1. Introduction
El Niño–Southern Oscillation (ENSO) is the dominant source of interannual climate variability on Earth (e.g., McPhaden et al. 2006). Its variability is associated with different anomalous sea surface temperature (SST) and convection patterns and amplitudes. Two opposite phases of ENSO have typically been referred to in the literature (e.g., Hoerling et al. 1997). During El Niño, anomalously warm SSTs and increased convection occur in the tropical eastern Pacific Ocean with suppressed convection over the Maritime Continent, whereas during La Niña, anomalously cool SSTs occur in the tropical eastern Pacific and anomalously warm SSTs and increased convection are shifted westward, relative to El Niño, toward the western Pacific. El Niño and La Niña are often monitored using the positive (warm) and negative (cool) values of the Niño-3.4 index, respectively, which represents the average equatorial SST anomalies across the Pacific (5°S–5°N, 170°–120°W; e.g., Trenberth 1997). However, ENSO events differ in amplitude, evolution, and spatial pattern (e.g., Ashok et al. 2007). Therefore, over the past decades, the traditional view of opposite El Niño and La Niña phases and impacts has evolved to recognize ENSO diversity (or flavors; e.g., Capotondi et al. 2015). Eastern Pacific (EP or canonical) El Niño events (e.g., Rasmusson and Carpenter 1982) peak in the east side of the Pacific basin, while Central Pacific (CP or Modoki) El Niño (e.g., Ashok et al. 2007) events occur farther westward. These ENSO types can be monitored by several indices, such as by averaging SST anomalies over the Niño-3 (5°S–5°N, 150°–90°W) and Niño-4 (5°S–5°N, 160°E–150°W) regions, respectively, but the convective centers do occur on a continuum.
ENSO exerts a profound influence on weather and climate in remote regions of the globe through atmospheric teleconnections (e.g., Trenberth et al. 1998; Liu and Alexander 2007; Stan et al. 2017; Yeh et al. 2018; Li et al. 2021, and references therein). These tropical to extratropical interactions mainly arise from stationary Rossby wave trains forced by tropical heating: changes in tropical convection due to ENSO induce upper-level divergence in the equatorial Pacific that advects mean vorticity out of the tropics, thus producing a Rossby wave source (RWS) in the subtropical westerlies, which subsequently initiates a poleward-propagating wave train (e.g., Hoskins and Karoly 1981; Sardeshmukh and Hoskins 1985). In the Southern Hemisphere (SH), the wave train associated with ENSO is known as the Pacific–South American (PSA) pattern (e.g., Karoly 1989; Mo and Ghil 1987; Kidson 1988; Mo and Paegle 2001). The amplitude of the PSA is larger during austral winter than summer (e.g., Karoly 1989) due to the presence of the strong subtropical jet (STJ), which promotes the development of a large RWS that can support stationary Rossby wave propagation and acts as a waveguide (e.g., Ambrizzi et al. 1995). Therefore, although ENSO is often in its developing phase during austral winter and does not mature until the following summer (e.g., Trenberth et al. 1998), a distinct PSA pattern can still be observed during winter (e.g., Karoly 1989; Mo and Higgins 1998). In addition to the zonally asymmetric PSA mechanism, the influence of ENSO is also communicated to the SH high latitudes through a zonally symmetric mechanism via modifications to the Hadley cell and STJ and their subsequent effects on the midlatitude transient eddies and eddy–mean flow zonal interactions (e.g., Seager et al. 2003; Lau et al. 2005; L’Heureux and Thompson 2006; Fogt et al. 2011; Schneider et al. 2012). However, this mechanism is thought to mainly operate during the austral summer season when the STJ and eddy-driven jets are merged.
It can be difficult to establish statistically significant asymmetries in the extratropical atmospheric circulation response to ENSO in the observational record due to large internal atmospheric variability (e.g., Deser et al. 2017; Garfinkel et al. 2019). However, studies have reported different teleconnections between ENSO flavors. For example, the upper-tropospheric response to La Niña is shifted westward compared to the response to El Niño because of the westward shift of the tropical convective anomalies (e.g., Hoerling et al. 1997; Cai et al. 2012). A similar difference also occurs between CP and EP El Niño events (e.g., Wilson et al. 2014; Ciasto et al. 2015). These different atmospheric responses for westward- and eastward-located ENSO heating also have different climate impacts, such as on precipitation, temperature, and sea ice (e.g., Barsugli and Sardeshmukh 2002; Power et al. 2006; Taschetto and England 2009; Cai et al. 2012; Ciasto et al. 2015; Freund et al. 2021).
While several studies have investigated the wintertime teleconnection from the tropical Pacific SST anomalies during ENSO to the SH mid- to high latitudes (e.g., Cai et al. 2011; Wilson et al. 2014; Ciasto et al. 2015; Li et al. 2015b; Yiu and Maycock 2019; Wang et al. 2022b), the cause of apparent differences in the extratropical atmospheric circulation response to the location and sign of the tropical heating anomalies have not been thoroughly investigated. A recent study by Wang et al. (2022b) suggested that differences in the extratropical responses to El Niño and La Niña can be linked to induced changes in the STJ, which then influences the position of the peak height anomalies via barotropic energy conversion (e.g., Simmons et al. 1983; Branstator 1985). However, this study used a linear dry model and therefore did not account for the role of transient eddy feedbacks, which are known, for instance, to play a first-order role for ENSO teleconnections into the NH (as reviewed below).
The possible role for transient eddy feedbacks to explain the difference between westward- and eastward-located ENSO heating is motivated by examining the Northern Hemisphere (NH) equivalent of the PSA, the Pacific–North American (PNA) pattern, which has been studied more extensively (e.g., Held et al. 1989; Hoerling et al. 1997; Li et al. 2006; Lin et al. 2007; Dai et al. 2017; Feng et al. 2017). An important role for a transient eddy feedback that acts to modify the direct response to tropical forcing has been identified (e.g., Held et al. 1989; Li et al. 2006; Lin et al. 2007): the tropically forced quasi-stationary wave interacts with the extratropical storm track and rearranges the synoptic eddies, thus inducing a positive feedback that further modifies the stationary wave response. Li et al. (2006) found that this storm-track mechanism could account for the different circulation responses to westward- and eastward-located tropical Pacific SST anomalies during boreal winter. This asymmetry between western and eastern Pacific heating is also supported by a waveguide effect provided by the midlatitude NH STJ (e.g., Hoskins and Ambrizzi 1993; Branstator 2002).
We build on earlier studies of the SH extratropical atmospheric circulation response to ENSO heating by investigating the daily evolution of the forced response to idealized prescribed diabatic heating in large-ensemble integrations with the Community Atmosphere Model version 5 (CAM5). The cause of a different response between westward- and eastward-located ENSO peak heating is explored by imposing diabatic heating in the equatorial western (170°E) and central/eastern (210°E) Pacific Ocean. We focus on the role of the STJ for influencing both the RWS and the propagation characteristics of the wave responses. We also test the role of the transient eddy feedback by contrasting the CAM5 results with a simplified linear general circulation model (SGCM). We will demonstrate that the extratropical response to western Pacific heating has larger zonal Rossby wave propagation due to the proximity to the STJ core, and the transient eddy feedback plays a primary role in establishing the time-mean response. In contrast, eastern Pacific heating produces larger meridional propagation and the transient eddy feedback has a secondary effect.
The observationally based datasets, models, and experiments are described in section 2. Section 3a explores observed ENSO teleconnections, section 3b presents the direct response to heating in CAM5, section 3c examines the role of the transient eddy feedback, and section 3d presents the SGCM results. A summary and conclusions are presented in section 4.
2. Data and numerical experiments
a. Observational and reanalysis datasets and analysis
Observed austral winter (June–August) teleconnections from ENSO into the SH are briefly reviewed using analyses during the satellite era from 1979 to 2019. We focus on the austral winter months, despite ENSO events not peaking until the end of the calendar year (e.g., Santoso et al. 2017), as one of our interests is on the role of the STJ for determining the RWS and tropical to extratropical wave propagation behavior. ENSO SST indices are computed using gridded (1°) monthly mean SST anomalies (deviations from the monthly climatology) from the merged Hadley–NOAA/Optimal Interpolation SST analyses (hereafter Hadley-OI; Hurrell et al. 2008). Following Deser et al. (2017), El Niño and La Niña winters are identified when the linearly detrended winter-mean Niño-3.4 index exceeds ±1 standard deviation, respectively. This leads to five El Niño (1982, 1987, 1997, 2002 and 2015) and four La Niña (1988, 1998, 1999 and 2000) samples. These years correspond to strong events based on the oceanic Niño index from the U.S. National Oceanic and Atmospheric Administration Climate Prediction Center (https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php). We also compare the differences between El Niño and La Niña to CP and EP El Niño. We use indices for CP and EP El Niño following Sullivan et al. (2016): CP El Niño = Niño-4normalized − 0.5 × Niño-3normalized and EP El Niño = Niño-3normalized − 0.5 × Niño-4normalized.
