¹³⁸³). The global direct radiative forcing for black carbon in seasonal snow and over sea ice is estimated to be 0.04 W m–2, but the effective forcing can be up to threefold greater at regional scales due to the enhanced albedo feedback triggered by the initial darkening (Bond et al., 2013). Lawrence et al. (2011)¹³⁹³ estimate the present-day radiative effect of black carbon and dust in land-based snow to be 0.083 W m–2, only marginally greater than the simulated 1850 effect (0.075 W m–2) due to offsetting effects from increased black carbon emissions and reductions in dust darkening (medium confidence). Kylling et al. (2018)¹³⁹⁴ estimate a surface radiative effect of 0.292 W m–2 caused by dust deposition (largely transported from Asia) to Arctic snow, approximately half of the black carbon central scenario estimate of Flanner et al. (2007)¹³⁹⁵. The forcing from brown carbon deposited in snow (associated with both combustion and secondary organic carbon) is estimated to be 0.09−0.25 W m–2, with the range due to assumptions of particle absorptivity (Lin et al., 2014¹³⁹⁶) (low confidence).

Precipitation remains a sparse and highly uncertain measurement over Arctic land areas: in situ datasets remain uncertain (Yang, 2014¹³⁹⁷) and are largely regional (Kononova, 2012¹³⁹⁸; Vincent et al., 2015¹³⁹⁹). Atmospheric reanalyses show increases in Arctic precipitation in recent decades (Lique et al., 2016¹⁴⁰⁰; Vihma et al., 2016¹⁴⁰¹), but there remains low confidence in reanalysis-based closure of the Arctic freshwater budget due to a wide spread between available reanalysis derived precipitation estimates (Lindsay et al., 2014¹⁴⁸⁴). Despite improved process understanding, estimates of sublimation loss during blowing snow events remain a key uncertainty in the mass budget of the Arctic snowpack (Sturm and Stuefer, 2013¹⁴⁸⁵).

Record high temperatures at ~10–20 m depth in the permafrost (near or below the depths affected by intra-annual fluctuation in temperature) have been documented at many long-term monitoring sites in the Northern Hemisphere circumpolar permafrost region (AMAP, 2017d¹³⁸⁶) (Figure 3.10) (very high confidence). At some locations, the temperature is 2°C–3°C higher than 30 years ago. During the decade between 2007 and 2016, the rate of increase in permafrost temperatures was 0.39°C ± 0.15°C for colder continuous zone permafrost monitoring sites, 0.20°C ± 0.10°C for warmer discontinuous zone permafrost, giving a global average of 0.29 ± 0.12°C across all polar and mountain permafrost (Biskaborn et al., 2019¹³⁸⁷). Relatively smaller increases in permafrost temperature in warmer sites indicate that permafrost is thawing with heat absorbed by the ice-to-water phase change, and as a result, the active layer may be increasing in thickness. In contrast to temperature, there is only medium confidence that active layer thickness across the region has increased. This confidence level is because decadal trends vary across regions and sites (Shiklomanov et al., 2012¹³⁸⁸) and because mechanical probing of the active layer can underestimate the degradation of permafrost in some cases because the surface subsides when ground ice melts and drains (Mekonnen et al., 2016¹³⁸⁹; AMAP, 2017d¹³⁹⁰; Streletskiy et al., 2017¹³⁹¹). Permafrost in the Southern Hemisphere polar region occurs in ice-free exposed areas (Bockheim et al., 2013¹³⁹²), 0.18% of the total land area of Antarctica (Burton-Johnson et al., 2016). This area is three orders of magnitude smaller than the 13–18 x 106 km2 area underlain by permafrost in the Northern Hemisphere terrestrial permafrost region (Gruber, 2012). Antarctic permafrost temperatures are generally colder (Noetzli et al., 2017¹⁴⁰³) and increased 0.37°C ± 0.10°C between 2007 and 2016 (Biskaborn et al., 2019¹⁴⁰⁴).

Permafrost thaw and loss of ground ice causes the land surface to subside and collapse into the volume previously occupied by ice, resulting in disturbance to overlying ecosystems and human infrastructure (Kanevskiy et al., 2013¹⁴⁰⁵; Raynolds et al., 2014¹⁴⁰⁶). Excess ice in permafrost is typical, varying for example from 40% of total volume in some sands up to 80–90% of total volume in fine-grained soil/sediments (Kanevskiy et al., 2013¹⁴⁰⁷). Ice rich permafrost areas where impacts of thaw could be greatest include the Yedoma deposits in Siberia, Alaska, and the Yukon in Canada, with ice divided between massive wedges interspersed with frozen soil/sediment containing pore ice and smaller ice features (Schirrmeister et al., 2011¹⁴⁰⁸; Strauss et al., 2017¹⁴⁰⁹). Other areas including, for example, Northwestern Canada, the Canadian Archipelago, the Yamal and Gydan peninsulas of West Siberia, and smaller portions of Eastern Siberia and Alaska contain buried glacial ice bodies of significant thickness and extent (Lantuit and Pollard, 2008; Leibman et al., 2011; Kokelj et al., 2017; Coulombe et al., 2019). The location and volume of ground ice integrated across the northern permafrost region (5.63–36.55 x 103 km3, equivalent to 2–10 cm sea level rise) is known with medium confidence and with no recent updates at the circumpolar scale (Zhang et al., 2008¹⁴¹⁰). 

The permafrost region represents a large, climate sensitive reservoir of organic carbon with the potential for some of this pool to be rapidly decayed and transferred to the atmosphere as CO2 and methane as permafrost thaws in a warming climate, thus accelerating the pace of climate change (Schuur et al., 2015¹⁴¹⁴). The current best mean estimate of total (surface plus deep) organic soil carbon (terrestrial) in the northern circumpolar permafrost region (17.8 x 106 km2 area) is 1460 to 1600 petagrams (medium confidence) (Pg; 1 Pg = 1 billion metric tonnes) (Schuur et al., 2018¹⁴¹⁵). All permafrost region soils estimated to 3 m in depth (surface) contain 1035 ± 150 Pg C (Tarnocai et al., 2009¹⁴¹⁶; Hugelius et al., 2014¹⁴¹⁷) (high confidence). Of the carbon in the surface, 800–1000 Pg C is perennially frozen, with the remainder contained in seasonally-thawed soils. The northern circumpolar permafrost region occupies only 15% of the total global soil area, but the 1035 Pg C adds another 50% to the rest of the 3 m soil carbon inventory (2050 Pg C for all global biomes excluding tundra and boreal; Jobbágy and Jackson, 2000¹⁴¹⁸; Schuur et al., 2015¹⁴¹⁹).

Substantial permafrost carbon exists below 3 m depth (medium confidence). Deep carbon (>3 m) has been best quantified for the Yedoma region of Siberia and Alaska, characterised by wind- and water-moved permafrost sediments tens of meters thick. The Yedoma region covers a 1.4 x 106 km2 area that remained ice-free during the last Ice Age (Strauss et al., 2013¹⁴²⁰) and accounts for 327–466 Pg C in deep sediment accumulations below 3 m (Strauss et al., 2017).

The current inventory has also highlighted additional carbon pools that are likely to be present but are so poorly quantified (low confidence) that they cannot yet be added into the number reported above. There are deep terrestrial soil/sediment deposits outside of the Yedoma region that may contain about 400 Pg C (Schuur et al., 2015¹⁴²¹). An additional pool is organic carbon remaining in permafrost but that is now submerged on shallow Arctic sea shelves that were formerly exposed as terrestrial ecosystems during the Last Glacial Maximum ~20,000 years ago (Walter et al., 2007¹⁴²³). This permafrost is degrading slowly due to seawater intrusion, and it is not clear what amounts of permafrost and organic carbon still remain in the sediment versus what has already been converted to greenhouse gases. A recent synthesis of permafrost extent for the Beaufort Sea shelf showed that most remaining subsea permafrost in that region exists near shore with much reduced area (high confidence) as compared to original subsea permafrost maps that outlined the entire 3 x 106 km2 shelf area (<120 m below sea level depth) that was formerly exposed as land (Ruppel et al., 2016¹⁴²⁴). These observations are supported by similar studies in the Siberian Arctic Seas (Portnov et al., 2013¹⁴²⁵), and by modelling that suggests that subsea permafrost would be thawed many meters below the seabed under current submerged conditions (Anisimov et al., 2012¹⁴²⁶; AMAP, 2017d¹⁴²⁷; Angelopoulos et al., 2019¹⁴²⁸).

Changes in temperature and precipitation act as gradual ‘press’ (i.e., continuous) disturbances that directly affect permafrost by modifying the ground thermal regime, as discussed in Section 3.4.1.2.1. Climate change can also modify the occurrence and magnitude of abrupt physical disturbances such as fire, and soil subsidence and erosion resulting from ice rich permafrost thaw (thermokarst). These ‘pulse’ (i.e., discrete) disturbances (Smith et al., 2009¹⁴²⁹) often are part of the ongoing disturbance and successional cycle in Arctic and boreal ecosystems (Grosse et al., 2011¹⁴³⁰), but changing rates of occurrence alter the landscape distribution of successional ecosystem states, with permafrost characteristics defined by the ecosystem and climate state (Kanevskiy et al., 2013¹⁴³¹). 

