1. Introduction
The understanding of Antarctic climate change has often been hindered by the spatial and temporal paucity of observations. However, it has recently been recognized that surface air temperature (SAT) has increased across many parts of the Antarctic, particularly over the Antarctic Peninsula and continental West Antarctica, but with marked spatial and temporal variability (e.g., Turner et al. 2005; Marshall 2007; Monaghan et al. 2008; Steig et al. 2009; Schneider et al. 2012a; Bromwich et al. 2013, and references therein). Given the potential implications for ice sheet mass balance, and thereby global sea level rise (e.g., Rignot et al. 2011), an in-depth understanding of the nature and causes of long-term Antarctic SAT trends is required.
Antarctic SAT trends during austral summer [December–February (DJF)] are dominated by rapid warming over the Antarctic Peninsula (e.g., Turner et al. 2005; Marshall 2007), and have been the focus of many investigations. This summertime warming has largely been attributed to a trend toward the positive phase of the southern annular mode (SAM), manifested as a pattern of negative (positive) pressure trends over the Antarctic continent (midlatitudes), and consequently enhancing warm air advection over much of the Antarctic Peninsula (Thompson and Solomon 2002; Marshall et al. 2006; Marshall 2007; Thompson et al. 2011; Simpkins et al. 2012). This positive SAM trend has been linked to stratospheric ozone depletion, and to a lesser degree, increased greenhouse gases (Arblaster and Meehl 2006; Thompson et al. 2011; Simpkins and Karpechko 2012). While stratospheric ozone depletion peaks during spring, coupling with the SAM is restricted to the summer season when the stratosphere and troposphere are dynamically linked owing to a decaying polar vortex (e.g., Thompson and Solomon 2002). Accordingly, SAM trends are also primarily constrained to the summer season, and, as such, the SAM is unable to explain atmospheric circulation and temperature trends during other times (Thompson et al. 2011; Simpkins et al. 2012).
However, several recent studies have identified that SAT warming signals also extend throughout much of continental West Antarctica during both austral winter [June–August (JJA)] and spring [September–November (SON)] (Steig et al. 2009; Schneider et al. 2012a). In fact, depending on the temperature reconstruction utilized, changes have been estimated to be ~0.5°–1.0°C decade−1, establishing West Antarctica as one of the fastest warming regions globally (Bromwich et al. 2013). These SAT trends have been increasingly linked to changes in low-latitude sea surface temperature (SST) and the corresponding impacts on the extratropical atmospheric circulation (Ding et al. 2011; Schneider et al. 2012a). For example, Ding et al. (2011) suggested that the wintertime increase in geopotential height over West Antarctica, and consequently the surface warming associated with warm air advection, may be part of a stationary Rossby wave train forced by higher SSTs in the central Pacific. By contrast, springtime atmospheric circulation trends are characterized by negative geopotential height anomalies over the high-latitude South Pacific, which similarly promotes warm air advection, and thereby positive SAT trends, over much of West Antarctica (Fig. 1a). A large proportion of these circulation trends have been connected to the modes of high-latitude atmospheric variability associated with the El Niño–Southern Oscillation (ENSO), representing the Pacific–South American (PSA) teleconnection patterns (Schneider et al. 2012a). Climatic change in West Antarctica is thus strongly sensitive to teleconnections associated with Pacific SST variability, as also identified in recent paleoclimate studies (Okumura et al. 2012; Steig et al. 2013).
Springtime (September–November) trends in (a) ERA-Interim 950-hPa wind (vectors), HadISST sea ice concentration (shading), and Antarctic station surface air temperature (colored dots); and (b) HadISST sea surface temperature, all calculated over 1979–2009. Black lines in (b) delineate the approximate ocean boundaries used for subsequent model experiments (SSTIND, SSTPAC, SSTATL, and SSTGLO; see Methods), with arrows signifying a straight southward extension to the Antarctic continent; A-B Seas in (a) refers to the Amundsen–Bellingshausen Seas.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Given such corroborating evidence that tropical SST trends may be driving, or contributing to, Antarctic SAT trends, it is prudent to establish a deeper understanding of tropical–extratropical interactions. On interannual time scales, Southern Hemisphere (SH) teleconnections associated with SST variability in the Pacific have been well documented (e.g., Karoly 1989; Mo and Higgins 1998; Garreaud and Battisti 1999; Ciasto and Thompson 2008; Ciasto and England 2011; Ding et al. 2011; Schneider et al. 2012a,b; Simpkins et al. 2012). In contrast, relatively few studies have examined the corresponding teleconnections arising from the Atlantic or Indian Oceans (exceptions include Haarsma and Hazeleger 2007; Luffman et al. 2010; Timmermann et al. 2010; Taschetto et al. 2011; Okumura et al. 2012; Taschetto and Ambrizzi 2012; Li et al. 2014), particularly beyond interannual time scales. Nevertheless, Luffman et al. (2010) suggest that long-term Indian Ocean warming likely has a minimal impact on the SH atmospheric circulation. Conversely, Okumura et al. (2012) identify links between Atlantic SST variability and the climate of West Antarctica, and argue that a change in the phase of this variability may have contributed to contemporary climate trends in this region, as also corroborated by Li et al. (2014). However, the dynamical mechanisms forcing this association remain largely unexplored, and as such, several questions remain.
Given the implications for SAT and broader-scale climatic change across Antarctica, this study aims to evaluate the extent to which atmospheric circulation trends in the SH extratropics may be forced by SST trends. Particular emphasis is placed on understanding teleconnections associated with the Atlantic, and thus we build upon the studies of Okumura et al. (2012) and Li et al. (2014). To this end, we use an atmospheric general circulation model (AGCM) in addition to observations to address the following: 1) what impact, if any, do SST trends in the Pacific, Indian, and Atlantic Oceans have on extratropical geopotential height trends in the SH between 1979–2009? and 2) what are the physical mechanisms governing an atmospheric teleconnection between the Atlantic and the SH extratropics? Owing to the springtime peak apparent in both the amplitude of SAT trends (e.g., Schneider et al. 2012a; Bromwich et al. 2013) and the teleconnections emanating from the Pacific and Atlantic Oceans (e.g., Haarsma and Hazeleger 2007; Jin and Kirtman 2009; Schneider et al. 2012b; Simpkins et al. 2012), subsequent analyses are restricted solely to austral spring (SON).