We construct the observed teleconnections using detrended winter-mean geopotential height and winds analyses from the Japanese 55-year Reanalysis (JRA-55; 1.25° grid), which is produced by the Japan Meteorological Agency (Kobayashi et al. 2015). We also use the gridded (2.5°) winter-mean precipitation analyses from the Global Precipitation Climatology Project (GPCP; Adler et al. 2003). The teleconnections are analyzed using composites for El Niño and La Niña and regressions for EP and CP El Niño. Statistical significance of the composited height anomalies for El Niño and La Niña is assessed using a two-tailed t test with five and four samples, respectively. Significance of the regressed anomalies is assessed assuming 40 degrees of freedom (i.e., each year is independent). We evaluated the sensitivity of the observed teleconnection signals in several ways, and they produced similar results, demonstrating that the teleconnections are robust. These sensitivity tests included reducing the threshold for selecting El Niño and La Niña events to half a standard deviation to increase the sample size, comparing the teleconnections in two other reanalysis products (ERA-Interim and the twentieth-century reanalysis) over the satellite era, using a different SST dataset (HadISST), and examining the longer JRA-55 record (1958–2019). Our main analysis approach to understanding the asymmetry in the response to western- and eastern-located heating is the model experiments, which are described in the next section.
b. Community Atmosphere Model
The response to prescribed tropical diabatic heating anomalies is examined using paired control and diabatic heating anomaly experiments with CAM5 (Neale et al. 2012). The design of the control and diabatic heating anomaly experiments is the same as described in Gillett et al. (2022). We use the version of CAM5 with horizontal resolution of 0.9° longitude × 1.25° latitude and 30 hybrid sigma–pressure levels. In both the control and diabatic heating anomaly experiments, prescribed SSTs and sea ice concentrations are from Hadley-OI and are set to repeating monthly SST and sea ice climatologies averaged over 1982–2001 to represent present-day climate (FC5 compset). All other boundary conditions (greenhouse gases, ozone, aerosols, etc.) are prescribed to be year 2000 conditions.
The control ensemble is generated by integrating CAM5 30 times for one year starting from different realizations of 1 January 2000 initial conditions. We then restart the model from each of the thirty 1 June, 1 July, and 1 August start dates (i.e., 90 in total) and apply a steady, local diabatic heating anomaly at the beginning of each control simulation and integrate for 30 days. The diabatic heating anomaly is imposed in the temperature tendency equation at each 30-min time step in the radiative heating subroutine in CAM5. We directly prescribe a diabatic heating anomaly instead of, for instance, an SST anomaly because the latter has been shown to not always produce a consistent location and magnitude of the atmospheric heating anomaly (e.g., Trenberth et al. 2015). However, limited experimentation reveals that the response to directly imposed heating and the response to imposed SST develop nearly identically (not shown). We form the daily evolution of the response by concatenating the daily difference of the 90 start dates with the applied forcing and the control without the applied forcing.
The prescribed diabatic heating anomalies have the shape of an elliptical squared cosine in the horizontal (e.g., Barsugli and Sardeshmukh 2002) with half-widths of 20° in longitude and 10° in latitude. The response is not very sensitive to the size of the patch (not shown). The heating anomalies follow a half sine wave in the vertical (e.g., Meehl et al. 2008) with maximum heating of 5 K day−1 occurring near 500 hPa and which goes to zero at the surface and 100 hPa. At the center of the patch, this heating rate produces the same vertically integrated latent heating as a ∼12 mm day−1 precipitation anomaly. This imposed heating amplitude is quite large, particularly because we examine the developing phase of ENSO. Therefore, the CAM5 response is likely larger than in the observed system; however, we compared various heating amplitudes (1.25, 2.5, 5, and 10 K day−1) and the response scales approximately linearly for positive imposed heating. The large amplitude and extent of the applied heating anomaly nonetheless help to ensure that the response is statistically significant.
The horizontal structure of the patch at 500 hPa and the patch area-averaged vertical profile (red curve) are displayed in Figs. 1a and 1b, respectively. Similar to Gillett et al. (2022), the model also generates its own diabatic heating anomaly in response to the imposed heating anomaly (see black curves in Fig. 1b). The magnitude of the induced heating is similar to that in Gillett et al. (2022) for an Indian Ocean heating anomaly and it is also primarily a result of induced moist convection (not shown): the imposed heating induces upward motion, which acts to increase precipitation and cloudiness. For the prescription of positive heating, the induced heating is typically twice as large (Fig. 1b). While our focus is on prescribed positive heating anomalies, for the prescription of negative heating the induced heating is small (discussed further below), because the induced negative rainfall anomaly is limited by the magnitude of the mean rainfall at a particular location.
Anomalous heating in the CAM5 experiments. (a) The shading and contours show the 500-hPa horizontal structure of the heating imposed at each time step (K day−1) for the patch centered at 170°E. The red cross represents the center of the 210°E patch. (b) Vertical structure of the heating imposed at each time step averaged over the patch (same for 170° and 210°E patches; red curve). The 30-day and patch-averaged total induced diabatic heating anomaly (solid black curve for 170°E patch and dashed black curve for 210°E patch). The total induced heating rate in CAM5 is the temperature tendency due to moist processes (DTCOND), vertical diffusion (DTV), longwave heating (QRL), solar heating (QRS), and orographic wave drag (TTGWORO).
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
We conduct two idealized experiments with a positive equatorial diabatic heating anomaly imposed at 170°E (shading in Fig. 1a) in the first experiment and at 210°E (red cross in Fig. 1a) in the second experiment. These longitudes are designed to mimic the location of the maximum convective anomalies that occur during westward-shifted ENSO events [i.e., like La Niña (Fig. 2a) or CP El Niño (Fig. 3a)] and eastward-shifted ENSO events [i.e., like canonical or EP El Niño (Figs. 2c and 3c)], respectively. Our goal is to focus on the sensitivity of the response to the location of the diabatic heating anomaly, rather than the sign. We emphasize that these are idealized experiments. For example, dipole heating might be more representative of the observed ENSO flavors (Figs. 2a,c and 3a,c) and La Niña might be better represented by an imposed negative heating (reduced convection) anomaly. However, as mentioned above, it is difficult to induce a strong response to an imposed negative heating anomaly because precipitation is positive definite (i.e., the rainfall anomaly is limited by the background rainfall rate). Furthermore, the horizontal and vertical structure of the total (imposed + induced) heating anomaly can be very different between positive and negative imposed heating or SST (not shown). The total heating anomaly can also be different for anomalies applied at different longitudes (note the small differences in the induced heating profiles for the 170° and 210°E experiments in Fig. 1b) due to differences in the background SST, which influences the ability of the anomaly to exceed the threshold for the occurrence of deep tropical convection (e.g., Lachlan-Cope and Connolley 2006; Ciasto et al. 2015; Taschetto et al. 2016). Therefore, we follow the approach of earlier studies (e.g., Li et al. 2006) by imposing a single positive heating patch in each experiment, which turns out to be a good compromise, as shown in section 3b by the good coherence between the observed and simulated teleconnection patterns and similarity between the anomalous induced heating profiles for the eastward- and westward-located heating anomalies (Fig. 1b).
Observed (1979–2019) austral winter (left) La Niña and (right) El Niño composite (top) precipitation (shading; mm day−1) and 250-hPa divergent wind (vectors; scale in bottom right; vectors with magnitude less than 0.75 m s−1 are not displayed) and (bottom) 250-hPa geopotential height (Z; shading; m) and horizontal Takaya and Nakamura (2001) wave activity flux (vectors; scale in bottom right; vectors with magnitude less than 0.5 m2 s−2 are not displayed) anomalies. La Niña and El Niño winters are defined when the Niño-3.4 index is less than −1 and greater than +1 standard deviation, respectively. The light green contours enclose geopotential height anomalies where p < 0.10 is calculated using a two-sided t test.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
Observed (1979–2019) austral winter regressions of (left) CP El Niño and (right) EP El Niño against (top) precipitation (shading; mm day−1) and 250-hPa divergent wind (vectors; scale in bottom right; vectors with magnitude less than 0.2 m s−1 are not displayed), and (bottom) 250-hPa geopotential height (Z; shading; m) and horizontal Takaya and Nakamura (2001) wave activity flux (vectors; scale in bottom right; vectors with magnitude less than 0.02 m2 s−2 are not displayed) anomalies. The light green contours enclose geopotential height anomalies where p < 0.10 is calculated using a two-sided t test.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
c. Linear simple general circulation model
To investigate the role of transient eddy feedbacks for enhancing and organizing the PSA pattern, we contrast the comprehensive CAM5 simulations to similar experiments conducted using the linear SGCM (Hall 2000). The SGCM is a dry primitive equation atmospheric model that is linearized about an “observed” three-dimensional austral winter basic state. Moist and other nonlinear processes (transient flux divergence of eddy vorticity and steady nonlinear flux divergence; e.g., Hendon 1986; Ting and Yu 1998) are therefore not included in these equations. The basic state is obtained from the time mean of a long integration of the nonlinear version of the model with a prescribed constant forcing calculated from observed daily data that acts to maintain a realistic climatology. The SGCM has horizontal resolution of ∼3.75° and 10 sigma–vertical levels. Experiments are conducted where a thermal forcing anomaly is imposed centered on the same location as the CAM5 experiments but has an elliptical squared cosine shape in the horizontal with half-widths of 40° in longitude and 11° in latitude (navy-blue contours in Fig. 9), and half sine wave in the vertical with peak heating of 4.5 K day−1 at the 0.35 sigma level, similar to the patches used in Lin and Brunet (2018; see their Fig. 2). The zonal half-width in the SGCM is double the width used in the CAM5 simulations; however, our conclusions are the same when the CAM5 half-width is increased (not shown). The SGCM results also do not appear to be sensitive to a reasonable change in the heating extent (not shown). The SGCM with the heating anomaly is integrated for 20 days and once daily output is analyzed.