Pulse disturbances often rapidly remove the insulating soil organic layer, leading to permafrost degradation (Gibson et al., 2018¹⁴²³). Of all pulse disturbance types, wildfire affects the most high-latitude land area annually at the continental scale. In some well-studied regions, there is high confidence that area burned, fire frequency and extreme fire years are higher now than the first half of the last century, or even the last 10,000 years (Kasischke and Turetsky, 2006¹⁴³³; Flannigan et al., 2009¹⁴³⁴; Kelly et al., 2013¹⁴³⁵; Hanes et al., 2019¹⁴³⁶). Recent climate warming has been linked to increased wildfire activity in the boreal forest regions in Alaska and western Canada where this has been studied (Gillett, 2004¹⁴³⁷; Veraverbeke et al., 2017¹⁴³⁸). Based on satellite imagery, an estimated 80,000 km2 of boreal area was burned globally per year from 1997 to 2011 (van der Werf et al., 2010¹⁴³⁹; Giglio et al., 2013¹⁴⁴⁰). Extreme fire years in northwest Canada during 2014 and Alaska during 2015 doubled the long-term (1997–2011) average area burned annually in this region (Canadian Forest Service, 2017), surpassing Eurasia to contribute 60% of the global boreal area burned (van der Werf et al., 2010¹⁴⁴¹; Randerson et al., 2012¹⁴⁴²; Giglio et al., 2013¹⁴⁴³). These extreme North American fire years were balanced by lower-than-average area burned in Eurasian forests, resulting in a 5% overall increase in global boreal area burned. The annual area burned in Arctic tundra is generally small compared to the forested boreal biome. In Alaska—the only region where estimates of burned area exist for both boreal forest and tundra vegetation types—tundra burning averaged approximately 270 km2 yr-1 during the last half century (French et al., 2015¹⁴⁴⁵), accounting for 7% of the average annual area burned throughout the state (Pastick et al., 2017¹⁴⁴⁶). There is high confidence that changes in the fire regime are degrading permafrost faster than had occurred over the historic successional cycle (Turetsky et al., 2011¹⁴⁴⁷; Rupp et al., 2016¹⁴⁴⁸; Pastick et al., 2017¹⁴⁴⁹), and that the effect of this driver of permafrost change is under-represented in the permafrost temperature observation network.

Abrupt permafrost thaw occurs when changing environmental and ecological conditions interact with geomorphological processes. Melting ground ice causes the ground surface to subside. Pooling or flowing water causes localised permafrost thaw and sometimes mass erosion. Together, these localised feedbacks can thaw through meters of permafrost within a short time, much more rapidly than would be caused by increasing air temperature alone. This process is a pulse disturbance to permafrost that can occur in response to climate, such as an extreme precipitation event (Balser et al., 2014¹⁴⁵⁰; Kokelj et al., 2015¹⁴⁵¹), or coupled with other disturbances such as wildfire that affects the ground thermal regime (Jones et al., 2015a¹⁴⁵²). There is medium confidence in the importance of abrupt thaw for driving change in permafrost at the circumpolar scale because it occurs at point locations rather than continuously across the landscape, but the risk for widespread change from this mechanism remains high because of the rapidity of change in these locations (Kokelj et al., 2017¹⁴⁵³; Nitze et al., 2018¹⁴⁵⁴). New research at the global scale has revealed that 3.6 x 106 km2, about 20% of the northern permafrost region, appears to be vulnerable to abrupt thaw (Olefeldt et al., 2016¹⁴⁵⁵).

Long-term in situ river ice records indicate that the duration of ice cover in Russian Arctic rivers decreased by 7–20 days between 1955 and 2012 (Shiklomanov and Lammers, 2014¹⁴⁵⁹) (high confidence). This is consistent with historical reductions in Arctic river ice cover derived from models (Park et al., 2015) and regional analysis of satellite data (Cooley and Pavelsky, 2016¹⁴⁶¹).

Analysis of satellite imagery between 2000 and 2013 identified a significant trend of earlier spring ice break-up across all regions of the Arctic (Šmejkalová et al., 2016¹⁴⁶²); independent satellite data showed approximately 80% of Arctic lakes experienced declines in ice cover duration during 2002–2015, due to both a later freeze-up and earlier break-up (Du et al., 2017¹⁴⁶³) (high confidence). There are indications that lake ice across Alaska has thinned in recent decades (Alexeev et al., 2016¹⁴⁶⁴), but ice thickness trends are not available at the pan-Arctic scale. Analysis of satellite data over northern Alaska show that approximately one-third of bedfast lakes (the entire water volume freezes by the end of winter) experienced a regime change to floating ice over the 1992–2011 period (Surdu et al., 2014¹⁴⁶⁵; Arp et al., 2015¹⁴⁶⁶). This can result in degradation of underlying permafrost (Arp et al., 2016¹⁴⁶⁷; Bartsch et al., 2017¹⁴⁶⁸). Lakes of the central and eastern Canadian High Arctic are transitioning from a perennial to seasonal ice regime (Surdu et al., 2016¹⁴⁶⁹).

A general trend of increasing discharge has been observed for large Siberian (Troy et al., 2012¹⁴⁷⁰; Walvoord and Kurylyk, 2016¹⁴⁷¹) and Canadian (Ge et al., 2013¹⁴⁷²; Déry et al., 2016¹⁴⁷³) rivers that drain to the Arctic Ocean (medium confidence). Between 1976 and 2017, trends are 3.3 ± 1.6% for Eurasian rivers and 2.0 ± 1.8% for North American rivers (Holmes et al., 2018¹⁴⁷⁴) (Figure 3.10). Extreme regional runoff events have also been identified (Stuefer et al., 2017¹⁴⁷⁵). An observed increase in baseflow in the North American (Walvoord and Striegl, 2007¹⁴⁷⁶; St. Jacques and Sauchyn, 2009) and Eurasian Arctic (Smith et al., 2007¹⁴⁷⁷; Duan et al., 2017¹⁴⁷⁸) over the last several decades is attributable to permafrost thaw and concomitant enhancement in groundwater discharge. The timing of spring season peak flow is generally earlier (Ge et al., 2013¹⁴⁷⁹; Holmes et al., 2015¹⁴⁸⁰). There is consistent evidence of decreasing summer season discharge for the Yenisei, Lena, and Ob watersheds in Siberia (Ye et al., 2003¹⁴⁸¹; Yang et al., 2004a¹⁴⁸²; Yang et al., 2004b¹⁴⁸³) and the majority of northern Canadian rivers (Déry et al., 2016¹⁴⁸⁴). Long-term records indicate water temperature increases (Webb et al., 2008¹⁴⁸⁵; Yang and Peterson, 2017¹⁴⁸⁶); attribution to rising air temperatures is complicated by the influence of reservoir regulation over Siberian regions (Liu et al., 2005¹⁴⁸⁷; Lammers et al., 2007¹⁴⁸⁸). Increases in discharge and water temperature in the spring season represent notable freshwater and heat fluxes to the Arctic Ocean (Yang et al., 2014¹⁴⁸⁹).

A large proportion of low-lying Arctic land areas are covered by lakes because permafrost limits surface water drainage and supports ponding even across areas with high moisture deficits (Grosse et al., 2013¹⁴⁹⁰). While thaw in continuous permafrost is linked to intensified thermokarst activity and subsequent ponding (resulting in lake/wetland expansion), observations of change in surface water coverage across the Arctic are regionally variable (Nitze et al., 2017¹⁴⁹¹; Ulrich et al., 2017¹⁴⁹²; Pastick et al., 2019¹⁴⁹³). In landscapes with degrading ice-wedge polygons, subsidence can reduce inundation, increase runoff, and decrease surface water (Liljedahl et al., 2016¹⁴⁹⁴; Perreault et al., 2017¹⁴⁹⁵). In discontinuous permafrost, thaw opens up pathways of subsurface flow, improving the connection among inland water systems which supports the drainage of lakes and overall reduction in surface water cover (Jepsen et al., 2013¹⁴⁹⁶). Enhanced subsurface connectivity from thaw in discontinuous permafrost serves tempers short-term lake fluctuations (Rey et al., 2019¹⁴⁹⁷).

There is high confidence that environmental drivers of Arctic surface water change are diverse and depend on local and regional factors such as permafrost properties and geomorphology (Nitze et al., 2018¹⁴⁹⁸). Thermokarst lake expansion has been observed in the continuous permafrost of northern Siberia (Smith et al., 2005¹⁴⁹⁹; Polishchuk et al., 2015¹⁵⁰⁰) and Alaska (Jones et al., 2011¹⁵⁰¹); surface water area reduction has been observed in discontinuous permafrost of central and southern Siberia (Smith et al., 2005¹⁵⁰²; Sharonov et al., 2012¹⁵⁰³), western Canada (Labrecque et al., 2009¹⁵⁰⁴; Carroll et al., 2011¹⁵⁰⁵; Lantz and Turner, 2015¹⁵⁰⁶) and interior Alaska (Chen et al., 2012¹⁵⁰⁷; Rover et al., 2012¹⁵⁰⁸). Increased evaporation from warmer/longer summers, decreased recharge due to reductions in snow melt volume, and dynamic processes such as ice-jam flooding (Chen et al., 2012¹⁵⁰⁹; Bouchard et al., 2013¹⁵¹⁰; Jepsen et al., 2015¹⁵¹¹) are important considerations for understanding observed surface water area change across the Arctic.

Satellite and model-derived estimates of evapotranspiration show increases across the Arctic (Rawlins et al., 2010¹⁵¹²; Liu et al., 2014¹⁵¹³; Liu et al., 2015b¹⁵¹⁴; Fujiwara et al., 2016¹⁵¹⁵; Suzuki et al., 2018¹⁵¹⁶) (medium confidence). Increases in the seasonal active layer thickness impact temporary water storage and thus runoff regimes in drainage basins. Formation of taliks underneath lakes and rivers may result in reconnection of surface with sub-permafrost ground water aquifers with varying hydrological consequences depending on local geological and hydraulic settings (Wellman et al., 2013¹⁵¹⁷).

Climate warming is expected to change the storage of carbon in vegetation and soils in northern regions, and net carbon transferred to the atmosphere as CO2 and methane acts as a feedback to accelerate global climate change. There is high confidence that the northern region acted as a net carbon sink as carbon accumulated in terrestrial ecosystems over the Holocene (Loisel et al., 2014¹⁵⁷⁵; Lindgren et al., 2018¹⁵⁷⁶). There is medium evidence with low agreement whether changing climate in the modern period has shifted these ecosystems into net carbon sources. Syntheses of ecosystem CO2 fluxes have alternately showed tundra ecosystems as carbon sinks or neutral averaged across the circumpolar region for the 1990s and 2000s (McGuire et al., 2012¹⁵⁷⁷), or carbon sources over the same time period (Belshe et al., 2013¹⁵⁷⁸). Both syntheses agree that the summer growing season is a period of net carbon uptake into terrestrial ecosystems (high confidence), and this uptake appears to be increasing as a function of vegetation density/biomass (Ueyama et al., 2013¹⁵⁷⁹). The discrepancy between these syntheses may be a result of CO2 release rates during the non-summer season that are now thought to be higher than previously estimated (high confidence) (Webb et al., 2016¹⁵⁸⁰) or the separation of upland and wetland ecosystem types, which was done in one synthesis but not the other. Moisture status is a primary control over ecosystem carbon sink/source strength with wetlands more often than not still acting as annual net carbon sinks even while methane is emitted (Lund et al., 2010¹⁵⁸¹). Recent aircraft measurements of atmospheric CO2 concentrations over Alaska showed that tundra regions of Alaska were a consistent net CO2 source to the atmosphere, whereas boreal forest regions were either neutral or net CO2 sinks for the period 2012–2014 (Commane et al., 2017¹⁵⁸²). That study region as a whole was estimated to be a net carbon source of 25 ± 14 Tg CO2-C yr-1 averaged over the land area of both biomes for the entire study period. For comparison to projected global emissions, this would be equivalent to a net source of 0.3 Pg CO2-C yr-1 assuming the Alaska study region (1.6 x 106 km2) could be scaled to the entire northern circumpolar permafrost region soil area (17.8 x 106 km2).