The outline of the paper is as follows. Section 2 describes the observational records utilized in this study and outlines the atmospheric model setup and experimental design. Section 3 analyzes the patterns of atmospheric circulation associated with SST trends in individual ocean basins. The atmospheric dynamics of an Atlantic–Antarctic teleconnection are diagnosed in section 4, and the cause of the Atlantic-related atmospheric circulation trend pattern examined in section 5, along with the climatic impacts of these trends. Finally, a summary and discussion are provided in section 6.
2. Data and numerical experiments
a. Observational and reanalysis datasets
To both motivate and validate subsequent model simulations, a suite of observational and reanalysis datasets are used throughout this investigation. Monthly-mean sea surface temperature and sea ice concentration (SIC) are taken from the Hadley Centre Sea Ice and Sea Surface Temperature (HadISST) dataset (Rayner et al. 2003). The data are available on a 1° × 1° latitude–longitude grid, and are derived using a blended analysis of in situ measurements and satellite retrievals. We additionally use 18 of the most temporally continuous records of observed Antarctic surface air temperature from the Reference Antarctic Data for Environmental Research (READER) archive (Turner et al. 2004), a collection of meteorological measurements obtained from Antarctic research and automatic weather stations. The 500-hPa geopotential height (Z500) and 950-hPa wind fields are taken from the Interim European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA-Interim) (Dee et al. 2011). Note that our results have also been repeated using alternative reanalysis products [e.g., National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) and NCEP–U.S. Department of Energy (DOE) Atmospheric Model Intercomparison Project phase 2 (AMIP-II) reanalyses], and remain qualitatively similar to those presented here. Due to the sparse and temporally limited observational record, analyses are restricted to the postsatellite era, 1979–2009, when data are more reliable, and spatial coverage is more complete over the SH mid to high latitudes (Bromwich and Fogt 2004). As described above, analyses are further constrained to austral spring (SON).
b. Atmospheric model setup, experimental design, and validation
Various numerical experiments are performed using the NCAR Community Atmosphere Model, version 3 (CAM3), a complete description of which can be found in Collins et al. (2006) and Hurrell et al. (2006). CAM3 has been used extensively in investigations of climate research relevant to this study, including tropical climate variability (e.g., Deser et al. 2006), Antarctic climate variability and change (e.g., Bracegirdle et al. 2008; Raphael and Holland 2006), and tropical–extratropical interactions (e.g., Okumura et al. 2012; Schneider et al. 2012a). For all simulations, CAM3 was configured at T42 horizontal spectral resolution (approximately 2.8° latitude × 2.8° longitude) with 26 hybrid sigma-pressure vertical levels.
Prior to assessing the relationships between SST trends in individual basins and the extratropical atmospheric circulation, it is prudent to assess the capability of CAM3 to simulate the observed pattern of springtime Z500 trends. To do so, CAM3 is forced with observed monthly varying SST over the global oceans between January 1978 and December 2009 using the HadISST dataset; sea ice conditions are further prescribed as a repeating pattern of monthly mean climatologies. This control experiment, termed SSTGLO, is integrated 12 times from different atmospheric initial conditions to account for the internal variability in the climate system. The ensemble mean is then assessed over austral spring during 1979–2009.
Figure 2 compares springtime Z500 trends from ERA-Interim (shading) and the ensemble mean from the SSTGLO simulations (contours). It can be seen that the spatial pattern of modeled Z500 trends bears similarity to ERA-Interim, as illustrated by the gray hatching, which marks where trends are consistent in sign. In particular, SSTGLO is successful in replicating the negative Z500 trends over the high-latitude South Pacific, congruous with the cyclonic wind pattern depicted in Fig. 1a. Furthermore, the observed positive Z500 structure over the central midlatitude Pacific is well captured by SSTGLO. Nevertheless, Fig. 2 also reveals discrepancies between simulated atmospheric circulation trends and their observational counterparts, largely in relation to latitudinal/longitudinal shifts in the location of Z500 structures. For example, relative to ERA-Interim, the modeled negative Z500 trends located over the South Pacific are shifted eastward toward the Amundsen Sea. Additionally, SSTGLO fails to capture the observed positive Z500 trends centered over the South Atlantic and directly south of Australia. Finally, the magnitude of trends is typically underestimated in SSTGLO. However, the modeled Z500 trends are only associated with the applied SST forcing, and other factors may also contribute to the observed Z500 trend pattern depicted in Fig. 2. Nonetheless, SSTGLO captures some notable aspects of the large-scale Z500 trends, making CAM3 a suitable tool to further investigate tropical–extratropical interactions over multidecadal time scales.
Springtime (September–November) Z500 trends from ERA-Interim (shading) and the SSTGLO model simulations (contours) calculated over 1979–2009. Solid (dashed) contours denote positive (negative) trends, and are drawn at intervals of 7 m. Hatching indicates where ERA-Interim and SSTGLO agree on the sign of the trend.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
While the similarities evident in Fig. 2 highlight that global SST trends likely play a role in forcing Z500 trends over the SH extratropics, this study aims to additionally separate and diagnose the impact of individual ocean basins. Thus, further idealized model simulations are performed that isolate SST trends in the Pacific (SSTPAC), Indian (SSTIND), and Atlantic (SSTATL) Oceans. For each of these experiments, CAM3 is forced with observed monthly varying SST between January 1978 and December 2009 in the named basin (see Fig. 1b for spatial domains), but linearly detrended SST elsewhere. In SSTPAC, for example, observed SST are prescribed throughout the Pacific Ocean (i.e., including trends and variability), whereas detrended SST are prescribed in both the Indian and Atlantic Oceans (i.e., only including variability); thus, in this instance, modeled atmospheric circulation changes can be linked to Pacific SST trends. To minimize spurious atmospheric responses, linear damping was applied at the domain boundaries over a 10° latitude–longitude band. As in SSTGLO, the average of 12 ensemble members initialized from different atmospheric initial conditions is analyzed. Note that additional experiments performed using a repeating cycle of climatological SST beyond the basin of interest, rather than detrended SST as described above, produce quantitatively similar results to those presented here (not shown).
