Atmospheric Variability and Sea-Ice Changes in the Southern Hemisphere

Abstract

The Antarctic sea ice concentration (SIC) plays a crucial role in global climate dynamics by influencing atmospheric and oceanic circulation. This study examines SIC variability and its relationship with major climate modes, including the El Niño-Southern Oscillation (ENSO), Pacific-South American (PSA) pattern, Southern Annular Mode (SAM), and Antarctic Dipole (ADP). Using NSIDC satellite-derived sea ice data and ERA5 reanalysis from 1980 to 2022, we analyzed SIC anomalies in the Weddell, Ross, and Bellingshausen and Amundsen (B&A) Seas, assessing their response to climatic forcings across different timescales. Our findings reveal strong linkages between SIC variability and large-scale atmospheric circulation. ENSO-related teleconnections drive a dipolar SIC response, with warming in the Pacific sector and cooling in the Atlantic during El Niño, and the opposite pattern during La Niña. PSA and ADP further modulate this response by altering Rossby wave propagation and heat fluxes, leading to significant SIC fluctuations. The ADP emerges as a dominant driver of interannual SIC anomalies, showing an out-of-phase relationship between the Atlantic and Pacific sectors of the Southern Ocean. Regional SIC trends exhibit contrasting patterns: the Ross Sea shows a significant positive SIC trend, while the B&A and Weddell Seas experience persistent negative anomalies due to enhanced meridional heat transport and stronger westerly winds. SAM strongly influences SIC, particularly in the Atlantic sector, with delayed responses of up to six months, likely due to ice-albedo feedbacks and ocean memory effects. These results enhance our understanding of Antarctic sea ice variability and its sensitivity to large-scale climate oscillations. Given the observed trends and ongoing climate change, further research is needed to assess how these processes will evolve under future warming scenarios. This study highlights the importance of continuous satellite observations and high-resolution climate modeling for improving projections of Antarctic sea ice behavior and its implications for the global climate system.

1. Introduction

The Antarctic sea ice concentration (SIC) plays an important role in global climate variability through dynamic and thermodynamic processes, as well as through feedback mechanisms between atmosphere and ocean circulations acting at different spatial and temporal scales [1,2]. SIC anomalies are characterized in the percentage of area covered and persist for several months [3]. The SIC as a component of the Earth’s system has a significant climatic impact through its influences on the planetary albedo and the insulating effect, being a contributing component of the global climate system (p.e., [4,5]).

Sea ice and its resulting meltwater serve as an effective insulating barrier, limiting vertical fluxes of heat, mass, and momentum between the ocean and atmosphere. This process is influenced by the Southern Ocean (SO) memory of sea ice anomalies, which persists for approximately four years in response to a maximum historical field applied over the Antarctic SIC and thickness [6]. This period is followed by another four years dominated by cold, fresh water at the SO surface, leading to increased buoyancy and subsurface warming due to insulating effects. These changes ultimately affect the formation of sea ice in the following year.

The strong relationship between SIC and teleconnection patterns has been discussed over the last decades [7,8,9,10,11,12,13]. These studies have shown that SIC variability has a great relation to different modes of climate variability, such as the Southern Annular Mode (SAM), El Niño-Southern Oscillation (ENSO) and Pacific-South American (PSA) pattern. More specifically, teleconnection indicates the relationship between local climatic anomalies and remote forcing [14]. Teleconnections trigger the propagation of waves and the transport of energy in the atmosphere and oceans, and spatial patterns are most evident in the medium and upper troposphere [15]. These patterns present a vertical barotropic structure with the same phase throughout the tropospheric column. In general, these modes of variability manifest themselves with centers of action that extend from the tropics towards the extratropics. The PSA and ADP modes are examples of the extratropical response associated with the ENSO tropical phenomenon [16].

The influences of ADP on the SIC have been investigated since satellite observations became available in the late 1970s [17]. The ADP is characterized as a permanent atmospheric wave with an out-of-phase relationship with Antarctic sea ice and surface fields between Antarctic sectors located in the center and east of the South Pacific and South Atlantic Oceans [8,9,17]. The influence of ENSO on the Antarctic occurs through the stationary Rossby wave, driven by the heating and cooling of sea water creating dipole patterns associated with SIC, sea level pressure (SLP), and surface air temperature (${\mathrm{T}}_{Tair}$) [17]. In a study related to Antarctic sea ice variability, ref. [18] show that when ENSO and SAM modes are activated, they contribute to a dipolar pattern with inverse anomalies in the Atlantic and Pacific sectors. Therefore, the main mode of climate variability in sea ice in the Southern Hemisphere is the ADP pattern.

As tropical forcing alters the meridional circulation in the South Pacific and South Atlantic sectors of the SO, a Rossby stationary wave is triggered and propagated towards extratropical latitudes when configured as an ADP pattern over high latitudes. The ADP presents a pattern similar to the atmospheric Antarctic Circumpolar Wave (ACW), which includes anomalies with opposite signals in the Pacific and Atlantic sectors of the SO occurring in the same period. The study in [9] found that the magnitude of dipole variability is considerably greater than the ACW, so the permanent ADP wave is influenced by remote connections. The atmospheric and oceanic anomalies related to SIC changes are advected by the Antarctic Circumpolar Current (ACC), which has consequences for the formation of the ACW. Additionally, the influence of ENSO on the Antarctic occurs through the stationary Rossby wave driven by the heating and cooling of sea water, creating the dipole patterns associated with SIC, sea level pressure, and surface air temperature [17].

As shown by [17], both remote and local climatic variability influence the SIC at different timescales. The strongest teleconnection between ENSO and ADP occurs on the interannual timescale. The author identified two main mechanisms responsible for the formation and maintenance of the ADP. One mechanism involves heat flow driven by the meridional circulation within the Southern Hemisphere’s Ferrel cell, while the other is associated with regional anomalous circulation induced by stationary vortices. The dynamical and thermodynamic processes that relate the influence of ENSO to the Antarctic SIC have been studied only in the Western Pacific Ocean [17]. However, the mechanisms that determine the local and remote variability in SIC in other Antarctic sectors remain unknown.

Although many studies have pointed out that the interannual variability in SIC is associated with patterns of teleconnections such as ENSO and PSA, in this study, we investigate the SIC variability and its relation to these metrics and also to the ADP and SAM, through an out-of-phase relationship between the SIC and the ${\mathrm{T}}_{Tair}$.

This paper aims to assess the influence of ENSO, PSA, and SAM patterns on the spatiotemporal variability in SIC in the Ross Sea, Bellingshausen and Amundsen (B&A), and Weddell sectors, as well as to provide a review of ADP behavior. The influence of ENSO and ADP regions becomes even more important when related to the trends of increasing and decreasing SIC on the Antarctic Peninsula (AP) east and west sectors. This paper is organized as follows: Section 2 describes the data and the methods. Section 3 discusses the results, showing the relationship between the SIC and the variability modes. Finally, a summary and discussion of the main outcomes are presented in Section 4.

