1. Introduction
The surface temperature has been rising significantly with increasing anthropogenic greenhouse gases (GHGs) since the Industrial Revolution, affecting the various components of Earth’s climate system and human activities (Hansen et al. 2010). The polar regions are most climatically sensitive to this effect as shown not only in observations but also by general circulation models (Houghton 2017). These polar regions warm particularly strongly, a robust feature called polar amplification (PA; Holland and Bitz 2003; Stuecker et al. 2018). In particular, the Arctic region warms 2–4 times more than the global mean in recent observations and model simulations, referred to as Arctic amplification (AA; Serreze et al. 2009; Pithan and Mauritsen 2014; Dai et al. 2019; Rantanen et al. 2022), with significant decreasing Arctic sea ice extent and Greenland ice sheet melting. However, there is a strong warming asymmetry between the Arctic and the Antarctic. The surface warming in the Antarctic has yet to be observed and is projected to be weaker than in the Arctic, with slightly increasing sea ice extent in the Antarctic (Marshall et al. 2015; Salzmann 2017; Singh et al. 2018; Li et al. 2021). The efficiency of the oceanic equatorward energy transport is revealed as the key to maintain Antarctic sea ice, which can be captured by higher-resolution models (Rackow et al. 2022).
Previous mass studies have tried to untangle the underlying mechanisms for AA. Both external forcing and internal climate feedback contribute to surface warming (Marotzke and Forster 2015). Various feedback diagnostic frameworks have been used to attribute surface warming to individual feedback processes, such as the radiative kernel method, partial radiative perturbation, climate feedback-response analysis method (CFRAM), online suppression method, and so on (Beer and Eisenman 2022; Taylor et al. 2022). Multiple processes contribute to AA, including surface albedo, lapse rate, cloud and water vapor feedbacks, and poleward energy transport (Goosse et al. 2018). The albedo feedback is typically a positive feedback because melting sea ice and snow lowers surface albedo, leading to increased absorption of shortwave radiation and amplified warming. AA is also found in climate models without sea ice feedback, which indicates that other processes also play important roles in AA (Alexeev et al. 2005; Cai 2005, 2006). The lapse-rate feedback is often positive because the stable stratification in polar regions suppresses vertical mixing and contributes to stronger warming near the surface than aloft (Boeke et al. 2021). The atmospheric heat transport into the Arctic from lower latitudes drives warming and thinner sea ice (Cai 2005, 2006; Park et al. 2015; Hegyi and Taylor 2017).
Driven by the seasonality of solar radiation in the polar region, feedbacks in the Arctic show significant seasonality, resulting in larger Arctic warming in boreal winter but small warming in summer (Screen and Simmonds 2010). The primary mechanism responsible for the winter amplification is the seasonal energy transfer mechanism (SETM). The sea ice retreat is strongest in boreal summer, as well as the increase in solar absorption by the Arctic Ocean. The surplus energy is temporally stored in the ocean in summer as the open water has greater heat capacity than sea ice, explaining why the actual summer warming in open water is far less than the warming in a hypothetical situation where the surface is still covered by sea ice and subject to the same amount of the additional solar energy absorption. In winter, the heat stored by newly opened water is then released, causing winter warming despite the lack of additional solar energy absorption in winter. Concurrently, the greater loss of energy from surface to atmosphere via enhanced upward surface turbulent and sensible heat fluxes. In other words, the increase of the thermal inertial from sea ice surface to open water surface in response to the anthropogenic radiative forcing acts to amplify the seasonal asymmetry of Arctic warming by suppressing the summer warming and amplifying the winter warming (Robock 1983; Dwyer et al. 2012; Sejas and Taylor 2023; Hahn et al. 2022; Hu et al. 2022).
Several hypotheses have been put forth to explain the weak warming in the Antarctic. The albedo feedback is also a positive contributor, but it is weaker than that in the Arctic due to the less melting of sea ice and snow (Manabe and Stouffer 1980). Ozone recovery over the Antarctic is associated with a widespread change in the Southern Hemisphere circulation and the Antarctic cooling (Thompson et al. 2011). The high elevation of the Antarctic Plateau is also considered as one of the origins of weak warming in the Antarctic (Kad et al. 2022). The interhemispheric asymmetries in ocean circulation, with sinking in the northern North Atlantic and upwelling around Antarctica, strongly influence the sea surface temperature response to GHG forcing, accelerating warming in the Arctic while delaying it in the Antarctic (Marshall et al. 2014; Chen and Tung 2018; Singh et al. 2018). The delayed warming of the Southern Ocean is attributed to strong heat absorption and the equatorward transport of surface water caused by the Antarctic Circumpolar Current (ACC; Armour et al. 2016). Hahn et al. (2020) suggest that the weaker Antarctic warming under CO2 forcing can be a consequence of the shallower, less intense climatological inversions due to limited atmospheric heat transport above the ice sheet elevation. The lapse-rate feedback in the Antarctic is weaker than that in the Arctic (Cai et al. 2021).
