1. Introduction
Ocean circulation in the tropical Pacific plays a vital role in the climate system, and its variability on seasonal to decadal time scales influences weather patterns and global climate (Xie and Philander 1994; Philander et al. 1987; Reverdin et al. 1994; Donguy and Meyers 1996). Its interaction with the overlying wind field in the equatorial band leads to El Niño–Southern Oscillation (ENSO) while off the equator it affects and is influenced by the intertropical convergence zone (ITCZ)—an intense zonal band of tropical precipitation. With rising global temperatures and changing tropical SST patterns, the question arises of how the ongoing anthropogenic climate change may affect ocean equatorial currents (Xie et al. 2010; Collins et al. 2010; Dai 2013; Li et al. 2019). A recent study suggests an acceleration of the global ocean surface currents since the early 1990s—an inference based on changes in integrated kinetic energy (Hu et al. 2020); however, uncertainties remain large due to differences in the measurements, internal variability (Peng et al. 2022), and sharp contrast between recent observations and future projections (e.g., Heede and Fedorov 2023). In this study, we will focus on the future changes in the two, perhaps most intriguing tropical Pacific currents—the North Equatorial Countercurrent (NECC) and the Equatorial Undercurrent (EUC)—as projected by phase 6 of the Coupled Model Intercomparison Project (CMIP6) (Eyring et al. 2016).
The NECC is a zonally oriented upper-ocean current flowing across the North Pacific, transporting on average over 20 Sv (1 Sv ≡ 106 m3 s−1) of warm water from the west Pacific warm pool to the relatively cold east (Wyrtki and Kendall 1967; Philander et al. 1987; Reverdin et al. 1994; Donguy and Meyers 1996; Yu et al. 2000; Kessler 2006; Li and Fedorov 2022). It is centered at ∼5°N in the western tropical Pacific but shifts slightly poleward as it flows eastward to approximately 7°–8°N in the central and eastern Pacific (Donguy and Meyers1996; Johnson et al. 2002). Generated by wind stress curl of the northeasterly trade winds, the NECC is a geostrophic current maintained by a balance between the Coriolis force and pressure gradient force due to meridional Ekman transport divergence (Sverdrup 1947). Its spatial pattern and intensity undergo pronounced seasonal changes with the migration of the ITCZ (Philander et al. 1987) as well as significant interannual changes with ENSO. As the ITCZ migrates northward during the summer–fall season, the NECC also shifts northward, resulting in an increase in its intensity. Conversely, during the winter–spring season, as the ITCZ moves southward, the NECC weakens.
Despite its importance in heat transport and ITCZ regulation (Masunaga and L’Ecuyer 2011; Li and Fedorov 2022), both ocean and coupled GCMs commonly simulate a too weak NECC (Brown and Fedorov 2008; Grima et al. 1999; Danabasoglu et al. 2014). A study comparing ocean GCMs shows almost no signals of NECC in the multimodel mean (Tseng et al. 2016). Insufficient model resolution, deficiency in wind stress, unsolved physics, and inadequate representation of ocean–atmosphere interactions are considered as possible explanations. A recent study suggests that positive feedback between the NECC and ITCZ, involving wind stress uurl–Advection–SST–precipitation (WASP feedback), modulates the intensity of both NECC and ITCZ (Li and Fedorov 2022). When this feedback is too weak, the NECC cannot reach its full strength. In turn, a weak NECC could be a potential factor in the double-ITCZ bias (Zhao and Fedorov 2020). Schematics of the Pacific upper-ocean currents are shown in Fig. 1.
Schematics of the upper-ocean zonal currents in the tropical Pacific. Thin black arrows in the zonal direction represent zonal winds. Heavy black arrows in the meridional direction indicate meridional Ekman transport due to the zonal winds. Curl < 0 and “High” indicate negative wind stress curl and positive SSH anomalies. Curl > 0 and “Low” indicate positive wind stress curl and negative SSH anomalies. Red and blue arrows in the vertical direction show off-equatorial Ekman pumping (downwelling) and Ekman suction (upwelling) off the equator. Curves along the cross section represent isotherms. The NEC and SEC (both westward) and the NECC and EUC (both eastward) are also shown.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
Future changes in the NECC are critical for the tropical climate system: not only does the NECC affect local temperature and precipitation changes along its path but also influences the behavior of the ITCZ and more generally the interhemispheric energy transport. Furthermore, understanding the mechanisms of the NECC better would help improve GCMs simulations.
Observations in the past few decades have revealed a narrowing of the ITCZ core accompanied by increased precipitation (Lau and Wu 2007; Adler et al. 2017). However, future projections from CMIP6 suggest that the ITCZ may intensify, become wider, and shift close to the equator (Mamalakis et al. 2021; Studholme et al. 2022). These trends in the ITCZ can be attributed to various mechanisms, including the “wet-getting-wetter” (Chou and Neelin 2004) or “warm-getting-wetter” (Xie et al. 2010) paradigms, atmospheric circulation response to increases in water vapor (Held and Soden 2006; Voigt and Shaw 2015), and changes in the radiative forcing of clouds (Voigt et al. 2014).
The EUC constitutes another crucial element of the tropical ocean circulation (Bryden and Brady 1985; Kessler et al. 1998; Sloyan et al. 2003). It feeds equatorial upwelling, playing an important role in CO2 outgassing and maintaining high biological productivity along the equator. Observations show the shoaling of the EUC as it flows eastward along the equator, confined within 2°N and 2°S (Johnson et al. 2002). Its core lies just slightly below the equatorial thermocline (e.g., Brown and Fedorov 2010), and its peak velocity reaches 1.15 m s−1 in the central Pacific near 125°W. It subsequently decreases to less than 0.5 m s−1 at around 93°W, west of the Galápagos Archipelago. The observed rate of EUC shoaling, estimated by linear regression between 165°E and 95°W, is 1.36 m per degree longitude (Johnson et al. 2002; Karnauskas et al. 2020). We note however that previous studies have suggested that, despite its limited extent, the Galápagos Archipelago strides on the equator, interfering with ocean measurements. It can also cause model bias due to the complicated structure of local divergence and upwelling (Karnauskas et al. 2012; Johnson et al. 2002).