2. Data and Methodology

In this study, the monthly data on the SIC from satellite-derived products were obtained from the Passive Microwave Sea Ice Dataset from the National Snow and Ice Data Center (NSIDC) [19,20], version 4 (https://nsidc.org/data/data-programs/noaa-nsidc, accessed on 10 February 2024), available from January 1980 to December 2022. The SIC is calculated by algorithms that use information provided by passive microwave sensors, and is defined as the fraction or percentage, greater than 15% for each pixel of the oceanic area covered by sea ice [21,22].

Due to the scarcity of direct measurements in the Antarctic continent and in the SO, reanalysis products are fundamental tools for studying the variability and atmospheric trends in Antarctica [23]. For analyzing the variability in atmospheric circulation, ERA5 monthly reanalysis data on ${\mathrm{T}}_{Tair}$, Sea Surface Temperature (SST), and 500-hPa geopotential height (gph) from 1980 January to 2022 December were obtained. The ERA5 reanalysis was developed from the new generation of products from the European Center for Medium-Range Weather Forecasts (ECMWF) [24]. The advantage of using the new database is that it has a number of improvements when compared to the predecessor, the ERA-Interim [25], and its horizontal resolution surpasses it, being approximately 30 km (T639), presenting an accurate data assimilation system and physical model with good performance (Table 1).

In order to examine the influence of teleconnections on Antarctic sea ice, it was necessary to define three regions of the SO that comprise the Ross Sea (160° E–130° W, 60° S–70° S), B&A Seas (130° W–70° W, 60° S–70° S), and Weddell Sea (60° W–10° W, 60° S–70° S) (Figure 1). The Weddell and B&A Seas locations are those closest to South America.

To assess long-term trends in the Antarctic SIC and ${\mathrm{T}}_{air}$ from 1980 to 2022, we applied a statistical framework that incorporates the Mann-Kendall test [26,27]. This methodology ensures the reliable detection of monotonic trends while minimizing the influence of outliers and data irregularities. Given the complexity of climate variability in the Southern Ocean, it is essential to use non-parametric methods that do not assume a normal data distribution. The statistical significance of the detected trends was assessed at a confidence level of $p\le 0.05$.

To correlate the modes of climate variability based on the SIC anomalies, we applied the technique of Empirical Orthogonal Function (EOF) to generate their respective Principal Components (PCs), that were assigned to SAM and PSA. The identification of both modes was based on monthly data of 500-hPa gph anomalies ERA5 reanalysis. The first dominant mode over the SO referring to SAM was extracted. Removing the zonal average of the eddy components allowed for the identification of the PSA mode as the first dominant mode [28,29].

The Oceanic Niño Index (ONI) was calculated using the quarterly average of the SST anomalies in the Niño_3.4 region (5° S–5° N and 170° W–120° W), which represents the central region of the equatorial Pacific [30,31]. The selection of ENSO episodes adopted the following criteria: SST anomalies ≥ 0.5 °C (≤−0.5 °C) for at least five consecutive quarters to characterize El Niño (La Niña) events (Table 2). Based on this selection criterion, it was possible to identify seven episodes for each ENSO phase, as well as seven neutral events. The years defined based on this criterion are shown in Table 2 and are in line with those discussed in [32,33].

The ADP sea ice concentration (${\mathrm{ADP}}_{sic}$) and ADP surface air temperature (${\mathrm{ADP}}_{Tair}$) indices were calculated by subtracting the monthly anomalies of SIC and surface air temperature over the Atlantic (40° W–60° W, 60° S–70° S) and Pacific (130° W–150° W, 60° S–70° S) sectors, as described in [17]. From the regions selected to capture the ADP signal, it was possible to verify the relationship between the phases and the correlation fields. The two ADP indices were calculated using data from different sources, specifically the NSIDC and ERA5 databases.

Three regions were defined in the areas of the Pacific and Atlantic of SO to investigate ADP patterns and their connections with ENSO by NSIDC data using composites of SIC. To identify the ADP signal in the composites, we separated the composite SIC anomalies. Each composite was made upon an average of seven years of El Niño, seven years of La Niña, and six ENSO neutral phases (Table 3). Additionally, the influence of the Southern Annular Mode (SAM) and the Pacific South America (PSA) pattern was considered by selecting seven years for each positive, negative, and neutral phase of these modes. These years of El Niño and La Niña composites revealed dipole patterns with a delay of six months between the Tropical Pacific and the Southern Ocean (SO) around the Antarctic Peninsula (AP), according to [8]. Similarly, variations in SIC anomalies were observed under different SAM and PSA phases, demonstrating their impact on the atmospheric and oceanic circulation in the region.

The quarterly spatial anomaly patterns of these variables were created by taking into account the average of seven intense El Niño events and seven intense La Niña events, and after subtracting the average of all El Niño and La Niña events, respectively. Likewise, composites were generated for SAM and PSA phases, assessing their role in modulating SIC anomalies. In this way, seven composites were examined for each ENSO, SAM, and PSA phase: September-October-November (SON₀) of the year preceding the event, December-January-February (DJF_0–1) of the transition year linked to the event, and March-April-May (MAM₁), June-July-August (JJA₁), and September-October-November (SON₁) of the year following the event. The quarterly anomalies were calculated by the difference between the quarterly values and the quarterly climatology for the period from 1980 to 2022. The inclusion of SAM and PSA in the composite analysis allowed for a more comprehensive understanding of the interactions between atmospheric variability and Antarctic sea ice dynamics.

3. Results and Discussion

3.1. Seasonal SIC Analyses

Figure 2a shows the seasonal variability in SIC for the three selected sectors of Antarctic seas (see Figure 1). A clear seasonal cycle is evident, with minimum SIC values being observed during the austral summer (December–February) and maximum values occurring in the austral winter (June–September). The Weddell Sea exhibits the highest SIC throughout the year, reaching almost 95% during peak months, while the Ross and B&A Seas exhibit lower SIC values, particularly in summer. Among the three regions, the B&A area has the lowest SIC, indicating greater seasonal variability and greater ice loss during the summer months. The seasonal progression of the SIC reflects the influence of atmospheric and oceanic forcing, with sea ice growth beginning in autumn (March–May) and reaching its maximum extent in winter. The consistently high SIC of the Weddell Sea suggests cooler surface waters and limited ocean heat flux, which contribute to a more persistent ice cover. In contrast, the B&A Seas exhibit a more dynamic ice cover, likely influenced by stronger ocean currents and warmer waters, which promote seasonal melting.