The uncertainty in PA projection is a prominent issue of polar warming. Differences in model physics, including parameterization schemes and model parameter estimations, are the most important sources of the uncertainty in the global warming projection (Bonan et al. 2021). Ye and Messori (2021) highlight the importance of constraining climate discrepancies in simulating recent and future climate variations. Large intermodel spread in surface warming appears at both poles in climate simulations (Hahn et al. 2021; Ye and Messori 2021). The albedo feedback is the major contributor to intermodel spread in both Arctic and Antarctic warming, followed by the lapse-rate feedback (Pithan and Mauritsen 2014). The moist poleward atmospheric heat transport also contributes largely to intermodel spread but is dampened by the dry poleward atmospheric heat transport (Hahn et al. 2021). Boeke and Taylor (2018) also point out that the intermodel spread in Arctic warming can be attributed to the difference in sea ice retreat regions. Hu et al. (2022) quantitatively proves that the intermodel spread in Arctic surface warming is mainly caused by the intermodel spread in sea ice melting in early summer and the associated processes involved in SETM. However, the intermodel spread in the warming in Antarctic is weaker than that in the Arctic (Taylor et al. 2022). Model discrepancies in climatological Antarctic sea ice area are a significant source of the intermodel spread in the Antarctic warming (Po-Chedley et al. 2018). Further investigation into the causes of warming asymmetry and its intermodel spread is still needed to reduce the uncertainties in climate simulations and future projections.
The CFRAM (Cai and Lu 2009) has been widely used to understand the change in temperature and its intermodel spread (Fan et al. 2021; Hu et al. 2022). Using this approach, we put the variables in the climate system on the same level and quantify their contributions to temperature changes. Especially for AA, a seasonal energy transfer mechanism (SETM) is proposed qualitatively by Chung et al. (2021), while Hu et al. (2022) extracts Arctic warming and its intermodel spread related to SETM quantitatively by using the CFRAM. It has been demonstrated that SETM plays an important role in Arctic warming and its intermodel spread. Is the SETM in the Antarctic as important as it is in the Arctic? An improved understanding of SETM will enhance our knowledge of the physical processes that drive the warming asymmetry between the two poles and its intermodel spread, thereby reducing the uncertainties in climate simulations and developing a better understanding of polar and global climate research.
The rest of the paper is organized as follows. Section 2 presents the data and methods applied. Section 3 describes the characteristics of warming asymmetry between the poles in CMIP6. Section 4 analyzes the source of multimodel ensemble (MME) warming asymmetry between the Arctic and the Antarctic. The intermodel spread in asymmetric Arctic and Antarctic warming is investigated in section 5. Conclusions are given in section 6.
2. Data and methods
The abrupt-4 × CO2 experiments from CMIP6 offer an opportunity to identify the major contributors of climate feedbacks to the asymmetric warming response to quadrupling CO2 and its intermodel spread, regardless of the influences from other external forcings and internal variability. In this study, we use CFRAM to decompose the simulated surface warming into partial surface warmings due to individual external forcing and internal feedback processes. The variables needed include skin and air temperatures, specific humidity, ozone, cloud cover, cloud liquid/ice water content, surface sensible and latent heat fluxes, downward/upward longwave radiative fluxes (LW) at the surface, downward/upward shortwave radiative fluxes (SW) at the surface, and downward SW and upward LW radiative fluxes at the top of the atmosphere. Eighteen CMIP6 models are available with the variables needed for our analysis, whose names and resolutions are summarized in Table 1. All climate data are downloaded from the CMIP6 (https://esgf-node.llnl.gov/search/cmip6/) archives. Then we adopt the model outputs from preindustrial (PI) and abrupt CO2 quadrupling (abrupt-4 × CO2) experiments, in which CO2 concentrations are quadrupled from the PI concentrations in the first years. The mean of the last 50 years in the PI runs of each model is calculated as the basic climate state. The abrupt-4 × CO2 experiments are integrated for 150 years and the mean of the last 20 years is adopted as the warming climate state. The difference between the warming and basic states is defined as the climate response to the CO2 increase. Then the model outputs are monthly averaged and have been interpolated to a 2.5° × 2.5° grid using the method of bilinear interpolation.
Details of the CMIP6 models used in this study.
By using CFRAM, the changes in surface temperature forced by CO2 from model outputs are decomposed into partial temperature changes (PTCs) due to the increase in CO2 concentration and individual radiative/nonradiative processes by solving the linearized infrared radiation transfer model. In the abrupt-4 × CO2 experiments, the increase in CO2 is the only external forcing to drive the changes in temperature and other feedback processes. The radiative feedback processes include surface albedo (AL), water vapor (WV), cloud shortwave (CLDS), and cloud longwave (CLDL) effects. The nonradiative processes comprise atmospheric horizontal large-scale circulation and vertical convection (ATM), oceanic circulation and surface heat storage (OCH), and surface turbulent heat fluxes (THF). The THF feedback is made of sensible heat and latent heat flux feedback (SH + LH). Moreover, the sources of intermodel spread in warming asymmetry between the two poles will be investigated via cross-model EOF analysis.