The permafrost soil carbon pool is climate sensitive and an order of magnitude larger than carbon stored in plant biomass (Schuur et al., 2018¹⁵⁸³) (very high confidence). Initial estimates were converging on a range of cumulative emissions from soils to the atmosphere by 2100, but recent studies have actually widened that range somewhat (Figure 3.11) (medium confidence). Expert assessment and laboratory soil incubation studies suggest that substantial quantities of C (tens to hundreds Pg C) could potentially be transferred from the permafrost carbon pool into the atmosphere under RCP8.5 (Schuur et al., 2013¹⁵⁸⁴; Schädel et al., 2014 ⁴). Global dynamical models supported these findings, showing potential carbon release from the permafrost zone ranging from 37–174 Pg C by 2100 under high emission climate warming trajectories, with an average across models of 92 ± 17 Pg C (mean ± SE) (Zhuang et al., 2006¹⁵⁸⁵; Koven et al., 2011¹⁵⁸⁶; Schaefer et al., 2011¹⁵⁸⁷; MacDougall et al., 2012¹⁵⁸⁸; Burke et al., 2013¹⁵⁸⁹; Schaphoff et al., 2013¹⁵⁹⁰; Schneider von Deimling et al., 2015¹⁵⁹¹). This range is generally consistent with several newer data-driven modelling approaches that estimated that soil carbon releases by 2100 (for RCP8.5) will be 57 Pg C (Koven et al., 2015¹⁵⁹²) and 87 Pg C (Schneider von Deimling et al., 2015¹⁵⁹³), as well as an updated estimate of 102 Pg C from one of the previous models (MacDougall and Knutti, 2016¹⁵⁹⁴). However, the latest model runs performed with either structural enhancements to better represent permafrost carbon dynamics (Burke et al., 2017a¹⁵⁹⁵), or common environmental input data (McGuire et al., 2016¹⁵⁹⁶) show similar soil carbon losses, but also indicate the potential for stimulated plant growth (nutrients, temperature/growing season length, CO2 fertilisation) to offset some (Kleinen and Brovkin, 2018¹⁵⁹⁷) or all of these losses, at least during this century, by sequestering new carbon into plant biomass and increasing carbon inputs into the surface soil (McGuire et al., 2018¹⁵⁹⁸). These future carbon emission levels would be a significant fraction of those projected from fossil fuels with implications for allowable carbon budgets that are consistent with limiting global warming, but will also depend on how vegetation responds (high confidence). Furthermore, there is high confidence that climate scenarios that involve mitigation (e.g., RCP4.5) will help to dampen the response of carbon emissions from the Arctic and boreal regions.

Northern ecosystems contribute significantly to the global methane budget, but there is low confidence about the degree to which additional methane from northern lakes, ponds, wetland ecosystems, and the shallow Arctic Ocean shelves is currently contributing to increasing atmospheric concentrations. Analyses of atmospheric concentrations in Alaska concluded that local ecosystems surrounding the observation site have not changed in the exchange of methane from the 1980s until the present, which suggests that either the local wetland ecosystems are responding similarly to other northern wetland ecosystems, or that increasing atmospheric methane concentrations in northern observation sites is derived from methane coming from mid-latitudes (Sweeney et al., 2016¹⁶⁰⁰). However, this contrasts with indirect integrated estimates of methane emissions from observations of expanding permafrost thaw lakes that suggest a release of an additional 1.6–5 Tg CH4 yr–1 over the last 60 years (Walter Anthony et al., 2014¹⁶⁰¹). At the same time, there is high confidence that methane fluxes at the ecosystem to regional scale have been under-observed, in part due to the low solubility of methane in water leading to ebullution (bubbling) flux to the atmosphere that is heterogeneous in time and space. Some new quantifications include: cold-season methane emissions that can be >50% of the annual budget of terrestrial ecosystems (Zona et al., 2016¹⁶⁰²); geological methane seeps that may be climate sensitive if permafrost currently serves as a cap preventing atmospheric release (Walter Anthony et al., 2012¹⁶⁰³; Ruppel and Kessler, 2016¹⁶⁰⁴; Kohnert et al., 2017¹⁶⁰⁵); estimates of shallow Arctic Ocean shelf methane emissions where the range of estimates based on methane concentrations in air and water has widened with more observations and now ranges from 3 Tg CH4 yr–1 (Thornton et al., 2016¹⁶⁰⁶) to 17 Tg CH4 yr–1 (Shakhova et al., 2013¹⁶⁰⁷). Observations such as these underlie the fact that source estimates for methane made from atmospheric observations are typically lower than methane source estimates made from upscaling of ground observations (e.g., Berchet et al., 2016), and this problem has not improved, even at the global scale, over several decades of research (Saunois et al., 2016¹⁶⁰⁸; Crill and Thornton, 2017¹⁶⁰⁹). 

In many of the dynamical model projections previously discussed, methane release is not explicitly represented because fluxes are small even though higher global warming potential of methane makes these emissions relatively more important than on a mass basis alone. Global models that do include methane show that emissions may already (from 2000 to 2012) be increasing at a rate of 1.2 Tg CH4 yr–1 in the northern region as a direct response to temperature (Riley et al., 2011¹⁶¹⁰; Gao et al., 2013¹⁶¹¹; Poulter et al., 2017¹⁶¹²). A model intercomparison study forecast northern methane emissions to increase from 18 Tg CH4 yr–1 to 42 Tg CH4 yr–1 under RCP8.5 by 2100 largely as a result of an increase in wetland extent (Zhang et al., 2017¹⁶¹³). However, projected methane emissions are sensitive to changes in surface hydrology (Lawrence et al., 2015¹⁶¹⁴) and a suite of models that were thought to perform well in high-latitude ecosystems showed a general soil drying trend even as the overall water cycle intensified (McGuire et al., 2018¹⁶¹⁵). Furthermore, most models described above do not include many of the abrupt thaw processes that can result in lake expansion, wetland formation, and massive erosion and exposure to decomposition of previously frozen carbon-rich permafrost, leading to medium confidence in future model projections of methane. Recent studies that addressed some of these landscape controls over future emissions projected increases in methane above the current levels on the order 10–60 Tg CH4 yr-1 under RCP8.5 by 2100 (Schuur et al., 2013¹⁶¹⁶; Koven et al., 2015¹⁶¹⁷; Lawrence et al., 2015¹⁶¹⁸; Schneider von Deimling et al., 2015¹⁶¹⁹; Walter Anthony et al., 2018¹⁶²⁰). These additional methane fluxes are projected to cause 40–70% of total permafrost-affected radiative forcing in this century even though methane emissions are much less than CO2 by mass (Schneider von Deimling et al., 2015¹⁶²¹; Walter Anthony et al., 2018¹⁶²²). As with total carbon emissions, there is high confidence that mitigation of anthropogenic methane sources could help to dampen the impact of increased methane emissions from the Arctic and boreal regions (Christensen et al., 2019¹⁶²³).

Warming induced reductions in the duration and extent of Arctic spring snow cover (Section 3.4.1.1) lower albedo because snow-free land reflects much less solar radiation than snow. The corresponding increase in net radiation absorption at the surface provides a positive feedback to global temperatures (Flanner et al., 2011¹⁶²⁴; Qu and Hall, 2014¹⁶²⁵; Thackeray and Fletcher, 2016¹⁶²⁶) (high confidence). Estimates of increases in global net solar energy flux due to snow cover loss range from 0.10–0.22 W m–2 (± 50%; medium confidence) depending on dataset and time period (Flanner et al., 2011¹⁶²⁷; Chen et al., 2015¹⁶²⁸; Singh et al., 2015¹⁶²⁹; Chen et al., 2016b¹⁶³⁰). Sources of uncertainty include the range in observed spring snow cover extent trends (Hori et al., 2017¹⁶³¹) and the influence of clouds on shortwave feedbacks (Sedlar, 2018¹⁶³²; Sledd and L’Ecuyer, 2019¹⁶³³). Terrestrial snow changes also affect the longwave energy budget via altered surface emissivity (Huang et al., 2018¹⁶³⁴). Climate model simulations show that changes in snow cover dominate land surface related positive feedbacks to atmospheric heating (Euskirchen et al., 2016¹⁶³⁵), but regional variations in surface albedo are also influenced by vegetation (Loranty et al., 2014¹⁶³⁶). There is evidence for positive sensitivity of surface temperatures to increased northern hemisphere boreal and tundra leaf area index, which contributes a positive feedback to warming (Forzieri et al., 2017¹⁶³⁷).

Changes in tundra vegetation can have important ecosystem effects, in particular on hydrology, carbon and nutrient cycling and surface energy balance, which together impact permafrost (e.g., Myers-Smith and Hik, 2013; Frost and Epstein, 2014¹⁶³⁸; Nauta et al., 2014¹⁶³⁹). Aside from physical impacts, changing vegetation influences the diversity and abundance of herbivores (e.g., Fauchald et al., 2017b; Horstkotte et al., 2017¹⁶⁴⁰) in the Arctic. The overall trend for tundra vegetation across the 36–year satellite period (1982–2017) shows increasing above ground biomass (greening) throughout a majority of the circumpolar Arctic (high confidence) (Xu et al., 2013a¹⁶⁴¹; Ju and Masek, 2016¹⁶⁴²; Bhatt et al., 2017¹⁶⁴³). Increasing greenness has been in some cases linked with shifts in plant species dominance away from graminoids (grasses and sedges) towards shrubs (high confidence) (Myers-Smith et al., 2015¹⁶⁴⁴). Within the overall trend of increases (greening), some tundra areas show declines (browning) (Bhatt et al., 2017).