3. Results
a. Observed ENSO teleconnections
We first review the observed winter-mean teleconnection patterns associated with ENSO. Figure 2 displays composite precipitation, and 250-hPa divergent wind, height and wave activity flux anomalies during La Niña and El Niño. The tropical convection anomalies observed during La Niña and El Niño are not simply the opposite sign that would be expected for a linear relationship: the maximum precipitation anomalies in the equatorial Pacific are weaker and occur farther westward for La Niña (Fig. 2a) than El Niño (Fig. 2c). This asymmetry has been linked to differences in the climatological SST over the equatorial Pacific and the threshold (total SST > 27.5°C) for the occurrence of deep convection: an SST anomaly with the same magnitude in the western and eastern Pacific will produce larger rainfall anomalies in the west due to the warmer climatological SST (e.g., Barsugli and Sardeshmukh 2002; Lachlan-Cope and Connolley 2006; Ciasto et al. 2015; Taschetto et al. 2016).
An asymmetry between El Niño and La Niña also extends into the SH extratropics. Enhanced convection near the central equatorial Pacific during El Niño is associated with a classical wave train (e.g., Hoskins and Karoly 1981) that appears to propagate poleward and eastward along a great circle path into the South Pacific (Fig. 2d). A second, weaker wave train also propagates southeastward from suppressed convection in the eastern Indian Ocean (e.g., Cai et al. 2011; Gillett et al. 2022). In contrast, the extratropical response during La Niña heating is more zonally elongated and aligned in the meridional direction in the South Pacific and at lower zonal wavenumber (Fig. 2b). The La Niña response is clearly not a simple inverse of the El Niño response; for example, both phases share an anticyclonic height anomaly to the south of Australia. A classical Rossby wave train also does not appear to emerge from the tropical Pacific in the La Niña case. It is therefore unclear how the height structure depicted in Fig. 2b is established if it is simply a Rossby wave response to tropical forcing.
We compare the teleconnection patterns during La Niña and El Niño to CP and EP El Niño. Figure 3 shows regressions onto the CP and EP El Niño indices. There is good resemblance of the extratropical height response between the two indices that have westward-shifted tropical heating anomalies i.e., La Niña (Fig. 2b) and CP El Niño (Fig. 3b), apart from the sign change due to opposite-signed precipitation anomalies (Figs. 2a and 3a), and between El Niño (Fig. 2d) and EP El Niño (Fig. 3d). However, the extratropical asymmetry between western- and eastern-located heating is more apparent for the opposite phases of ENSO (Fig. 2). For example, CP El Niño is possibly associated with more wave train–like wave activity flux (Fig. 3b) than EP El Niño (Fig. 3d). The difference is also less apparent when CP and EP El Niño are simply defined using the Niño-4 and Niño-3 indices, respectively (see e.g., Fu et al. 2013; Ciasto et al. 2015). Nonetheless, these apparent different in the extratropical responses associated with westward- and eastward-located equatorial Pacific heating motivate us to explore the development of the responses and possible causes of these differences with our idealized model experiments.
b. CAM5 response
To understand the development of the observed time-mean response to diabatic heating anomalies that are located in the western Pacific and in the central/eastern Pacific, we explore its development by examining the daily evolution of the 250-hPa height anomalies in response to suddenly switching on equatorial diabatic heating at 170°E (left column in Fig. 4) and 210°E (right column in Fig. 4) using the 90-member paired ensemble from CAM5. For all model analysis, we present the ensemble-mean difference between the paired diabatic heating anomaly and control experiments so that the anomalies reflect a forced response to the imposed tropical forcing (i.e., by averaging over 90 different start dates the effects of internal variability are largely removed). We focus on the daily evolution during the first two weeks because the atmospheric adjustment to a tropical heating anomaly develops on the time scale of 1–2 weeks (Jin and Hoskins 1995): in the first week the tropical and midlatitude response are established and in the second week the high-latitude response is established and propagation back toward the equator occurs.
Geopotential height anomalies (m) at 250 hPa on (a) day 3, (b) day 5, (c) day 8, (d) day 12, and (e) for the day 17–30 mean from the CAM5 170°E experiment minus control. (f)–(j) As in (a)–(e), but for the 210°E experiment. The navy-blue contours in (a) and (f) show the horizontal structure of the prescribed heating.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
Initially (day 3), the upper-tropospheric height response to diabatic heating imposed at 170°E (Fig. 4a) is confined to the tropics and subtropics with the anomalies taking the form of a typical Matsuno-Gill (1966; Gill 1980) response: in the upper-troposphere anticyclonic anomalies appear to the west of the heating in both hemispheres, which are stronger in the winter SH, and an equatorial Kelvin wave propagates eastward away from the heating, circling the equator in around 10 days, consistent with a dry first baroclinic wave (e.g., Wheeler and Nguyen 2015). Opposite-signed anomalies develop in the equatorial lower troposphere (see Figs. S1a–e in the online supplemental material), indicating that the forced response in the tropics has a deep baroclinic structure. By day 5 (Fig. 4b), the anticyclonic anomaly south of the imposed heating near 25°S has grown in amplitude and an alternating pattern of low to high anomalies have developed toward its east, with additional low to high anomalies evident by day 8 (Fig. 4c). These anomalies also appear to propagate back toward the equator near South America (Fig. 4c). This initial wave train, primarily strung out in the zonal direction, is a shallow response that is not evident in the lower troposphere (Figs. S1a–e), confirming that it travels primarily eastward along the waveguide of the westerly STJ (see Fig. 5a, which shows the observed 250-hPa zonal wind climatology during austral winter), which is confined to the mid- to upper troposphere (e.g., Gillett et al. 2021). By day 12 (Fig. 4d), anomalies also appear farther south at higher latitudes, which also have an equivalent-barotropic structure (Figs. S1a–e), while the more zonally oriented wave train along the STJ waveguide appears to have encircled the globe. The time-mean response (average of days 17–30) exhibits a distinctly different spatial structure as compared to the clear wave train patterns evident during days 3–12, having become more zonally elongated and meridionally stacked with larger zonal scale. This time-mean response now closely resembles the observed La Niña composite (cf. Fig. 4e with Fig. 2b but with sign reversed due to the negative precipitation anomalies associated with La Niña) and the CP El Niño regression (Fig. 3b) depicted using JRA-55. We tested that the height response is not very sensitive to the size of the imposed patch by elongating the ellipse half-width to 45° and 90° longitude and a similar response was obtained (not shown).
Observed climatological (a) zonal wind (m s−1), (b) meridional gradient of mean absolute vorticity (
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
The initial tropical response to diabatic heating imposed at 210°E is similar to that at 170°E but is translated 40° eastward and is slightly weaker (Fig. 4 right column), presumably because the heating is imposed in a region of cooler climatological SSTs so that the additional induced diabatic heating is weaker. We confirmed that the total induced diabatic heating anomaly for the 210°E experiment averaged over the patch and the 30-day period is slightly weaker than that of the 170°E experiment (a 14% reduction; Fig. 1b). As for the heating at 170°E, an initial Kelvin wave along the equator is also seen but appears to propagate at a slower eastward speed, taking around 17 days to circle the globe. Also, in contrast to the forcing at 170°E, the fast and shallow response along the STJ is absent, instead replaced by larger meridional propagation and a more poleward arcing wave train (Fig. 4g–i). The wave train evolves little from day 12 onward, with the time-mean response (Fig. 4j) having a similar structure as at day 12 (Fig. 4i). This wave train is reminiscent of a classical Rossby wave train propagating poleward and eastward from a tropical source (e.g., Hoskins and Karoly 1981). This response to the eastern Pacific heating also resembles the time-mean wintertime response to a steady canonical El Niño SST anomaly simulated by Yiu and Maycock (2019; see their Fig. 3d), which induces a maximum OLR anomaly just east of the date line (see their Fig. 4b). The contrasting responses to western- and eastern-located imposed diabatic heating anomalies seen here may reflect differences in the strength and location of the RWS and subsequent wave refraction because of the proximity of the STJ. Differences in the development of transient eddy feedbacks may also develop because of the differences in the proximity of the heating locations to the extratropical storm track. We address each of these in turn.
We first address the RWS generated by the two different locations of the imposed heating. Here, we approximate the RWS by its leading two linearized components: S1 and S2 (e.g., Sardeshmukh and Hoskins 1985; Qin and Robinson 1993). We previously showed that the linearized RWS is a good approximation for the total RWS (Gillett et al. 2022).