3. Characteristics of the warming asymmetry in CMIP6
Shown in Fig. 1a are the latitudinal profiles of surface warmings of individual CMIP6 simulations. The nearly exact overlapping of the total warming in MME (red curve) with its counterpart derived from the CFRAM decomposition (black line) is the evidence of the robustness of the CFRAM decomposition. It is seen that all these simulations exhibit the characteristics of PA, particularly the pattern of enhanced surface warming in the Arctic (60°–90°N). These features are consistent with findings by previous studies based on observation and model simulations (Taylor et al. 2013; Stuecker et al. 2018; Smith et al. 2019). The weakest warming occurs at 60°S along the Antarctic Circumpolar Current (ACC), which is considered the main cause of weakened surface warming in the Southern Ocean (Oke and England 2004; Hutchinson et al. 2013; Marshall et al. 2015; Armour et al. 2016). The intermodel spread in the warming in the Arctic is greater than that in the Antarctic, as indicated by the vertical width of shadings in Fig. 1a.
Surface warmings of the MME from model simulation (red solid lines) and CFRAM (black solid lines) and individual models (shadings). (a) Annual and zonal means (K); (b) as in (a), but normalized by the global-mean values; (c) seasonal patterns of the Arctic (60°–90°N) warmings; and (d) seasonal patterns of the Antarctic (60°–90°S) warmings. The solid red lines are derived from the model outputs and the solid black lines are derived from the CFRAM analysis. The light orange shadings show the full intermodel spread in the 18 CMIP6 models, whereas the dark orange shadings show the spread from the 25th to 75th percentiles.
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
To highlight the polar amplification and its hemispheric asymmetry, we calculate the zonal-mean surface warming normalized by the global-mean surface warming for individual models (Fig. 1b). This is following the definition of PA index, namely that the PA index is defined as the ratio of the Arctic (Antarctic) mean surface warming to the global mean. It is seen that, in model simulation, the Arctic warming is 2.5–3.5 times greater than the global-mean warming, while the Antarctic warming is only about 1.1–2.0 times as large as the global mean value. Thus, the Arctic warming is 2 times stronger than the Antarctic warming. The surface warming in the Arctic and the Antarctic (Figs. 1c,d) shows a similar seasonal pattern with maximum warming in winter but minimum warming in summer. In addition, the seasonal contrast of surface warming in the Antarctic is weaker than that in the Arctic. Both poles exhibit significant intermodel spread in surface warming, especially during the cold months.
4. Attribution of MME warming asymmetry
One of the most important characteristics in the polar regions is the seasonality in climatology and climate change (Holland and Bitz 2003; Sejas et al. 2014). As shown in Figs. 1c and 1d, the summer warmings over the Arctic and Antarctic are similar (about 5 K). The asymmetric warming between the two poles mainly occurs in winter. The CFRAM allows us to investigate the root sources of the asymmetric winter warming via calculating PTCs month by month.
Figure 2 shows the seasonal cycles of the zonal-mean PTCs due to individual processes (Figs. 2a–h) and the total surface temperature changes (Fig. 2i). The albedo feedback is the main process for the summer warming (Fig. 2a). Most of the summer warming due to sea ice melting is suppressed by the ocean heat storage term (Fig. 2g) due to the increase in effective heat capacity as sea ice declines. Such increase of effective heat capacity from sea ice surface to open water surface acts to suppress the summer warming and amplify the winter warming by temporally withholding the warming in summer and then releasing it in winter (Robock 1983; Dwyer et al. 2012; Sejas and Taylor 2023; Hahn et al. 2022; Hu et al. 2022). Furthermore, there is a strong negative correlation between the PTCs of ocean heat storage and turbulent heat flux terms (−0.939 in the Arctic and −0.946 in the Antarctic). This indicates that the greater energy loss via turbulent heat fluxes acts to damp the surface warming in winter as shown in Fig. 2h. The aforementioned processes in the Arctic have been summarized by Hu et al. (2022) as the seasonal energy transfer mechanism (SETM). Figure 2 illustrates that SETM also works in the Antarctic albeit with weaker amplitude.
Seasonal cycles of the zonal-mean PTCs (K) due to individual processes including changes in (a) surface albedo, (b) water vapor, (c),(d) cloud shortwave and longwave effects, (e) quadrupled CO2 concentration, (f) atmospheric dynamics, (g) oceanic circulation and surface heat storage terms, and (h) surface turbulence sensible and latent heat fluxes. (i) The sum of (a) to (h), namely the total surface warming. Black dots indicate that the area has passed the 90% significance test.
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
Feedback processes in the atmosphere also contribute to the asymmetric winter amplification, especially the cloud longwave and shortwave effects (Figs. 2c,d). Cloud shortwave feedback (Fig. 2c) is a cooling effect annually at both poles except in summer in the middle and high latitudes (60°–70°N) of the Northern Hemisphere. And it amplifies the asymmetric winter warming by inducing a stronger surface cooling in the Antarctic in summer. The cloud longwave process (Fig. 2d) over the Arctic causes a larger warming in boreal winter relative to that in boreal summer but causes uniform warming over the Antarctic throughout the whole year. The PTCs due to the atmospheric dynamic term in winter are larger than the PTCs in summer, further enhancing winter amplification in the Arctic and the Antarctic. However, atmospheric dynamics does not contribute much to the warming asymmetry as it warms both poles approximately the same. Note that the warming on the Antarctic continent primarily results from atmospheric heat transport. Water vapor contributes to the polar warming uniformly throughout the year in the Antarctic, but a stronger warming in the Arctic.