During summer (DJF) (Figure 2b), the SIC is at its minimum extent, with the ice cover largely restricted to high latitudes near the Antarctic coast. The lowest SIC values (10–30%) were observed at the outer ice edges, while areas of higher concentration (>70%) were limited to specific coastal regions, particularly in the Weddell and Ross Seas. This pattern reflects the strong seasonal melting of Antarctic sea ice due to increased solar radiation and atmospheric temperatures during the austral summer. In autumn (MAM)(Figure 2c), the SIC begins to expand, extending further from the coast. SIC values increase significantly, with regions of 50–70% concentration covering wider areas, particularly in the Weddell, Ross, and Amundsen Seas. Winter (JJA) (Figure 2d) displays the maximum extent of sea ice, with the highest SIC values observed over a much larger area around Antarctica. Most of the Southern Ocean is covered by ice, with extensive regions exhibiting an SIC above 80%. The regions of greatest concentration are found in the Weddell and Ross Seas, where sea ice is most persistent. Spring (SON) (Figure 2e) represents the beginning of sea ice retreat as temperatures rise. The distribution of the SIC remains extensive but begins to decline, particularly along the ice margins. Meanwhile, central regions maintain high concentrations (>80%), and areas closer to the Antarctic Circumpolar Current show signs of melting, leading to a reduction in the SIC.

The sea ice extent has been assessed since 1973, when such data became available for the first time [34]. The incidence of solar radiation is responsible for sea ice melting, either by direct surface heating, or by direct heating from the surface layer of the ocean, which leads to basal melting and its edges melting. It is worth mentioning that the Weddell and Ross Sea sectors have the highest concentrations of sea ice, and the B&A Sea sector the lowest concentration. It is evident that the Weddell Sea sector contains the highest concentrations of sea ice among the other sectors. This is due to a regional cyclonic circulation, defined as the Weddell Gyre [35], that converges the sea ice on the coast by Ekman transport, making sea ice thicker by deformation [36].

3.2. Seasonal Sea Ice Area and Air Temperature Trends

Figure 3 shows seasonal spatial climate trends in ${\mathrm{T}}_{Tair}$ (right panel) and SIC (left panel) in Antarctica. During the austral summer and autumn (Figure 3a,b), negative SIC trends are observed over the B&A Seas, while positive trends are observed over the Ross and Weddell Sea sectors [37,38]. The decrease of SIC near the west of AP is due to the combination of a low pressure anomalous field and higher temperatures over the B&A Seas sector. This favors the transport of meridional heat from subtropical to extratropical latitude regions. As previously discussed by [39], the authors attribute the observed changes in the Antarctic SIC to stratospheric ozone depletion, which has led to a non-annular atmospheric circulation response, influencing sea ice extent through changes in surface wind patterns and ocean-atmosphere interactions; the declining SIC observed in recent decades corroborates to a dipolar pattern between the AP and the Ross Sea SIC. In contrast, [40] emphasize the role of increasing CO₂ concentrations as the dominant driver of large-scale changes in Southern Hemisphere circulation and sea ice variability. While ozone depletion has been a significant forcing mechanism, particularly in strengthening the SAM and modulating regional atmospheric circulation, recent analyses suggest that effective CO₂ forcing has played a more dominant role in long-term SIC trends. In winter and spring (Figure 3c,d), a significant SIC decrease around Weddell Bay may be related to the attenuation of long-wave radiation, as well as to southern winds from low latitudes [41]. Climatologically, the SIC variability is influenced by an advection of southern cold air over the western Ross Sea sector (related to the Ross Gyre) and a northern warm advection over the Amundsen Sea sector (related to the Amundsen Low) [37,42]. The ocean dynamics drive the advections of the cold air in the Ross sector, working as a mechanism acting on the SIC decrease (in summer) and increase (in winter) [43].

The changes in the Antarctic SIC are associated with the phases of ENSO through remote interactions with the Central East Equatorial Pacific SST. As shown by observational data from the last four decades, there is a considerable increase in the Ross Sea SIC, and a positive trend (Figure 3a–d). This significant positive trend in the SIC in west Antarctica is associated with the SST variability in the Equatorial Pacific, which triggered the Rossby wave propagation [12,44].

Spatial changes in the ${\mathrm{T}}_{Tair}$ in SO are evident over the last decades, as shown in (Figure 3). Over the past decades, there has been a statistically negative trend, being more evident in the southern summer (Figure 3e). A cooling trend over the past 40 years was observed in the SO, with positive anomalies in the middle latitudes and negative anomalies in the high latitudes. The negative anomalies of ${\mathrm{T}}_{air}$ around the Antarctic are associated with the positive phases of the SAM mode [45]. According to [11], the influence of ENSO was strong in some warm and cold events; however, it can be weakened, depending on the random phase relationship between PSA and SAM. In addition, it was found that over the past few decades, the SST in the Equatorial Pacific basin was the main factor in modulating both the PSA and SAM modes. The positive trend of ${\mathrm{T}}_{Tair}$ anomalies between the eastern and western sectors of the AP in autumn is due to the southern heat transport from low to high latitudes (Figure 3f). Anomalous circulation in the southeastern Pacific Ocean causes an anomalous southern heat flow towards the Equator (Pole) on the surface of the Ross-Amundsen Seas, as well as in the Bellingshausen and Weddell (B&W) Seas [15]. From winter to spring (Figure 3g,h), a tendency of increasing ${\mathrm{T}}_{Tair}$ in the Weddell Sea is observed, while in the B&A and Ross Seas have dominated a temperature decrease trend. This pattern is strongly related to the remote interaction between La Niña and PSA. The surface air heating (cooling) coincides with a reduction (increase) in the SIC on the Weddell Sea (B&A) (Figure 3g and Figure 3h), respectively.

Additional mechanisms related to changes in the Ross sector are also present to ensure that the positive phase of the SAM reduces the heat exchange during continental cold air towards the Ross sea ice shelf. It has been shown that the increase in sea ice over the Ross Sea contributes to the deepening of the Amundsen Sea’s low [39,46]. In fact, sea ice variations can influence atmospheric circulation through changes in surface albedo and surface flows of heat, moisture, and momentum. In the austral winter, for instance, the effect of albedo is suppressed by the attenuated effect of solar radiation.

3.3. Correlation Between Climatic Modes and SIC

In this section, we investigate the spatial correlations between the ENSO, SAM, PSA, and ADP modes and the SIC anomalies (Figure 4). During the summer, the SIC was positively (negatively) correlated in the Weddell Sea sector (Ross west-B&A) when analyzing the ENSO and PSA modes (Figure 4a,j), so this is an out-of-phase relationship when compared to SAM (Figure 4e). As seen by this correlation analysis, positive SIC anomalies over the Weddell Sea sector are associated with positive ENSO events. This suggests that the SIC is related to the ENSO and PSA variability, exhibiting anomalies of the opposite sign in the Atlantic and Pacific sectors. The spatial correlation between the PSA mode and the SIC during the spring and southern summer (Figure 4h,i) has a dipolar pattern between the west and east of the Antarctic Peninsula. This indicates the influence of the positive phase of the PSA, in which low atmospheric pressure anomalies are positive in the Ross and negative across the B&A seas. There is an intensification of positive correlation over the Ross and B&A Seas and the negative in Weddell Sea, from autumn to winter (Figure 4j,k). In spring, a symmetrical relationship between the ENSO and PSA modes is observed (Figure 4d,l), with the SIC showing a moderately positive correlation between the western and eastern regions of the Antarctic Peninsula, while exhibiting an opposite correlation with the SAM pattern (Figure 4h). It is evident that the Ross Sea sector exhibits a moderate positive correlation with the SAM in the summer (Figure 4e), but in the other seasons, it becomes almost insignificant. On the other hand, the eastern and western sectors of the Peninsula correlate negatively with the intraseasonal variability in the SIC anomaly (Figure 4e–h).

