Author/s Kyle Clem1, Rob Massom2, Sharon Stammerjohn3, Phillip Reid4 Victoria University of Wellington, New Zealand Australian Antarctic Division, Australian Antarctic Program Partnership, and Australian Research Council Australian Centre of Excellence in Antarctic Science, Tasmania, Australia Institute of Arctic and Alpine Research, University of Colorado Boulder, USA Australian Bureau of Meteorology and Australian Antarctic Program Partnership, Tasmania, Australia Brief Overview Antarctic sea ice, in the form of immobile coastal “fast ice” and the more extensive moving pack ice (see Antarctic Sea Ice #1), supports one of the most extensive and productive ecosystems on Earth and is crucial to the structure and function of Southern Ocean marine ecosystems that are highly attuned to its presence and seasonal rhythms. Changes in Antarctic sea-ice coverage and seasonality, thickness (and snow cover depth) and properties have wide-ranging consequences with cascading effects across food chains. These effects include: change in the phenology of phytoplankton and ice algal blooms; shifts in species composition, distribution and abundance, leading to trophic mismatches in both time and space that impact ecosystem structure and function; changes in the breeding and foraging distribution of sea ice-obligate predators such as Adélie penguins; and the incursion of sub-Antarctic and/or invasive warmer-climate marine species. Looking to the future, sea-ice coverage is predicted to significantly decrease by the end of this century in response to anthropogenic warming (see Antarctic Sea Ice #3), leading to significant reductions in ice-associated primary production and sea ice-dependent species – including Antarctic krill, Antarctic Silverfish, Adélie and Emperor penguins, Weddell and other pack-ice seals, and southern minke and other whale species. Detailed Overview Biological and Ecological Roles of Antarctic Sea Ice Antarctic sea ice is one of the most extensive, seasonally variable, and productive habitats on Earth (Arrigo and Thomas 2004). The ice and its snow cover create a unique and highly-dynamic environment at the atmosphere-ocean interface – and over an area of the Southern Ocean that varies from ~2-4 million km2 in summer to ~19-20 million km2 in winter with a distinct annual cycle (see Antarctic Sea Ice #1). Although relatively thin (a few centimetres to metres thick), this ice substrate entails a crucially-important home, shelter/refuge, breeding platform and food source for a plethora of biota that are highly adapted to, and reliant on, its presence and seasonal rhythms (Massom and Stammerjohn 2010). These range from microbes (including microscopic algae and bacteria) through pelagic herbivores, such as krill, to fish, seabirds (including penguins), seals and whales (Thomas 2017). At the base of the food web, Antarctic sea ice is a fundamental driver of Southern Ocean primary production (Lizotte 2001). The ice forms a major habitat for ice algae and other microbial species that proliferate at high concentrations within and on the underside of the sea ice compared to the underlying water column (Arrigo 2017; Caron et al. 2017). This is due to a combination of: 1) an availability of sufficient key nutrients such as nitrate and silicate (Meiners and Michel 2017); and 2) enhanced exposure to relatively high light levels at/near the ocean surface. Particularly high concentrations of algae typically occur in the permeable and porous sea ice bottom or “skeletal layer” where there is a steady supply of nutrient-rich seawater (Arrigo 2017). Elsewhere within the sea ice, nutrient replenishment over time occurs via convection in microscopic inter-connected brine channels that permeate the ice matrix (Meiners and Michel 2017), by flooding of the sea ice surface when a heavy snow cover submerges the surface below the sea surface, and by upward brine rejection during bottom sea-ice formation (Wakatsuchi and Ono 1983; Fripiat et al. 2017). Interior algal communities, while typically lower in biomass than underside communities, can also make a significant contribution to sea-ice primary production (Meiners et al. 2012). The thickness and properties (e.g. density and particle size) of the snow cover also determine the intensity and spectral composition of light available for primary production both within the sea ice and just below the sea ice cover (Perovich 1990; Arndt et al. 2017). As the sea ice melts back each spring-summer, the release of ice algae, freshwater, and nutrients (including iron, an important micro-nutrient) into the upper water column fuels the formation of intense phytoplankton blooms both seaward of the retreating sea ice edge (Smith and Nelson 1986) and within open coastal polynyas (Arrigo et al. 2008). Blooms can also occur throughout the pack ice, but these remain unresolved by satellite ocean-colour sensors (Massom et al. 2006). The ice algae in turn represent a critical food source for pelagic herbivores such as Antarctic krill (Euphausia superba; krill hereafter), especially during times of the year when other food is extremely scarce e.g., winter (Massom et al. 2006; Meyer et al. 2017). The sea ice underside environment also forms an important refuge for juvenile krill that are often found within nooks and crannies of the rafted sea ice cover (Hamner et al. 1989; Frazer 2002). Krill populations are of crucial importance as a keystone