To accurately analyze the contributions of different processes to the seasonal warming pattern in both the Arctic and the Antarctic, we calculate the average warming for each season [spring, namely MAM (SON) in the Arctic (Antarctic); summer, JJA (DJF) in the Arctic (Antarctic); autumn, SON (MAM) in the Arctic (Antarctic); and winter, DJF (JJA) in the Arctic (Antarctic)]. It is seen from Fig. 3i that the warming asymmetry under CO2 forcing is most pronounced during winter with a maximum difference of approximately 7 K, whereas in summer it is less than 3 K. For individual feedback, the key components are the albedo feedback in summer, the heat storage term due to the increase in the effective heat capacity from sea ice to open water, and surface turbulent heat fluxes and they contribute significantly to the large warming asymmetry. The substantial loss of sea ice in the Arctic and steady ice and snow cover in the Antarctic continent amplifies the warming asymmetry between the two polar regions (Fig. 3a; Manabe and Stouffer 1980; Sévellec et al. 2017). Ocean heat storage term also exacerbates this asymmetry in autumn and winter (Fig. 3g) because of the equatorward heat transport from the Southern Ocean caused by the ACC, resulting in reduced heat release during the cold months (Armour et al. 2016). Additionally, the energy loss due to turbulent heat flux term suppresses the warming asymmetry throughout the year (Fig. 3h). The warming asymmetry is also intensified by water vapor feedback (Fig. 3b). It is worth pointing out that the summer cloud shortwave effect positively contributes to the asymmetry but is partially offset by the cloud longwave feedback (Figs. 3c,d).
Seasonal and regional average PTCs (K) due to individual processes including changes in (a) surface albedo, (b) water vapor, (c),(d) cloud shortwave and longwave effects, (e) quadrupled CO2 concentration, (f) atmospheric dynamics, (g) oceanic circulation and surface heat storage terms, and (h) surface turbulence sensible and latent heat fluxes. (i) The sum of (a) to (h). The orange (blue) bars represent the contribution from the Arctic (Antarctic), and the gray bars are for the difference in warming between the two regions. Solid bars represent the spread of PTCs respectively. The x axis of each subplot represents the four seasons, where “Spr” represents spring, MAM (SON) in the Arctic (Antarctic); “Sum” represents summer, JJA (DJF) in the Arctic (Antarctic); “Aut” represents autumn, SON (MAM) in the Arctic (Antarctic); and “Win” represents winter, DJF (JJA) in the Arctic (Antarctic).
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
5. Intermodel spread in Arctic and Antarctic warming
To gain more insights into the primary contributors to the intermodel spread in the asymmetric Arctic and Antarctic warming, a cross-model EOF analysis is applied to reveal the dominant seasonal pattern of the intermodel spread in total warming. The EOF analysis of the warming in the Arctic and Antarctic is carried out separately. The first two modes account for 78.5% and 9.2% (a total of 87.7%) variance of the intermodel spread in the total Arctic warming, and for 90.2% and 2.9% (a total of 93.1%) variance in the Antarctic. The first mode is dominant at both poles and thus we just focus on the first mode in the following discussion. For easy reference, we denote the EOF1 for the Arctic warming as EOF1_N and that for the Antarctic warming as EOF1_S and their corresponding principal components as PC1_N and PC1_S, respectively. Shown in Fig. 4 are the spatial patterns of EOF1_N and EOF1_S and their corresponding principal components (standardized), namely PC1_N and PC1_S. It is seen that the greatest intermodel spread appears over oceans in cold months: the Arctic Ocean in January–March (JFM; Fig. 4a) and the Southern Ocean in May–July (MJJ; Fig. 4c). The amplitude of the intermodel spread in the Arctic is much larger than that in the Antarctic. These characteristics suggest that the processes contributing to winter surface warming would be the critical factor for the intermodel spread in the polar warming.
EOF analysis of intermodel spreads of seasonal cycles of total temperature changes in the Arctic and the Antarctic. (a),(c) The latitude–time pattern (K) of the first EOF mode in the Arctic and the Antarctic, respectively, and (b),(d) the principal components (dimensionless) of the first EOF mode. The x axis in (b) and (d) represents the sequential model names as listed in Table 1.
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
Figure 5 is a scatterplot of PC1_N versus PC1_S. There is a prominent positive correlation (about 0.52) between PC1_N and PC1_S, which exceeds the 95% confidence level of statistical significance. The close relationship between PC1_N and PC1_S means that the model with a stronger (weaker) warming in the Arctic also has a stronger (weaker) warming in the Antarctic, which is consistent with the intermodel spread in the zonal and annual mean surface warming simulated by CMIP5 models in Hu et al. (2020). Because the intermodel spread of the Arctic warming captured by the EOF1_N is much stronger than that of the Antarctic warming captured by the EOF1_S (Fig. 4a vs Fig. 4c), it is expected that the intermodel spread of the warming asymmetry between the two polar regions can be captured by their differences, despite the positive correlation between PC1_N and PC1_S. To confirm this, we show in Fig. 6 the total Arctic-minus-Antarctic warming of each model (blue bars) and its decomposition into the portion that is projected on the EOF1 modes (red bars) and the residual part (yellow bars). Almost all models have positive blue bars except SAM0-UNICON, which means that the Arctic warms more than the Antarctic in all models except for SAM0-UNICON. The box plot on the right side of Fig. 6 provides a summary of the intermodel spread in the total warming asymmetry and its projection on the first EOF mode. The non-EOF-related portion of the Arctic-minus-Antarctic warmings (yellow bars) shows a relatively uniform positive value of about 4.5 K across all 18 models, which is very close to the MME value of the total Arctic-minus-Antarctic warming. The EOF1 portion of the Arctic-minus-Antarctic warmings (red bars), on the other hand, has positive values for some models and negative values for the other models and the median value of the red bars is close to zero. Therefore, the two first EOF modes, one for the intermodel spread in the Arctic warming (EOF1_N) and the other for the Antarctic warming (EOF1_S), also account for most of the intermodel spread in the asymmetric warming between the Arctic and Antarctic.