The correlation coefficients are significant, with a 95% confidence level. Therefore, this relates the variability in the SIC to being promoted by the PSA. The ADP (${\mathrm{ADP}}_{sic}$ and ${\mathrm{ADPT}}_{Tair}$) indices are significantly correlated with SIC anomalies, and the structure of the ADP pattern becomes more evident, with anomalies of opposite signs in the Pacific and Atlantic sectors (Figure 4m–p and Figure 4q–t), respectively.

Table 4 shows the correlation coefficients between the PSA, Niño_3.4, and ADP indices with the SIC anomalies of the three sectors of the SO. The results show that the correlation between PSA and SIC over the Weddell Sea is positive during the winter and negative between the spring and summer. Although the correlated coefficients are low, B&A is negatively correlated between winter and spring, whereas in Ross there is a positive correlation.

The relationship between the SIC and the modes of variability was investigated through the time lag of the correlation coefficients, as shown in Figure 5. The correspondence between the equatorial region of the Pacific and the SO is evident when related to the variability in the SIC (Figure 5a). It is observed that the Niño_3.4 index does not exhibit a lag in the Ross Sea. However, the variability in the Weddell SIC is positively correlated, with values of up to 0.3 at lag = −4 months. In the B&A region, the correlation between Niño_3.4 and the SIC is weaker compared to the Weddell Sea sector of the Antarctic Peninsula, with a value of approximately 0.2. The increase in the SIC leads over the Ross, B&A, and Weddell sectors with a lag of one month (lag = −1 month) and is negatively correlated with the SAM mode (Figure 5b). This suggests that the increase in the SIC in these sectors is driven by the SIC anomalies of the spring. However, the positive correlation between the SAM and SIC in the Weddell sector is low, with 0.2 and almost 0.0 for Ross and B&A. Therefore, the increase in ice cover over the Atlantic is only evident after 6 months, as this is a response to the SAM and the approach of winter. The maximum correlation between the PSA and SIC in the Ross Sea occurs without any lag and is negative for the B&A Sea around 3 months (Figure 5c). In the Weddell Sea, the lag is positive between 1 and 8 months, and it indicates that an increase in the SIC over this sector will occur almost eight months after the wave propagates. However, the correlations among these sectors remain below 0.4. The correlation between the ${\mathrm{ADP}}_{SIC}$ index is largely dominated in the Ross Sea, with a coefficient greater than 0.5 and a positive lag of about 0–2 months (Figure 5d).

In the Weddell and B&A Seas, the correlation coefficient ranges from −0.8 to 0.4, with no observed lag. The direct influence of temperature in the SO with sea ice can be seen with the ${\mathrm{ADP}}_{Tair}$ index (Figure 5e). ${\mathrm{ADP}}_{Tair}$ is negatively correlated over the Ross and B&A sectors, and occurs with a positive lag of about 0–2 months. For the Weddell sector, the maximum correlation occurs with a lag of 2 months and with a coefficient greater than 0.4. It is suggested that these negative correlations of SIC anomalies over the B&A and Ross sectors are driven by the exchange of energy between atmosphere and ocean. On the other hand, as it progresses from July–December, the increase in ice cover remains amplified over the Atlantic sector, as shown by the positive correlation coefficients. Therefore, as shown above, the ENSO, SAM, and PSA patterns contribute to the intraseasonal variability in the SIC anomalies, and are also associated with the ${\mathrm{ADP}}_{sic}$ and ${\mathrm{ADP}}_{Tair}$ pattern anomalies.

3.4. Composites of Sea Ice Concentration During Neutral Years and Phases of Climate Variability Modes

The defining characteristic of the ADP is the dipole signature between the Weddell and B&A Seas. Therefore, the anomalies for the ENSO composites are presented through a polar stereographic projection in the SO (Figure 6). During the spring (SON₀) in neutral years (Figure 6a), changes in the SIC are noticeable, with a dipolar pattern manifested between the B&A and Weddell Seas, with the strongest being over the Indian Ocean and the Ross Sea, in response to a wavenumber-2 pattern. The largest correlations were mainly contained in the sectors with the greatest SIC variations, exhibiting an out-of-phase relationship for the sectors around the AP for El Niño (Figure 6b,e,h,k) and La Niña (Figure 6c,f,i,l,o) events. In Figure 6k, the dipolarity of the SIC is observed between the western and eastern regions of the Peninsula in winter (JJA₁). This may be related to the sea ice’s response to a wavenumber-3 pattern during a warm event, and therefore becomes more evident around Antarctica [47]. In La Niña events, the ADP signature is not well defined between the seasons that precede the event, i.e., spring, summer, and autumn (Figure 6c,f,i). The ADP pattern is noted six months after beginning of the La Niña event, with positive anomalies of the Seas B&A and Ross SIC. Negative anomalies are positioned in the Weddell Sea sector. The ADP pattern, in the cold phase, lasts in JJA₁ (Figure 6l), being intensified until the end of the SON₁ (Figure 6o). In general, the ADP pattern seems to be present when an out-of-phase relationship is observed between the time series of the SIC anomalies of the B&A, Ross, and the Weddell Sea regions. The SIC anomalies in the ADP region generally persist for three to four seasons after an ENSO event [17].

The out-of-phase relationship is more evident during extreme La Niña events, whereas in El Niño years, it is observed only in 1992–1993 and 2002–2003. Following the ADP standard concept (${\mathrm{ADP}}_{Tair}$), the time series would be expected to show positive SIC anomalies on the Weddell Sea sector during the El Niño event. The anomalies for both reanalysis, and for the observed data, however, are negative in that sector. The location and size of the selected areas play an important role in the statistical analyses. The area defined for the Weddell Sea sector shows anomalies that propagate eastwards in response to the negative anomalies that occur in the west sector (windward), which influences the average obtained for the area.

Figure 7 shows the seasonal composites of the Antarctic SIC anomalies in relation to the SAM phases. The spatial distribution of the anomalies was analyzed for different seasonal stages, including the periods before, during, and after SAM events.

Before the onset of a SAM event (SON₀, Figure 7a–c), the neutral phase shows minor SIC anomalies, indicating a more stable ice cover. However, during SAM+, an increase in SIC is observed in the Atlantic sector, particularly in the Weddell Sea, while the Ross Sea experiences ice loss. In contrast, SAM− exhibits the opposite pattern, with SIC reductions in the Weddell Sea and an increase in the Ross Sea. These findings are consistent with previous studies highlighting the modulation of sea ice by westerly winds. During the transition phase (DJF_0–1, Figure 7d–f), SIC anomalies become more pronounced. SAM+ is associated with significant sea ice loss in the Bellingshausen and Ross Seas, attributed to increased warm air advection and enhanced oceanic heat uptake. Conversely, SAM− supports the persistence of SIC in the Amundsen Sea, with a corresponding decline in the Weddell Sea. This phase demonstrates the sensitivity of Antarctic sea ice to atmospheric forcing during summer months, as highlighted in previous ENSO-related studies. In the fall following SAM events (MAM₁, Figure 7g–i), SIC anomalies remain, indicating a delayed response of the Antarctic climate system. Under SAM+ conditions, SIC loss continues in the Pacific sector, with modest ice expansion in the Atlantic. SAM− exhibits a reversal of this pattern, further reinforcing the persistence of teleconnection effects beyond the immediate event timeframe.