3. Modeled SH atmospheric response to SST trends
Figure 3 displays springtime Z500 trends from SSTGLO (Fig. 3a), SSTPAC (Fig. 3b), SSTATL (Fig. 3c), and SSTIND (Fig. 3d); solid-red (dashed-blue) contours denote positive (negative) trends, and hatching highlights where 9 of 12 ensemble members agree on the sign of the trend, thus functioning as a measure of robustness across model integrations. Although geopotential height trends are presented only at 500 hPa, these are found to be largely consistent with trends evaluated at all levels of the troposphere (not shown), demonstrating an equivalent barotropic response over the mid to high southern latitudes. Significant differences in both sign and structure of Z500 trends are seen across the SST experiments, particularly between SSTPAC/SSTIND and SSTATL. For example, SSTPAC (Fig. 3b) is characterized by a robust positive pressure pattern extending over much of the South Pacific and into southern South America. This positive lobe is coupled with pronounced negative Z500 trends spanning across southern Australia and New Zealand. Similarly, SSTIND (Fig. 3d) is dominated by distinct positive trends centered over the Amundsen–Bellingshausen Seas, with less coherent structures observed elsewhere. Both SSTPAC and SSTIND thus simulate Z500 trends of opposite sign to SSTGLO over the high-latitude South Pacific (cf. Figs. 2, 3b, and 3d).
Springtime (September–November) Z500 trends for the various SST-forced atmospheric model experiments calculated over 1979–2009 (see text for explanation of forcing). Positive (negative) trends are denoted by solid-red (dashed-blue) contours, drawn at intervals of 7 m; the zero contour has been omitted. Hatching indicates where 9 of 12 ensemble members agree on the sign of the trend.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Interestingly, the South Pacific negative pressure center seen in SSTGLO and reanalyses is only reproduced when SST trends are applied in the Atlantic Ocean (Fig. 3c). In fact, the spatial pattern of Z500 trends associated with SSTATL projects very strongly onto that of SSTGLO, such that large-scale consistency in trend structures, magnitudes, and sign are observed (cf. Figs. 3a and 3c). The spatial correlation between these two trend patterns is 0.76. However, subtle regional differences are also apparent when comparing SSTATL to SSTGLO. For example, the negative Z500 trends located over the South Pacific are extended farther northward into South America, and the positive Z500 trends cover a larger area extending over Australia and the subtropical Indian and Pacific Oceans. Nevertheless, the similarities between SSTATL and SSTGLO suggests that SST trends in the Atlantic may play a significant role in forcing contemporary springtime atmospheric circulation trends in the SH extratropics.
Figure 3 thus demonstrates that the Z500 trend structures associated with SSTGLO more closely reflect the impact of SST trends in the Atlantic, rather than the Pacific or Indian Oceans (cf. Figs. 3). Given that Pacific SST variability is known to strongly impact the extratropical atmosphere (e.g., Karoly 1989; Ciasto and Thompson 2008; Simpkins et al. 2012), it is somewhat surprising that SSTPAC does not display stronger similarities with SSTGLO. This may be related to the reduced amplitude of Pacific SST trends in comparison to the Atlantic (Fig. 1b), or to the lack of ocean–atmosphere coupling inherent in AGCM experiments. Furthermore, it is also important to note that SSTATL, SSTPAC, and SSTIND do not combine to reproduce the spatial pattern of Z500 trends modeled by SSTGLO (not shown). As illustrated by Fig. 4, it is seen that the summed precipitation response for the individual basin experiments is considerably larger than that of SSTGLO, suggesting that nonlinearities in convective precipitation, and the associated tropical–extratropical atmospheric dynamics, may play a role in driving the nonlinear Z500 trends observed in Fig. 3. Moreover, it must be remembered that complex interbasin atmosphere–ocean interactions (e.g., Wang 2006; Keenlyside and Latif 2007; Kushnir et al. 2010; Timmermann et al. 2010; Ding et al. 2012; Luo et al. 2012; Santoso et al. 2012, and references therein) also complicate the diagnosis of tropical–extratropical interactions, particularly when isolating the impact of individual basins as in these idealized simulations. Regardless, the similarities between SSTGLO and SSTATL in Fig. 3 highlights that Atlantic SST trends may be a key factor influencing atmospheric circulation trends, motivating further investigation. As such, the next section determines the underlying atmospheric dynamics of the relatively little-explored Atlantic teleconnections to the SH extratropics.
Springtime (September–November) convective precipitation trends from (a) SSTGLO; (b) the sum of SSTPAC, INDSSTIND, and SSTATL; and (c) the difference between (a) and (b).
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
4. Atlantic teleconnections to the SH extratropics: Atmospheric dynamics
To diagnose the atmospheric dynamics governing Atlantic SST teleconnections to the SH extratropics, we perform daily resolution perturbation experiments using CAM3, hereafter referred to as ATLPERT. For these simulations, an SST anomaly is superimposed on the background climatological forcing across the tropical Atlantic Ocean, and the resulting atmospheric response is tracked through time. This SST perturbation, as illustrated in Fig. 5, represents the spatial pattern of total springtime SST trends linearly tapered between 30° and 40° latitude for both hemispheres in the Atlantic, and chosen as previous studies highlight the importance of tropical latitudes in forcing extratropical responses (Okumura et al. 2012; Schneider et al. 2012a; Li et al. 2014). We perform 100 ATLPERT simulations, wherein each member was initiated from different atmospheric conditions starting in September, and subsequently integrated for 60 days to give an end date in October or November. In each case, the Atlantic SST anomaly was held constant for the duration of simulation, while the global oceans followed a cycle of daily interpolated climatological SST. Control simulations were performed as above, but without the superimposed SST anomaly. The approximate springtime response to the anomalous Atlantic forcing was then determined by subtracting the individual control from the corresponding perturbation experiment, and analyzing the 100-member ensemble mean. In what follows, all figures illustrate the average response to the ATLPERT over days 1–60 unless otherwise stated.
Sea surface temperature perturbation used for the ATLPERT experiments (see text for details). Shading corresponds to the total springtime (September–November) SST trends observed in the tropical Atlantic over 1979–2009, with tapering at latitudes >30° in both hemispheres.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Figure 6 illustrates various tropical anomalies from the ATLPERT experiments. In response to the applied SST forcing, upward vertical motion is initiated (Fig. 6a), generating anomalous convective precipitation across the equatorial Atlantic, but with maxima clearly discernible over the Caribbean Sea and off the coast of Guyana/Suriname/French Guiana (Fig. 6b, shading). By continuity, intensified surface westerlies are observed over ~0°–120°W (Fig. 6b, vectors), driving an analogous pattern of surface convergence (not shown), which may be further enhanced by the introduced Atlantic/Pacific SST gradient. Corresponding upper-level divergence structures are also apparent (Fig. 6a, vectors; Fig. 6c, shading) and, as expected, project strongly onto the interrelated patterns of precipitation anomalies (cf. Figs. 6b,c, shading), given the relationship to latent heat release during deep convection. Consequently, poleward flowing divergent wind anomalies emerge in the upper troposphere (Fig. 6c, vectors). These are primarily constrained to ~30°–120°W, highlighting a regional response emanating from areas of strongest divergence, but representing a significant meridional perturbation to the large-scale circulation of the atmosphere.