Scatterplot of PC1_N of the Arctic vs PC1_S of the Antarctic. The legend shows the individual CMIP6 models used. The solid gray line shows the linear regression between the two principal components. The numbers in the bottom right are their correlation (R) and p value (P).
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
Annual means of the Arctic-minus-Antarctic warming (K) of individual models with blue bars for the total warming, red bars for the EOF1_related portion, and yellow bars for the non-EOF1_related portion. The boxes plotted on the right show their spreads.
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
Figure 7 shows the regressed seasonal patterns of individual PTCs against the PC1_N (for the Arctic warming) and PC1_S (for the Antarctic warming). The result indicates that the only major process that contributes to the large intermodel spread in PTCs in warm months both in the Arctic and in the Antarctic is the albedo (Fig. 7a), which is suppressed by the ocean heat storage term (Fig. 7g), consistent with the findings of Hu et al. (2022). Chung et al. (2021) show that the maximum ocean heat release in winter is located in marginal sea ice regions. In late spring and summer, associated with the retreat of sea ice is the increasing of the effective heat capacity of the ocean surface layer, which temporally withholds the surface warming. Negative values of PTC-OCH mean that the temporal withholding of the heat acts to reduce the intermodel spread in late spring and early summer. On the other hand, due to the intense heat release during winter, the oceanic term contributes positively to the intermodel spread in warming by amplifying surface warming, which is partially dampened by surface turbulent sensible and latent heat fluxes (Fig. 7h). These characteristics suggest that the intermodel spread in the strength of SETM would be the key factor for the intermodel spread in the asymmetric warming. In addition, strong intermodel spread in polar warming is further enhanced by cloud longwave effects (Fig. 7d) and atmospheric dynamic processes (Fig. 7f) in winter. Associated with more heat release from oceans in cold months are less sea ice refreezing, stronger positive cloud longwave feedback, and stronger poleward atmospheric heat transport. Water vapor feedback is a small positive contributor to intermodel spread all year around. The contributions of CO2 and cloud shortwave effect are negligible compared to other processes. According to previous research, the high elevations of the Antarctic enhance the warming asymmetry between the two poles by influencing the atmospheric heat transport toward the poles (Salzmann 2017; Hahn et al. 2020). The warming contribution from ATM is stronger in the Arctic. Consequently, the intermodel spread in warming due to ATM is larger in the Arctic than in the Antarctic.
Regressed seasonal cycles of the zonal mean of partial surface temperature changes (K) due to individual processes against PC1_N shown in Fig. 4b (for the Arctic warming) and PC1_S in Fig. 4d (for the Antarctic warming). Partial temperature changes due to feedbacks of (a) albedo, (b) atmospheric water vapor, (c) shortwave effect of clouds, (d) longwave effects of clouds, (e) the concentration of CO2, (f) atmospheric heat transport, (g) oceanic circulation and ocean heat storage term, and (h) surface turbulence sensible and latent heat fluxes. Black dots indicate the 95% confidence level of statistical significance.
Citation: Journal of Climate 36, 23; 10.1175/JCLI-D-23-0118.1
6. Conclusions
In this study, we investigate the main sources of the asymmetric Arctic and Antarctic warming and its intermodel spread under the forcing of abrupt-4 × CO2 from the CMIP6 by using the climate feedback-response analysis method (CFRAM), a framework for calculating the contributions of climate feedbacks to total polar warming. The results demonstrate that the seasonal energy transfer mechanism (SETM) plays a primary role in both Arctic and Antarctic MME surface warmings. In response to the CO2 forcing, there is more melting of sea ice in the summer months, which in turn results in an increase in the effective heat capacity of the oceans. The increase of the effective heat capacity temporally withholds part of the summer warming due to more absorption of solar energy. As a result, the summer warming in the polar regions is weaker than the annual mean. The temporally stored heat in summer is released in the cold months, which is the main contributor to the pronounced PA, although part of the heat released in the cold months gets lost to the atmosphere above via the strengthening of the sensible and latent heat fluxes from the oceans to the atmosphere.