Winter composites reveal that SAM+ leads to intensified SIC decrease in the Amundsen sector (JJA₁, Figure 7j–l), suggesting continued heat transport anomalies linked to the SAM phase. The impacts of SAM are evident even a year after the event (SON₁, Figure 7m–o). SAM+ remains associated with a reduced SIC in the Pacific sector, particularly in the Bellingshausen and Ross Seas, while an opposite pattern is observed in the Atlantic. SAM− continues to favor ice retention in the Pacific sector but reduces SIC in the Weddell Sea, suggesting a persistent, seasonally lagged response.

The neutral phase shows minor SIC anomalies (SON₀, Figure 8a), indicating stable sea ice conditions, before the onset of a PSA event. During PSA+ (SON₀, Figure 8b), SIC reductions are observed in the B&A Seas, while PSA− promotes ice expansion in these areas (SON₀, Figure 8c). This pattern suggests an early atmospheric influence on sea ice formation and melting, modulating preexisting oceanic conditions. During summer (DJF_0–1, Figure 8d), PSA-related anomalies become more pronounced. PSA+ is associated with intensified SIC loss in the B&A Seas due to enhanced warm air advection and oceanic heat flux (Figure 8e).

In contrast, PSA+ (Figure 8f) fosters an increased SIC in the Pacific sector while promoting ice retreat in the Weddell Sea. This asymmetric response highlights the PSA’s ability to modulate air-sea interactions during peak summer. Following PSA events (MAM₁, Figure 8g–i), SIC anomalies persist into autumn, indicating the extended influence of atmospheric conditions. PSA+ leads to continued SIC reductions in the Ross and Amundsen Seas (Figure 8h), whereas PSA− supports ice retention in these regions (Figure 8i). This seasonal persistence suggests a lagged response of sea ice to PSA induced atmospheric forcing, which could be linked to delayed oceanic adjustments.

During winter (JJA₁, Figure 8j–l), PSA-related anomalies remain evident. PSA+ is associated with substantial sea ice loss in the Pacific sector, reinforcing earlier seasonal trends (Figure 8k). Conversely, PSA− promotes SIC expansion in the Ross and Bellingshausen Seas (Figure 8l), indicating prolonged atmospheric forcing. The long-term effects of PSA are evident in the subsequent spring (SON₀, Figure 8m–o). PSA+ continues to suppress the SIC in the Bellingshausen and Ross Seas (Figure 8n), whereas PSA− supports ice expansion in these regions (Figure 8o). This suggests that PSA-related variability has a lasting impact on Antarctic sea ice, potentially influencing multi-year trends and ice-albedo feedback mechanisms.

4. Conclusions

The results of this investigation show that the dynamic mechanisms associated with the periodic and inactive phases of the ADP mode were analyzed and discussed. The SIC’s variability in west and east Antarctica sectors is linked to changes in the surface temperature of the Equatorial Pacific Ocean that interact with the SO circulation patterns. As mentioned above, the strong warming along the west of the AP is associated with changes in the variability in the Equatorial Pacific and in response to Rossby wave propagation [12,44]. The SIC anomalies and atmospheric variable correlations suggest that such influences occur in association with ENSO and PSA forcings. This link is suggested to stem from persistent sea ice anomalies associated with an ADP pattern. First, we seek to understand the links between the SIC seasonal and intraseasonal variability with the less dominant modes of variability in the SO (such as PSA and ADP). Subsequently, the ADP signal and its structure pattern were evaluated in terms of sea ice fields anomalies for the intense El Niño and La Niña events.

As shown by [48], the response of the Antarctic sea ice to the SAM pattern has a maximum lag between 1 and 5 months, which is associated with the albedo-ice feedback mechanism. In the current study, however, the SIC’s response to SAM showed 6 months of maximum lag, but only in the Atlantic. In the Ross and B&A Sea sectors, the SAM was negatively correlated and the persistence of sea ice until summer was attributed to the SIC anomalies from spring. Another important fact is that the intensification of the westerly winds associated with the positive phase of the SAM reinforced the decrease in the surface temperature and the increase in the SIC anomalies in the Weddell sector in summer, and was attributed to the interactive mechanisms between low-pressure areas on Antarctica and the high pressure at subtropical latitudes [45,49].

The analysis of the NSIDC V4 dataset showed similar patterns of [17], with the climate composites related to the SIC anomalies resembling the ADP signal. The results show that the ADP is the main interannual pattern of variability in the Antarctic sea ice. It has been related to the variability in the Equatorial Pacific Ocean SST anomalies resulting from ENSO events [8] and atmospheric forcing associated with Antarctic Oscillation, wavenumber-3 near-stationary patterns, and PSA.

The selected ENSO years and defined areas for a possible establishment of the ADP pattern become even more important when related to one of the main objectives of this research, that was, to verify the relationship with teleconnection patterns. The patterns found on the composites of the spatial SIC anomalies show that the results might vary depending on the phase analyzed by ENSO.

This analysis demonstrates the substantial influence of the SAM and PSA on Antarctic sea ice dynamics, revealing distinct spatial patterns that persist across multiple seasons. The results highlight that SAM+ and PSA+ lead to a reduction in the SIC in the Pacific sector and an increase in the Atlantic, whereas SAM− and PSA− exhibit the reverse effect. Furthermore, the observed lagged response suggests a strong oceanic memory in the Southern Ocean, reinforcing the importance of continued monitoring and high-resolution climate modeling. Given ongoing climate change, future studies should further investigate how SAM− and PSA− related variability will interact with long-term warming trends and their implications for Antarctic ice stability. This perspective aligns with [40], who demonstrated that regime shifts in Southern Hemisphere circulation, driven largely by anthropogenic CO₂ forcing, have already altered large-scale atmospheric and oceanic patterns, influencing SIC variability and reinforcing the need for improved predictive models.

Future work will concentrate on the analysis of multiple reanalysis data sources and statistical methods to verify the ADP pattern and magnitude, evaluating the main variability modes of the sea ice field and atmospheric variables considering only neutral years, and verifying if the ADP manifests in the B&A and Weddell seas. The natural variability in the climate has played an important role in recent decades. It is clear that some sectors of Antarctica have undergone a strong warming in the last few decades. Although, there are gaps in the knowledge that hinder our understanding of future projections of climate change, preventive studies can be strategic, and help us to understand. The reduction in Antarctic sea ice affects the increase in freshwater discharge, which can impact the deceleration of the southern Atlantic overturning circulation (AMOC) in the SO [50] and Antarctic marine ecosystems [51]. Studies have identified a persistent rise in temperature on the west coast of the AP during autumn and winter over the past few decades [52,53]. This significant warming has an impact on the local ecosystem, which has favored a reduction in crab reproduction and increased the competition for krill due to the reduction in sea ice [54]. Dynamic and thermodynamic processes that occur in the oceans, mainly the oceanic circulation of deep waters, can explain this persistence, but for further confirmation further study is required.