Sixty-day average (a) vertical velocity (shading) and U-W wind (vectors) anomalies averaged over 5°N–5°S, (b) convective precipitation (shading) and 950-hPa wind (vectors), and (c) 200-hPa divergence (shading) and divergent wind (vectors) anomalies from the ATLPERT experiments. In (a) red (blue) shading denotes downward (upward) flow, and to accentuate vertical motion, W vector anomalies have been scaled by a factor of 300.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Through mass balance, the Hadley circulation also exhibits changes in association with the ATLPERT experiments. Consistent with the regional divergent wind response (Fig. 6c), these Hadley circulation modifications are also primarily regional, but represent a large enough disturbance to impact zonal-mean meridional streamfunctions (Fig. 7). Climatologically (Fig. 7, contours), the general circulation is dominated by an anticlockwise flowing SH Hadley cell, in accord with the seasonal location of the intertropical convergence zone during austral spring. In response to the Atlantic thermal forcing and the subsequent initiation of rising motion, surface convergence, and upper-level divergence (i.e., the establishment of an anomalous zonal Walker circulation; Fig. 6), a pronounced intensification of the Hadley cell is observed (Fig. 7, shading); while a northward extension of the ascending branch is also apparent, the magnitude of the anomalies is an order less than the climatology so that the total expansion is modest. Changes to the other overturning cells are limited. It can thus be interpreted that the subsequent extratropical response is likely a consequence of perturbations to the Hadley cell, which provides the mechanism by which tropical signals are transmitted to the extratropics. Specifically, the intensified overturning of the SH Hadley cell enhances upper-level convergence and subsidence at the descending branch (Fig. 6c); these features are again regional in character, and are most clearly expressed over the eastern Pacific. This upper-level convergence, and the associated development of anomalous vorticity forcing, will have significant implications for the initiation of extratropical atmospheric Rossby waves.
Sixty-day average zonal-mean (over all longitudes) meridional streamfunction anomalies from the ATLPERT experiments (shading). Contours denote the climatological streamfunctions from the daily control simulations, and are drawn at intervals of 2 × 1010 kg s−1. Solid (dashed) contours denote anticlockwise (clockwise) flow.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Sixty-day average (a) Rossby wave source, (b) vortex stretching, and (c) vorticity advection anomalies at 200 hPa from the ATLPERT experiments. See text for details of terms.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
In response to the substantial upper-level divergence (Fig. 6c), several pronounced RWS features are apparent over the Caribbean Sea despite the relatively weak planetary vorticity. Nonetheless, background climatological conditions will likely inhibit Rossby wave development from these locations owing to the lack of an associated waveguide (Hoskins and Ambrizzi 1993; Lee et al. 2009). Of considerable note to this study, however, is the distinct negative RWS region that emerges at the convergent boundary of the intensified Hadley cell in the east Pacific (~20°–40°S, ~70°–130°W). In this instance, it is these extratropical sources, driven remotely through changes in the local Hadley circulation, that initiate the development of Rossby waves.
Here, Rossby wave evolution is examined by tracking Z500 anomalies averaged over days 1–3 (Fig. 9a), 4–6 (Fig. 9b), and 7–12 (Fig. 9c): note the contrasting color axes. In response to the thermal forcing, negative pressure anomalies initially develop over the tropical Atlantic Ocean (Fig. 9a), forcing an anomalous zonal Walker circulation (Fig. 6a). By days 3–5, however, these tropically sourced signals have been transferred to the extratropics via subsequent changes to the local Hadley circulation (Fig. 7), matching the time scales noted by Tyrrell et al. (1996). The resultant vorticity forcing, and thus RWS (Fig. 8), subsequently initiates a Rossby wave, as clearly expressed as a pattern of positive and negative Z500 anomalies located east of South America and south of Africa, respectively (Fig. 9b). Over time, these Z500 anomalies strengthen in magnitude and begin to propagate eastward with the climatological flow of the subtropical jet (Fig. 9c), allowing extratropical anomalies to be transferred circumglobally. These eastward-propagating features are further identifiable in a time–longitude Hovmöller analysis of Z500 anomalies averaged over 30°–60°S (Fig. 10). In particular, negative Z500 anomalies are simulated over the high-latitude South Pacific (i.e., in the vicinity of the Amunsden–Bellingshausen Seas), with implications for the climate of West Antarctica.
The Z500 anomalies from the ATLPERT experiments averaged over days (a) 1–3, (b) 4–6, and (c) 7–12. Note the contrasting color scales.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
Time–longitude Hovmöller diagram of Z500 anomalies averaged over 30°–60°S from the ATLPERT experiments.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
A dynamical link has therefore been identified between perturbations in tropical Atlantic SST and atmospheric circulation changes in the SH extratropics, as summarized schematically in Fig. 11. Specifically, thermal forcing in the tropical Atlantic drives changes to the zonal Walker circulation, whereby the corresponding anomalous vertical velocities and upper-level divergence subsequently produce an intensification of the local Hadley circulation. In doing so, upper-level convergence is enhanced at the descending branch of the Hadley cell, which consequently becomes a source of Rossby waves that propagate with the climatological mean flow. Consistent with previous studies, we therefore find that the local Hadley circulation, which itself is perturbed through an anomalous zonal Walker circulation, provides the key dynamical connection between the tropics and extratropics, allowing for the development of Rossby waves well removed from the tropical source of disturbance (Sardeshmukh and Hoskins 1988; Dréevillon et al. 2003; Hoskins and Ambrizzi 1993; Rasmusson and Mo 1993; Tyrrell et al. 1996).
Schematic diagram outlining how tropical Atlantic SST variability teleconnects to the Southern Hemisphere extratropics.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
5. The connection between Atlantic SST and climatic change in the SH extratropics
While the physical processes connecting tropical Atlantic SST variability to the extratropical atmospheric circulation have been established (e.g., as summarized in Fig. 11), it remains to be seen how these dynamics relate to the Z500 trend structure simulated by SSTATL (Fig. 3c) and by deduction, how Atlantic SST trends may have influenced the observed pattern of circulation change over the past three decades. Here, we therefore synthesize the dynamical information gained from section 4 and place it in the context of the SSTATL experiments.