We use a cross-model EOF analysis to identify the key processes that drive the intermodel spread in the asymmetric Arctic and Antarctic warming. It is found that the intermodel spread can be adequately captured by the intermodel spread in the seasonal pattern of the first EOF mode for each of the two polar regions. The maximum of the intermodel spread appears over oceans in the cold months. The regressed seasonal cycles of individual PTCs over the Arctic and the Antarctic against their corresponding PCs of the EOF1 modes indicate that the intermodel strength of the SETM is responsible for the intermodel spread in the asymmetric warming. Large intermodel spread in the albedo feedback in summer is largely suppressed by the ocean heat storage term. Hence, there is a little intermodel spread in the two polar regions in the warm months. The intermodel spread in the amount of release of heat from the oceans is the main source for the intermodel spread in the surface warming in the cold months. The intermodel spread in the asymmetric polar warming is primarily driven by the intermodel spreads of the oceanic dynamic and heat storage term and albedo feedback but suppressed by the intermodel spread in the surface LH and SH heat fluxes. All these three processes are parts of the SETM, reconfirming the dominant role of SETM in leading to the intermodel spread in asymmetric polar warming. In addition, water vapor and atmospheric heat transport strengthen the intermodel spread in polar warming with a more prominent effect in the Antarctic.
Acknowledgments.
We thank the editor Dr. Yi Deng and two anonymous reviewers for their constructive comments leading to the improvement of the manuscript. This study was supported by the National Key Research and Development Program of China (2019YFA0607000), the National Natural Science Foundation of China (42222502 and 42075028), the Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) (Grant SML2021SP302), and the Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies (Grant 2020B1212060025).
Data availability statement.
All data used in this study are derived from the model simulations produced by the Coupled Model Intercomparison Project version 6 (CMIP6), which are archived and freely accessible at https://www.wcrp-climate.org/wgcm-cmip/wgcm-cmip6. The CMIP6 model simulations used in this study are listed in Table 1.
REFERENCES
Alexeev, V. A., P. L. Langen, and J. R. Bates, 2005: Polar amplification of surface warming on an aquaplanet in “ghost forcing” experiments without sea ice feedbacks. Climate Dyn., 24, 655–666, https://doi.org/10.1007/s00382-005-0018-3.
Armour, K. C., J. Marshall, J. R. Scott, A. Donohoe, and E. R. Newsom, 2016: Southern Ocean warming delayed by circumpolar upwelling and equatorward transport. Nat. Geosci., 9, 549–554, https://doi.org/10.1038/ngeo2731.
Beer, E., and I. Eisenman, 2022: Revisiting the role of the water vapor and lapse rate feedbacks in the Arctic amplification of climate change. J. Climate, 35, 2975–2988, https://doi.org/10.1175/JCLI-D-21-0814.1.
Boeke, R. C., and P. C. Taylor, 2018: Seasonal energy exchange in sea ice retreat regions contributes to differences in projected Arctic warming. Nat. Commun., 9, 5017, https://doi.org/10.1038/s41467-018-07061-9.
Boeke, R. C., P. C. Taylor, and S. A. Sejas, 2021: On the nature of the Arctic’s positive lapse-rate feedback. Geophys. Res. Lett., 48, e2020GL091109, https://doi.org/10.1029/2020GL091109.
Bonan, D. B., F. Lehner, and M. M. Holland, 2021: Partitioning uncertainty in projections of Arctic sea ice. Environ. Res. Lett., 16, 044002, https://doi.org/10.1088/1748-9326/abe0ec.
Cai, M., 2005: Dynamical amplification of polar warming. Geophys. Res. Lett., 32, L22710, https://doi.org/10.1029/2005GL024481.
Cai, M., 2006: Dynamical greenhouse-plus feedback and polar warming amplification. Part I: A dry radiative-transportive climate model. Climate Dyn., 26, 661–675, https://doi.org/10.1007/s00382-005-0104-6.
Cai, M., and J. Lu, 2009: A new framework for isolating individual feedback processes in coupled general circulation climate models. Part II: Method demonstrations and comparisons. Climate Dyn., 32, 887–900, https://doi.org/10.1007/s00382-008-0424-4.
Cai, S., P.-C. Hsu, and F. Liu, 2021: Changes in polar amplification in response to increasing warming in CMIP6. Atmos. Oceanic Sci. Lett., 14, 100043, https://doi.org/10.1016/j.aosl.2021.100043.
Chen, X., and K.-K. Tung, 2018: Global surface warming enhanced by weak Atlantic overturning circulation. Nature, 559, 387–391, https://doi.org/10.1038/s41586-018-0320-y.
Chung, E.-S., K.-J. Ha, A. Timmermann, M. F. Stuecker, T. Bodai, and S.-K. Lee, 2021: Cold-season Arctic amplification driven by Arctic Ocean-mediated seasonal energy transfer. Earth’s Future, 9, e2020EF001898, https://doi.org/10.1029/2020EF001898.
Dai, A., D. Luo, M. Song, and J. Liu, 2019: Arctic amplification is caused by sea-ice loss under increasing CO2. Nat. Commun., 10, 121, https://doi.org/10.1038/s41467-018-07954-9.
Dwyer, J. G., M. Biasutti, and A. H. Sobel, 2012: Projected changes in the seasonal cycle of surface temperature. J. Climate, 25, 6359–6374, https://doi.org/10.1175/JCLI-D-11-00741.1.