The EUC does not migrate seasonally like the NECC; however, its intensity follows an annual cycle—it tends to be stronger during the Northern Hemisphere summer and weaker in winter, in contrast to the changes of South Equatorial Current (SEC) which lies above. While the NECC and the EUC flow against the dominant winds, the SEC is forced directly by surface winds and flows westward in the upper 100 m of the ocean. The EUC is also closely linked to the Walker circulation and ENSO events. Studies show that the current disappeared during the strong 1997–98 and 1982–83 El Niño events due to the basinwide adjustment of sea surface slope caused by strong westerly wind anomalies in the western and central Pacific (e.g., Firing et al. 1983).
In fully coupled global climate models (GCMs), the ability of a particular model to replicate the EUC serves as a useful diagnostic of the model’s performance in the equatorial region. In coupled models, the EUC depends on the simulated wind stress and a wide range of physical processes and parameterizations including horizontal and vertical turbulent diffusion, entrainment, and horizontal eddies (Maes et al. 1997; Pedlosky 1988; Yu and Schopf 1997; Brown et al. 2007). Adequate simulation of the EUC and its relation to wind forcing in coupled models can be essential for simulating realistic ENSO dynamics (Hayashi et al. 2020). However, like the NECC, the EUC is also underestimated by GCMs, with its peak velocity reaching on average only 2/3rds of the observations (Karnauskas et al. 2020).
In this paper, we investigate the response of the tropical Pacific Ocean currents and the ITCZ to the anthropogenic warming using CMIP6 twenty-first century simulations. Specifically, we analyze how future changes in the ocean wind stress, sea level height, and thermal structure will affect the NECC, the EUC, and the SEC, within the SSP585 scenario (methods), and also compare the results to recently observed decadal trends.
2. Methods
a. Observations
We use the Ocean Surface Currents Analysis–Real Time (OSCAR) data across the broader equatorial Pacific between 1993 and 2022, which are based on satellite datasets, and use a simplified physical model of an upper ocean turbulent mixed layer including a geostrophic term, a wind-driven term and thermal wind adjustment (Lagerloef et al. 1999; Bonjean and Lagerloef 2002; Santiago-Mandujano and Firing 1990).
b. CMIP6 simulations
We use the CMIP6 dataset and compare the historical simulations and the Shared Socioeconomic Pathway 5-8.5 (SSP585 for brevity) scenarios, representing the present climate and a global warming scenario, respectively. The SSP585 experiment is an SSP-based representative concentration pathway (RCP) scenario with high radiative forcing, which is comparable to the RCP8.5 global forcing pathway of CMIP5. According to this scenario, it is expected that the global mean surface temperature relative to preindustrial levels will rise by approximately 4.7°–5.1°C with an average of 8.5 W m2 global warming delivered by the increasing carbon concentration by the end of the twenty-first century (O’Neill et al. 2016). For our comparison, annual mean variables are calculated from 1930 to 1980 for historical run and from 2050 to 2100 for SSP585 run. The two experiments are referred as “historical” and “SSP585,” respectively.
Each model has only one ensemble member with the variant id “r1i1p1f1” in the CMIP6 dataset. The four letters refer to realization, initialization, physics, and forcing. The numbers are indices for particular model configurations. A more comprehensive introduction of the CMIP6 Data Reference Syntax can be found at http://goo.gl/v1drZl. We compute annual mean values of relevant variables (e.g., precipitation rate) for each model. The standard deviation is estimated using 50 annual mean data points for each model.
There are 18 CMIP6 models available with historical and SSP585 experiments and all the variables needed. However, five of them simulate a too weak NECC with its maximum annual mean zonal velocity less than 0.1 m s−1. Therefore, only 13 best models are used to construct multimodel means, including ACCESS-ESM1-5, BCC-CSM2-MR, CAS-ESM2-0, CESM2-WACCM, CMCC-CM2-SR5, CMCC-ESM2, CanESM5, FGOALS-g3, FIO-ESM-2-0, MIROC6, MRI-ESM2-0, NorESM2-MM, and TaiESM1.
3. Results
As the first step, we examine the projected changes in ITCZ intensity, position, and structure, as it is critical for wind stress curl that drives the NECC. We also analyze overall changes in wind stress, important for all equatorial currents. Figures 2a and 2b show the mean precipitation and latitude of the ITCZ in the Pacific Ocean in the historical (blue) and SSP585 (orange) experiments for each model considered. All models simulate an increase in precipitation, and nearly all of them (except NorESM2-MM) indicate an equatorward shift of the ITCZ in the SSP585 scenario relative to the historical experiment. The mean position of the ITCZ and the SPCZ for the historical experiment is shown in Fig. 2c (white lines), which is based on the centroid of precipitation. A too strong precipitation band in the southeastern Pacific parallel to the ITCZ, related to the double-ITCZ model bias (Mechoso et al. 1995; Tian and Dong 2020; Li and Fedorov 2022), remains a pronounced feature of the CMIP6 models.