of the high-latitude Southern Ocean food web (Bluhm et al. 2017). In addition to sustaining krill, sea ice algae can sink to the seabed when the sea ice melts, and this serves as the primary food source for coastal benthic (seabed) habitats (Clark et al. 2017). Moving up the food chain, Antarctic mammal and bird species rely on sea ice for foraging, breeding, resting, escaping predators, and moulting. Four seal species (i.e., Crabeater Lobodon carcinophaga], Leopard Hydrurga leptonyx, Ross Ommatophoca rossii and Weddell Leptonychotes weddellii) are closely tied to sea ice (with the Weddell Seal also breeding on coastal fast ice), while the southern elephant seal (Mirounga leonina) and Antarctic fur seal (Arctocephalus gazella) migrate seasonally to the Antarctic sea ice zone to feed (Siniff et al. 2008; Bester et al. 2017). Antarctic minke (Balaenoptera bonaerensis) and blue (Balaenoptera musculus) whales, the killer whale (Orcinus orca), humpback whale (Megaptera novaeangliae) and the southern bottlenose whale (Hyperoodon planifrons) also depend on the Antarctic sea ice zone for foraging (Thomisch et al. 2016; Bester et al. 2017; Andrews-Goff et al. 2018). Of Antarctic bird species, four are “ice obligate” (reliant on sea ice year-round), namely Adélie (Pygoscelis adeliae) and Emperor (Aptenodytes forsteri) penguins and Antarctic (Thalassoica antarctica) and snow (Pagodroma nivea) petrels (Ainley et al. 2017). In addition and with the exception of one land-based colony in East Antarctica (Wienecke 2012), Emperor penguins breed on coastal fast ice, which must remain in place from at least May to December for successful chick rearing, fledging and moulting to occur (Massom et al. 2009; Fretwell et al. 2012). At the same time, however, too extensive and persistent fast ice can substantially reduce breeding success (Massom et al. 2009), due to the increased distance to open water resulting in fewer forays to feed chicks. Adélie and Emperor penguins must also remain near or within sea ice-covered areas during summer (where air temperatures remain cool), as they begin to experience heat stress at temperatures greater than 2°C (Ainley et al. 2017). Particularly high concentrations of Antarctic seabird and marine mammal species occur both: (i) at the ice edge and in the marginal ice zone or MIZ (the highly-dynamic outer part of the sea ice zone that is strongly affected by wind and waves and is characterised by an unconsolidated sea ice cover consisting of smaller floes); and (ii) in coastal polynyas. Both the MIZ and polynyas support high productivity in upper trophic levels (Ainley et al. 1998; Karnovsky et al. 2007), while recurrent and persistent polynyas (Massom et al. 1998) and extensive networks of leads (episodic linear openings) also enable apex predators to remain far within the sea ice zone in winter as well as in other seasons (Massom 1988). Challenges: Change and Variability Much of our current understanding of the impacts of sea ice change and variability – which displays strong regional differences around Antarctica (see Antarctic Sea Ice #3) – comes from the western Antarctic Peninsula. This is due in large part to the establishment there of coordinated cross-disciplinary monitoring initiatives such as the Palmer Long-Term Ecological Research (LTER) program (Smith et al. 1995; pal.lternet.edu). Declining sea-ice extent and annual duration along the western Antarctic Peninsula since the late-1970s (Stammerjohn and Maksym 2017; see also Fig. 1 in Antarctic Sea Ice #3) has had major and cascading effects on food-web structure and function and on biodiversity, with substantial impacts on primary production, krill, fish, birds and marine mammals (Ducklow et al. 2013; Henley et al. 2019). At the base of the food chain, there has been a significant reduction in primary production in the sea-ice environment (McClintock et al. 2008) and a shift to smaller phytoplankton species that are a less suitable food source for krill (Schloss et al. 2012). The response of krill to regional change in sea-ice seasonality has been complex, with populations in the northern Antarctic Peninsula sector showing significant decreases (Atkinson et al. 2019), while krill populations in the mid-to-southwestern Antarctic Peninsula region have exhibited no significant long-term directional change (Steinberg et al. 2015). Quetin and Ross (2009) have shown that the later the timing of annual sea-ice formation and advance, the lower the food availability for krill, and the lower the growth rates and predicted survival rates of larval/juvenile krill. There has also been a decrease of Silverfish in the diet of predators in the western Antarctic Peninsula (Emslie and Patterson 2007), in concert with an apparent decline in larval Silverfish abundance along the western Antarctic Peninsula (Quetin and Ross 2009; Corso et al. 2022). Silverfish, the most abundant fish in Antarctic coastal pelagic waters and a major food source of Adélie and Emperor penguins, rely on sea ice for spawning and as a nursery (Vacchi et al. 2004), and are vulnerable to sea ice change (Mintenbeck and Torres 2017). In terms of higher vertebrate species, there has been a regional decrease in Weddell seals on the western Antarctic Peninsula due to a reduction in summer fast-ice areas during the breeding season (Costa et al. 2010). In addition, southward shifts in the range of southern elephant seal populations