Author Contributions

Conceptualization, C.D.G., C.K.P., F.B.J., C.B.C., V.S. and A.C.; methodology, C.D.G., C.K.P., F.B.J. and C.B.C.; Formal analysis, C.D.G. and F.B.J.; Data curation, C.D.G. and C.K.P.; Writing—original draft, C.D.G.; Writing—review & editing, C.D.G., L.P.P., C.K.P., F.B.J., C.B.C., V.S. and A.C.; Visualization, L.P.P., C.B.C., V.S. and A.C.; Supervision, L.P.P. and C.K.P.; Project administration, L.P.P.; Funding acquisition, L.P.P. All authors read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

All data are available upon request to the corresponding author.

Conflicts of Interest

The Authors declare that there is no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Table 1

Comparison of the characteristics of the reanalysis data products of the ERA-Interim e ERA5.

Reanalysis	Era-Interim	ERA5
Time interval	1979–2020	1979–present
Domain	Global	Global
Model resolution	TL255 (80 km)	TL639 (31 km)
Spatial resolution	0.75° × 0.75°	0.25° × 0.25°
Vertical levels	Hybrid-Sigma (60 levels)	Hybrid-Sigma (137 levels)
Assimilation scheme	4D-Var	4D-Var
Reference	[25]	[24]

Table 2

List of ENSO classification scheme and index based on SST anomalies (°C) from average in the Niño_3.4 region.

Event	Niño3.4 Index	Category
El Niño	0.5 to 0.9	Weak
	1.0 to 1.4	Moderate
	≥1.5	Strong
La Niña	−0.5 to −0.9	Weak
	−1.0 to −1.4	Moderate
	≤−1.5	Strong

Table 3

Selection of positive, negative, and neutral ENSO, SAM, and PSA events for this study based on ERA5 SST reanalysis (1980–2022).

Category	ENSO	SAM	PSA
Neutral	1981/1982	1980/1981	1981/1982
	1989/1990	1983/1984	1986/1987
	1990/1991	1991/1992	1987/1988
	1993/1994	1998/1999	2000/2001
	2003/2004	2001/2002	2004/2005
	2012/2014	2007/2008	2012/2013
	2019/2020	2021/2022	2019/2020
Positive	1982/1983	1981/1982	1983/1984
	1986/1987	1985/1986	1986/1987
	1992/1993	1995/1996	1991/1992
	1997/1998	2000/2001	1994/1995
	2002/2003	2010/2011	2003/2004
	2015/2016	2017/2018	2013/2014
	2018/2020	2020/2021	2018/2019
Negative	1984/1985	1980/1981	1983/1984
	1988/1989	1984/1985	1985/1986
	1995/1996	1989/1990	1988/1989
	1999/2000	1994/1995	1998/1999
	2007/2008	2009/2010	2009/2010
	2011/2012	2014/2015	2015/2016
	2020/2021	2019/2020	2021/2022

Table 4

The correlation coefficients between the PSA, NINO_3.4, and ADP indices with the SIC of three sectors of the Southern Ocean for summer, autumn, winter and spring.

	DJF				MAM
	PSA	Niño3.4	${\mathbf{ADP}}_{\mathit{SIC}}$	${{\mathbf{ADP}}_{\mathit{T}}}_{\mathit{air}}$	PSA	Niño3.4	${\mathbf{ADP}}_{\mathit{SIC}}$	${{\mathbf{ADP}}_{\mathit{T}}}_{\mathit{air}}$
Weddell	−0.16	0.24	−0.87	0.38	0.07	0.22	−0.82	0.43
B&A	0.07	−0.16	0.29	−0.07	−0.11	0.10	0.42	−0.21
Ross	0.26	0.40	0.53	−0.37	−0.17	−0.16	0.39	−0.42
	JJA				SON
Weddell	0.40	0.16	−0.75	0.68	−0.43	0.16	−0.75	0.60
B&A	−0.26	0.21	0.42	−0.26	−0.04	−0.01	0.27	−0.11
Ross	−0.01	−0.27	0.68	−0.43	0.30	−0.41	0.52	−0.29

References

1. Lemke, P.; Trinkl, E.W.; Hasselmann, K. Stochastic Dynamic Analysis of Polar Sea Ice Variability. J. Phys. Oceanogr.; 1980; 10, pp. 2100-2120. [DOI: https://dx.doi.org/10.1175/1520-0485(1980)010<2100:SDAOPS>2.0.CO;2]

2. Walsh, J.E. The role of sea ice in climatic variability: Theories and evidence. Atmos.-Ocean; 1983; 21, pp. 229-242. [DOI: https://dx.doi.org/10.1080/07055900.1983.9649166]

3. Stocker, T. Observations: Cryosphere. Climate Change 2013. The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, 2014; pp. 317-382.

4. Parkinson, C.L. Southern Ocean sea ice and its wider linkages: Insights revealed from models and observations. Antarct. Sci.; 2004; 16, 387. [DOI: https://dx.doi.org/10.1017/S0954102004002214]

5. Pellichero, V.; Sallée, J.B.; Chapman, C.C.; Downes, S.M. The Southern Ocean meridional overturning in the sea-ice sector is driven by freshwater fluxes. Nat. Commun.; 2018; 9, 1789. [DOI: https://dx.doi.org/10.1038/s41467-018-04101-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29724994]

6. Parise, C.K.; Pezzi, L.P.; Hodges, K.I.; Justino, F. The Influence of Sea Ice Dynamics on the Climate Sensitivity and Memory to Increased Antarctic Sea Ice. J. Clim.; 2015; 28, pp. 9642-9668. [DOI: https://dx.doi.org/10.1175/JCLI-D-14-00748.1]

7. Ghil, M.; Mo, K. Intraseasonal Oscillations in the Global Atmosphere. Part II: Southern Hemisphere. J. Atmos. Sci.; 1991; 48, pp. 780-790. [DOI: https://dx.doi.org/10.1175/1520-0469(1991)048<0780:IOITGA>2.0.CO;2]

8. Yuan, X.; Martinson, D.G. Antarctic Sea Ice Extent Variability and Its Global Connectivity. J. Clim.; 2000; 13, pp. 1697-1717. [DOI: https://dx.doi.org/10.1175/1520-0442(2000)013<1697:ASIEVA>2.0.CO;2]

9. Yuan, X.; Martinson, D.G. The Antarctic dipole and its predictability. Geophys. Res. Lett.; 2001; 28, pp. 3609-3612. [DOI: https://dx.doi.org/10.1029/2001GL012969]

10. Yuan, X.; Li, C. Climate modes in southern high latitudes and their impacts on Antarctic sea ice. J. Geophys. Res.; 2008; 113, pp. 6-91. [DOI: https://dx.doi.org/10.1029/2006JC004067]