Projections for tropical precipitation. (a, b) Precipitation rate and latitude of the ITCZ in the Pacific Ocean in the historical (blue) and SSP585 (orange) experiments for 13 CMIP6 models. Dots and bars indicate average values and one standard deviation for each model/experiment, respectively. The Pacific ITCZ precipitation rate is defined as annual mean precipitation averaged between 0°–15°N and 130°E–90°W, and its latitude is defined as the centroid of the precipitation. (c) Multimodel mean precipitation map for the historical experiment and (d) the difference between the SSP585 and historical experiments. The multimodel mean centroids of the ITCZ and SPCZ in the historical experiments are shown as white lines in (c) and black lines in (d). The precipitation centroids in the SSP585 experiments are shown as red lines in (d). (e) Zonal mean profiles of multimodel mean precipitation averaged between 130°E and 90°W.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
With warming, we observe a general increase in precipitation rates in the entire equatorial band roughly between 8°S and 9°N (Fig. 2d). The strongest precipitation increase occurs over the western Pacific warm pool around the date line and over the eastern Pacific warm pool along 7°N, east of 130°W (also see Fig. 3a). Higher precipitation rates extend over zonal, roughly symmetric bands on both sides of the equator. The largest changes in precipitation are confined between the ITCZ and the SPCZ locations (defined via precipitation centroids), which implies that both precipitation bands shift toward the equator in the SSP585. To the north of the ITCZ, precipitation increases in the central and western Pacific but decreases in the eastern Pacific. To the south of the SPCZ, precipitation decreases in the central and eastern Pacific. The SPCZ shift toward the equator is greater since precipitation in the central southern Pacific decreases the most, presumably due to the suppressed warming and strengthening of southeasterly trade winds in this region in the SSP585 experiments (see Fig. 3a), which is related to the wind–evaporation–SST (WES) feedback (Xie and Philander 1994; Heede et al. 2021). Precipitation anomalies for each model are shown in Fig. S1 in the online supplemental material.
Projected changes in SST, SSH, precipitation and wind stress patterns in the tropical Pacific. (a) Anomalies in SST (colors), precipitation (contours), and wind stress (vectors) in the SSP585 relative to the historical experiment. An average SST increase of 2.49°C is subtracted to show the regional pattern. The blue, white, and red lines refer to positive, zero, and negative values of precipitation changes with an interval of 1 mm day−1. (b) Anomalies in SSH (colors) and wind stress (vectors).
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
These spatial changes result in several district changes in the meridional profile of the topical Pacific precipitation: the ITCZ intensifies, broadens in both directions, and undergoes a slight equatorward shift (Fig. 2e). The SPCZ strengthens, broadens, and shifts toward the equator as well, but this broadening occurs only on its equatorward flank. There is also a clear increase in precipitation rates on the equator. These results are generally consistent with those of Studholme et al. (2022), who used a different selection of models.
The double-ITCZ bias appears to become worse in the SSP585 experiments with an even stronger precipitation band extending toward the coast of Peru (Figs. 2d,e). Whether this strengthening of the southern precipitation band is a physical feature of global warming in the tropics or a consequence of the initial model bias remains unclear.
Many of the aforementioned changes in tropical precipitation are related to the development of the eastern equatorial warming pattern along the equator in the Pacific (i.e., El Niño–like mean conditions). This pattern implies a pronounced weakening of the easterly winds along the equator (Fig. 3a)—a signature of a weaker Walker circulation. Anomalies in SST, SSH, precipitation, and wind stress for each model are shown in Figs. S2 and S3.
Next, we discuss the projected changes in the NECC (Fig. 4). First, we show the multimodel mean surface zonal velocity field (Fig. 4c) with the black line indicating the core of NECC in the historical experiments. Four major currents, namely, the NEC, NECC, SEC, and South Equatorial Countercurrent (SECC) are clearly seen.
Projections for equatorial surface currents. (a), (b) Zonal surface velocity and latitude of the NECC in the historical (blue) and SSP585 (orange) experiments for 13 CMIP6 models. Dots and bars indicate average values and one standard deviation for each model/experiment. The NECC velocity is defined as the maximum zonal mean velocity averaged over 150°E–120°W and its latitude as the latitude at which the maximum velocity is observed. (c) Multimodel mean surface zonal velocity (colors) and wind stress (vectors) in the historical experiment, and (d) the difference between the SSP585 and historical experiments. The multimodel mean latitude of NECC in the historical experiments is shown as black lines in (c) and (d). The mean latitude of NECC in the SSP585 experiments is shown as the red line in (d). (e), (f) Zonal mean profiles of multimodel mean surface zonal velocity and wind stress curl averaged between 130°E and 90°W.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
Among the 13 models considered, 12 models (except NorESM2-MM) exhibit an equatorward shift of the NECC in the SSP585 scenario compared to historical experiment (Fig. 4b). The only model (NorESM2-MM) showing a poleward shift is the one that also simulates a poleward shift of the ITCZ, while all other models generate an equatorward shift of both ITCZ and NECC. The latitude of the NECC’s core is closely linked to the location of the ITCZ which marks a local minimum in zonal wind stress. The resulting wind stress curl creates divergence of the horizontal Ekman transport, leading to meridional variations in SSH which maintain the NECC. This explains the NECC shifts in the same direction as the ITCZ (see in Fig. 1). It is noteworthy that the NECC exhibits a very strong interannual variability due to ENSO, as reflected in Fig. 4b. Consequently, the equatorward shifts of the NECC do not seem to be of great statistical significance; however, the consistency among the models supports the robustness of our conclusion.
Among the same 13 models, 11 models simulate a weaker NECC while 2 models (NorESM2-MM and TaiESM1) generate a slightly stronger NECC under global warming (Fig. 4a). Why is the majority of models generating a weaker NECC while the ITCZ becomes more intense in these models? We discuss this next.
Surface maps of anomalies in zonal velocity in the tropical band are shown in Fig. 4d. The NEC and SEC are weakened due to zonal wind stress reduction caused by a weaker zonal SST gradient along the equator and hence a weaker Walker circulation. Additionally, the NECC shifts toward the equator following an increase in zonal velocity on its southern flank but a decrease on its northern flank. The shift of the NECC is greater in the eastern Pacific than in the central western Pacific, which mirrors ITCZ changes. Figures 4e and 4f show the meridional profile of surface zonal velocity and wind stress curl averaged between 130°E and 90°W. The general reduction of velocity and the equatorward shift of the NECC are evident in Fig. 4e, and the weakening of the NECC favorable wind stress curl is evident in Fig. 4f. The shift of the NECC toward the equator also implies smaller values of the Coriolis parameter within the current and therefore a stronger convergence of Ekman transport, which would result in a stronger NECC. However, this effect appears to be weaker than the effect of wind stress changes. Zonal velocity anomalies for each model are shown in Fig. S4.