have occurred (Costa et al. 2010) along with decreases in their northern range (McIntyre et al. 2011). Inter-annual variability in sea-ice concentration and seasonality has been shown to affect the foraging behaviour of southern elephant seals (Labrousse et al. 2017), with males foraging longer in coastal polynyas during years of low sea-ice concentration and females foraging longer in pack ice during years of early sea-ice advance and high sea ice concentration. Understanding such phenological relationships is crucial to determining (predicting) how high-level predators will respond to future climate change and/or variability. In recent decades, regional loss of sea-ice habitat and associated change in prey availability has led to a decline in ice-obligate Adélie penguins close to Palmer Station on the western Antarctic Peninsula, which are being replaced by Gentoo penguins (Pygoscelis papua) that are ice-tolerant species adapted to warmer conditions (McClintock et al. 2008; Ducklow et al. 2013). In contrast to the western Antarctic Peninsula (and the western Ross Sea, where populations have shown variable trends in recent decades e.g., Lyver et al. (2014)), East Antarctic Adélie penguin populations have increased since the 1960s and almost doubled since the 1980s (Southwell et al. 2015). This illustrates the complexity of environmental and associated biological change and variability around Antarctica – and underscores the need to better understand the association between sea ice physics and biota at multiple trophic levels, and their regional dependence. Of particular importance is the concept of optimal sea ice physical conditions (“habitat optimum”) for given vertebrate species e.g., Adélie penguins (Fraser and Trivelpiece 1996; Smith et al. 1999), with species, food webs and ecosystems showing non-linear responses to sea ice change once certain critical thresholds/tipping points are crossed (Gutt et al. 2021). The complex nature of ecosystem response to sea ice variability and change is further illustrated by the effects of extreme events (Massom et al. 2006). In the austral spring-summer of 2001/2, a major and persistent atmospheric circulation anomaly off the western Antarctic Peninsula led to a prolonged period of warm and moist northwesterly winds (an “atmospheric river”) that caused unusually rapid and early seasonal sea ice retreat and extreme wind-driven sea ice compaction against the Peninsula. These conditions resulted in simultaneous dynamic ice thickening and melt while also delivering anomalously-high snowfall. The extreme wind-driven sea-ice compaction and warm, wavy conditions led to high primary productivity (Massom et al. 2006) that supported higher-than-normal krill abundance (Steinberg et al. 2015) and a statistically-significant higher krill recruitment the following year (Saba et al. 2014). In spite of this, the persistent lack of open water (leads) due to the high level of ice compaction, combined with intense late-season snowfalls and melt (causing catastrophic flooding at penguin nests), led to the largest single-season Adélie penguin breeding failure in the 30-year observational record near Palmer Station – an effect that was still felt 10 years later (Fraser et al. 2013; Fountain et al. 2016). The scenario described by Massom et al. (2006) shows how the ecological effects of rapid and extreme sea ice loss/change can be both positive and negative, with these effects also having regional dependence around Antarctica (Massom and Stammerjohn 2010; Fountain et al. 2016). It further underscores the importance of extreme events that, while being short-lived, can have long-lasting ecological consequences i.e., adverse sea ice environmental conditions can breach critical biological thresholds (Gutt et al. 2021). Understanding the wider impacts of extreme atmospheric and oceanic events (including marine heatwaves (Montie et al. 2020)) on the Antarctic marine cryosphere and ecosystems is a research priority in Antarctic and Southern Ocean science (Kennicutt et al. 2015), given that such events are anticipated to become more frequent in a warming climate (Meredith et al. 2019). The Future It is anticipated that many aspects of the wider Antarctic ecosystem will be negatively affected by predicted sea ice loss over coming decades (see Antarctic Sea Ice #3), through a reduction in breeding and foraging habitat and a decrease and/or relocation in Southern Ocean primary production and associated prey availability for higher trophic levels (Constable et al. 2014; Meredith et al. 2019; Rintoul et al. 2018; Steiner et al. 2021). Changes to key phenological relationships between primary (algal) production and sea ice extent and seasonality (advance, retreat and duration) will likely lead to mismatches between the life cycles of ice-associated herbivores and ice algal food availability (Bluhm et al. 2017). Moreover, the habitat of Antarctic krill (a key prey species for penguins, seals and whales) is projected to contract southwards (with medium confidence) under future climate change scenarios (Atkinson et al. 2019; Meredith et al. 2019; Veytia et al. 2020). Looking to apex predators, it has been estimated that the abundance and distribution of both Crabeater and Weddell seals will be negatively affected by future reductions in sea-ice extent and duration/persistence as well as changing types of sea ice – given the impacts on the