Figure 6a shows the 250-hPa S1 RWS (pink contours) and 250-hPa S2 RWS (green contours), and the outgoing longwave radiation (OLR; shading) and 250-hPa divergent wind (vectors) anomalies on day 3 of the CAM5 experiment for the imposed heating located at 170°E. Enhanced tropical convection (reduced OLR) occurs near the location of the prescribed forcing, which promotes upward motion (not shown) and divergent outflow at upper levels toward the SH subtropics, thus generating a large positive S1 (advection) RWS (anticyclonic tendency) in the subtropics directly south of the heating anomaly. A dipole pattern in S2 (convergence) RWS occurs in the subtropics, with negative but weaker S2 anomaly to the south and west of the positive S1 anomaly and a similar magnitude positive S2 anomaly occurring to the northeast of New Zealand. These S2 anomalies result from the compensating convergence (divergence) on the poleward edge of the divergent outflow (inflow) from the tropical heating anomalies. The RWS anomalies in response to the heating at 210°E (Fig. 6b) follow a similar pattern but are shifted eastward and noticeably weaker than for the 170°E experiment. In both experiments, there is compensation of the anticyclonic S1 anomaly directly south of the heating by the cyclonic S2 anomaly, but the S1 anomaly dominates and provides the overall anticyclonic forcing for the teleconnections [and the observed teleconnection (not shown)].
(a) Day 3 outgoing longwave radiation (OLR; shaded; W m−2), 250-hPa divergent wind (vectors; scale in bottom right; vectors with magnitude less than 0.5 m s−1 are not displayed), 250-hPa Rossby wave source (RWS) advection S1 term (magenta contours; positive solid and negative dashed; interval: 0.15 × 10−11 s−2 from −1.5 to 1.5 × 10−11 s−2, and zero contours are omitted), and 250-hPa RWS stretching term S2 (green contours; positive solid and negative dashed; interval: 0.15 × 10−11 s−2 from −1.5 to 1.5 × 10−11 s−2, and zero contours are omitted,) anomalies for the CAM5 170°E experiment minus control. In the SH, positive RWS indicates anticyclonic tendency and negative RWS indicates cyclonic tendency. The RWS fields have been smoothed with the Hoskins spectral filter. (b) As in (a), but for the 210°E experiment.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
We show in Fig. 5a the winter-mean 250-hPa zonal wind
The different wave propagation characteristics resulting from the western- and eastern-located Pacific heating anomalies can also be explained by variations in the mean-state zonal wind, which affects the group velocity of the Rossby waves. The total stationary Rossby wavenumber
The STJ waveguide is more pronounced to the west of 210°E, hence we expect that the wave train emerging from the heating centered at 170°E will be initially more trapped in the waveguide than is the wave train emerging from the heating centered at 210°E where the waveguide is weaker and less confined. Hence, more pronounced zonal propagation occurs for the 170°E forcing (Figs. 4b–e), which is positioned closer in longitude to the core of the STJ. Similar results are obtained if the 170°E heating anomaly is positioned at 160°E or 180°, with the subtropical waveguide trapping effect becoming more pronounced for farther-westward-located heating closer to the STJ entrance (Gillett et al. 2022). However, away from the STJ in the central Pacific (Fig. 5a), where
The steady-state extratropical response for eastern Pacific heating (Fig. 4j) propagates in a typical Hoskins and Karoly (1981) manner and can be understood from the direct dispersion of a tropically forced Rossby wave. However, the zonally elongated meridional dipole structure of the response for western Pacific heating (Fig. 4e) appears to not be able to be completely understood from linear Rossby dispersion theory. In the next section, we therefore examine the impact of transient eddy feedbacks on the structure of the extratropical response.
c. Role of the transient eddy feedback
At the same time that the direct linear response to tropical forcing is being established, and extending to higher latitudes, interactions between the transient eddies and the developing wave train can act to enhance the existing height anomalies and even modify their structure (e.g., Held et al. 1989; Li et al. 2006). For context, we provide an estimate of the time-mean distribution of the transient eddies by examining the root-mean-square (RMS) amplitude of high-pass-filtered 250-hPa geopotential height from JRA-55 averaged for austral winter (Fig. 7a). The high-pass filter is calculated by retaining periods of less than 7 days using a Lanczos filter with 31 weights (Duchon 1979). Maxima of the high-pass-filtered height amplitude is indicative of the location of the climatological storm track. The strongest transient eddy amplitude is collocated with the midlatitude eddy-driven jet over the South Atlantic and Indian Oceans (∼50°S; see overlaid contours), with little evidence of a localized maximum of transient eddy activity associated with the STJ (see also, e.g., Nakamura and Shimpo 2004).
(a) Root-mean-square (RMS) amplitude of high-pass-filtered 250-hPa geopotential height (m; i.e., an indicator of the storm track) during austral winter from JRA-55. The climatological austral winter zonal wind at 250 hPa is overlaid (interval of 10 m s−1; westerly winds are solid, easterly winds are dashed, and zero contours are omitted). Composite RMS anomalies during (b) La Niña and (e) El Niño from JRA-55. Regressed RMS anomalies during (c) CP and (f) EP El Niño from JRA-55. RMS anomalies for the day 16–30 mean of (d) the CAM5 170°E experiment minus control and (g) the 210°E experiment minus control. The RMS anomalies have been smoothed with the Hoskins spectral filter.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
To understand how the Rossby wave train can act to perturb the transient eddies, we examine storm-track (RMS) anomalies in the observations and CAM5 experiments (Fig. 7). Figures 7b and 7e show composites of the observed transient eddy activity anomalies using JRA-55 at the 250-hPa level during La Niña and El Niño winters, respectively. Focusing on the Pacific Ocean sector, La Niña is associated with enhanced transient eddy activity (i.e., a strengthening of the storm track) across the entire South Pacific Ocean basin (Fig. 7b), which occurs on the equatorward flank of the observed high-latitude anticyclonic anomaly (Fig. 2b) that would act to accelerate the eddy-driven jet in the South Pacific. There is also a positive storm-track anomaly in the southern Indian Ocean, also in conjunction with the observed cyclonic height anomaly there (Fig. 2b). In contrast, El Niño is associated with a suppressed storm-track anomaly in the eastern South Pacific (Fig. 7e), which occurs on the equatorward flank of the high-latitude anticyclonic anomaly that acts to decelerate the eddy-driven jet (Fig. 2d). Figures 7c and 7f show regressions onto the CP El Niño and EP El Niño indices, respectively. The storm-track anomalies during CP El Niño (Fig. 7c) clearly resemble those during La Niña (but with the sign reversed), such that CP El Niño is associated with a broad suppressed storm-track anomaly across the South Pacific. The EP El Niño anomalies (Fig. 7f) are broadly similar to the El Niño anomalies, but the center of the suppressed storm-track anomaly over the South Pacific is shifted slightly westward and is less confined to the eastern Pacific basin.
In the CAM5 experiments, we calculate the equivalent RMS high-pass-filtered height for every 5-day non-overlapping period (days 6–10, 11–15, 16–20, 21–25, and 26–30) by first removing the 5-day time-mean for each period. We do this for each ensemble member individually and then average across members and difference the experiment and control. Such an approach to isolate the transient eddies was used earlier by Lin et al. (2007). A similar pattern of anomalies is obtained if we simply remove the 7-day running mean as in Gillett et al. (2022) but this approach helps to smooth the data and highlight the signal. Figures 7d and 7g show the day 16–30 mean [which is first calculated separately for the three periods (16–20, 21–25, and 26–30) and then averaged] RMS anomaly in the CAM5 170° and 210°E experiments, respectively. Over the Pacific Ocean, the 170°E experiment is associated with a zonally elongated meridional dipole storm-track anomaly, i.e., an enhanced storm-track anomaly in the subtropical Pacific and a suppressed storm-track anomaly in the midlatitudes (Fig. 7d). The respective anomalies for the 210°E experiment are shifted eastward toward South America (Fig. 7g). The key features of the extratropical storm-track anomalies in the CAM5 experiments agree well with the observed anomalies (Figs. 7b,c,e,f) but are subtropically amplified.
Figures 8a and 8f show the observed height tendency produced by the transient eddy anomalies at the 250-hPa level during composited La Niña and El Niño winters, respectively. The height tendency is stronger during La Niña (Fig. 8b) and is coherent with the height anomalies (Fig. 2b), with a strong zonally symmetric cyclonic tendency over the southern high latitudes that is largest over the South Pacific, and a localized anticyclonic tendency in the central South Pacific. These height tendency anomalies agree with the implied westerly acceleration across the South Pacific (anticyclonic circulation around the positive height tendency and cyclonic circulation around the negative height tendency; Fig. 7b), and induce a positive feedback (i.e., weakening the storm track causes less momentum to converge into it, which further weakens the storm track etc.). In comparison to La Niña, the height tendency during El Niño is weaker and less zonally symmetric (Fig. 8f). Like the subtropical storm-track anomaly (Fig. 7e), the subtropical negative height tendency anomaly during El Niño is also confined to the western Pacific Ocean.
The 250-hPa geopotential height tendency by transient vorticity flux convergence (m day−1). Composite transient eddy feedback anomalies during (a) La Niña and (f) El Niño from JRA-55. Regressed transient eddy feedback anomalies during (b) CP and (g) EP El Niño from JRA-55. Transient eddy feedback in the CAM5 170°E experiment minus control on (c) days 6–10, (d) days 11–15, and for the (e) day 16–30 mean. (g)–(i) As in (c)–(e), but for the 210°E experiment. The eddy feedback fields have been smoothed with the Hoskins spectral filter.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
Figures 8b and 8g show regressions of the observed CP El Niño and EP El Niño indices against the height tendency. Over the Pacific Ocean, there is a clear agreement between the westward-shifted ENSO events (La Niña and CP El Niño) and the eastward-shifted events (El Niño and EP El Niño). CP El Niño is associated with a meridional dipole height tendency, with a cyclonic anomaly in the midlatitudes and an anticyclonic anomaly farther south (Fig. 8b), while the EP El Niño height tendency has a different structure, with a strong anticyclonic anomaly in the high-latitude western Pacific and a weaker cyclonic anomaly in the subtropical-to-midlatitude eastern Pacific (Fig. 8g).