In contrast to these future projections, surface current data from OSCAR suggest an intensification of the NECC and SEC in the past several decades. The mean state of the ocean and climate trends from 1993 to 2022 is plotted in Fig. 5. This recent strengthening of the surface equatorial currents is consistent with the observed multidecadal strengthening of the Walker circulation and the cooling of the eastern equatorial Pacific since the 1990s, whose mechanisms are intensely debated (e.g., Heede and Fedorov 2023 and references therein). Nevertheless, the broad expectation is that the recent cooling of the eastern equatorial Pacific will eventually be replaced by an enhanced eastern equatorial warming pattern corresponding to a weaker Walker circulation (Heede and Fedorov 2022), leading to the changes in equatorial currents suggested by our study.
OSCAR datasets. (a), (b) Mean state and trends of zonal surface velocity for the time span from 1993 to 2022. The trend is calculated using a linear regression at each grid point. Note the strengthening of the NECC and the SEC, which is consistent with the recent decadal strengthening of the Walker circulation (e.g., Heede and Fedorov 2023) but is opposite to the future projections.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
Next, we focus on changes in the EUC. The velocity and depth of EUC are displayed in Figs. 6a and 6b. The EUC characteristic zonal velocity is defined as the maximum velocity at 140°W, and current’s depth is taken as the depth where the maximum velocity occurs. Although the EUC velocity and depth vary along the equator, they show generally similar changes at different longitudes. All 13 models simulate a stronger and shoaling EUC in the SSP585 experiment relative to the historical run, and the difference is statistically significant as shown by the error bars.
Projections for the EUC. (a), (b) Characteristic zonal velocity and depth of the EUC at 140°W in the historical (blue) and SSP585 (orange) experiments for 13 CMIP6 models. Dots and bars indicate average values and one standard deviation for each experiment. The EUC velocity is defined as the maximum velocity at 140°W and its core’s depth as the depth where the maximum velocity is observed. (c) The vertical structure of the multimodel mean zonal velocity at 140°W in the historical experiment as a function of latitude and depth and (d) the difference between the SSP585 and the historical experiments. (e) Multimodel mean zonal velocity averaged between 2°N and 2°S along the equator as a function of longitude and depth in the historical experiment and (f) the difference between the SSP585 and the historical experiments. The depth of the EUC core in the historical and SSP585 experiments is shown as black and red lines, respectively. (g) The vertical profiles of the zonal velocity on the equator at 140°W in the two experiments.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
The depth–latitude structure of the multimodel mean zonal velocity and its anomalies are plotted in Figs. 6c and 6d. Although all four surface equatorial currents weaken, the velocity within the EUC increases at the top and decreases at the bottom, indicating a shoaling of the core of the EUC. This is confirmed by the multimodel mean zonal velocity field along the equator (averaged between 2°N and 2°S) (see Figs. 6e and 6f). Positive values in Fig. 6e mark the EUC, which moves to shallower depths as it flows to the east. The core of EUC moves from ∼200-m depth at 160°E to ∼100-m depth at 120°W, flowing just slightly below the equatorial thermocline. The anomalies of zonal velocity in Fig. 6f have a strong dipole pattern with eastward anomalies at the top and westward anomalies at the bottom. The red line is the core of EUC in the SSP585 experiments and is around 20–30 m above the initial depth. This is because ocean surface warms more than the subsurface ocean, and then a stronger vertical stratification is formed due to density changes. We also find a greater shoaling in the west than in the east, indicating a smaller east–west incline of the EUC. The vertical profile of the zonal velocity at 140°W in Fig. 6g also indicates a strengthened and raised EUC as well as a weakened SEC above it. The vertical structure of zonal velocity anomalies along the equator for each model is shown in Fig. S5.
Further, we find that the net increase in horizontal water convergence into the eastern equatorial Pacific due to the stronger EUC and weaker SEC is balanced by a greater net outgoing transport by the subtropical cells (Figs. S6 and S7, also see Fig. 3 in Brown and Fedorov 2010). Regarding the strength of the subtropical cells, there is no general agreement across climate models on future projections for the Pacific subtropical cells total mass transport (Capotondi and Qiu 2023; McPhaden and Zhang 2002; Wang and Cane 2011; Heede et al. 2021; Graffino et al. 2021; Han et al. 2024). The relatively robust result is the weakening of the northern cell, but the strengthening of the southern cell (due to the strengthening of extratropical winds in the Southern Hemisphere). As a result, the projected changes in the total STC mass transport by the century end for realistic climate warming scenarios are generally small.
The main physical driver of the EUC is the eastward pressure gradient force associated with the west–east SSH gradient along the equator. The higher SSH in the west is maintained by easterly trade winds. However, the intensity of the EUC becomes stronger in the SSP585 scenario, even if the zonal SSH gradient becomes weaker (Fig. 3b). The zonal profiles of multimodel mean SSH and wind stress on the equator are shown in Figs. 7a and 7b. The black, red, and blue curves refer to the historical scenario, SSP585 scenario, and their difference, respectively. The zonal SSH gradient is reduced in SSP585 due to a weaker easterly wind stress. This calls for an investigation into what causes the strengthening of the EUC, as described below.
Changes in SSH and wind stress along the equator. (a), (b) Multimodel mean SSH and zonal wind stress on the equator in the historical (black) and SSP585 (red) scenarios and the difference (blue) between the two. SSH zonal mean value for the historical simulation is removed from both experiments. (c) Changes in west–east SSH gradient (blue) in the SSP855 relative to the historical experiment. These changes are decomposed into steric components due to local temperature (orange) and salinity (green) changes as well as a small residual (red). The SSH gradient is defined as SSH averaged over 2°N–2°S and 160°E–180° minus SSH averaged over 2°N–2°S and 120°–100°W. A standard equation of state for seawater density is used (McDougall and Barker 2011).