To establish the approximate springtime Z500 structures associated with propagating Rossby waves (Fig. 10), Z500 anomalies from the ATLPERT experiments are averaged over days 1–60. Figure 12 compares the resulting time-mean anomalies (shading) with the corresponding Z500 trends simulated by SSTATL (contours; as in Fig. 3c). Despite the contrasting experimental design, Fig. 12 illustrates that the two Z500 patterns possess many similarities. For example, both display negative Z500 anomalies in the vicinity of the South Pacific, along with positive anomalies spanning the Indian and east Pacific Oceans. Several shifts in the location of Z500 structures are also apparent, as evidenced by the more-southerly extension of positive ATLPERT Z500 anomalies over the South Atlantic compared to the trends. These differences may simply be attributed to the contrasting boundary conditions used for the two experiments, that is, constant tropical SST forcing in ATLPERT, in contrast to time-varying SST forcing over all Atlantic latitudes–longitudes in SSTATL. Nevertheless, spatial correlations between the two Z500 structures are 0.50. As such, it can be suggested that the Z500 trend pattern associated with SSTATL may emerge in relation to the time-averaged impact of propagating Rossby waves over the spring season.
Sixty-day average Z500 anomalies from the ATLPERT experiments (shading), and springtime (September–November) Z500 trends from the SSTATL experiments calculated over 1979–2009 (contours). Solid (dashed) contours denote positive (negative) trends, and are drawn at intervals of 7 m; the zero contour has been omitted.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
To determine whether the SSTATL Z500 trends result from enhanced Rossby wave activity, trends are calculated for each of the components responsible for their initiation (i.e., each stage of Fig. 11). Figure 13 illustrates the springtime trends for (Fig. 13a) vertical velocity (shading) and zonal-vertical velocity (U-W) wind (vectors) averaged over 5°N–5°S, (Fig. 13b) convective precipitation (shading) and 950-hPa wind (vectors), (Fig. 13c) 200-hPa divergence (shading) and divergent wind (vectors), and (Fig. 13d) the Rossby wave source at 200 hPa from the SSTATL experiments. In each case it is seen that the pattern of trends projects strongly onto the corresponding anomalies associated with the ATLPERT experiments (cf. Figs. 6, 8, and 13), highlighting that similar atmospheric dynamics likely force both Z500 patterns.
Springtime (September–November) trends in (a) vertical velocity (shading) and U-W wind (vectors) averaged over 5°N–5°S, (b) convective precipitation (shading) and 950-hPa wind (vectors), (c) 200-hPa divergence (shading) and divergent wind (vectors), and (d) Rossby wave source from the SSTATL experiments, all calculated over 1979–2009: (e)–(h) the equivalent trends from the SSTGLO experiments.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
In particular, Atlantic warming causes marked changes to the equatorial zonal circulation, manifested as trends toward enhanced upward vertical velocities (Fig. 13a), increased precipitation (Fig. 13b, shading), surface convergence (not shown), and thus upper-level divergence (Fig. 13c, shading), over Central America and the tropical Atlantic. Accordingly, strengthened poleward-flowing divergent winds emerge over the east Pacific–South American continent (Fig. 13c, vectors), which in turn intensify the local Hadley circulation (not shown), enhance upper-level convergence at the descending branch (Fig. 13c, shading), and magnify RWS activity at this location (Fig. 13d). The right-hand panels of Fig. 13 illustrate the equivalent trends for the SSTGLO experiments, and correspond strongly to those of SSTATL. This similarity suggests that enhanced Atlantic SST (Fig. 1b), and the resulting impact of heightened Rossby wave activity, likely drives the simulated atmospheric circulation changes seen in both SSTATL and SSTGLO, and thus explains their resemblance in Z500 structures (cf. Figs. 3a and 3c). Furthermore, the similarity between SSTGLO and reanalyses (Fig. 2) indicates that observed Z500 trends are likely to be heavily influenced by teleconnections emanating from the tropical Atlantic. By contrast, SSTPAC and SSTIND display conflicting patterns (not shown), demonstrating that they likely play a lesser role in forcing the global response; for example, the spatial correlation between SSTGLO precipitation trends and SSTPAC is 0.26, compared to 0.15 for SSTIND, and 0.70 for SSTATL.
Given the established impacts on the atmospheric circulation, Atlantic-related teleconnections also have broader-scale climatic implications, driven largely by thermal advection associated with the corresponding wind changes (e.g., Ding et al. 2011; Okumura et al. 2012; Schneider et al. 2012a; Li et al. 2014). Figure 14 displays springtime SAT (shading) and surface wind (vectors) trends from the SSTATL experiments, and observed SAT trends from Antarctic research stations (colored dots); note that the spatial structure of SSTATL-related trends is similar at all levels of the troposphere, demonstrating an equivalent barotropic response (not shown). In association with the negative Z500 trends simulated over the high-latitude South Pacific (Fig. 3c), a pattern of cyclonic wind trends are established over this region (Fig. 14, vectors). As a result, anomalous onshore (northerly) winds drive warm air advection, and thus positive temperature trends, over the Antarctic Peninsula and eastern West Antarctic (Fig. 14, shading), coincident with the observed pattern of SAT trends (Fig. 14, colored dots). While the magnitude of modeled temperature trends is approximately half those of observations, it must be remembered that complex feedback mechanisms, unresolved by an AGCM, may accentuate the magnitude of observed temperature trends. Conversely, offshore (southerly) winds enhance cold air advection over the Ross and Amundsen Sea regions, driving negative temperature trends that are largely absent from SAT reconstructions (e.g., Schneider et al. 2012a). However, this cooling trend is consistent with the observed pattern of sea ice expansion evident in the Ross Sea (Fig. 1a, shading) (e.g., Parkinson and Cavalieri 2012; Stammerjohn et al. 2012; Simpkins et al. 2013). Through driving changes in atmospheric circulation, teleconnections emanating from the tropical Atlantic may therefore play a prominent role in forcing climatic change in Antarctica, particularly in relation to the positive springtime temperature trends observed in the Antarctica Peninsula.
Springtime (September–November) trends in Antarctic station surface air temperature (colored dots), and modeled surface wind (vectors) and air temperature (shading) from the SSTATL experiments, all calculated over 1979–2009. Note the contrasting color axes for the observed and modeled temperature trends.