Fan, H., X. Hu, S. Yang, Y.-S. Choi, and Y.-K. Lee, 2021: Process-based analysis of relative contributions to the multi-model warming projection over East Asia. Climate Dyn., 56, 2729–2747, https://doi.org/10.1007/s00382-020-05614-7.
Fu, Q., and K. N. Liou, 1992: On the correlated k-distribution method for radiative transfer in nonhomogeneous atmosphere. J. Atmos. Sci., 49, 2139–2156, https://doi.org/10.1175/1520-0469(1992)049<2139:OTCDMF>2.0.CO;2.
Fu, Q., and K. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci., 50, 2008–2025, https://doi.org/10.1175/1520-0469(1993)050<2008:POTRPO>2.0.CO;2.
Goosse, H., and Coauthors, 2018: Quantifying climate feedbacks in polar regions. Nat. Commun., 9, 1919, https://doi.org/10.1038/s41467-018-04173-0.
Hahn, L. C., K. C. Armour, D. S. Battisti, A. Donohoe, A. G. Pauling, and C. M. Bitz, 2020: Antarctic elevation drives hemispheric asymmetry in polar lapse rate climatology and feedback. Geophys. Res. Lett., 47, e2020GL088965, https://doi.org/10.1029/2020GL088965.
Hahn, L. C., K. C. Armour, M. D. Zelinka, C. M. Bitz, and A. Donohoe, 2021: Contributions to polar amplification in CMIP5 and CMIP6 models. Front. Earth Sci., 9, 710036, https://doi.org/10.3389/feart.2021.710036.
Hahn, L. C., K. C. Armour, D. S. Battisti, I. Eisenman, and C. M. Bitz, 2022: Seasonality in Arctic warming driven by sea ice effective heat capacity. J. Climate, 35, 1629–1642, https://doi.org/10.1175/JCLI-D-21-0626.1.
Hansen, J., R. Ruedy, M. Sato, and K. Lo, 2010: Global surface temperature change. Rev. Geophys., 48, RG4004, https://doi.org/10.1029/2010RG000345.
Hegyi, B. M., and P. C. Taylor, 2017: The regional influence of the Arctic oscillation and Arctic dipole on the wintertime Arctic surface radiation budget and sea ice growth. Geophys. Res. Lett., 44, 4341–4350, https://doi.org/10.1002/2017GL073281.
Holland, M. M., and C. M. Bitz, 2003: Polar amplification of climate change in coupled models. Climate Dyn., 21, 221–232, https://doi.org/10.1007/s00382-003-0332-6.
Houghton, J., 2017: The science of global warming. Climatic Change, 188, 3–13, https://doi.org/10.1179/isr.2001.26.4.247.
Hu, X., H. Fan, M. Cai, S. A. Sejas, P. Taylor, and S. Yang, 2020: A less cloudy picture of the inter-model spread in future global warming projections. Nat. Commun., 11, 4472, https://doi.org/10.1038/s41467-020-18227-9.
Hu, X., Y. Liu, Y. Kong, and Q. Yang, 2022: A quantitative analysis of the source of inter‐model spread in Arctic surface warming response to increased CO2 concentration. Geophys. Res. Lett., 49, e2022GL100034, https://doi.org/10.1029/2022GL100034.
Hutchinson, D. K., M. H. England, A. Santoso, and A. M. Hogg, 2013: Interhemispheric asymmetry in transient global warming: The role of Drake Passage. Geophys. Res. Lett., 40, 1587–1593, https://doi.org/10.1002/grl.50341.
Kad, P., M. T. Blau, K.-J. Ha, and J. Zhu, 2022: Elevation-dependent temperature response in early Eocene using paleoclimate model experiment. Environ. Res. Lett., 17, 114038, https://doi.org/10.1088/1748-9326/ac9c74.
Li, X., and Coauthors, 2021: Tropical teleconnection impacts on Antarctic climate changes. Nat. Rev. Earth Environ., 2, 680–698, https://doi.org/10.1038/s43017-021-00204-5.
Manabe, S., and R. J. Stouffer, 1980: Sensitivity of a global climate model to an increase of CO2 concentration in the atmosphere. J. Geophys. Res., 85, 5529–5554, https://doi.org/10.1029/JC085iC10p05529.
Marotzke, J., and P. M. Forster, 2015: Forcing, feedback and internal variability in global temperature trends. Nature, 517, 565–570, https://doi.org/10.1038/nature14117.
Marshall, J., K. C. Armour, J. R. Scott, Y. Kostov, U. Hausmann, D. Ferreira, T. G. Shepherd, and C. M. Bitz, 2014: The ocean’s role in polar climate change: Asymmetric Arctic and Antarctic responses to greenhouse gas and ozone forcing. Philos. Trans. Roy. Soc., A372, 20130040, https://doi.org/10.1098/rsta.2013.0040.
Marshall, J., J. R. Scott, K. C. Armour, J.-M. Campin, M. Kelley, and A. Romanou, 2015: The ocean’s role in the transient response of climate to abrupt greenhouse gas forcing. Climate Dyn., 44, 2287–2299, https://doi.org/10.1007/s00382-014-2308-0.
Oke, P. R., and M. H. England, 2004: Oceanic response to changes in the latitude of the Southern Hemisphere subpolar westerly winds. J. Climate, 17, 1040–1054, https://doi.org/10.1175/1520-0442(2004)017<1040:ORTCIT>2.0.CO;2.