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
First, our analysis indicates that in most models the SSH gradient reduction is largely controlled by changes in temperature compensated by changes in salinity (Fig. 7c). The eastern Pacific warms more leading to a lighter seawater density and a higher SSH there shown in Figs. 8a and 8b, decreasing the west–east SST gradient. On the other hand, precipitation increase more in the west, decreasing surface salinity there more, and partially compensating the effect of temperature shown in Figs. 8c and 8d. In fact, a greater SSH gradient in SSP585 developing in a couple of models is due to the overcompensation from surface salinity.
The vertical structure of temperature and salinity in the simulations. (a), (b) Multimodel mean vertical temperature structure along the equator averaged between 2°N and 2°S in the historical experiment and the corresponding changes in the SSP585 experiment. The multimodel mean core of the EUC in the historical experiment is shown as the black line. The core is positioned just slightly below the thermocline. (c), (d) As in (a) and (b), but for salinity.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
Further, considering zonal momentum balance along the equator, a stronger zonal pressure gradient force and/or weaker friction could be the two major factors leading to a faster EUC. Figure 9c shows a local decrease in the multimodel mean zonal pressure gradient along the equator, which is caused by the reduction in the west–east SSH gradient as we just discussed. However, although the pressure gradient decreases almost everywhere in the upper ocean, the core of the EUC shoals, moving up toward higher values of the pressure gradient (Figs. 9a,b). Therefore, the net effect of the shoaling is to increase slightly the pressure gradient force acting on the EUC (Fig. 9d). In fact, the zonal pressure gradient along the core of the EUC increases in the western and central Pacific but decreases a little in the east. Zonal pressure gradient profiles for each model are shown in Fig. S8.
Changes in the zonal pressure gradient. (a), (b) Multimodel mean zonal pressure gradient averaged over 2°N–2°S for the historical and SSP585 scenarios and (c) the difference between the two. Black and red lines denote the core of the EUC in the historical and SSP585 scenarios, respectively. (d) Longitudinal profiles of the zonal pressure gradient along the core of the EUC in the historical (black) and SSP585 (red) experiments. The zonal pressure gradient is defined as dp/dx, where p is the pressure and x is the distance along the equator.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
Next, we investigate changes in vertical eddy viscosity. Since vertical viscosity is not a standard output in most of the CMIP6 models, here we only present a detailed analysis based on the IPSL-CM6A-LR model, using it as an example. It is important however that this model’s zonal velocity changes and vertical profiles are similar to the multimodel mean across CMIP6. Anomalies in vertical viscosity for SSP585 show alternating negative and positive signals as depth increases from the ocean surface (Fig. 10a). Critically, viscosity along the EUC path decreases so that the vertical friction term in the momentum equation is reduced. Figure 10b shows consistently smaller viscosity values in the SSP585 experiment than that in historical experiment at almost all longitudes.
Changes in vertical eddy viscosity, buoyancy frequency, and vertical shear. (a) Differences in vertical eddy viscosity averaged over 2°N–2°S along the equator between the SSP585 and historical scenarios in the IPSL-CM6A-LR model (SSP minus historical). The cores of the EUC in the two experiments are shown as the red and black lines, respectively. (b) Longitudinal profiles of vertical eddy viscosity along the core of the EUC in the historical (black) and SSP585 (red) experiments. (c), (d) Vertical profiles of buoyancy frequency and zonal velocity shear in the historical (black) and SSP585 (red) experiments and their difference (blue) at 140°W.
Citation: Journal of Climate 37, 24; 10.1175/JCLI-D-23-0738.1
The NEMO ocean component in IPSL-CM6A-LR implements a turbulent kinetic energy scheme to calculate the vertical mixing of tracers and momentum (Blanke and Delecluse 1993; Gaspar et al. 1990). Although this makes it difficult to compute eddy viscosity offline, it is still largely controlled by local Richardson numbers and higher Richardson numbers imply higher eddy viscosity. As the Richardson number is given by the squared ratio of the buoyancy frequency to vertical shear, next we will analyze the vertical profiles of these two variables to further understand the reduction of eddy viscosity.
Similar to the pressure gradient analysis, the viscosity change along the core of the EUC represents a combined effect of the current’s shoaling and local change. The EUC flows at depths with the largest vertical temperature gradient and hence largest buoyancy frequency. The center of the EUC rises from 135-m depth in the historical experiment to around 110-m depth in the SSP585 experiment. The buoyancy frequency increases mainly due to a stronger stratification shown in Fig. 10b. Meanwhile, the intensity of the vertical shear does increase locally, but these changes are negligible when considering shoaling effect as depicted in Fig. 10c. The strong negative signal near the surface results from a weaker wind stress, which might then trigger a subsequence of redistribution of density, zonal velocity, viscosity, and other variables simultaneously, leading to a stronger and shallower EUC in the SSP585 scenario. Using a simpler Richardson number–dependent mixing scheme following the approach of Pacanowski and Philander (1981) suggests qualitatively similar changes in eddy viscosity (not shown).
4. Discussion
Equatorial ocean currents, tightly linked to the atmospheric Walker circulation, as well as the ITCZ and SPCZ position and intensity, play important roles in the tropical ocean dynamics and climate. Projections of future changes in these currents due to anthropogenic global warming are of great interest because of their potential impacts on climate and ecosystems. For example, the NECC transports warm water to the colder eastern Pacific, influencing the ITCZ and hence interhemispheric energy transport, while the EUC plays an important role in equatorial upwelling and ENSO. In this paper, we have compared these projections, based on the SSP585 scenario, against the historical simulations in CMIP6. We have also compared the future projection to the recently observed decadal trends.