The spatial variation in greening and browning trends in tundra are also not consistent over time (decadal scale) and can vary across landform/ecosystem types (Lara et al., 2018¹⁶⁴⁵), suggesting interactions between the changing environment and the biological components of the system that control these trends. There is high confidence that increases in summer, spring and winter temperatures lead to tundra greening, as well as increases in growing season length (e.g., Vickers et al., 2016; Myers-Smith and Hik, 2018¹⁶⁴⁶) that are in part linked to reductions in Arctic Ocean sea ice cover (Bhatt et al., 2017¹⁶⁴⁷; Macias-Fauria et al., 2017¹⁶⁴⁸). Other factors that stimulate tundra greening include increases in snow water equivalent and soil moisture (Westergaard-Nielsen et al., 2017¹⁶⁴⁹), increases in active layer thickness (via nutrient availability or changes in moisture), changes in herbivore activity, and to a lesser degree, human use of the land (e.g., Salmon et al., 2016; Horstkotte et al., 2017; Martin et al., 2017; Yu et al., 2017). Research on tundra browning is more limited but suggests causal mechanisms that include changes in winter climate—specifically reductions in snow cover due to winter warming events that expose tundra to subsequent freezing and desiccation—, insect and pathogen outbreaks, increased herbivore grazing and ground ice melting and subsidence that increases surface water (Phoenix and Bjerke, 2016; Bjerke et al., 2017) (medium confidence).

Projections of tundra vegetation distribution across the Arctic by 2050 in response to changing environmental conditions suggest that the areal extent of most tundra types will decrease by at least 50% (Pearson et al., 2013) (medium confidence). Woody shrubs and trees are projected to expand to cover 24–52% of the current tundra region by 2050, or 12–33% if tree dispersal is restricted. Adding to this, the expansion of fire into tundra that has not experienced large-scale disturbance for centuries causes large reductions in soil carbon stocks (Mack et al., 2011), shifts in vegetation composition and productivity (Bret-Harte et al., 2013), and can lead to widespread permafrost degradation (Jones et al., 2015a) at faster rates than would occur by changing environmental conditions alone. In tundra regions, graminoid (grasses and sedges) tundra is projected to be replaced by more flammable shrub tundra in future climate scenarios, and tree migration into tundra could further increase fuel loading (Pastick et al., 2017) (medium confidence).

Similar to tundra, boreal forest vegetation shows trends of both greening and browning over multiple years in different regions across the satellite record (Beck and Goetz, 2011; Ju and Masek, 2016) (high confidence). Here, patterns of changing vegetation are a result of direct responses to changes in climate (temperature, precipitation and seasonality) and other driving factors for vegetation (nutrients, disturbance) similar to what has been reported in tundra. While boreal forest may expand at the northern edge (Pearson et al., 2013), climate projections suggest that it could diminish at the southern edge and be replaced by lower biomass woodland/shrublands (Koven, 2013; Gauthier et al., 2015) (medium confidence). Furthermore, changes in fire disturbance are leading to shifts in landscape distribution of early and late successional ecosystem types, which is also a major factor in satellite trends. Fires that burn deeply into the organic soil layer can alter permafrost stability, hydrology and vegetation. Loss of the soil organic layer exposes mineral soil seedbeds (Johnstone et al., 2009), leading to recruitment of deciduous tree and shrub species that do not establish on organic soil (Johnstone et al., 2010). This recruitment has been shown to shift post-fire vegetation to alternate successional trajectories (Johnstone et al., 2010). Model projections suggest that Alaskan boreal forest soon may cross a point where recent increases in fire activity have made deciduous stands as abundant as spruce stands on the landscape (Mann et al., 2012) (medium confidence). This projected trend of increasing deciduous forest at the expense of evergreen forest is mirrored in Russian and Chinese boreal forests as well (Shakhova et al., 2013; Shuman et al., 2015; Wu et al., 2017) (medium confidence).

Reindeer and caribou (Rangifer tarandus), through their numbers and ecological role as a large-bodied herbivores, are a key driver of Arctic ecology. The seasonal migrations that characterise Rangifer link the coastal tundra to the continental boreal forests for some herds, while others live year-round on the tundra. Population estimates and trends exist for most herds, and indicate that pan-Arctic migratory tundra Rangifer have declined from about 5 million in the 1990s to about 2 million in 2017 (Gunn, 2016¹⁶⁸²; Fauchald et al., 2017a¹⁶⁸³) (high confidence). Numbers have recently increased for two Alaska herds and the Porcupine caribou herd straddling Yukon and Alaska is at a historic high.

There is low confidence in understanding the complex drivers of observed Rangifer changes. Hunting and predation (the latter exacerbated by modification of the landscape for exploration and resource extraction; Dabros et al., 2018¹⁶⁸⁴) increase in importance as populations decline. Climate strongly influences productivity: extremes in heat, drought, winter icing and snow depth reduce Rangifer survival (Mallory and Boyce, 2017¹⁶⁸⁵). Changes in the timing of sea ice formation have direct effects on risks during Rangifer migration via inter-island movement and connection to the mainland (Poole et al., 2010¹⁶⁸⁶). Summer warming is changing the composition of tundra plant communities, modifying the relationship between climate, forage and Rangifer (Albon et al., 2017¹⁶⁸⁷), which also impacts other Arctic species such as musk ox (Ovibos moschatus) (Schmidt et al., 2015¹⁶⁸⁸). As polar trophic systems are highly connected (Schmidt et al., 2017¹⁶⁸⁹), changes will propagate through the ecosystem with effects on other herbivores such as geese and voles, as well as predators such as wolves (Hansen et al., 2013¹⁶⁹⁰; Klaczek et al., 2016¹⁶⁹¹).

In northern Fennoscandia, there are approximately 600,000 semi-domesticated reindeer. Lichen rangelands are key to sustaining reindeer carrying capacity, with variable response to climate change: enhanced summer precipitation increases lichen biomass, while an increase in winter precipitation lowers it (Kumpula et al., 2014¹⁶⁹²). Fire disturbance reduces the amount of pasture available for domestic reindeer and increases predation on herding lands (Lavrillier and Gabyshev, 2017¹⁶⁹³). Later ice formation on waterbodies can impact herding activities (Turunen et al., 2016¹⁶⁹⁴). Ice formation from rain-on-snow events is associated with population changes including cases of catastrophic mass starvation (Bartsch et al., 2010¹⁶⁹⁵; Forbes et al., 2016¹⁶⁹⁶), but there is no evidence of trends in rain-on-snow events (Cohen et al., 2015¹⁶⁹⁷; Dolant et al., 2017¹⁶⁹⁸).

Management of keystone species requires an understanding of pathogens and disease in the context of climate warming, but evidence of changing patterns across northern ecosystems (spanning terrestrial, aquatic, and marine environments) is hindered by an incomplete picture of pathogen diversity and distribution (Hoberg, 2013¹⁶⁹⁹; Jenkins et al., 2013¹⁷⁰⁰; Cook et al., 2017¹⁷⁰¹). Among ungulates, it is virtually certain that the emergence of disease attributed to nematode pathogens has accelerated since 2000 in the Canadian Arctic islands and Fennoscandia (Kutz et al., 2013¹⁷⁰²; Hoberg and Brooks, 2015¹⁷⁰³; Laaksonen et al., 2017¹⁷⁰⁴; Kafle et al., 2018¹⁷⁰⁵). Discovery of the pathogenic bacterium Erysipelothrix rhusiopathiae has been linked to massive and widespread mortality among muskoxen from the Canadian Arctic Archipelago; loss of >50% of the population since 2010 may be attributable to disease interacting with extreme temperature events, although unequivocal links to climate have not been established (Kutz et al., 2015¹⁷⁰⁶; Forde et al., 2016a¹⁷⁰⁷; Forde et al., 2016b¹⁷⁰⁸). Anthrax is projected to expand northward in response to warming, and resulted in substantial mortality events for reindeer on the Yamal Peninsula of Russia in 2016 with mobilisation of bacteria possibly from a frozen reindeer carcass or melting permafrost (Walsh et al., 2018¹⁷⁰⁹). In concert with climate forcing, pathogens are very likely responsible for increasing mortality in Arctic ungulates (muskox, caribou/reindeer) and alteration of transmission patterns in marine food chains, broadly threatening sustainability of subsistence and commercial hunting and fishing and safety of traditional foods for northern cultures at high latitudes (Jenkins et al., 2013¹⁷¹⁰; Kutz et al., 2014¹⁷¹¹; Hoberg et al., 2017¹⁷¹²).

Climate-driven changes in seasonal ice and permafrost conditions influence water quality (high confidence). Shortened duration of freshwater ice cover (more light absorption, increased nutrient input) is expected to result in higher primary productivity (Hodgson and Smol, 2008¹⁷¹³; Vincent et al., 2011¹⁷¹⁴; Griffiths et al., 2017b¹⁷¹⁵) and may also encourage greater methane emissions from Arctic lakes (Greene et al., 2014¹⁷¹⁶; Tan and Zhuang, 2015¹⁷¹⁷). Thaw slumps, active layer detachments and peat plateau collapse affect surface water connectivity (Connon et al., 2014¹⁷¹⁸) and enhance sediment, particulate and solute fluxes in river and stream networks (Kokelj et al., 2013¹⁷¹⁹). The transfer of enhanced nutrients from land to water (driven by active layer thickening and thermokarst processes; Abbott et al., 2015¹⁷²⁰; Vonk et al., 2015¹⁷²¹) has been linked to heightened autotrophic productivity in freshwater ecosystems (Wrona et al., 2016¹⁷²²). Still, there is low confidence in the influence of permafrost changes on dissolved organic carbon, because of competing mechanisms that influence carbon export. Permafrost thaw could contribute to the mobilisation of previously frozen organic carbon (Abbott et al., 2014¹⁷²³; Wickland et al., 2018¹⁷²⁴; Walvoord et al., 2019¹⁷²⁵) thereby enhancing both particulate and dissolved organic carbon export to aquatic systems. Increased delivery of this dissolved carbon from enhanced river discharge to the Arctic Ocean (Section 3.4.3.1.2) can exacerbate regionally extreme aragonite undersaturation of shelf waters (Semiletov et al., 2016¹⁷²⁶) driven by ocean uptake of anthropogenic CO2 (Section 3.2.1.2.4). Conversely, reduced dissolved organic carbon export could accompany permafrost thaw as (1) water infiltrates deeper and has longer residence times for decomposition (Striegl et al., 2005¹⁷²⁷) and (2) the proportion of groundwater (typically lower in dissolved organic carbon and higher in DIC than runoff) to total streamflow increases (Walvoord and Striegl, 2007¹⁷²⁸). Increased thermokarst also has the potential to impact freshwater cycling of inorganic carbon (Zolkos et al., 2018¹⁷²⁹).