The tropical OLR anomalies are larger during El Niño (Fig. 2c), which should lead to a height response with a larger amplitude than the La Niña response. However, the maximum amplitude of the height anomalies is comparable between La Niña and El Niño (Figs. 2a,c). The height tendency is overall stronger during La Niña than during El Niño (Figs. 8b,f), indicating the important role of the transient eddy feedback in strengthening the height anomalies during La Niña and a lesser effect on the El Niño–induced height anomalies, which appear to be largely tropically forced. In agreement, the height tendency associated with CP El Niño is also slightly stronger than that associated with EP El Niño.
To understand the temporal development of the transient eddy feedback, the bottom three rows of Fig. 8 display the height tendency in the CAM5 experiments on days 6–10 and days 11–15 and then for the day 16–30 mean. The induced height tendency is weak and not well-organized during the first 10 days of the integration for both the 170°E (Fig. 8c) and 210°E (Fig. 8g) experiments. From days 11–15, the feedback strengthens in the 170°E experiment with a cyclonic height tendency over the subtropical Pacific and an anticyclonic height tendency over the mid- to high latitudes (Fig. 8d). The pattern of the time-mean transient eddy feedback response for the 170°E experiment (Fig. 8e) well-resembles the equivalent figures based on the JRA-55 (Figs. 8a,b). The circumglobal nature of the wave propagation in this experiment (Figs. 4a–e) due to its closer proximity to the STJ core (Fig. 5a) enables the storm track to be perturbed both locally in the South Pacific and remotely in the southern Atlantic and Indian Oceans (Fig. 7d), where the climatological storm track is largest (Fig. 7a), resulting in an annular response in the mid- to high latitudes (Fig. 8e). The difference in the height tendency in the 210°E experiment is subtle (cf. Fig. 8i with Fig. 8e). At mid- to high latitudes, the response has a similar pattern to the 170°E experiment (anticyclonic tendency) but is clearly weaker. At subtropical latitudes, however, the main cyclonic tendency is shifted eastward and is more localized to the west of South America, acting to intensify the cyclonic height anomaly there (Fig. 4j), compared to the subtropical anomalies in the 170°E experiment that are more zonal and span most of the Pacific basin. Li et al. (2006) also reported a subtle difference between the transient eddy feedback for the NH extratropical response to western and eastern tropical Pacific forcing linked to the proximity of the heating anomaly to the STJ core. The authors further used idealized transient eddy forcing anomaly experiments to support this result.
A key point from the CAM5 panels in Fig. 8 is that the transient eddy feedback anomalies do not have a strong influence on the height anomalies until after at least 10 days. This demonstrates that the transient height response depicted in Fig. 4 in the first 10 days represents the direct linear response to the imposed heating. The western and eastern Pacific heating experiments have reasonably similar patterns up to about day 5 (Figs. 4b,g), after which point different propagation paths arise due to reflective waveguide effects, as discussed in section 3b. Later in the integration, from days 11–15, the transient eddy feedback strengthens, particularly in the 170°E experiment, causing the height anomalies to obtain a very different structure to the anomalies in the 210°E experiment which depict the typical Hoskins–Karoly pattern. This suggests that the observed atmospheric circulation response to westward-shifted tropical Pacific heating anomalies like La Niña or CP El Niño cannot be obtained without a feedback between the forced stationary wave and transient eddies, as previously reported in the NH for the PNA-like response to western tropical Pacific forcing (e.g., Li et al. 2006).
Due to the weaker transient eddy feedback and different structure of this feedback over the Pacific Ocean for the observed El Niño response (Fig. 8f), the EP El Niño response and the 210°E CAM5 experiment (Figs. 8g–i), we interpret the transient eddy feedback to be of secondary importance for the 210°E response. It appears to simply enhance the height anomalies after they have already been established, rather than act to establish a completely different structure. We verified that the 210°E experiment does not require larger total heating in order to induce a transient eddy feedback by running an experiment where the maximum amplitude of the imposed diabatic heating is increased to 10 K day−1. This larger heating almost doubled the total diabatic heating anomaly (averaged over the patch and 30-day period) but the structure of the height anomalies was essentially the same (not shown). The structure of the transient eddy feedback in the 170°E experiment is also similar when the imposed heating amplitude is reduced (not shown). Lim et al. (2013) and Wilson et al. (2016) previously showed that there is a weaker feedback between the transient eddies and mean flow during EP ENSO events compared to CP events. Yu et al. (2015) also found that CP ENSO can induce both a regionally confined response (i.e., the PSA pattern) and an annular response [the Southern Annular Mode (SAM)], but EP ENSO can only induce a PSA response.
d. Comparison with a linearized simple model
We further examine the possible role of a transient eddy feedback for organizing the extratropical response to western tropical Pacific heating by comparing to a similar experiment in the SGCM, which expressly excludes this feedback. The transient eddy feedback is one component that is not incorporated in this model; however, we acknowledge that other nonlinear processes are also excluded, and which could possibly be important such as nonlinear advection (e.g., Hendon 1986; Ting and Yu 1998).
Figure 9 shows the evolution of the 250-hPa height anomalies in the SGCM. The tropical-subtropical response to the imposed heating at both 170° and 210°E develops very quickly; the subtropical anticyclonic height anomaly south of the imposed heating and a cyclonic anomaly to its east are already established by day 3 (Figs. 9a,f) and the equatorial Kelvin wave rapidly circles the globe in less than one week. In contrast to the CAM5 experiments, there are only subtle differences between the two SGCM experiments: the mid- to high-latitude height anomalies over the eastern South Pacific appear to tilt toward the east/South America in the 210°E experiment (Fig. 9j) but in the 170°E experiment they are flattened (Fig. 9e). The height responses in the SGCM experiments are also subtropically amplified compared to the CAM5 experiments, which are midlatitude amplified. This is likely related to the enhanced STJ waveguide effect in the SGCM due to the extended STJ (Figs. S2d,f). Overall, however, the two SGCM experiments appear to be a zonal translation of the other, with both exhibiting a very similar structure at day 12 (Figs. 9d,i) as for days 17–20 (Figs. 9e,j).
Geopotential height anomalies (m) at 250 hPa on (a) day 3, (b) day 5, (c) day 8, (d) day 12, and (e) for the day 17–20 mean from the SGCM 170°E experiment minus control. (f)–(j) As in (a)–(e), but for the 210°E experiment. The navy-blue contours in (a) and (f) show the horizontal structure of the prescribed forcing.
Citation: Journal of Climate 36, 8; 10.1175/JCLI-D-22-0159.1
We display in Fig. S3, the day 3 RWS and divergent wind in the SGCM, like in Fig. 4 for CAM5, to emphasize that the linear model captures the correct RWS pattern and also that that the 170°E subtropical anticyclonic S1 RWS anomaly is stronger than the equivalent 210°E anomaly. The corresponding anticyclonic subtropical height anomaly is thus stronger in the former experiment.
To first order, these SGCM results suggest that due to the lack of a transient eddy feedback in the linearized SGCM, the height response for western Pacific heating cannot obtain the same realistic structure as in the observations (Figs. 2b and 3b) and in the CAM5 simulations (Fig. 3e). The extratropical height anomalies in both the 170° and 210°E SGCM experiments (Fig. 9) are also weaker than the respective subtropical height anomalies and are weaker than the mid- to high-latitude response in the observed system (Figs. 2 and 3) and CAM5 experiments (Fig. 4). There is no contribution from the transients in the mid- to high latitudes in the SGCM, which could explain why the mid- to high-latitude response is weaker in the SGCM. This further supports the role of a transient eddy feedback that acts to amplify these anomalies, as shown in Fig. 8 for JRA-55 and CAM5.
However, the similar structure and amplitude of the height response between the 170° and 210°E experiments could also be associated with the more zonally symmetric negative Ks region in the SGCM compared to the observations (cf. Fig. 6c with Fig. S2f), such that the extratropical response is less sensitive to the longitude of the applied tropical forcing. Nonetheless, these results are consistent with the inferred primary role of the transient eddy feedback for molding the response to the heating at 170°E, producing its zonally elongated meridionally dipole structure that only emerges after about 2 weeks.
4. Discussion and conclusions
This study set out to understand the mechanisms behind the apparent differences in the austral winter extratropical wave train response to westward-shifted ENSO events [like what occurs during La Niña (Fig. 2a) and CP El Niño (Fig. 3a)] and eastward-shifted ENSO events [like canonical or EP El Niño (Figs. 2b and 3b)]. Idealized spinup experiments conducted using a 90-member atmospheric GCM ensemble enabled us to examine the evolution of the atmospheric response to suddenly switching on steady tropical diabatic heating in the equatorial western (located west of the date line at 170°E) and eastern (located east of the date line at 210°E) Pacific Ocean, so to better understand the underlying dynamics. These results extend previous studies of the teleconnection from ENSO into the SH extratropics that focused on the time-independent equilibrium response or did not consider the role of nonlinear processes (e.g., Cai et al. 2011; Wilson et al. 2014; Ciasto et al. 2015; Yiu and Maycock 2019; Wang et al. 2022b).