Specifically, SSP585 simulations suggest that both the ITCZ and SPCZ will shift toward the equator and become more intense, with more precipitation along the equator especially within the western Pacific warm pool. The Walker circulation is expected to weaken due to enhanced warming over the cold tongue in the eastern Pacific. The resulting wind stress reduction will lead to slower equatorial surface currents including the NEC and SEC. Even with a more intense ITCZ, one could also expect a slower NECC due to a reduction in local wind stress curl. Additionally, the NECC will move closer to the equator following the ITCZ.
The EUC is expected to shoal due to stronger vertical stratification induced by ocean surface warming. However, its zonal velocity will increase even as the west–east SSH gradient and the corresponding zonal pressure gradient become weaker due to the weakening of the Walker circulation. As the EUC shoals, it will experience a greater pressure gradient due to the east–west SSH contrast, which will compensate the general weakening of the pressure gradient. A further analysis of available data from one of the CMIP6 models shows that the vertical viscosity is reduced in the core of the EUC, which should lead to a higher velocity of this current. In turn, the reduction of viscosity is consistent with higher Richardson numbers due to changes in vertical shear and stratification.
The future strengthening and shoaling of the EUC have been noted in previous studies, especially in the context of EUC water transport changes. For example, Luo et al. (2009) showed that water transport from the subtropics to the tropics could increase via western boundary pathways but decrease via interior pathways, due to a westward expansion of the high potential vorticity zone. They pointed to a stronger New Guinea Coastal Undercurrent (NGCU) as a potential cause of a stronger EUC. Sen Gupta et al. (2012) then showed that the projected intensification of the south-easterly trade winds and the corresponding off-equatorial wind stress curl change in the Southern Hemisphere could lead to a stronger NGCU and hence greater water input into the EUC. Stellema et al. (2022) further investigated the source waters for the EUC, focusing on the combined impacts of the Mindanao Current (MC), NGCU, and local wind stress. However, they also acknowledged the impact of local buoyancy forcing. Peng et al. (2022) forced the MIT general circulation model (MITgcm) with CMIP6 ensemble mean changes in wind stress, SST, and salinity. They showed that in the absence of surface wind stress changes, surface warming caused the EUC to shoal and explained their result with enhanced vertical stratification, which is consistent with our study.
However, none of the aforementioned studies have explained the future anomalies of the EUC via local momentum balance. Nor did they discuss the reduction of vertical friction as the potential mechanism of the EUC strengthening. This explains our focus on the maximum zonal velocity changes of the EUC, rather than on water volume transport used in previous studies. Another related finding of our study is that, unlike anomalies in the transport, the maximum velocity increases in the SSP585 experiments at all longitudes (Fig. S9).
Uncertainties in CMIP6 future projections should be taken into account when interpreting these results. In contrast to the future projections, observational and reanalysis datasets indicate intensifying NECC and SEC in the past several decades, likely resulting from the pronounced strengthening of the Walker circulation. These recent atmospheric decadal trends have been attracting a lot of attention in the context of how the tropical Pacific as a whole is responding to global warming (Heede and Fedorov 2023 and references therein).
Moreover, in all models considered, the simulated NECC is still too weak, suggesting potential deficiencies in the models’ underlying physical mechanisms. The model atmospheric horizontal resolution is relatively coarse, on the order of 1°, which can also affect future projections. Only one model within CMIP6 (IPSL) provided the viscosity data, necessary for the analysis of the EUC changes. Ultimately, more detailed studies using high-resolution coupled models may be needed to resolve some of these issues.
Acknowledgments.
This study has been supported by NASA (80NSSC21K0558 and 80NSSC20K1634) and NOAA (NA20OAR4310377). Additional support is provided by the ARCHANGE Project (ANR-18-MPGA-0001, France).
Data availability statement.
The Ocean Surface Current Analysis–Real Time (OSCAR) dataset used in this study is openly available from the NASA Physical Oceanography Distributed Active Archive Center at https://doi.org/10.5067/OSCAR-10D01 as cited in Bonjean and Lagerloef (2002). The CMIP6 dataset is also available from World Climate Research Programme at https://esgf-node.llnl.gov/projects/cmip6/ as cited in Eyring et al. (2016) and Balaji et al. (2018).
REFERENCES
Adler, R. F., G. Gu, M. Sapiano, J.-J. Wang, and G. J. Huffman, 2017: Global precipitation: Means, variations and trends during the satellite era (1979–2014). Surv. Geophys., 38, 679–699, https://doi.org/10.1007/s10712-017-9416-4.
Balaji, V., and Coauthors, 2018: Requirements for a global data infrastructure in support of CMIP6. Geosci. Model Dev., 11, 3659–3680, https://doi.org/10.5194/gmd-11-3659-2018.
Blanke, B., and P. Delecluse, 1993: Variability of the tropical Atlantic Ocean simulated by a general circulation model with two different mixed-layer physics. J. Phys. Oceanogr., 23, 1363–1388, https://doi.org/10.1175/1520-0485(1993)023<1363:VOTTAO>2.0.CO;2.
Bonjean, F., and G. S. E. Lagerloef, 2002: Diagnostic model and analysis of the surface currents in the tropical Pacific Ocean. J. Phys. Oceanogr., 32, 2938–2954, https://doi.org/10.1175/1520-0485(2002)032<2938:DMAAOT>2.0.CO;2.
Brown, J. N., and A. V. Fedorov, 2008: Mean energy balance in the tropical Pacific Ocean. J. Mar. Res., 66 (1), 1–23, https://doi.org/10.1357/002224008784815757.
Brown, J. N., and A. V. Fedorov, 2010: Estimating the diapycnal transport contribution to warm water volume variations in the tropical Pacific Ocean. J. Climate, 23, 221–237, https://doi.org/10.1175/2009JCLI2347.1.