Enhanced subsurface water fluxes resulting from permafrost degradation has consequences for inorganic natural and anthropogenic constituents. Emerging evidence suggests large natural stores of mercury (Schuster et al., 2018¹⁷³⁰; St Pierre et al., 2018¹⁷³¹) and other trace elements in permafrost (Colombo et al., 2018¹⁷³²) may be released upon thaw, thereby having effects (largely unknown at this point) on aquatic ecosystems. In parallel, increased development activity in the Arctic is likely to lead to enhanced local sources of anthropogenic chemicals of emerging Arctic concern, including siloxanes, parabens, flame retardants, and per- and polyfluoroalkyl substances (AMAP, 2017c¹⁷³³). For legacy pollutants, there is high confidence that black carbon and persistent organic pollutants (e.g., hexachlorocyclohexanes, polycyclic aromatic hydrocarbons, and polychlorinated biphenyls) can be transferred downstream and affect water quality (Hodson, 2014¹⁷³⁴). Lakes can become sinks of these contaminants, while floodplains can be contaminated (Sharma et al., 2015).

There is high confidence that habitat loss or change due to climate change impact Arctic fishes. Thinning ice on lakes and streams changes the overwintering habitat for aquatic fauna by impacting winter water volumes and dissolved oxygen levels (Leppi et al., 2016¹⁷³⁵). Surface water loss, reduced surface water connectivity among aquatic habitats, and changes to the timing and magnitude of seasonal flows (Section 3.4.1.2) result in a direct loss of spawning, feeding, or rearing habitats (Poesch et al., 2016¹⁷³⁶). Changes to permafrost landscapes have reduced freshwater habitat available for fish and other aquatic biota, including aquatic invertebrates upon which the fish depend for food (Chin et al., 2016¹⁷³⁷). Gullying deepens channels (Rowland et al., 2011¹⁷³⁸; Liljedahl et al., 2016¹⁷³⁹) that otherwise may connect lake habitats occupied by fishes. This can lead to the loss of surface water connectivity, limit fish access to key habitats, and lower fish diversity (Haynes et al., 2014¹⁷⁴⁰; Laske et al., 2016¹⁷⁴¹). Small connecting stream channels, which are vulnerable to drying, provide necessary migratory pathways for fishes, allowing them to access spawning and summer rearing grounds (Heim et al., 2016¹⁷⁴²; McFarland et al., 2017¹⁷⁴³).

Changes to the timing, duration and magnitude of high surface flow events in early and late summer threaten Arctic fish dispersal and migration activities (Heim et al., 2016¹⁷⁴⁴) (high confidence). Timing of important life history events such as spawning can become mismatched with changing stream flows (Lique et al., 2016¹⁷⁴⁵). There is regional evidence that migration timing has shifted earlier and winter egg incubation temperature has increased for pink salmon (Oncorhynchus gorbuscha), directly related to warming (Taylor, 2007¹⁷⁴⁶). While long-term, pan-Arctic data on run timing of fishes are limited, phenological shifts could create mismatches with food availability or habitat suitability in both marine and freshwater environments for anadromous species, and in freshwater environments for freshwater resident species. Changes to the Arctic growing season (Xu et al., 2013a¹⁷⁴⁷) increase the risk of drying of surface water habitats and pose a potential mismatch in seasonal availability of food in rearing habitats.

Freshwater systems across the Arctic are relatively shallow, and thus are expected to warm (high confidence). This may make some surface waters inhospitably warm for cold water species such as Arctic Grayling (Thymallus arcticus) and whitefishes (Coregonus spp.), or may increase the risk of Saprolegnia fungus that appears to have recently spread rapidly, infecting whitefishes at much higher rates in Arctic Alaska than noted in the past (Sformo et al., 2017¹⁷⁴⁸). High infection rates may be driven by stress or nutrient enrichment from thawing permafrost, which increases pathogen virulence with fish (Wedekind et al., 2010¹⁷⁴⁹). Warmer water and longer growing seasons will also affect food abundance because invertebrate life histories and production are temperature and degree-day dependent (Régnière et al., 2012¹⁷⁵⁰). Increased nutrient export from permafrost loss (Frey et al., 2007¹⁷⁵¹), facilitated by warmer temperatures, will likely increase food resources for consumers, but the impact on lower trophic levels within food webs is not clearly understood.

Impacts of climate change on food and water security in the Arctic can be severe in regions where infrastructure (including ice roads), travel and subsistence practices are reliant on elements of the cryosphere such as snow cover, permafrost and freshwater or sea ice (Cochran et al., 2013¹⁸²⁸; Inuit Circumpolar Council-Alaska (ICC-AK), 2015).

There is high confidence in indicators that food insecurity risks are on the rise for Indigenous Arctic peoples. Food is strongly tied to culture, identity, values and ways of life (Donaldson et al., 2010¹⁸²⁹; Cunsolo Willox et al., 2015¹⁸³⁰; Inuit Circumpolar Council-Alaska (ICC-AK), 2015); thus, impacts to food security go beyond access to food and physical health. Food systems in northern communities are intertwined with northern ecosystems because of subsistence hunting, fishing and gathering activities. Environmental changes to animal habitat, population sizes and movement mean that culturally important food species may no longer be found within accessible ranges or familiar areas (Parlee and Furgal, 2012¹⁸³¹; Rautio et al., 2014¹⁸³²; Inuit Circumpolar Council-Alaska (ICC-AK), 2015; Lavrillier et al., 2016¹⁸³³) (Section 3.4.3.2.2). This impacts negatively the accessibility of culturally important local food sources (Lavrillier, 2013; Rosol et al., 2016¹⁸³⁴) that make important contributions to a nutritious diet (Donaldson et al., 2010¹⁸³⁵; Hansen et al., 2013¹⁸³⁶; Dudley et al., 2015¹⁸³⁷). Longer open water seasons and poorer ice conditions on lakes impact fishing options (Laidler, 2012¹⁸³⁸) and waterfowl hunting (Goldhar et al., 2014¹⁸³⁹). Permafrost warming and increases in active layer thickness (Section 3.4.1.3) reduce the reliability of permafrost for natural refrigeration. In some cases these changes have reduced access to, and consumption of, locally resourced food and can result in increased incidence of illness (Laidler, 2012¹⁸⁴⁰; Cochran et al., 2013¹⁸⁴¹; Cozzetto et al., 2013¹⁸⁴²; Rautio et al., 2014¹⁸⁴³; Beaumier et al., 2015¹⁸⁴⁴). These consequences of climate change are intertwined with processes of globalisation, whereby complex social, economic and cultural factors are contributing to a dietary transformation from locally resourced foods to imported market foods across the Arctic (Harder and Wenzel, 2012¹⁸⁴⁵; Parlee and Furgal, 2012¹⁸⁴⁶; Nymand and Fondahl, 2014¹⁸⁴⁷; Beaumier et al., 2015¹⁸⁴⁸). Limiting exposures to zoonotic, foodborne and waterborne pathogens (Section 3.4.3.2.2) depends on accurate and comprehensive data on species diversity, biology and distribution and pathways for invasion (Hoberg and Brooks, 2015¹⁸⁴⁹; Kafle et al., 2018¹⁸⁵⁰).

There is high confidence that changes to travel conditions impact food security through access to hunting grounds. Shorter snow cover duration (Section 3.4.1.1), and changes to snow conditions (such as density) make travel more difficult and dangerous (Laidler, 2012¹⁸⁵¹; Ford et al., 2019¹⁸⁵²). Changes in dominant wind direction and speed reduce the reliability of traditional navigational indicators such as snow drifts, increasing safety concerns (Ford and Pearce, 2012¹⁸⁵³; Laidler, 2012¹⁸⁵⁴; Ford et al., 2013¹⁸⁵⁵; Clark et al., 2016b¹⁸⁵⁶). Permafrost warming, increased active layer thickness and landscape instability (Section 3.4.1.3), fire disturbance and changes to water levels (Section 3.4.1.2) impact overland navigability in summer (Goldhar et al., 2014¹⁸⁵⁷; Brinkman et al., 2016¹⁸⁵⁸; Dodd et al., 2018¹⁸⁵⁹).

There is high confidence that both risks and opportunities arise for coastal communities with changing sea ice and open water conditions. Of particular concern for coastal communities is landfast sea ice (Section 3.3.1.1.5), which creates an extension of the land in winter that facilitates travel (Inuit Circumpolar Council Canada, 2014¹⁸⁶⁰). The floe edge position, timing and dynamics of freeze-up and break-up, sea ice stability through the winter, and length of the summer open water season are important indicators of changing ice conditions and safe travel (Gearheard et al., 2013¹⁸⁶¹; Eicken et al., 2014¹⁸⁶²; Baztan et al., 2017¹⁸⁶³). Warming water temperature, altered salinity profiles, snow properties, changing currents and winds all have consequences for the use of sea ice as a travel or hunting platform (Hansen et al., 2013¹⁸⁶⁴; Eicken et al., 2014¹⁸⁶⁵; Clark et al., 2016a¹⁸⁶⁶). More leads (areas of open water), especially in the spring, can mean more hunting opportunities such as whaling off the coast of Alaska (Hansen et al., 2013¹⁸⁶⁷; Eicken et al., 2014¹⁸⁶⁸). In Nunavut, a floe edge closer to shore improves access to marine mammals such as seals or narwhal (Ford et al., 2013¹⁸⁶⁹). However, these conditions also hamper access to coastal or inland hunting grounds (Hansen et al., 2013¹⁸⁷⁰; Durkalec et al., 2015¹⁸⁷¹), have increased potential for break-off events at the floe edge (Ford et al., 2013¹⁸⁷²), or can result in decreased presence (or total absence) of ice-associated marine mammals with an absence of summer sea ice (Eicken et al., 2014¹⁸⁷³).