Citation: Journal of Climate 27, 13; 10.1175/JCLI-D-13-00615.1
6. Summary and discussion
Several recent studies have identified a link between trends in tropical SST and large-scale circulation changes over the SH extratropics, prompting further investigation into tropical–extratropical interactions beyond interannual time scales. Here, we use a suite of idealized numerical experiments performed with the CAM3 AGCM to document the springtime relationships between SST trends in the Pacific, Indian, and Atlantic Oceans and the SH extratropical atmospheric circulation. Particular emphasis was given to diagnosing the impact of the Atlantic Ocean, a teleconnection which has received little attention in the literature to date. The key conclusions from this study include the following.
a. SST trends have likely played a role in forcing springtime atmospheric circulation trends in the SH extratropics
Forcing CAM3 with observed SST over 1979–2009 (SSTGLO) captures several notable features of Z500 trends seen in reanalyses (Fig. 2), suggesting that such structures may be driven, at least partially, by global SST trends. Separating the SST influence of the Pacific, Indian, and Atlantic Ocean basins reveals differences in the associated atmospheric teleconnection patterns, particularly in regard to the sign of changes over the high-latitude South Pacific (Fig. 3). Both Pacific (SSTPAC) and Indian Ocean (SSTIND) experiments, for example, invoke a positive Z500 trend in this location, promoting a contrasting pattern of circulation trends to those seen in the reanalyses. It is only when SST trends in the Atlantic Ocean are included (SSTATL) that a negative Z500 trend is simulated over the South Pacific, bearing a marked similarity to the structures associated with SSTGLO. As such, Atlantic SST trends are suggested to have influenced the observed atmospheric circulation trends in the SH extratropics.
b. Atlantic teleconnections to the SH extratropics are driven via changes to the zonal tropical circulation, corresponding perturbations to the local Hadley cell, and the subsequent initiation of atmospheric Rossby waves
Further AGCM experiments (ATLPERT) were used to diagnose the atmospheric dynamics controlling Atlantic–Antarctic teleconnections. As summarized schematically in Fig. 11, increased tropical Atlantic SST establishes marked changes to the zonal Walker circulation, expressed as anomalous upward motion (Fig. 6a), enhanced precipitation (Fig. 6b), and upper-level divergence (Fig. 6c) across the tropical Atlantic, changes which may be further enhanced by the introduction of an SST gradient between the Pacific and Atlantic. By continuity, the upper-level divergent winds induce anomalous poleward flow (Fig. 6c), which intensifies the local Hadley circulation (Fig. 7) and enhances convergence at the descending branch (Fig. 6c). This, in turn, favors the initiation of Rossby waves at the convergent boundary (Fig. 8), which subsequently propagate and strengthen in time (Figs. 9 and 10) with implications for the Antarctic climate. Thus, consistent with previous studies (e.g., Dréevillon et al. 2003; Rasmusson and Mo 1993; Tyrrell et al. 1996), the Hadley circulation (itself modified through an anomalous zonal circulation) provides a direct link between the tropics and extratropics, and consequently, a mechanism by which Rossby waves can be established far from the tropical heat source in the Atlantic.
c. Enhanced Rossby wave activity forced by Atlantic SST trends may have influenced climatic change in the SH extratropics
Springtime Z500 trends simulated by SSTATL represent the time-averaged impact of increased and/or strengthened Rossby waves (Figs. 12 and 13) associated with higher SST in the Atlantic Ocean (Fig. 1b). The similarities among the Z500 structures of SSTATL, SSTGLO, and reanalyses (Figs. 2 and 3), therefore suggests that observed Z500 trends may also be related, at least in part, to Atlantic SST trends. Owing to changes in regional atmospheric circulation, Atlantic teleconnections also greatly impact the Antarctic climate. In particular, the cyclonic wind trends simulated over the high-latitude South Pacific promote warm air advection over the Antarctic Peninsula, driving a pattern of positive temperature trends similar to observations (Fig. 14). Teleconnections associated with Atlantic SST variability may thus represent a significant mechanism driving climatic change in the SH extratropics.
These results add to the growing body of evidence that suggests that tropical trends are key factors forcing climatic change over the SH high latitudes (e.g., Ding et al. 2011; Okumura et al. 2012; Schneider et al. 2012a; Li et al. 2014). In particular, we emphasize the importance of (the little explored) Atlantic teleconnections in driving springtime Z500, and thereby SAT, trends over the South Pacific and West Antarctic, building upon similar conclusions made by Okumura et al. (2012) and Li et al. (2014). In doing so, these results contrast with previous studies that largely attribute springtime trends to the Pacific. Schneider et al. (2012a), for instance, relate geopotential height trends to higher SSTs in the tropical/subtropical Pacific Ocean, and a resulting increase in the Rossby wave train associated with the PSA teleconnection patterns. However, our model simulations produce Z500 trends of opposite sign to observations when forced with Pacific SST variability (SSTPAC, Fig. 3b). While contrasting methodologies make it difficult to discern the cause of such differences, it is clear that further work is needed to clarify the relationships between trends in tropical SST and extratropical Z500; for example, the relative role of the Atlantic and Pacific remains uncertain, as do the impacts of interbasin interactions and the function of atmosphere–ocean coupling. Moreover, while simulations performed using the NCAR Community Atmosphere Model, version 4, reproduce the main findings of this investigation (see Li et al. 2014), future work is needed to validate the robustness of these results across independent AGCMs, particularly given that the climatic impacts of Atlantic SST variability can be highly model dependent (e.g., Hodson et al. 2010).
Regardless, this study demonstrates that Atlantic SST trends need to be considered when diagnosing the mechanisms forcing climatic change in the SH extratropics. As such, understanding multidecadal SST variability [linked namely to the Atlantic multidecadal oscillation (AMO, Deser et al. 2010)], and future projections of Atlantic SST under increased greenhouse gas concentrations, could offer improved scope for prediction of longer-term trends in Antarctica. Nevertheless, it is important to note that other factors will continue to affect the SH extratropical climate in conjunction with the SST-forced changes described here. For example, both coupled atmosphere–ocean–sea ice feedbacks (unresolved in these AGCM experiments) and natural variability may have contributed to the observed trends over 1979–2009, and will likely continue to do so under an evolving climate.
Acknowledgments
The authors thank the three anonymous reviewers for their helpful and valuable comments. GRS was supported by a University of New South Wales University International Postgraduate Award. LMC was supported by the Research Council of Norway through the Earthclim (207711/E10) project. This work was also supported the Australian Research Council (ARC), including the ARC Centre of Excellence for Climate System Science. Computer time was awarded under the Merit Allocation Scheme on the NCI National Facility at the ANU. The use of the NCAR CAM3 model is gratefully acknowledged, along with the institutions responsible for providing the observational and reanalysis datasets.