Brown, J. N., J. S. Godfrey, and R. Fiedler, 2007: A zonal momentum balance on density layers for the central and eastern equatorial Pacific. J. Phys. Oceanogr., 37, 1939–1955, https://doi.org/10.1175/JPO3090.1.
Bryden, H. L., and E. C. Brady, 1985: Diagnostic model of the three-dimensional circulation in the upper equatorial Pacific Ocean. J. Phys. Oceanogr., 15, 1255–1273, https://doi.org/10.1175/1520-0485(1985)015<1255:DMOTTD>2.0.CO;2.
Capotondi, A., and B. Qiu, 2023: Decadal variability of the Pacific shallow overturning circulation and the role of local wind forcing. J. Climate, 36, 1001–1015, https://doi.org/10.1175/JCLI-D-22-0408.1.
Chou, C., and J. D. Neelin, 2004: Mechanisms of global warming impacts on regional tropical precipitation. J. Climate, 17, 2688–2701, https://doi.org/10.1175/1520-0442(2004)017<2688:MOGWIO>2.0.CO;2.
Collins, M., and Coauthors, 2010: The impact of global warming on the tropical Pacific Ocean and El Niño. Nat. Geosci., 3, 391–397, https://doi.org/10.1038/ngeo868.
Dai, A., 2013: Increasing drought under global warming in observations and models. Nat. Climate Change, 3, 52–58, https://doi.org/10.1038/nclimate1633.
Danabasoglu, G., and Coauthors, 2014: North Atlantic simulations in Coordinated Ocean-ice Reference Experiments phase II (CORE-II). Part I: Mean states. Ocean Modell., 73, 76–107, https://doi.org/10.1016/j.ocemod.2013.10.005.
Donguy, J. R., and G. Meyers, 1996: Mean annual variation of transport of major currents in the tropical Pacific Ocean. Deep-Sea Res. I, 43, 1105–1122, https://doi.org/10.1016/0967-0637(96)00047-7.
Eyring, V., S. Bony, G. A. Meehl, C. A. Senior, B. Stevens, R. J. Stouffer, and K. E. Taylor, 2016: Overview of the Coupled Model Intercomparison Project phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev., 9, 1937–1958, https://doi.org/10.5194/gmd-9-1937-2016.
Firing, E., R. Lukas, J. Sadler, and K. Wyrtki, 1983: Equatorial undercurrent disappears during 1982–1983 El Niño. Science, 222, 1121–1123, https://doi.org/10.1126/science.222.4628.1121.
Gaspar, P., Y. Grégoris, and J.-M. Lefevre, 1990: A simple eddy kinetic energy model for simulations of the oceanic vertical mixing: Tests at station Papa and long‐term upper ocean study site. J. Geophys. Res., 95, 16 179–16 193, https://doi.org/10.1029/JC095iC09p16179.
Graffino, G., R. Farneti, and F. Kucharski, 2021: Low-frequency variability of the Pacific Subtropical Cells as reproduced by coupled models and ocean reanalyses. Climate Dyn., 56, 3231–3254, https://doi.org/10.1007/s00382-021-05639-6.
Grima, N., A. Bentamy, K. Katsaros, Y. Quilfen, P. Delecluse, and C. Levy, 1999: Sensitivity of an oceanic general circulation model forced by satellite wind stress fields. J. Geophys. Res., 104, 7967–7989, https://doi.org/10.1029/1999JC900007.
Han, X., J. Feng, Y. Lu, and D. Hu, 2024: Variability of the Pacific subtropical cells under global warming in CMIP6 models. J. Oceanol. Limnol., 42, 24–40, https://doi.org/10.1007/s00343-023-2315-2.
Hayashi, M., F.-F. Jin, and M. F. Stuecker, 2020: Dynamics for El Niño-La Niña asymmetry constrain equatorial-Pacific warming pattern. Nat. Commun., 11, 4230, https://doi.org/10.1038/s41467-020-17983-y.
Heede, U. K., and A. V. Fedorov, 2022: Colder eastern equatorial Pacific and stronger Walker cell in the early 21st century: Isolating the forced response to global warming. ESS Open Archive, 10512304.1, https://doi.org/10.1002/essoar.10512304.1.
Heede, U. K., and A. V. Fedorov, 2023: Colder eastern equatorial Pacific and stronger Walker circulation in the early 21st century: Separating the forced response to global warming from natural variability. Geophys. Res. Lett., 50, e2022GL101020, https://doi.org/10.1029/2022GL101020.
Heede, U. K., A. V. Fedorov, and N. J. Burls, 2021: A stronger versus weaker Walker: Understanding model differences in fast and slow tropical Pacific responses to global warming. Climate Dyn., 57, 2505–2522, https://doi.org/10.1007/s00382-021-05818-5.
Held, I. M., and B. J. Soden, 2006: Robust responses of the hydrological cycle to global warming. J. Climate, 19, 5686–5699, https://doi.org/10.1175/JCLI3990.1.
Hu, S., J. Sprintall, C. Guan, M. J. McPhaden, F. Wang, D. Hu, and W. Cai, 2020: Deep-reaching acceleration of global mean ocean circulation over the past two decades. Sci. Adv., 6, eaax7727, https://doi.org/10.1126/sciadv.aax7727.
Johnson, G. C., B. M. Sloyan, W. S. Kessler, and K. E. McTaggart, 2002: Direct measurements of upper ocean currents and water properties across the tropical Pacific during the 1990s. Prog. Oceanogr., 52, 31–61, https://doi.org/10.1016/S0079-6611(02)00021-6.
Karnauskas, K. B., G. C. Johnson, and R. Murtugudde, 2012: An equatorial ocean bottleneck in global climate models. J. Climate, 25, 343–349, https://doi.org/10.1175/JCLI-D-11-00059.1.
Karnauskas, K. B., J. Jakoboski, T. M. S. Johnston, W. B. Owens, D. L. Rudnick, and R. E. Todd, 2020: The Pacific Equatorial Undercurrent in three generations of global climate models and glider observations. J. Geophys. Res. Oceans, 125, e2020JC016609, https://doi.org/10.1029/2020JC016609.