Park, H.-S., S. Lee, Y. Kosaka, S.-W. Son, and S.-W. Kim, 2015: The impact of Arctic winter infrared radiation on early summer sea ice. J. Climate, 28, 6281–6296, https://doi.org/10.1175/JCLI-D-14-00773.1.
Pithan, F., and T. Mauritsen, 2014: Arctic amplification dominated by temperature feedbacks in contemporary climate models. Nat. Geosci., 7, 181–184, https://doi.org/10.1038/ngeo2071.
Po-Chedley, S., K. C. Armour, C. M. Bitz, M. D. Zelinka, B. D. Santer, and Q. Fu, 2018: Sources of intermodel spread in the lapse rate and water vapor feedbacks. J. Climate, 31, 3187–3206, https://doi.org/10.1175/JCLI-D-17-0674.1.
Rackow, T., S. Danilov, H. F. Goessling, H. H. Hellmer, D. V. Sein, T. Semmer, D. Sidorenko, and T. Jung, 2022: Delayed Antarctic sea-ice decline in high-resolution climate change simulations. Nat. Commun., 13, 637, https://doi.org/10.1038/s41467-022-28259-y.
Rantanen, M., A. Y. Karpechko, A. Lipponen, K. Nordling, O. Hyvärinen, K. Ruosteenoja, T. Vihma, and A. Laaksonen, 2022: The Arctic has warmed nearly four times faster than the globe since 1979. Commun. Earth Environ., 3, 168, https://doi.org/10.1038/s43247-022-00498-3.
Robock, A., 1983: Ice and snow feedbacks and the latitudinal and seasonal distribution of climate sensitivity. J. Atmos. Sci., 40, 986–997, https://doi.org/10.1175/1520-0469(1983)040<0986:IASFAT>2.0.CO;2.
Salzmann, M., 2017: The polar amplification asymmetry: Role of Antarctic surface height. Earth Syst. Dyn., 8, 323–336, https://doi.org/10.5194/esd-8-323-2017.
Screen, J. A., and I. Simmonds, 2010: Increasing fall-winter energy loss from the Arctic Ocean and its role in Arctic temperature amplification. Geophys. Res. Lett., 37, L16707, https://doi.org/10.1029/2010GL044136.
Sejas, S. A., and P. C. Taylor, 2023: The role of sea ice in establishing the seasonal Arctic warming pattern. Environ. Res. Climate, 2, 035008, https://doi.org/10.1088/2752-5295/ace20f.
Sejas, S. A., M. Cai, A. Hu, G. A. Meehl, W. Washington, and P. C. Taylor, 2014: Individual feedback contributions to the seasonality of surface warming. J. Climate, 27, 5653–5669, https://doi.org/10.1175/JCLI-D-13-00658.1.
Serreze, M. C., A. P. Barrett, J. C. Stroeve, D. N. Kindig, and M. M. Holland, 2009: The emergence of surface-based Arctic amplification. Cryosphere, 3, 11–19, https://doi.org/10.5194/tc-3-11-2009.
Sévellec, F., A. V. Fedorov, and W. Liu, 2017: Arctic sea-ice decline weakens the Atlantic meridional overturning circulation. Nat. Climate Change, 7, 604–610, https://doi.org/10.1038/nclimate3353.
Singh, H. A., O. A. Garuba, and P. J. Rasch, 2018: How asymmetries between Arctic and Antarctic climate sensitivity are modified by the ocean. Geophys. Res. Lett., 45, 13 031–13 040, https://doi.org/10.1029/2018GL079023.
Smith, D. M., and Coauthors, 2019: The Polar Amplification Model Intercomparison Project (PAMIP) contribution to CMIP6: Investigating the causes and consequences of polar amplification. Geosci. Model Dev., 12, 1139–1164, https://doi.org/10.5194/gmd-12-1139-2019.
Stuecker, M. F., and Coauthors, 2018: Polar amplification dominated by local forcing and feedbacks. Nat. Climate Change, 8, 1076–1081, https://doi.org/10.1038/s41558-018-0339-y.
Taylor, P. C., M. Cai, A. Hu, J. Meehl, W. Washington, and G. J. Zhang, 2013: A decomposition of feedback contributions to polar warming amplification. J. Climate, 26, 7023–7043, https://doi.org/10.1175/JCLI-D-12-00696.1.
Taylor, P. C., and Coauthors, 2022: Process drivers, inter-model spread, and the path forward: A review of amplified Arctic warming. Front. Earth Sci., 9, 758361, https://doi.org/10.3389/feart.2021.758361.
Thompson, D. W. J., S. Solomon, P. J. Kushner, M. H. England, K. M. Grise, and D. J. Karoly, 2011: Signatures of the Antarctic ozone hole in Southern Hemisphere surface climate change. Nat. Geosci., 4, 741–749, https://doi.org/10.1038/ngeo1296.
Ye, K., and G. Messori, 2021: Inter-model spread in the wintertime Arctic amplification in the CMIP6 models and the important role of internal climate variability. Global Planet. Change, 204, 103543, https://doi.org/10.1016/j.gloplacha.2021.103543.