availability of both prey (notably krill) and suitable resting and breeding habitat (Siniff et al. 2008; Bester et al. 2017). Both Adélie and Emperor penguins and Antarctic and snow petrels are similarly vulnerable to future sea ice loss (Ainley et al. 2017). Any loss of fast ice is particularly impactful on the Weddell seal, given its dependence on fast ice as a site for pupping (LaRue et al. 2019). For the Emperor penguin, the loss of stable fast ice may result in their extinction by the end of this century, given their strong reliance on it for a platform to incubate their egg and raise their chick (Fretwell et al. 2012). Late formation of fast ice, failure to form and/or unusually early seasonal breakup at specific breeding locations significantly reduces the chance of successful breeding and fledging (Jouventin 1975; Massom et al. 2009). For global temperature increases of 2°C above pre-industrial levels, Emperor penguin colonies north of 70°S, that is ∼50% of colonies (40% of the breeding population), are projected to decrease or disappear (Ainley et al. 2010). Under high-emission scenarios of 3-4°C of warming, 80% of Emperor colonies are projected to become quasi-extinct (individuals are still alive but extinction is inevitable) by 2100 (Jenouvrier et al. 2020). Accordingly, Trathan et al. (2020) argue that Emperor penguins should be considered for reclassification by the IUCN Red List as vulnerable or even endangered. In the case of further warming and sea ice loss, commercially-exploitable fishes, such as the southern blue whiting, could extend their distribution range into the Antarctic (Agnew et al. 2003). At the same time, the projected loss of sea ice in coming decades would also lead to decreased abundance of Silverfish and krill, two key prey items for penguins and other predators (Corso et al. 2022; Atkinson et al. 2019). The loss of sea ice would also lead to increased prevalence of open water and possible future expansion of the krill fishery (Bester et al. 2017). Indeed, reduced winter sea ice coverage has already resulted in a southward shift of the krill fishery in the South Atlantic (Kawaguchi and Nicol, 2020). To date and as stated above, much of our understanding of the ecological effects of sea-ice change comes from the western Antarctic Peninsula where rapid sea ice loss occurred during the past four decades (see Antarctic Sea Ice #3). The effects of other changes in sea ice extent and duration in other Antarctic sectors e.g., increases in the western Ross Sea, are less understood – as are the wider effects of extreme events (see above). This highlights the importance of establishing coordinated long-term joint physical-biological-biogeochemical monitoring programmes such as the Palmer LTER program (Smith et al. 1995; pal.lternet.edu) in other regions of the circum-Antarctic sea-ice zone. This would enable important regional intercomparison of marine ecosystem response to sea ice change and/or variability, and greater understanding of regional dependencies. Also critical to a more robust model-based prediction of the likely responses of Southern Ocean ecosystems to future change and/or variability in the Antarctic sea ice environment is: Improved understanding of (and reduced uncertainty in) the ecological relationships between sea ice, primary production and key species (e.g., Reiss et al. 2017; Southern Ocean United Nations Decade), and “optimal” sea ice conditions for given species and how these vary in space and time (Massom and Stammerjohn 2010; Gutt et al. 2021); Understanding how Antarctic marine species and ecosystems are affected by multiple physical stressors (including sea ice change, ocean warming, extreme heatwave and precipitation events, ocean acidification (McMinn 2017), and upper-ocean freshening affecting stratification), and how these stressors interact and what their combined effects are (Gutt et al. 2021); Understanding the effects of ice-sheet change on the Antarctic coastal sea ice environment and habitat e.g., enhanced ice-shelf meltwater input to the upper ocean as it affects sea ice conditions (Bintanja et al. 2013); increased iceberg production and its effects on fast ice and polynyas (Massom and Stammerjohn 2010); and increased availability of key micronutrients such as dissolved iron for enhanced phytoplankton growth (primary production) in coastal polynyas (Dinniman et al. 2020); and Reducing the large uncertainties in current climate model predictions of regional and seasonal changes in the Antarctic sea ice in coming decades and in response to a warming climate (Meredith et al. 2019) (see Antarctic Sea Ice #1 and #3). Conclusion Antarctic sea ice plays a crucial role in the structuring and function of Southern Ocean ecosystems. Changing sea-ice conditions are already resulting in observable changes in the distribution and behaviours of several sea ice-dependent and -tolerant species, with cascading effects across Antarctic marine ecosystems. Future loss of sea ice habitat, and increased occurrence of extreme atmospheric and oceanic events, is predicted to have wide-ranging effects on these ecosystems. Determining the likely nature of these impacts, however, requires much improved knowledge of the relationships between sea ice physics and biota (including primary production) and how Antarctic marine species and ecosystems are affected by multiple physical stressors across the sea ice system.