REFERENCES
Arblaster, J. M., and G. A. Meehl, 2006: Contributions of external forcings to southern annular mode trends. J. Climate, 19, 2896–2905, doi:10.1175/JCLI3774.1.
Bracegirdle, T. J., W. M. Connolley, and J. Turner, 2008: Antarctic climate change over the twenty first century. J. Geophys. Res., 113, 3103, doi:10.1029/2007JD008933.
Bromwich, D. H., and R. L. Fogt, 2004: Strong trends in the skill of the ERA-40 and NCEP–NCAR reanalyses in the high and midlatitudes of the Southern Hemisphere, 1958–2001. J. Climate, 17, 4603–4619, doi:10.1175/3241.1.
Bromwich, D. H., J. P. Nicolas, A. J. Monaghan, M. A. Lazzara, L. M. Keller, G. A. Weidner, and A. B. Wilson, 2013: Central West Antarctica among the most rapidly warming regions on Earth. Nat. Geosci., 6, 139–145, doi:10.1038/ngeo1671.
Ciasto, L. M., and D. W. J. Thompson, 2008: Observations of large-scale ocean atmosphere interaction in the Southern Hemisphere. J. Climate, 21, 1244–1259, doi:10.1175/2007JCLI1809.1.
Ciasto, L. M., and M. H. England, 2011: Observed ENSO teleconnections to Southern Ocean SST anomalies diagnosed from a surface mixed layer heat budget. Geophys. Res. Lett., 38, L09701, doi:10.1029/2011GL046895.
Collins, W. D., and Coauthors, 2006: The Community Climate System Model version 3 (CCSM3). J. Climate, 19, 2122–2143, doi:10.1175/JCLI3761.1.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, doi:10.1002/qj.828.
Deser, C., A. Capotondi, R. Saravanan, and A. S. Phillips, 2006: Tropical Pacific and Atlantic climate variability in CCSM3. J. Climate, 19, 2451–2481, doi:10.1175/JCLI3759.1.
Deser, C., M. A. Alexander, S.-P. Xie, and A. S. Phillips, 2010: Sea surface temperature variability: Patterns and mechanisms. Annu. Rev. Mar. Sci., 2, 115–143, doi:10.1146/annurev-marine-120408-151453.
Ding, H., N. S. Keenlyside, and M. Latif, 2012: Impact of the equatorial Atlantic on the El Niño–Southern Oscillation. Climate Dyn., 38, 1965–1972, doi:10.1007/s00382-011-1097-y.
Ding, Q., E. J. Steig, D. S. Battisti, and M. Küttel, 2011: Winter warming in West Antarctica caused by central tropical Pacific warming. Nat. Geosci., 4, 398–403, doi:10.1038/ngeo1129.
Dréevillon, M., C. Cassou, and L. Terray, 2003: Model study of the North Atlantic region atmospheric response to autumn tropical Atlantic sea-surface-temperature anomalies. Quart. J. Roy. Meteor. Soc., 129, 2591–2611, doi:10.1256/qj.02.17.
Garreaud, R. D., and D. S. Battisti, 1999: Interannual (ENSO) and interdecadal (ENSO-like) variability in the Southern Hemisphere tropospheric circulation. J. Climate, 12, 2113–2123, doi:10.1175/1520-0442(1999)012<2113:IEAIEL>2.0.CO;2.
Haarsma, R. J., and W. Hazeleger, 2007: Extratropical atmospheric response to equatorial Atlantic cold tongue anomalies. J. Climate, 20, 2076–2091, doi:10.1175/JCLI4130.1.
Hodson, D. L. R., R. T. Sutton, C. Cassou, N. Keenlyside, Y. Okumura, and T. Zhou, 2010: Climate impacts of recent multidecadal changes in Atlantic Ocean sea surface temperature: A multimodel comparison. Climate Dyn., 34, 1041–1058, doi:10.1007/s00382-009-0571-2.
Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50, 1661–1671, doi:10.1175/1520-0469(1993)050<1661:RWPOAR>2.0.CO;2.
Hurrell, J. W., J. J. Hack, A. S. Phillips, J. Caron, and J. Yin, 2006: The dynamical simulation of the Community Atmosphere Model version 3 (CAM3). J. Climate, 19, 2162–2183, doi:10.1175/JCLI3762.1.
Jin, D., and B. P. Kirtman, 2009: Why the Southern Hemisphere ENSO responses lead ENSO. J. Geophys. Res.,114, D23101, doi:10.1029/2009JD012657.
Karoly, D. J., 1989: Southern Hemisphere circulation features associated with El Niño–Southern Oscillation events. J. Climate, 2, 1239–1252, doi:10.1175/1520-0442(1989)002<1239:SHCFAW>2.0.CO;2.
Keenlyside, N. S., and M. Latif, 2007: Understanding equatorial Atlantic interannual variability. J. Climate, 20, 131–142, doi:10.1175/JCLI3992.1.
Kushnir, Y., R. Seager, M. Ting, N. Naik, and J. Nakamura, 2010: Mechanisms of tropical Atlantic SST influence on North American precipitation variability. J. Climate, 23, 5610–5628, doi:10.1175/2010JCLI3172.1.
Lee, S.-K., C. Wang, and B. E. Mapes, 2009: A simple atmospheric model of the local and teleconnection responses to tropical heating anomalies. J. Climate, 22, 272–284, doi:10.1175/2008JCLI2303.1.
Li, X., D. M. Holland, E. P. Gerber, and C. Yoo, 2014: Impacts of the north and tropical Atlantic Ocean on the Antarctic Peninsula and sea ice. Nature, 505, 538–542, doi:10.1038/nature12945.
Luffman, J. J., A. S. Taschetto, and M. H. England, 2010: Global and regional climate response to late twentieth-century warming over the Indian Ocean. J. Climate, 23, 1660–1674, doi:10.1175/2009JCLI3086.1.
Luo, J.-J., W. Sasaki, and Y. Masumoto, 2012: Indian Ocean warming modulates Pacific climate change. Proc. Natl. Acad. Sci. USA, 109, 18 701–18 706 , doi:10.1073/pnas.1210239109.
Marshall, G. J., 2007: Half-century seasonal relationships between the Southern Annular Mode and Antarctic temperatures. Int. J. Climatol., 27, 373–383, doi:10.1002/joc.1407.