11. Yu, J.Y.; Paek, H.; Saltzman, E.S.; Lee, T. The early 1990s change in ENSO-PSA-SAM relationships and its impact on Southern Hemisphere climate. J. Clim.; 2015; 28, pp. 9393-9408. [DOI: https://dx.doi.org/10.1175/JCLI-D-15-0335.1]

12. Turney, C.S.; Klekociuk, A.; Fogwill, C.J.; Zunz, V.; Goosse, H.; Parkinson, C.L.; Marzinelli, E. Brief communication: Changing mid-twentieth century Antarctic sea ice variability linked to tropical forcing. Cryosphere Discuss.; 2017; preprint

13. Isaacs, F.E.; Renwick, J.A.; Mackintosh, A.N.; Dadic, R. ENSO modulates summer and autumn sea ice variability around Dronning Maud Land, Antarctica. J. Geophys. Res. Atmos.; 2021; 126, e2020JD033140. [DOI: https://dx.doi.org/10.1029/2020JD033140]

14. Coelho, C.A.; de Oliveira, C.P.; Ambrizzi, T.; Reboita, M.S.; Carpenedo, C.B.; Campos, J.L.P.S.; Tomaziello, A.C.N.; Pampuch, L.A.; Custódio, M.D.S.; Dutra, L.M.M. et al. The 2014 southeast Brazil austral summer drought: Regional scale mechanisms and teleconnections. Clim. Dyn.; 2016; 46, pp. 3737-3752. [DOI: https://dx.doi.org/10.1007/s00382-015-2800-1]

15. Liu, Z.; Alexander, M. Atmospheric bridge, oceanic tunnel, and global climatic teleconnections. Rev. Geophys.; 2007; 45, pp. 1-34. [DOI: https://dx.doi.org/10.1029/2005RG000172]

16. Yuan, X.; Kaplan, M.R.; Cane, M.A. The interconnected global climate system—A review of tropical—polar teleconnections. J. Clim.; 2018; 31, pp. 5765-5792. [DOI: https://dx.doi.org/10.1175/JCLI-D-16-0637.1]

17. Yuan, X. ENSO-related impacts on Antarctic sea ice: A synthesis of phenomenon and mechanisms. Antarct. Sci.; 2004; 16, 415. [DOI: https://dx.doi.org/10.1017/S0954102004002238]

18. Holland, M.M.; Bitz, C.M.; Hunke, E.C. Mechanisms forcing an Antarctic dipole in simulated sea ice and surface ocean conditions. J. Clim.; 2005; 18, pp. 2052-2066. [DOI: https://dx.doi.org/10.1175/JCLI3396.1]

19. Stroeve, J.; Meier, W.N. Sea Ice Trends and Climatologies from SMMR and SSM/I-SSMIS, Version 3; NASA National Snow and Ice Data Center Distributed Active Archive Center: Boulder, CO, USA, 2018.

20. Stroeve, J.; Liston, G.E.; Buzzard, S.; Zhou, L.; Mallett, R.; Barrett, A.; Tschudi, M.; Tsamados, M.; Itkin, P.; Stewart, J.S. A Lagrangian snow evolution system for sea ice applications (SnowModel-LG): Part II—Analyses. J. Geophys. Res. Ocean.; 2020; 125, e2019JC015900. [DOI: https://dx.doi.org/10.1029/2019JC015900]

21. Cavalieri, D.J.; Crawford, J.P.; Drinkwater, M.R.; Eppler, D.T.; Farmer, L.D.; Jentz, R.R.; Wackerman, C.C. Aircraft active and passive microwave validation of sea ice concentration from the Defense Meteorological Satellite Program Special Sensor Microwave Imager. J. Geophys. Res. Ocean.; 1991; 96, pp. 21989-22008. [DOI: https://dx.doi.org/10.1029/91JC02335]

22. Meier, W.N.; Stewart, J.S.; Windnagel, A.; Fetterer, F.M. Comparison of hemispheric and regional sea ice extent and area trends from NOAA and NASA passive microwave-derived climate records. Remote Sens.; 2022; 14, 619. [DOI: https://dx.doi.org/10.3390/rs14030619]

23. Bromwich, D.H.; Nicolas, J.P.; Monaghan, A.J. An assessment of precipitation changes over Antarctica and the Southern Ocean since 1989 in contemporary global reanalyses. J. Clim.; 2011; 24, pp. 4189-4209. [DOI: https://dx.doi.org/10.1175/2011JCLI4074.1]

24. Hersbach, H.; Bell, B.; Berrisford, P.; Hirahara, S.; Horányi, A.; Muñoz-Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; Schepers, D. et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc.; 2020; 146, pp. 1999-2049. [DOI: https://dx.doi.org/10.1002/qj.3803]

25. Dee, D.P.; Uppala, S.M.; Simmons, A.J.; Berrisford, P.; Poli, P.; Kobayashi, S.; Andrae, U.; Balmaseda, M.; Balsamo, G.; Bauer, d.P. et al. The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Q. J. R. Meteorol. Soc.; 2011; 137, pp. 553-597. [DOI: https://dx.doi.org/10.1002/qj.828]

26. Cooper, M. A non-parametric test for increasing trend. Educ. Psychol. Meas.; 1975; 35, pp. 303-306. [DOI: https://dx.doi.org/10.1177/001316447503500207]

27. Kendall, M.G. Rank Correlation Methods; 1st ed. C. Griffin & Company: London, UK, 1948; 160p.

28. Mo, K.C. Relationships between low-frequency variability in the Southern Hemisphere and sea surface temperature anomalies. J. Clim.; 2000; 13, pp. 3599-3610. [DOI: https://dx.doi.org/10.1175/1520-0442(2000)013<3599:RBLFVI>2.0.CO;2]

29. Cavalcanti, I.F.A.; Shimizu, M.H. Climate fields over South America and variability of SACZ and PSA in HadGEM2-ES. Am. J. Clim. Change; 2012; 1, pp. 132-144. [DOI: https://dx.doi.org/10.4236/ajcc.2012.13011]

30. Trenberth, K.E. The Definition of El Niño. Bull. Am. Meteorol. Soc.; 1997; 78, pp. 2771-2778. [DOI: https://dx.doi.org/10.1175/1520-0477(1997)078<2771:TDOENO>2.0.CO;2]

31. Trenberth, K.E.; Stepaniak, D.P. Indices of El Niño evolution. J. Clim.; 2001; 14, pp. 1697-1701. [DOI: https://dx.doi.org/10.1175/1520-0442(2001)014<1697:LIOENO>2.0.CO;2]

32. Huang, B.; Thorne, P.W.; Banzon, V.F.; Boyer, T.; Chepurin, G.; Lawrimore, J.H.; Zhang, H.M. Extended reconstructed sea surface temperature, version 5 (ERSSTv5): Upgrades, validations, and intercomparisons. J. Clim.; 2017; 30, pp. 8179-8205. [DOI: https://dx.doi.org/10.1175/JCLI-D-16-0836.1]