Part of the difference in the response to the western- and eastern-located tropical diabatic heating anomalies can be explained from linear Rossby wave initiation and propagation theory (e.g., Hoskins and Karoly 1981; Sardeshmukh and Hoskins 1985). Western Pacific heating, closer to the core of the strong wintertime STJ, induces a large subtropical RWS that triggers a stationary wave train. This wave train is initially trapped along the STJ waveguide and so rapidly results in a response that extends around the entire SH hemisphere. In contrast, eastern Pacific heating, farther away from the STJ core, produces a weaker RWS and a Rossby wave train that is more able to escape to the extratropics, resulting in anomalies with larger meridional orientation that are regionally confined.
The time-mean extratropical response for eastern Pacific heating can be understood simply from Hoskins–Karoly Rossby wave dispersion; however, the zonally elongated time-mean response for western Pacific heating requires an additional mechanism. We examined the evolution of the barotropic feedback of the transient eddies onto the mean height anomaly and found that this feedback intensified from around day 10 after circumglobal propagation perturbed the storm track in both the Indian and Pacific basins and thus promoted a zonal elongation of the height anomalies at mid- to high latitudes. This suggests that the structure of the extratropical height response to western Pacific heating cannot be obtained without a feedback between the forced stationary wave and transient eddies. In contrast, the wave train forced by eastern Pacific heating appeared to be less modified by this feedback, with the main effect being an intensification of the wave train that is established by week 2. This important distinction for the role of the transient eddy feedback for western- and eastern-located tropical Pacific heating builds on previous studies that examined links between ENSO and the SAM (which is maintained by a positive feedback from the transient eddies in the storm track; e.g., Ding et al. 2012; Lim et al. 2013; Yu et al. 2015; Wilson et al. 2016). By exploring the evolution of the response, we demonstrated that this feedback modifies the direct tropically forced response about 10 days after the heating is switched on, and we could therefore distinguish the direct tropically forced response from the nonlinear indirect response.
To first order, experimentation with the linearized SGCM appears to support this mechanism as it does not capture a realistic response for the western Pacific heating due to the lack of transient eddy feedback (noting that other nonlinear processes are also excluded from the linear model equations). However, mean-state biases in the SGCM and stronger dissipative behavior at the subtropical turning point (where Ks is undefined) may mean that a realistic response may not be supported, with or without a transient eddy feedback. This possibility reemphasizes the importance of accurately representing the mean state when simulating teleconnections, which has implications for seasonal predictions. For example, CMIP6 models still struggle to maintain the separation between the subtropical and midlatitude jets during austral winter, exhibiting a substantial equatorward bias in the midlatitude jet position (e.g., Simpson et al. 2020). Further targeted experimentation such as with an imposed idealized local transient eddy vorticity source in the midlatitudes (e.g., Li et al. 2006) in a linear storm-track model would help to further elucidate the role of an extratropical transient eddy feedback for organizing the structure of the wave train.
Similar waveguide behavior also occurs in the NH during winter (e.g., Branstator 2002; Li et al. 2006; Branstator and Teng 2017). A transient eddy feedback has also been reported to control the NH extratropical response to tropical Pacific forcing (Li et al. 2006). Unlike NH studies that have explored the ENSO teleconnection during boreal winter when ENSO is in its mature stage (e.g., Trenberth et al. 1998), we have examined the teleconnection during austral winter when the STJ is strongest, but ENSO is only in its developing stage. However, a distinct PSA pattern associated with tropical convection can still be observed in this season (e.g., Karoly 1989; Mo and Higgins 1998). We speculate that similar results and mechanisms may operate during austral spring when the STJ is still present. Cai et al. (2012) reported a similar asymmetry of upper-tropospheric height anomalies between El Niño and La Niña during spring. The link between SAM and ENSO is also stronger in this season (e.g., Lim et al. 2013), potentially leading to stronger transient eddy feedbacks.
Although there is good coherence between the observed (Figs. 2 and 3) and CAM5 simulated (Fig. 4) teleconnections, these experiments are idealized. The strongly asymmetric height anomalies between El Niño and La Niña (Figs. 1b,d) could arise from other factors not tested here. For example, due to a contribution from opposite-signed heating to the west of the tropical precipitation anomaly (Figs. 1a,c), or due to the different effects of a negative La Niña heating anomaly on the zonal wind and jet waveguide, as suggested by the slightly weaker asymmetry between CP and EP El Niño events (Fig. 3) than between La Niña and El Niño events (Fig. 2). Other factors and mechanisms, aside from tropical convection related to ENSO and transient eddies, that we have not considered could also influence the observed wave train structures. For example, decadal and multidecadal variability in the Atlantic (e.g., Li et al. 2014, 2015a) and Pacific Oceans (e.g., Clem and Fogt 2015) also influence the South Pacific region. Finally, future work could also explore the impact anthropogenic warming may have on these teleconnections associated with changes in tropical convection during ENSO events (e.g., Cai et al. 2021) and the extratropical atmospheric circulation (e.g., Lu et al. 2008; Chenoli et al. 2017; Wang et al. 2022a).
Acknowledgments.
We acknowledge support from the Australian Research Council (ARC) through the Centre of Excellence for Climate Extremes (CE170100023). Z.E.G. was supported by an Australian Government Research Training Program Scholarship and an ARC Centre of Excellence for Climate Extremes/Australian Bureau of Meteorology PhD Top-up scholarship. J.M.A. was partially supported by the Regional and Global Model Analysis component of the Earth and Environmental System Modeling Program of the U.S. Department of Energy’s Office of Biological and Environmental Research via National Science Foundation IA 1947282. Computing resources and services were provided by the National Computational Infrastructure, which is supported by the Australian government. Additional computing resources were provided through the ARC LIEF Grant LE200100040. The NCAR Command Language (http://www.ncl.ucar.edu) version 6.6.2 was used for data analysis and visualization. We thank Christine Chung and Roseanna McKay for their comments on an earlier version of the manuscript, and Cristiana Stan and two anonymous reviewers for reviewing the manuscript.
Data availability statement.
The JRA-55 dataset is available through the Collaborative REAnalysis Technical Environment – Intercomparison Project (CREATE-IT) website (https://esgf-node.llnl.gov/projects/create-ip/). The merged Hadley–NOAA/Optimal Interpolation SST dataset is available from the NCAR/UCAR Geoscience Data Exchange (https://doi.org/10.5065/r33v-sv91). GPCP data are available from the NOAA/OAR/ESRL PSL website (https://psl.noaa.gov/data/gridded/data.gpcp.html). Model output is available upon request.
REFERENCES
Adler, R. F., and Coauthors, 2003: The version-2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979–present). J. Hydrometeor., 4, 1147–1167, https://doi.org/10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Ambrizzi, T., B. J. Hoskins, and H.-H. Hsu, 1995: Rossby wave propagation and teleconnection patterns in the austral winter. J. Atmos. Sci., 52, 3661–3672, https://doi.org/10.1175/1520-0469(1995)052<3661:RWPATP>2.0.CO;2.
Ashok, K., S. K. Behera, S. A. Rao, H. Weng, and T. Yamagata, 2007: El Nino Modoki and its possible teleconnection. J. Geophys. Res., 112, C11007, https://doi.org/10.1029/2006JC003798.
Barsugli, J. J., and P. D. Sardeshmukh, 2002: Global atmospheric sensitivity to tropical SST anomalies throughout the Indo-Pacific basin. J. Climate, 15, 3427–3442, https://doi.org/10.1175/1520-0442(2002)015<3427:GASTTS>2.0.CO;2.
Branstator, G., 1985: Analysis of general circulation model sea-surface temperature anomaly simulations using a linear model. Part I: Forced solutions. J. Atmos. Sci., 42, 2225–2241, https://doi.org/10.1175/1520-0469(1985)042<2225:AOGCMS>2.0.CO;2.
Branstator, G., 2002: Circumglobal teleconnections, the jet stream waveguide, and the North Atlantic Oscillation. J. Climate, 15, 1893–1910, https://doi.org/10.1175/1520-0442(2002)015<1893:CTTJSW>2.0.CO;2.
Branstator, G., and H. Teng, 2017: Tropospheric waveguide teleconnections and their seasonality. J. Atmos. Sci., 74, 1513–1532, https://doi.org/10.1175/JAS-D-16-0305.1.
Cai, W., P. van Rensch, T. Cowan, and H. H. Hendon, 2011: Teleconnection pathways of ENSO and the IOD and the mechanisms for impacts on Australian rainfall. J. Climate, 24, 3910–3923, https://doi.org/10.1175/2011JCLI4129.1.
Cai, W., P. van Rensch, T. Cowan, and H. H. Hendon, 2012: An asymmetry in the IOD and ENSO teleconnection pathway and its impact on Australian climate. J. Climate, 25, 6318–6329, https://doi.org/10.1175/JCLI-D-11-00501.1.
Cai, W., and Coauthors, 2021: Changing El Niño–Southern Oscillation in a warming climate. Nat. Rev. Earth Environ., 2, 628–644, https://doi.org/10.1038/s43017-021-00199-z.
Capotondi, A., and Coauthors, 2015: Understanding ENSO diversity. Bull. Amer. Meteor. Soc., 96, 921–938, https://doi.org/10.1175/BAMS-D-13-00117.1.