Many northern communities rely on ponds, streams and lakes for drinking water (Cochran et al., 2013¹⁸⁷⁴; Goldhar et al., 2013¹⁸⁷⁵; Nymand and Fondahl, 2014¹⁸⁷⁶; Daley et al., 2015¹⁸⁷⁷; Dudley et al., 2015¹⁸⁷⁸; Masina et al., 2019¹⁸⁷⁹), so there is high confidence that projected changes in hydrology will impact water supply (Section 3.4.2.2). Surface water is vulnerable to thermokarst disturbance and drainage, as well as bacterial contamination, the risks of which are increased by warming ground and water temperatures (Cozzetto et al., 2013¹⁸⁸⁰; Goldhar et al., 2013¹⁸⁸¹; Dudley et al., 2015¹⁸⁸²; Masina et al., 2019¹⁸⁸³). Icebergs or old multi-year ice are important sources of drinking water for some coastal communities, so reduced accessibility to stable sea ice conditions affects local water security. Small remote communities have limited capacity to respond quickly to water supply threats, which amplifies vulnerabilities to water security (Daley et al., 2015¹⁸⁸⁴).

Culture and knowledge

Spending time on the land is culturally important for indigenous communities (Eicken et al., 2014¹⁸⁸⁵; Durkalec et al., 2015¹⁸⁸⁶). There is high confidence that daily life is influenced by changes to ice freeze-up and break-up (rivers/lakes/sea ice), snow onset/melt, vegetation phenology, and related wildlife/fish/bird behaviour (Inuit Circumpolar Council-Alaska (ICC-AK), 2015). Inter-generational knowledge transmission of associated values and skills is also influenced by climate change because younger generations do not have the same level of experience or confidence with traditional indicators (Ford, 2012¹⁸⁸⁷; Parlee and Furgal, 2012¹⁸⁸⁸; Eicken et al., 2014¹⁸⁸⁹; Pearce et al., 2015¹⁸⁹⁰). Climate-driven changes undermine confidence in indigenous knowledge holders in regards to traditional indicators used for safe travel and navigation (Parlee and Furgal, 2012¹⁸⁹¹; Golovnev, 2017¹⁸⁹²; Ford et al., 2019¹⁸⁹³).

Economics

The Arctic mixed economy is characterised by a combination of subsistence activities, and employment and cash income. There is low confidence about the extent and nature of impact of climate change on local subsistence activities and economic opportunities across the Arctic (e.g., hunting, fishing, resource extraction, tourism and transportation; see Section 3.2.4) because of high variability between communities (Harder and Wenzel, 2012¹⁸⁹⁴; Cochran et al., 2013¹⁸⁹⁵; Clark et al., 2016b¹⁸⁹⁶; Fall, 2016¹⁸⁹⁷; Ford et al., 2016¹⁸⁹⁸; Lavrillier et al., 2016¹⁸⁹⁹). Longer ice-free travel windows in Arctic seas could lower the costs of access and development of northern resources (delivering supplies and shipping resources to markets) and thus, may contribute to increased opportunities for marine shipping, commercial fisheries, tourism and resource development (Sections 3.2.4.2, 3.2.4.3) (Ford et al., 2012¹⁹⁰⁰; Huskey et al., 2014¹⁹⁰¹; Overland et al., 2017¹⁹⁰²). This has important implications for economic development, particularly in relation to local employment opportunities but also raises concerns of detrimental impacts on animals, habitat and subsistence activities (Cochran et al., 2013¹⁹⁰³; Inuit Circumpolar Council-Alaska (ICC-AK), 2015).

For many polar residents, especially Indigenous peoples, the physical environment underpins social determinants of well-being, including physical and mental health. Changes to the environment impact most dimensions of health and well-being (Parlee and Furgal, 2012¹⁹⁰⁴; Ostapchuk et al., 2015¹⁹⁰⁵). Climate change consequences in polar regions (Sections 3.3.1.1, 3.4.1.2) have impacted key transportation routes (Gearheard et al., 2006¹⁹⁰⁶; Laidler, 2006¹⁹⁰⁷; Ford et al., 2013¹⁹⁰⁸; Clark et al., 2016a¹⁹⁰⁹) and pose increased risk of injury and death during travel (Durkalec et al., 2014¹⁹¹⁰; Durkalec et al., 2015¹⁹¹¹; Clark et al., 2016b¹⁹¹²; Driscoll et al., 2016¹⁹¹³).

Foodborne disease is an emerging concern in the Arctic because warmer waters, loss of sea ice (Section 3.3.1.1) and resultant changes in contaminant pathways can lead to bioaccumulation and biomagnification of contaminants in key food species. While many hypothesised foodborne diseases are not well studied (Parkinson and Berner, 2009¹⁹¹⁴), foodborne gastroenteritis is associated with shellfish harvested from warming waters (McLaughlin et al., 2005¹⁹¹⁵; Young et al., 2015¹⁹¹⁶). Mercury presently stored in permafrost (Schuster et al., 2018¹⁹¹⁷) has potential to accumulate in aquatic ecosystems.

Climate change increases the risk of waterborne disease in the Arctic via warming water temperatures and changes to surface hydrology (Section 3.4.1.2) (Parkinson and Berner, 2009¹⁹¹⁸; Brubaker et al., 2011¹⁹¹⁰; Dudley et al., 2015¹⁹²⁰). After periods of rapid snowmelt, bacteria can increase in untreated drinking water, with associated increases in acute gastrointestinal illness (Harper et al., 2011¹⁹²¹). Consumption of untreated drinking water may increase duration and frequency of exposure to local environmental contaminants (Section 3.4.3.2.3) or potential waterborne diseases (Goldhar et al., 2014¹⁹²²; Daley et al., 2015¹⁹²³). The potential for infectious gastrointestinal disease is not well understood, and there are concerns in relation to the safety of storage containers of raw water in addition to the quality of the source water itself (Goldhar et al., 2014¹⁹²⁴; Wright et al., 2018¹⁹²⁵; Masina et al., 2019¹⁹²⁶).

Climate change has negatively affected place attachment via hunting, fishing, trapping and traveling disruptions, which have important mental health impacts (Cunsolo Willox et al., 2012¹⁹²⁷; Durkalec et al., 2015¹⁹²⁸; Cunsolo and Ellis, 2018¹⁹²⁹). The pathways through which climate change impacts mental wellness in the Arctic varies by gender (Bunce and Ford, 2015¹⁹³⁰; Ostapchuk et al., 2015¹⁹³¹; Bunce et al., 2016¹⁹³²) and age (Petrasek-MacDonald et al., 2013¹⁹³³; Ostapchuk et al., 2015¹⁹³⁴). Emotional impacts of climate-related changes in the environment were significantly higher for women compared to men, linked to concern for family members (Ostapchuk et al., 2015¹⁹³⁵). However, men are also vulnerable due to gendered roles in subsistence and cultural activities (Bunce and Ford, 2015¹⁹³⁶). In coastal areas, sea ice means freedom for travel, hunting and fishing, so changes in sea ice affect the experience of and connection with place. In turn, this influences individual and collective mental/emotional health, as well as spiritual and social vitality according to relationships between sea ice use, culture, knowledge and autonomy (Cunsolo Willox et al., 2013a¹⁹³⁷; Cunsolo Willox et al., 2013b¹⁹³⁸; Gearheard et al., 2013¹⁹³⁹; Durkalec et al., 2015¹⁹⁴⁰; Inuit Circumpolar Council-Alaska (ICC-AK), 2015).

Permafrost is undergoing rapid change (Section 3.4.1.2), creating challenges for planners, decision makers and engineers (AMAP, 2017d¹⁹⁴¹). The observed changes in the ground thermal regime (Romanovsky et al., 2010¹⁹⁴²; Romanovsky et al., 2017¹⁹⁴³; Biskaborn et al., 2019¹⁹⁴⁴) threaten the structural stability and functional capacities of infrastructure, in particular that which is located on ice rich frozen ground. Extensive summaries of construction damages along with adaptation and mitigation strategies are available (Larsen et al., 2014¹⁹⁴⁵; Dore et al., 2016¹⁹⁴⁶; AMAP, 2017d¹⁹⁴⁷; Pendakur, 2017¹⁹⁴⁸; Shiklomanov et al., 2017a¹⁹⁴⁹; Shiklomanov et al., 2017b¹⁹⁵⁰; Vincent et al., 2017¹⁹⁵¹).

Projections of climate and permafrost suggest that a wide range current infrastructure will be impacted by changing conditions (medium confidence). A circumpolar study found that approximately 70% of infrastructure (residential, transportation and industrial facilities), including over 1200 settlements (~40 with population more than 5000) are located in areas where permafrost is projected to thaw by 2050 under RCP4.5 (Hjort et al., 2018¹⁹⁵²). Regions associated with the highest hazard are in the thaw-unstable zone characterised by relatively high ground-ice content and thick deposits of frost susceptible sediments (Shiklomanov et al., 2017b¹⁹⁵³). By 2050, these high hazard environments contain one-third of existing pan-Arctic infrastructure. Onshore hydrocarbon extraction and transportation in the Russian Arctic are at risk: 45% of the oil and natural gas production fields in the Russian Arctic are located in the highest hazard zone.

In a regional study of the state of Alaska, cumulative expenses projected for climate-related damage to public infrastructure totalled USD 5.5 billion between 2015 and 2099 under RCP8.5 (Melvin et al., 2017¹⁹⁵⁴). The top two causes of damage related costs were projected to be road flooding from increased precipitation, and building damage associated with near-surface permafrost thaw. These costs decreased by 24% to USD 4.2 billion for the same time frame under RCP4.5, indicating that reducing greenhouse gas emissions globally could lessen damages (Figure 3.13). In a related study that included these costs and others, as well as positive gains from climate change in terms of a reduction in heating costs attributable to warmer winter, annual net costs were still USD 340–700 million, or 0.6–1.3% of Alaska’s GDP, suggesting that climate change costs will outweigh positive benefits, at least for this region (Berman and Schmidt, 2019¹⁹⁵⁵).