33. Patricola, C.M.; O’Brien, J.P.; Risser, M.D.; Rhoades, A.M.; O’Brien, T.A.; Ullrich, P.A.; Collins, W.D. Maximizing ENSO as a source of western US hydroclimate predictability. Clim. Dyn.; 2020; 54, pp. 351-372. [DOI: https://dx.doi.org/10.1007/s00382-019-05004-8]

34. Streten, N.A.; Pike, D.J. Characteristics of the broadscale Antarctic sea ice extent and the associated atmospheric circulation 1972–1977. Arch. Meteorol. Geophys. Bioclimatol. Ser. A; 1980; 29, pp. 270-299. [DOI: https://dx.doi.org/10.1007/BF02247767]

35. Kumar, A.; Yadav, J.; Mohan, R. Seasonal sea-ice variability and its trend in the Weddell Sea sector of West Antarctica. Environ. Res. Lett.; 2021; 16, pp. 24-46. [DOI: https://dx.doi.org/10.1088/1748-9326/abdc88]

36. Sigmond, M.; Fyfe, J.C. Has the ozone hole contributed to increased Antarctic sea ice extent?. Geophys. Res. Lett.; 2010; 37, L18502. [DOI: https://dx.doi.org/10.1029/2010GL044301]

37. Turner, J.; Hosking, J.S.; Marshall, G.J.; Phillips, T.; Bracegirdle, T.J. Antarctic sea ice increase consistent with intrinsic variability of the Amundsen Sea Low. Clim. Dyn.; 2016; 46, pp. 2391-2402. [DOI: https://dx.doi.org/10.1007/s00382-015-2708-9]

38. Hosking, J.S.; Orr, A.; Marshall, G.J.; Turner, J.; Phillips, T. The influence of the Amundsen-Bellingshausen Seas low on the climate of West Antarctica and its representation in coupled climate model simulations. J. Clim.; 2013; 26, pp. 6633-6648. [DOI: https://dx.doi.org/10.1175/JCLI-D-12-00813.1]

39. Turner, J.; Comiso, J.C.; Marshall, G.J.; Lachlan-Cope, T.A.; Bracegirdle, T.; Maksym, T.; Meredith, M.P.; Wang, Z.; Orr, A. Non-annular atmospheric circulation change induced by stratospheric ozone depletion and its role in the recent increase of Antarctic sea ice extent. Geophys. Res. Lett.; 2009; 36, L08502. [DOI: https://dx.doi.org/10.1029/2009GL037524]

40. O’Kane, T.J.; Frederiksen, J.S.; Frederiksen, C.S.; Horenko, I. Beyond the First Tipping Points of Southern Hemisphere Climate. Climate; 2024; 12, 81. [DOI: https://dx.doi.org/10.3390/cli12060081]

41. Morioka, Y.; Engelbrecht, F.; Behera, S.K. Role of Weddell Sea ice in South Atlantic atmospheric variability. Clim. Res.; 2017; 74, pp. 171-184.

42. Justino, F.; Lindemann, D.; Kucharski, F.; Wilson, A.; Bromwich, D.; Stordal, F. Oceanic response to changes in the WAIS and astronomical forcing during the MIS31 superinterglacial. Clim. Past; 2017; 13, pp. 1081-1095. [DOI: https://dx.doi.org/10.5194/cp-13-1081-2017]

43. Lee, S.K.; Volkov, D.L.; Lopez, H.; Cheon, W.G.; Gordon, A.L.; Liu, Y.; Wanninkhof, R. Wind-driven ocean dynamics impact on the contrasting sea-ice trends around West Antarctica. J. Geophys. Res. Ocean.; 2017; 122, pp. 4413-4430. [DOI: https://dx.doi.org/10.1002/2016JC012416]

44. Smith, J.A.; Andersen, T.J.; Shortt, M.; Gaffney, A.M.; Truffer, M.; Stanton, T.P.; Vaughan, D.G. Sub-ice-shelf sediments record history of twentieth-century retreat of Pine Island Glacier. Nature; 2017; 541, pp. 77-80. [DOI: https://dx.doi.org/10.1038/nature20136] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27880756]

45. Comiso, J.C.; Gersten, R.A.; Stock, L.V.; Turner, J.; Perez, G.J.; Cho, K. Positive trend in the Antarctic sea ice cover and associated changes in surface temperature. J. Clim.; 2017; 30, pp. 2251-2267. [DOI: https://dx.doi.org/10.1175/JCLI-D-16-0408.1]

46. Jacobs, S. Observations of change in the Southern Ocean. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci.; 2006; 364, pp. 1657-1681. [DOI: https://dx.doi.org/10.1098/rsta.2006.1794] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16782605]

47. Baba, K.; Renwick, J. Aspects of intraseasonal variability of Antarctic sea ice in austral winter related to ENSO and SAM events. J. Glaciol.; 2017; 63, pp. 838-846. [DOI: https://dx.doi.org/10.1017/jog.2017.49]

48. Sen, G.A.; Matthew, H.E. Coupled ocean-atmosphere-ice response to variations in the southern annular mode. J. Clim.; 2006; 19, pp. 4457-4486.

49. Screen, J.A.; Bracegirdle, T.J.; Simmonds, I. Polar climate change as manifest in atmospheric circulation. Curr. Clim. Change Rep.; 2018; 4, pp. 383-395. [DOI: https://dx.doi.org/10.1007/s40641-018-0111-4]

50. De Lavergne, C.; Palter, J.B.; Galbraith, E.D.; Bernardello, R.; Marinov, I. Cessation of deep convection in the open Southern Ocean under anthropogenic climate change. Nat. Clim. Change; 2014; 4, pp. 278-282. [DOI: https://dx.doi.org/10.1038/nclimate2132]

51. Leung, S.; Cabré, A.; Marinov, I. A latitudinally banded phytoplankton response to 21st century climate change in the Southern Ocean across the CMIP5 model suite. Biogeosciences; 2015; 12, pp. 5715-5734. [DOI: https://dx.doi.org/10.5194/bg-12-5715-2015]

52. Clem, K.R.; Fogt, R.L. South Pacific circulation changes and their connection to the tropics and regional Antarctic warming in austral spring, 1979–2012. J. Geophys. Res. Atmos.; 2015; 120, pp. 2773-2792. [DOI: https://dx.doi.org/10.1002/2014JD022940]

53. Clem, K.R.; Renwick, J.A.; McGregor, J.; Fogt, R.L. The relative influence of ENSO and SAM on Antarctic Peninsula climate. J. Geophys. Res. Atmos.; 2016; 121, pp. 9324-9341. [DOI: https://dx.doi.org/10.1002/2016JD025305]

54. Trivelpiece, W.Z.; Hinke, J.T.; Miller, A.K.; Reiss, C.S.; Trivelpiece, S.G.; Watters, G.M. Variability in krill biomass links harvesting and climate warming to penguin population changes in Antarctica. Proc. Natl. Acad. Sci. USA; 2011; 108, pp. 7625-7628. [DOI: https://dx.doi.org/10.1073/pnas.1016560108] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21482793]