Marshall, G. J., A. Orr, N. P. M. van Lipzig, and J. C. King, 2006: The impact of a changing Southern Hemisphere Annular Mode on Antarctic Peninsula summer temperatures. J. Climate, 19, 5388–5404, doi:10.1175/JCLI3844.1.
Mo, K. C., and R. W. Higgins, 1998: The Pacific–South American modes and tropical convection during the Southern Hemisphere winter. Mon. Wea. Rev., 126, 1581–1596, doi:10.1175/1520-0493(1998)126<1581:TPSAMA>2.0.CO;2.
Monaghan, A. J., D. H. Bromwich, W. Chapman, and J. C. Comiso, 2008: Recent variability and trends of Antarctic near-surface temperature. J. Geophys. Res.,113, D04105, doi:10.1029/2007JD009094.
Okumura, Y. M., D. Schneider, C. Deser, and R. Wilson, 2012: Decadal–interdecadal climate variability over Antarctica and linkages to the tropics: Analysis of ice core, instrumental, and tropical proxy data. J. Climate, 25, 7421–7441, doi:10.1175/JCLI-D-12-00050.1.
Parkinson, C. L., and D. J. Cavalieri, 2012: Antarctic sea ice variability and trends, 1979-2010. The Cryosphere, 6, 871–880, doi:10.5194/tc-6-871-2012.
Raphael, M. N., and M. M. Holland, 2006: Twentieth century simulation of the Southern Hemisphere climate in coupled models. Part 1: Large scale circulation variability. Climate Dyn., 26, 217–228, doi:10.1007/s00382-005-0082-8.
Rasmusson, E. M., and K. Mo, 1993: Linkages between 200-mb tropical and extratropical circulation anomalies during the 1986–89 ENSO cycle. J. Climate, 6, 595–616, doi:10.1175/1520-0442(1993)006<0595:LBMTAE>2.0.CO;2.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, doi:10.1029/2002JD002670.
Rignot, E., I. Velicogna, M. R. van den Broeke, A. Monaghan, and J. Lenaerts, 2011: Acceleration of the contribution of the Greenland and Antarctic ice sheets to sea level rise. Geophys. Res. Lett.,38, L05503, doi:10.1029/2011GL046583.
Santoso, A., M. H. England, and W. Cai, 2012: Impact of Indo-Pacific feedback interactions on ENSO dynamics diagnosed using ensemble climate simulations. J. Climate, 25, 7743–7763, doi:10.1175/JCLI-D-11-00287.1.
Sardeshmukh, P. D., and B. J. Hoskins, 1988: The generation of global rotational flow by steady idealized tropical divergence. J. Atmos. Sci., 45, 1228–1251, doi:10.1175/1520-0469(1988)045<1228:TGOGRF>2.0.CO;2.
Schneider, D. P., C. Deser, and Y. Okumura, 2012a: An assessment and interpretation of the observed warming of West Antarctica in the austral spring. Climate Dyn., 38, 323–347, doi:10.1007/s00382-010-0985-x.
Schneider, D. P., Y. Okumura, and C. Deser, 2012b: Observed Antarctic interannual climate variability and tropical linkages. J. Climate, 25, 4048–4066, doi:10.1175/JCLI-D-11-00273.1.
Simpkins, G. R., and A. Y. Karpechko, 2012: Sensitivity of the southern annular mode to greenhouse gas emission scenarios. Climate Dyn., 38, 563–572, doi:10.1007/s00382-011-1121-2.
Simpkins, G. R., L. M. Ciasto, D. W. J. Thompson, and M. H. England, 2012: Seasonal relationships between large-scale climate variability and Antarctic sea ice concentration. J. Climate, 25, 5451–5469, doi:10.1175/JCLI-D-11-00367.1.
Simpkins, G. R., L. M. Ciasto, and M. H. England, 2013: Observed variations in multidecadal Antarctic sea ice trends during 1979–2012. Geophys. Res. Lett., 40, 3643–3648, doi:10.1002/grl.50715.
Stammerjohn, S., R. Massom, D. Rind, and D. Martinson, 2012: Regions of rapid sea ice change: An inter-hemispheric seasonal comparison. Geophys. Res. Lett.,39, L06501, doi:10.1029/2012GL050874.
Steig, E. J., D. P. Schneider, S. D. Rutherford, M. E. Mann, J. C. Comiso, and D. T. Shindell, 2009: Warming of the Antarctic ice-sheet surface since the 1957 International Geophysical Year. Nature, 457, 459–462, doi:10.1038/nature07669.
Steig, E. J., and Coauthors, 2013: Recent climate and ice-sheet changes in West Antarctica compared with the past 2,000 years. Nat. Geosci., 6, 372–375, doi:10.1038/ngeo1778.
Taschetto, A. S., and T. Ambrizzi, 2012: Can Indian Ocean SST anomalies influence South American rainfall? Climate Dyn., 38, 1615–1628, doi:10.1007/s00382-011-1165-3.
Taschetto, A. S., A. Sen Gupta, H. H. Hendon, C. C. Ummenhofer, and M. H. England, 2011: The contribution of Indian Ocean sea surface temperature anomalies on Australian summer rainfall during El Niño events. J. Climate, 24, 3734–3747, doi:10.1175/2011JCLI3885.1.
Thompson, D. W. J., and S. Solomon, 2002: Interpretation of recent southern hemisphere climate change. Science, 296, 895–899, doi:10.1126/science.1069270.
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, doi:10.1038/ngeo1296.
Timmermann, A., and Coauthors, 2010: Towards a quantitative understanding of millennial-scale Antarctic warming events. Quat. Sci. Rev., 29, 74–85, doi:10.1016/j.quascirev.2009.06.021.
Turner, J., and Coauthors, 2004: The SCAR READER Project: Toward a high-quality database of mean Antarctic meteorological observations. J. Climate, 17, 2890–2898, doi:10.1175/1520-0442(2004)017<2890:TSRPTA>2.0.CO;2.
Turner, J., and Coauthors, 2005: Antarctic climate change during the last 50 years. Int. J. Climatol., 25, 279–294, doi:10.1002/joc.1130.
Tyrrell, G. C., D. J. Karoly, and J. L. McBride, 1996: Links between tropical convection and variations of the extratropical circulation during TOGA COARE. J. Atmos. Sci., 53, 2735–2748, doi:10.1175/1520-0469(1996)053<2735:LBTCAV>2.0.CO;2.
Wang, C., 2006: An overlooked feature of tropical climate: Inter-Pacific-Atlantic variability. Geophys. Res. Lett.,33, L12702, doi:10.1029/2006GL026324.