Winter roads (snow covered ground and frozen lakes) are distinct from the infrastructure considered earlier, but have a strong influence on the reliability and costs of transportation in some remote northern communities and industrial development sites (Parlee and Furgal, 2012¹⁹⁵⁶; Huskey et al., 2014¹⁹⁵⁷; Overland et al., 2017¹⁹⁵⁸). For these communities, changing lake and river levels and the period of safe ice cover all affect the duration of use of overland travel routes and inland waterways, with associated implications for increased travel risks, time, and costs (Laidler, 2012¹⁹⁵⁹; Ford et al., 2013¹⁹⁶⁰; Goldhar et al., 2014¹⁹⁶¹). There have been recent instances of severely curtailed ice road shipping seasons due to unusually warm conditions in the early winter (Sturm et al., 2017¹⁹⁶²). While the impact of human effort on the maintenance of winter roads is difficult to quantify, a reduction in the operational time window due to winter warming is projected (Mullan et al., 2017¹⁹⁶³).

The Arctic Council

International cooperation on issues related to climate change in the Arctic mainly occurs at the Arctic Council (herein ‘the Council’), and consequently in important areas of its mandate: the (marine) environment and scientific research (Morgera and Kulovesi, 2016²¹⁷¹; Tesar et al., 2016a²¹⁷²; Wehrmann, 2016²¹⁷³; Young, 2016²¹⁷⁴). The Council is composed of eight Arctic States and six Permanent Participants representing organisations of Arctic Indigenous peoples. Observers status is open to: non-Arctic states, intergovernmental and inter-parliamentary organisations, global and regional non-governmental organisations (NGOs). The Council is an example of cooperation through soft law, a unique institutional body that does not possess a legal personality and is neither an international law nor a completely state-centric institution. However, it is acting state-centric and increasingly operating in a context of the Arctic affected by a changing climate, globalisation and transnationalism (Baker and Yeager, 2015²¹⁷⁵; Cassotta et al., 2015²¹⁷⁶; Pincus and Speth, 2015²¹⁷⁷) (medium confidence). In 2013, China, South Korea, Italy, Japan, India and Singapore joined France, Germany, the Netherlands, Poland, Spain and the UK as Observers states to the Arctic Council; Switzerland was granted Observer status in 2017. Non-Arctic States are stimulating the Council towards adopting a new approach for Arctic governance that would leave greater space for their participation.

Despite lacking the role to enact hard law, three binding agreements were negotiated under the auspices of the Council (in its task forces), the latest of which is the Agreement on Enhancing International Arctic Scientific Cooperation, which is an indication that the Council is preparing a regulatory role to respond to climate change in the Arctic using hard-law instruments (Morgera and Kulovesi, 2016²¹⁷⁸; Shapovalova, 2016²¹⁷⁹). Through organising the Task Force on Black Carbon and Methane (Morgera and Kulovesi, 2016²¹⁸⁰), the Council has catalysed action on short-lived climate forcers as the task force was followed by the adoption in 2015 of the Arctic Council Framework for Action on Enhanced Black Carbon and Methane Emission Reductions. In this non-legally binding agreement, Arctic States lay out a common vision for national and collective action to accelerate decline in black carbon and methane emissions (Shapovalova, 2016²¹⁸¹). The Council thereby moved from merely assessing problems to attempting to solve them (Baker and Yeager, 2015²¹⁸²; Young, 2016²¹⁸³; Koivurova and Caddell, 2018²¹⁸⁴). While mitigation of global emissions from fossil fuels requires global cooperation, progress with anthropogenic emissions of short-term climate forcers (such as black carbon and methane) may be achieved through smaller groups of countries (Aakre et al., 2018²¹⁸⁵). However, even though the Council has also embraced the Ecosystem Approach, it does not have a mandate to manage climate-related risks and impacts, or apply a precautionary approach, on fisheries issues.

Several studies have shown that the Council has the potential to enhance internal coherence in the current, fragmented landscape of multi-regulatory governance by providing integrated leadership. However, it is about as likely as not that the Council could play a strong role in combatting global climate problems and operating successfully within the climate transnational context unless it goes through restructuring and reconfiguration (Stokke, 2013²¹⁸⁶; Baker and Yeager, 2015²¹⁸⁷; Pincus and Speth, 2015²¹⁸⁸; Cassotta et al., 2016²¹⁸⁹; Tesar et al., 2016a²¹⁹⁰; Wehrmann, 2016²¹⁹¹; Young, 2016²¹⁹²; Koivurova and Caddell, 2018²¹⁹³). 

The future of the governance of the changing Arctic Ocean, including the role of the Council will also depend on the implications of the development for a new agreement on the Conservation and Sustainable use of Marine Biodiversity of Areas beyond National Jurisdictions under the UNCLOS (Baker and Yeager, 2015²¹⁹⁴; De Lucia, 2017; Nengye et al., 2017²¹⁹⁵; Koivurova and Caddell, 2018²¹⁹⁶) (medium confidence).

The Antarctic Treaty System

The Antarctic Treaty System (ATS) is the collective term for the Antarctic Treaty and related agreements. The ATS regulates international relations with respect to Antarctica. 54 countries have acceded to the Treaty and 29 of them participate in decision making as Consultative Parties. 27 countries are Party to the CCAMLR, and 40 have ratified the Protocol on Environmental Protection to the Antarctic Treaty. The importance of understanding, mitigating and adapting to the impacts of changes to the Southern Ocean and Antarctic cryosphere has been realised by all of the major bodies responsible for governance in the Antarctic region (south of 60°S). The Antarctic Treaty Consultative Parties agreed that a Climate Change Response Work Programme would address these matters (ATCM, 2016²¹⁹⁷). This led to the establishment of the Subsidiary Group of the Committee for Environmental Protection on Climate Change Response (ATCM, 2017²¹⁹⁸). By contrast, consensus is currently limiting work programme-level responses to climate change by CCAMLR (2017a), while opportunities exist to incorporate climate concerns into mechanisms for implementation and monitoring aimed to conserve ecosystems and the environment (Brooks et al., 2018²¹⁹⁹)

The Antarctic Treaty Consultative Parties, through the Committee for Environmental Protection (CEP) and its Subsidiary Group of the Committee for Environmental Protection on Climate Change Response, continue to work closely with the Scientific Committee on Antarctic Research, the Council of Managers of National Antarctic Programs, the International Association of Antarctica Tour Operators and other NGOs to understand, mitigate and adapt to impacts associated with changes to the Southern Ocean and Antarctic cryosphere. Understanding, mitigating and adapting to climate change are among the key priorities identified for research in the region (Kennicutt et al., 2014a²²⁰⁰; Kennicutt et al., 2014b²²⁰¹) and nationally funded bilateral and multilateral projects are underway.

Several studies show that informal actors of the Arctic can influence decision making process of the Council and shift the Council towards more cooperation with different actors to enhance the co-production of knowledge (Duyck, 2011²²⁰²; Makki, 2012²²⁰³; Keil and Knecht, 2017²²⁰⁴). Recently, non-state observers at the Council, such as the World Wide Fund for Nature  and the Circumpolar Conservation Union have played a role in raising awareness on climate change responses and contributing to the work of the Council’s Working Groups and Expert Groups (Keil and Knecht, 2017²²⁰⁵). 

Within the Antarctic Treaty System, several non-state actors play a major role in providing advice and influencing the governance of Antarctica and the Southern Ocean. Among the most prominent actors are formal observers such as the Scientific Committee on Antarctic Research, and invited experts such as the International Association of Antarctica Tour Operators and the Antarctic and Southern Ocean Coalition. At meetings of CCAMLR, the Scientific Committee’s 2009 report on Antarctic Climate Change and the Environment (Turner et al., 2009²²⁰⁶) precipitated an Antarctic Treaty Meeting of Experts on Climate Change in 2010 (Antarctic Treaty Meeting of Experts, 2010²²⁰⁹). The outcomes of the meeting led the Antarctic Treaty’s Committee for Environmental Protection to develop a Climate Change Response Work Programme (ATCM, 2017²²¹⁰)

Community-based monitoring (CBM) in the Arctic has emerged as a practice of great interest because of its potential to link western ways of knowing with local knowledge and indigenous knowledge (Retter et al., 2004; Johnson et al., 2015a; Johnson et al., 2015b; Kouril et al., 2016; AMAP, 2017a; Williams et al., 2018). In several CBM programs, innovative approaches using the internet, mobile phones, hand-held information devices, and camera-equipped GPS units are capturing, documenting and communicating local observations of change (Brubaker et al., 2011; Brubaker et al., 2013). The integration of community observations with instrument-based observations and its use in research has proven challenging, with technical and cultural issues (Griffith et al., 2018). Execution of CBM programs in the Arctic has also proven to be labour intensive and difficult to sustain, requiring long-term financial support, agreements specifying data ownership, sufficient human capital, and in some cases, the involvement of boundary organisations that provide technical support (Pulsifer et al., 2012; Eicken et al., 2014) and link CBM with governance (CAFF, 2015b; Robards et al., 2018). As is the case in all knowledge production, power relationships (i.e., who decides what is a legitimate observation, who has access to resources for involvement and who benefits) have been challenging where the legitimacy of local knowledge and indigenous knowledge is questioned (e.g., Pristupa et al., 2018). There is high agreement and limited evidence that CBM facilitates knowledge co-production and resilience building. More analyses of Arctic communities and their institutional capabilities related to CBM are needed to evaluate the potential of these observation systems, and experimentation and innovation may help determine how CBM can more effectively inform decision making beyond the community (Johnson et al., 2015a; Johnson et al., 2015b) (medium confidence).

Regime shifts are especially important in polar regions where there are limited data and where rapid directional change suggests the possibility of crossing thresholds that may dramatically alter the flow of ecosystem services (ARR, 2016). Better understanding of the thresholds and dynamics of regime shifts (i.e., SES state changes) is especially important for resilience building (ARR, 2016; Biggs et al., 2018; Rocha et al., 2018) (high confidence). While polar regime shifts have been documented (Biggs et al., 2018), most are poorly understood and rarely predictable (Rocha et al., 2018) (high confidence). Moreover, the focus on Arctic regime shifts to date has been on almost entirely on biophysical state changes that impact social systems. A limited number of studies have examined social regime shifts and fewer the feedbacks of social regimes shifts on ecosystems (Gerlach et al., 2017). Future needs for advancing knowledge of regime shifts include: 1) continued and refined updating of details on past regimes shifts, 2) structured comparative analysis of these phenomena to ascertain common patterns and variation, 3) greater investment in research resources on potential large-scale regime shifts, and 4) great attention on how social and economic change may affect ecosystems (ARR, 2016; Biggs et al., 2018).