Chenoli, S. N., M. Y. A. Mazuki, J. Turner, and A. Abu Samah, 2017: Historical and projected changes in the Southern Hemisphere sub-tropical jet during winter from the CMIP5 models. Climate Dyn., 48, 661–681, https://doi.org/10.1007/s00382-016-3102-y.
Ciasto, L. M., G. R. Simpkins, and M. H. England, 2015: Teleconnections between tropical Pacific SST anomalies and extratropical Southern Hemisphere climate. J. Climate, 28, 56–65, https://doi.org/10.1175/JCLI-D-14-00438.1.
Clem, K. R., and R. L. Fogt, 2015: South Pacific circulation changes and their connection to the tropics and regional Antarctic warming in austral spring, 1979–2012. J. Geophys. Res. Atmos., 120, 2773–2792, https://doi.org/10.1002/2014JD022940.
Dai, Y., S. B. Feldstein, B. Tan, and S. Lee, 2017: Formation mechanisms of the Pacific–North American teleconnection with and without its canonical tropical convection pattern. J. Climate, 30, 3139–3155, https://doi.org/10.1175/JCLI-D-16-0411.1.
Deser, C., I. R. Simpson, K. A. McKinnon, and A. S. Phillips, 2017: The Northern Hemisphere extratropical atmospheric circulation response to ENSO: How well do we know it and how do we evaluate models accordingly? J. Climate, 30, 5059–5082, https://doi.org/10.1175/JCLI-D-16-0844.1.
Ding, Q., E. J. Steig, D. S. Battisti, and J. M. Wallace, 2012: Influence of the tropics on the southern annular mode. J. Climate, 25, 6330–6348, https://doi.org/10.1175/JCLI-D-11-00523.1.
Duchon, C. E., 1979: Lanczos filtering in one and two dimensions. J. Appl. Meteor. Climatol., 18, 1016–1022, https://doi.org/10.1175/1520-0450(1979)018<1016:LFIOAT>2.0.CO;2.
Feng, J., W. Chen, and Y. Li, 2017: Asymmetry of the winter extra-tropical teleconnections in the Northern Hemisphere associated with two types of ENSO. Climate Dyn., 48, 2135–2151, https://doi.org/10.1007/s00382-016-3196-2.
Fogt, R. L., D. H. Bromwich, and K. M. Hines, 2011: Understanding the SAM influence on the South Pacific ENSO teleconnection. Climate Dyn., 36, 1555–1576, https://doi.org/10.1007/s00382-010-0905-0.
Freund, M. B., A. G. Marshall, M. C. Wheeler, and J. N. Brown, 2021: Central Pacific El Niño as a precursor to summer drought-breaking rainfall over southeastern Australia. Geophys. Res. Lett., 48, e2020GL091131, https://doi.org/10.1029/2020GL091131.
Fu, J.-J., M.-H. Zhang, Z. Han, and S. Li, 2013: Sensitivity difference in the extratropical atmosphere to two types of El Nino events. Atmos. Oceanic Sci. Lett., 6, 355–359, https://doi.org/10.1080/16742834.2013.11447107.
Garfinkel, C. I., I. Weinberger, I. P. White, L. D. Oman, V. Aquila, and Y.-K. Lim, 2019: The salience of nonlinearities in the boreal winter response to ENSO: North Pacific and North America. Climate Dyn., 52, 4429–4446, https://doi.org/10.1007/s00382-018-4386-x.
Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106, 447–462, https://doi.org/10.1002/qj.49710644905.
Gillett, Z. E., H. H. Hendon, J. M. Arblaster, and E.-P. Lim, 2021: Tropical and extratropical influences on the variability of the Southern Hemisphere wintertime subtropical jet. J. Climate, 34, 4009–4022, https://doi.org/10.1175/JCLI-D-20-0460.1.
Gillett, Z. E., H. H. Hendon, J. M. Arblaster, H. Lin, and D. Fuchs, 2022: On the dynamics of Indian Ocean teleconnections into the Southern Hemisphere during austral winter. J. Atmos. Sci., 79, 2453–2469, https://doi.org/10.1175/JAS-D-21-0206.1.
Hall, N. M. J., 2000: A simple GCM based on dry dynamics and constant forcing. J. Atmos. Sci., 57, 1557–1572, https://doi.org/10.1175/1520-0469(2000)057<1557:ASGBOD>2.0.CO;2.
Held, I. M., S. W. Lyons, and S. Nigam, 1989: Transients and the extratropical response to El Niño. J. Atmos. Sci., 46, 163–174, https://doi.org/10.1175/1520-0469(1989)046<0163:TATERT>2.0.CO;2.
Hendon, H. H., 1986: The time-mean flow and variability in a nonlinear model of the atmosphere with tropical diabatic forcing. J. Atmos. Sci., 43, 72–89, https://doi.org/10.1175/1520-0469(1986)043<0072:TTMFAV>2.0.CO;2.
Hoerling, M. P., A. Kumar, and M. Zhong, 1997: El Niño, La Niña, and the nonlinearity of their teleconnections. J. Climate, 10, 1769–1786, https://doi.org/10.1175/1520-0442(1997)010<1769:ENOLNA>2.0.CO;2.
Hoskins, B. J., and D. J. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 1179–1196, https://doi.org/10.1175/1520-0469(1981)038<1179:TSLROA>2.0.CO;2.
Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50, 1661–1671, https://doi.org/10.1175/1520-0469(1993)050<1661:RWPOAR>2.0.CO;2.
Hurrell, J. W., J. J. Hack, D. Shea, J. M. Caron, and J. Rosinski, 2008: A new sea surface temperature and sea ice boundary dataset for the Community Atmosphere Model. J. Climate, 21, 5145–5153, https://doi.org/10.1175/2008JCLI2292.1.
Jin, F., and B. J. Hoskins, 1995: The direct response to tropical heating in a baroclinic atmosphere. J. Atmos. Sci., 52, 307–319, https://doi.org/10.1175/1520-0469(1995)052<0307:TDRTTH>2.0.CO;2.
Karoly, D. J., 1989: Southern Hemisphere circulation features associated with El Niño–Southern Oscillation events. J. Climate, 2, 1239–1252, https://doi.org/10.1175/1520-0442(1989)002<1239:SHCFAW>2.0.CO;2.
Kidson, J. W., 1988: Interannual variations in the Southern Hemisphere circulation. J. Climate, 1, 1177–1198, https://doi.org/10.1175/1520-0442(1988)001<1177:IVITSH>2.0.CO;2.
Kobayashi, S., and Coauthors, 2015: The JRA-55 reanalysis: General specifications and basic characteristics. J. Meteor. Soc. Japan, 93, 5–48, https://doi.org/10.2151/jmsj.2015-001.
Lachlan-Cope, T., and W. Connolley, 2006: Teleconnections between the tropical Pacific and the Amundsen–Bellinghausens Sea: Role of the El Niño–Southern Oscillation. J. Geophys. Res., 111, D23101, https://doi.org/10.1029/2005JD006386.
Lau, N.-C., 1988: Variability of the observed midlatitude storm tracks in relation to low-frequency changes in the circulation pattern. J. Atmos. Sci., 45, 2718–2743, https://doi.org/10.1175/1520-0469(1988)045<2718:VOTOMS>2.0.CO;2.
Lau, N.-C., A. Leetmaa, M. J. Nath, and H.-L. Wang, 2005: Influences of ENSO-induced Indo–Western Pacific SST anomalies on extratropical atmospheric variability during the boreal summer. J. Climate, 18, 2922–2942, https://doi.org/10.1175/JCLI3445.1.
L’Heureux, M. L., and D. W. J. Thompson, 2006: Observed relationships between the El Niño–Southern Oscillation and the extratropical zonal-mean circulation. J. Climate, 19, 276–287, https://doi.org/10.1175/JCLI3617.1.
Li, S., M. P. Hoerling, S. Peng, and K. M. Weickmann, 2006: The annular response to tropical Pacific SST forcing. J. Climate, 19, 1802–1819, https://doi.org/10.1175/JCLI3668.1.
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, https://doi.org/10.1038/nature12945.
Li, X., E. P. Gerber, D. M. Holland, and C. Yoo, 2015a: A Rossby wave bridge from the tropical Atlantic to West Antarctica. J. Climate, 28, 2256–2273, https://doi.org/10.1175/JCLI-D-14-00450.1.
Li, X., D. M. Holland, E. P. Gerber, and C. Yoo, 2015b: Rossby waves mediate impacts of tropical oceans on west Antarctic atmospheric circulation in austral winter. J. Climate, 28, 8151–8164, https://doi.org/10.1175/JCLI-D-15-0113.1.
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.
Lim, E.-P., H. H. Hendon, and H. Rashid, 2013: Seasonal predictability of the southern annular mode due to its association with ENSO. J. Climate, 26, 8037–8054, https://doi.org/10.1175/JCLI-D-13-00006.1.
Lin, H., and G. Brunet, 2018: Extratropical response to the MJO: Nonlinearity and sensitivity to the initial state. J. Atmos. Sci., 75, 219–234, https://doi.org/10.1175/JAS-D-17-0189.1.
Lin, H., J. Derome, and G. Brunet, 2007: The nonlinear transient atmospheric response to tropical forcing. J. Climate, 20, 5642–5665, https://doi.org/10.1175/2007JCLI1383.1.