Well-crafted and effectively communicated indicators of polar geophysical, ecological and human systems have the potential to make complex issues more easily understood by society, including local residents and policy makers seeking to assess the implication of climate change (Petrov et al., 2016; Carson and Sommerkorn, 2017) (medium confidence). Having indicators of change is no guarantee they will be used; access to information, awareness of changing conditions, and the motivation to act are also important (e.g., van der Linden et al., 2015).

Indicators of the state of polar geophysical systems, biodiversity, ecosystems and human well-being are monitored as part of polar programs. For example, indicators are reported by the Arctic Council working groups Arctic Monitoring and Assessment Programme and Conservation of Arctic Flora and Fauna (e.g., Odland et al., 2016; CAFF, 2017; Box et al., 2019), the International Arctic Social Science Association (e.g., AHDR, 2014), the CCAMLR Ecosystem Monitoring Programme (e.g., Reid et al., 2005) and the Southern Ocean Observing System (e.g., Meredith et al., 2013).

There is limited development of indicators of social-ecological resilience (Jarvis et al., 2013; Carson and Sommerkorn, 2017). As well, indicators of human adaptive capacity are typically based on qualitative case studies with limited quantitative data, and thus have limited comparability and generalisability (Ford and King, 2013; Petrov et al., 2016; Berman et al., 2017) (high confidence). The identification and on-going use of indicators of social-ecological resilience are theoretically best achieved through highly participatory processes that engage stakeholders of a locale, with those processes potentially resulting in self-reflection and actions that improve adaptive capacity (Quinlan et al., 2016; Carson and Sommerkorn, 2017), however, this is untested empirically (low confidence).

Participatory scenario analysis is a quickly evolving and widely used practice in polar regions, and has proven particularly useful for supporting climate adaptation at multiple scales when it uses a social-ecological perspective (ARR, 2016; AMAP, 2017a²²⁷³; Crépin et al., 2017²²⁷⁴; Planque et al., 2019²²⁷⁵) (medium confidence). While there are technical dimensions in scenario analysis and planning (e.g., the building of useful simulation models that capture and communicate nuanced social-ecological system dynamics such as long-fuse big bang processes, pathological dynamics, critical thresholds, and unforeseen processes (Crépin et al., 2017), there are also creative aspects, such as the use of art to help in the visualisation of possible future (e.g., Planque et al., 2019). 

Participatory scenario analysis has been applied to various problem areas related to climate change responses in the polar regions. Applications demonstrate the utility of the practice for identifying possible local futures that consider climate change or socioeconomic pathways (e.g., in Alaska, Ernst and van Riemsdijk, 2013; and in Eurasian reindeer-herding systems, van Oort et al., 2015; Nilsson et al., 2017²²⁷⁶) and interacting drivers of change (e.g., in Antarctica; Liggett et al., 2017²²⁷⁷). Scenario analysis proved helpful for stakeholders with different expertise and perspectives to jointly develop scenarios to inform ecosystem-based management strategies and adaptation options (e.g., in the Barents region; Nilsson et al., 2017²²⁷⁸; Planque et al., 2019²²⁷⁹) and to identify research needs (e.g., in Alaska; Vargas-Moreno et al., 2016²²⁸⁰), including informing and applying climate downscaling efforts (e.g., in Alaska; Ernst and van Riemsdijk, 2013²²⁸¹). 

A review of scenario analysis in the Arctic, however, found that while the practice is widespread and many are using best practice methods, less than half scenarios programs incorporated climate projections and that those utilising a backcasting approach had higher local participation than those only using forecasting (Flynn et al., 2018²²⁸²). It noted that integrating different knowledge systems and attention to cultural factors influence program utility and acceptance. Planque et al. (2019)²²⁸³ also found that most participating stakeholders had limited experience using scenario analysis, suggesting the importance of process methods for engaging stakeholders when exploring possible, likely, and desirable futures. The long-term utility of this practice in helping stakeholders engage with each other to envision possible futures and be forward thinking in decision making will depend on the science of climate projections, further development of decision support systems to inform decision makers, attention to cultural factors and worldview, as well as refinement of processes that facilitate participants’ dialogue (medium confidence).

Structured decision making (SDM) is an emerging practice used with stakeholders to identify alternative actions, evaluate trade-offs, and inform decisions in complex situations (Gregory et al., 2012²²⁸⁴). Few SDM processes have been undertaken in polar regions, with most as exploratory demonstration projects led by researchers. These have included indigenous residents and researchers identifying trade-offs and actions related to subsistence harvesting in a changing environment (Christie et al., 2018²²⁸⁵) stakeholder interviews to show how a ‘triage method’ can link community monitoring with community needs and wildlife management priorities (Wheeler et al., 2018²²⁸⁶), and the application of multi-criteria decision analysis to address difficult decisions related to mining opportunities in Greenland (Trump et al., 2018²²⁸⁷). The Decision Theater North at the University of Alaska is also being explored as an innovative method of decision support (Kofinas et al., 2016²²⁸⁸). SDM may have potential in creating climate resilience pathways in polar regions (low confidence), but there is currently limited experience with its application.

‘Adaptive Ecosystem Governance’ differs from conventional resource management or integrated ecosystem management (Chapin III et al., 2009²²⁹³; Chapin III et al., 2010²²⁹⁴; Chapin III et al., 2015²²⁹⁵), with a strong focus on trajectories of change (i.e., emergence), implying that maintaining ecosystems in a state of equilibrium is not possible (Biggs et al., 2012²²⁹⁶; ARR, 2016). This approach strengthens response options by maintaining or increasing resource diversity (to support human adaptation) and biological diversity (to support ecosystem adaptation) (Biggs et al., 2012²²⁹⁷; Chapin III et al., 2015²²⁹⁸; Quinlan et al., 2016²²⁹⁹) (high confidence). Adaptive ecosystem governance emphasises iterative social learning processes of observing, understanding and acting with collaborative partnerships, such as adaptive co-management arrangements currently used in regions of the Arctic (Armitage et al., 2009²³⁰⁰; Dale and Armitage, 2011²³⁰¹; Chapin III et al., 2015²³⁰²; Arp et al., 2019²³⁰³). This approach is also currently realised through adaptive management of Arctic fisheries in Alaska that combines annual measures and within-season provisions informed by assessments of future ecosystem trends (Section 3.5.2.1), and the use of simulation models with Canadian caribou co-management boards to assess the cumulative effects of proposed land use change with climate change (Gunn et al., 2011²³⁰⁴; Russell, 2014a²³⁰⁵; Russell, 2014b²³⁰⁶). Linking these regional efforts to pan-polar programs can enhance resilience building cross multiple scales (e.g., Gunn et al., 2013) (medium confidence).

Shifts in the distribution, abundance and human use of species and populations due to climate-induced cryosphere and ocean change, concurrent with land use changes, increase the risks to ecosystem health and biodiversity (Kaiser et al., 2015²³⁰⁷). Building resilience in these challenging conditions follows from spatial planning for biodiversity that links multiple scales and considers how impacts to ecosystems may materialise in social-ecological systems elsewhere (Bengtsson et al., 2003²³⁰⁸; Cumming, 2011²³⁰⁹; Allen et al., 2016²³¹⁰). Developing pathways for spatial resilience in polar regions involves systematic planning and designating networks of protected areas to protect connected tracts of representative habitats, and biologically and ecologically significant features (Ban et al., 2014²³¹¹). Protected area networks that combine both spatially rigid and spatially flexible regimes with climate refugia can support ecological resilience to climate change by maintaining connectivity of populations, foodwebs, and the flow of genes across scales (McLeod et al., 2009²³¹²). This approach reduces direct pressures on biodiversity, and thus gives biological communities, populations and ecosystems the space to adapt (Nyström and Folke, 2001²³¹³; Hope et al., 2013²³¹⁴; Thomas and Gillingham, 2015²³¹⁵) (medium confidence). Networks of protected areas are now being planned (Solovyev et al., 2017²³¹⁶) and implemented (Juvonen and Kuhmonen, 2013²³¹⁷) in the marine and terrestrial Arctic, respectively; expanding the terrestrial protected area network in Antarctica is discussed (Coetzee et al., 2017²³¹⁸). The planning of protected area networks in polar regions is currently an active topic of international collaboration in both polar regions (Arctic Council, 2015b²³¹⁹; CCAMLR, 2016a²³²⁰; Wenzel et al., 2016²³²¹). Designating marine protected area networks contributes to achieving Sustainable Development Goal 14 and the Aichi Targets of the CBD but is often contested due to competing interests for marine resources.

Incorporating measures of ecosystem services into assessments is key in integrating environmental, economic, and social policies that build resilience to climate change in polar regions (CAFF, 2015a²³²²; Malinauskaite et al., 2019²³²³; Sarkki and Acosta García, 2019²³²⁴) (high confidence). Currently, there is limited recognition of the wide range of benefits people receive from polar ecosystems and a lack of management tools that demonstrate their benefits in decision-making processes (CAFF, 2015a²³²⁵). The concept of ecosystem services is increasingly used in the Arctic, yet there continues to be significant knowledge gaps in mapping, valuation, and the study of the social implications of changes in ecosystem services. There are few Arctic examples of the application of ecosystem services in management (Malinauskaite et al., 2019²³²⁶). A strategy of ecosystem stewardship, therefore, is to maintain a continued flow of ecosystem services, recognising how their benefits provide incentives for preserving biodiversity, while also ensuring options for sustainable development and ecosystem-based adaptation (Chapin III et al., 2015²³²⁷; Guerry et al., 2015²³²⁸; Díaz et al., 2019²³²⁹). Arctic stewardship opportunities at landscape, seascape, and community scales to a great extent lie in supporting culturally engrained (often traditional indigenous) values of respect for land and animals, and reliance on the local environment through the sharing of knowledge and power between local users of renewable resources and agencies responsible for managing resources (Mengerink et al., 2017²³³⁰) (high confidence). In the Antarctic, ecosystem stewardship is dependent on international formally-defined and informally-enacted cooperation, and the recognition of its service to the global community (Section 3.5.3.2).