Liu, Z., and M. Alexander, 2007: Atmospheric bridge, oceanic tunnel, and global climatic teleconnections. Rev. Geophys., 45, RG2005, https://doi.org/10.1029/2005RG000172.
Lu, J., G. Chen, and D. M. W. Frierson, 2008: Response of the zonal mean atmospheric circulation to El Niño versus global warming. J. Climate, 21, 5835–5851, https://doi.org/10.1175/2008JCLI2200.1.
Matsuno, T., 1966: Quasi-geostrophic motions in the equatorial area. J. Meteor. Soc. Japan, 44, 25–43, https://doi.org/10.2151/jmsj1965.44.1_25.
McPhaden, M. J., S. E. Zebiak, and M. H. Glantz, 2006: ENSO as an integrating concept in Earth science. Science, 314, 1740–1745, https://doi.org/10.1126/science.1132588.
Meehl, G. A., J. M. Arblaster, G. Branstator, and H. van Loon, 2008: A coupled air–sea response mechanism to solar forcing in the Pacific region. J. Climate, 21, 2883–2897, https://doi.org/10.1175/2007JCLI1776.1.
Mo, K. C., and M. Ghil, 1987: Statistics and dynamics of persistent anomalies. J. Atmos. Sci., 44, 877–902, https://doi.org/10.1175/1520-0469(1987)044<0877:SADOPA>2.0.CO;2.
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, https://doi.org/10.1175/1520-0493(1998)126<1581:TPSAMA>2.0.CO;2.
Mo, K. C., and J. N. Paegle, 2001: The Pacific–South American modes and their downstream effects. Int. J. Climatol., 21, 1211–1229, https://doi.org/10.1002/joc.685.
Nakamura, H., and A. Shimpo, 2004: Seasonal variations in the Southern Hemisphere storm tracks and jet streams as revealed in a reanalysis dataset. J. Climate, 17, 1828–1844, https://doi.org/10.1175/1520-0442(2004)017<1828:SVITSH>2.0.CO;2.
Neale, R. B., and Coauthors, 2012: Description of the NCAR Community Atmosphere Model (CAM 5.0). NCAR Tech. Note NCAR/TN-486+STR, 274 pp., www.cesm.ucar.edu/models/cesm1.0/cam/docs/description/cam5_desc.pdf.
Power, S., M. Haylock, R. Colman, and X. Wang, 2006: The predictability of interdecadal changes in ENSO activity and ENSO teleconnections. J. Climate, 19, 4755–4771, https://doi.org/10.1175/JCLI3868.1.
Qin, J., and W. A. Robinson, 1993: On the Rossby wave source and the steady linear response to tropical forcing. J. Atmos. Sci., 50, 1819–1823, https://doi.org/10.1175/1520-0469(1993)050<1819:OTRWSA>2.0.CO;2.
Rasmusson, E. M., and T. H. Carpenter, 1982: Variations in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation/El Niño. Mon. Wea. Rev., 110, 354–384, https://doi.org/10.1175/1520-0493(1982)110<0354:VITSST>2.0.CO;2.
Santoso, A., M. J. Mcphaden, and W. Cai, 2017: The defining characteristics of ENSO extremes and the strong 2015/2016 El Niño. Rev. Geophys., 55, 1079–1129, https://doi.org/10.1002/2017RG000560.
Sardeshmukh, P. D., and B. J. Hoskins, 1984: Spatial smoothing on the sphere. Mon. Wea. Rev., 112, 2524–2529, https://doi.org/10.1175/1520-0493(1984)112<2524:SSOTS>2.0.CO;2.
Sardeshmukh, P. D., and B. J. Hoskins, 1985: Vorticity balances in the tropics during the 1982–83 El Niño‐Southern Oscillation event. Quart. J. Roy. Meteor. Soc., 111, 261–278, https://doi.org/10.1256/smsqj.46801.
Schneider, D. P., Y. Okumura, and C. Deser, 2012: Observed Antarctic interannual climate variability and tropical linkages. J. Climate, 25, 4048–4066, https://doi.org/10.1175/JCLI-D-11-00273.1.
Seager, R., N. Harnik, Y. Kushnir, W. Robinson, and J. Miller, 2003: Mechanisms of hemispherically symmetric climate variability. J. Climate, 16, 2960–2978, https://doi.org/10.1175/1520-0442(2003)016<2960:MOHSCV>2.0.CO;2.
Simmons, A. J., J. M. Wallace, and G. W. Branstator, 1983: Barotropic wave propagation and instability, and atmospheric teleconnection patterns. J. Atmos. Sci., 40, 1363–1392, https://doi.org/10.1175/1520-0469(1983)040<1363:BWPAIA>2.0.CO;2.
Simpson, I. R., and Coauthors, 2020: An evaluation of the large-scale atmospheric circulation and its variability in CESM2 and other CMIP models. J. Geophys. Res. Atmos., 125, e2020JD032835, https://doi.org/10.1029/2020JD032835.
Stan, C., D. M. Straus, J. S. Frederiksen, H. Lin, E. D. Maloney, and C. Schumacher, 2017: Review of tropical-extratropical teleconnections on intraseasonal time scales. Rev. Geophys., 55, 902–937, https://doi.org/10.1002/2016RG000538.
Sullivan, A., J.-J. Luo, A. C. Hirst, D. Bi, W. Cai, and J. He, 2016: Robust contribution of decadal anomalies to the frequency of central-Pacific El Niño. Sci. Rep., 6, 38540, https://doi.org/10.1038/srep38540.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Taschetto, A. S., and M. H. England, 2009: El Niño Modoki impacts on Australian rainfall. J. Climate, 22, 3167–3174, https://doi.org/10.1175/2008JCLI2589.1.
Taschetto, A. S., R. R. Rodrigues, G. A. Meehl, S. McGregor, and M. H. England, 2016: How sensitive are the Pacific–tropical North Atlantic teleconnections to the position and intensity of El Niño-related warming? Climate Dyn., 46, 1841–1860, https://doi.org/10.1007/s00382-015-2679-x.
Ting, M., and L. Yu, 1998: Steady response to tropical heating in wavy linear and nonlinear baroclinic models. J. Atmos. Sci., 55, 3565–3582, https://doi.org/10.1175/1520-0469(1998)055<3565:SRTTHI>2.0.CO;2.
Trenberth, K. E., 1997: The definition of El Niño. Bull. Amer. Meteor. Soc., 78, 2771–2777, https://doi.org/10.1175/1520-0477(1997)078<2771:TDOENO>2.0.CO;2.
Trenberth, K. E., G. W. Branstator, D. Karoly, A. Kumar, N.-C. Lau, and C. Ropelewski, 1998: Progress during TOGA in understanding and modeling global teleconnections associated with tropical sea surface temperatures. J. Geophys. Res., 103, 14 291–14 324, https://doi.org/10.1029/97JC01444.
Trenberth, K. E., Y. Zhang, and J. T. Fasullo, 2015: Relationships among top-of-atmosphere radiation and atmospheric state variables in observations and CESM. J. Geophys. Res. Atmos., 120, 10 074–10 090, https://doi.org/10.1002/2015JD023381.
Wang, Y., G. Huang, K. Hu, W. Tao, H. Gong, K. Yang, and H. Tang, 2022a: Understanding the eastward shift and intensification of the ENSO teleconnection over South Pacific and Antarctica under greenhouse warming. Front. Earth Sci., 10, 916624, https://doi.org/10.3389/feart.2022.916624.
Wang, Y., G. Huang, K. Hu, W. Tao, X. Li, H. Gong, L. Gu, and W. Zhang, 2022b: Asymmetric impacts of El Niño and La Niña on the Pacific–South America teleconnection pattern. J. Climate, 35, 1825–1838, https://doi.org/10.1175/JCLI-D-21-0285.1.
Wheeler, M. C., and H. Nguyen, 2015: Tropical meteorology and climate: Equatorial waves. Encyclopedia of Atmospheric Sciences, 2nd ed. G. R. North, J. A. Pyle, and F. Zhang, Eds., Academic Press, 102–112.
Wilson, A. B., D. H. Bromwich, K. M. Hines, and S.-H. Wang, 2014: El Niño flavors and their simulated impacts on atmospheric circulation in the high southern latitudes. J. Climate, 27, 8934–8955, https://doi.org/10.1175/JCLI-D-14-00296.1.
Wilson, A. B., D. H. Bromwich, and K. M. Hines, 2016: Simulating the mutual forcing of anomalous high southern latitude atmospheric circulation by El Niño flavors and the southern annular mode. J. Climate, 29, 2291–2309, https://doi.org/10.1175/JCLI-D-15-0361.1.
Yeh, S.-W., and Coauthors, 2018: ENSO atmospheric teleconnections and their response to greenhouse gas forcing. Rev. Geophys., 56, 185–206, https://doi.org/10.1002/2017RG000568.
Yiu, Y. Y. S., and A. C. Maycock, 2019: On the seasonality of the El Niño teleconnection to the Amundsen Sea region. J. Climate, 32, 4829–4845, https://doi.org/10.1175/JCLI-D-18-0813.1.
Yu, J.-Y., H. Paek, E. S. Saltzman, and T. Lee, 2015: The early 1990s change in ENSO–PSA–SAM relationships and its impact on Southern Hemisphere climate. J. Climate, 28, 9393–9408, https://doi.org/10.1175/JCLI-D-15-0335.1.