Kessler, W. S., 2006: The circulation of the eastern tropical Pacific: A review. Prog. Oceanogr., 69, 181–217, https://doi.org/10.1016/j.pocean.2006.03.009.
Kessler, W. S., L. M. Rothstein, and D. Chen, 1998: The annual cycle of SST in the eastern tropical Pacific, diagnosed in an ocean GCM. J. Climate, 11, 777–799, https://doi.org/10.1175/1520-0442(1998)011<0777:TACOSI>2.0.CO;2.
Lagerloef, G. S. E., G. T. Mitchum, R. B. Lukas, and P. P. Niiler, 1999: Tropical Pacific near‐surface currents estimated from altimeter, wind, and drifter data. J. Geophys. Res., 104, 23 313–23 326, https://doi.org/10.1029/1999JC900197.
Lau, K.-M., and H.-T. Wu, 2007: Detecting trends in tropical rainfall characteristics, 1979–2003. Int. J. Climatol., 27, 979–988, https://doi.org/10.1002/joc.1454.
Li, Z., and A. V. Fedorov, 2022: Coupled dynamics of the North Equatorial Countercurrent and Intertropical Convergence Zone with relevance to the double-ITCZ problem. Proc. Natl. Acad. Sci. USA, 119, e2120309119, https://doi.org/10.1073/pnas.2120309119.
Li, Z., Y. Luo, N. Arnold, and E. Tziperman, 2019: Reductions in strong upwelling‐favorable wind events in the Pliocene. Paleoceanogr. Paleoclimatol., 34, 1931–1944, https://doi.org/10.1029/2019PA003760.
Luo, Y., L. M. Rothstein, and R.-H. Zhang, 2009: Response of Pacific subtropical‐tropical thermocline water pathways and transports to global warming. Geophys. Res. Lett., 36, L04601, https://doi.org/10.1029/2008GL036705.
Maes, C., G. Madec, and P. Delecluse, 1997: Sensitivity of an equatorial Pacific OGCM to the lateral diffusion. Mon. Wea. Rev., 125, 958–971, https://doi.org/10.1175/1520-0493(1997)125<0958:SOAEPO>2.0.CO;2.
Mamalakis, A., and Coauthors, 2021: Zonally contrasting shifts of the tropical rain belt in response to climate change. Nat. Climate Change, 11, 143–151, https://doi.org/10.1038/s41558-020-00963-x.
Masunaga, H., and T. S. L’Ecuyer, 2011: Equatorial asymmetry of the east Pacific ITCZ: Observational constraints on the underlying processes. J. Climate, 24, 1784–1800, https://doi.org/10.1175/2010JCLI3854.1.
McDougall, T. J., and P. M. Barker, 2011: Getting started with TEOS-10 and the Gibbs Seawater (GSW) oceanographic toolbox. SCOR/IAPSO WG127, 28 pp., https://www.teos-10.org/pubs/Getting_Started.pdf.
McPhaden, M. J., and D. Zhang, 2002: Slowdown of the meridional overturning circulation in the upper Pacific Ocean. Nature, 415, 603–608, https://doi.org/10.1038/415603a.
Mechoso, C. R., and Coauthors, 1995: The seasonal cycle over the tropical Pacific in coupled ocean–atmosphere general circulation models. Mon. Wea. Rev., 123, 2825–2838, https://doi.org/10.1175/1520-0493(1995)123<2825:TSCOTT>2.0.CO;2.
O’Neill, B. C., and Coauthors, 2016: The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6. Geosci. Model Dev., 9, 3461–3482, https://doi.org/10.5194/gmd-9-3461-2016.
Pacanowski, R. C., and S. G. H. Philander, 1981: Parameterization of vertical mixing in numerical models of tropical oceans. J. Phys. Oceanogr., 11, 1443–1451, https://doi.org/10.1175/1520-0485(1981)011<1443:POVMIN>2.0.CO;2.
Pedlosky, J., 1988: Entrainment and the termination of the equatorial undercurrent. J. Phys. Oceanogr., 18, 880–886, https://doi.org/10.1175/1520-0485(1988)018<0880:EATTOT>2.0.CO;2.
Peng, Q., S.-P. Xie, D. Wang, R. X. Huang, G. Chen, Y. Shu, J.-R. Shi, and W. Liu, 2022: Surface warming–induced global acceleration of upper ocean currents. Sci. Adv., 8, eabj8394, https://doi.org/10.1126/sciadv.abj8394.
Philander, S. G. H., W. J. Hurlin, and A. D. Seigel, 1987: Simulation of the seasonal cycle of the tropical Pacific Ocean. J. Phys. Oceanogr., 17, 1986–2002, https://doi.org/10.1175/1520-0485(1987)017<1986:SOTSCO>2.0.CO;2.
Reverdin, G., C. Frankignoul, E. Kestenare, and M. J. McPhaden, 1994: Seasonal variability in the surface currents of the equatorial Pacific. J. Geophys. Res., 99, 20 323–20 344, https://doi.org/10.1029/94JC01477.
Santiago-Mandujano, F., and E. Firing, 1990: Mixed-layer shear generated by wind stress in the central equatorial Pacific. J. Phys. Oceanogr., 20, 1576–1582, https://doi.org/10.1175/1520-0485(1990)020<1576:MLSGBW>2.0.CO;2.
Sen Gupta, A., A. Ganachaud, S. McGregor, J. N. Brown, and L. Muir, 2012: Drivers of the projected changes to the Pacific Ocean equatorial circulation. Geophys. Res. Lett., 39, L09605, https://doi.org/10.1029/2012GL051447.
Sloyan, B. M., G. C. Johnson, and W. S. Kessler, 2003: The Pacific cold tongue: A pathway for interhemispheric exchange. J. Phys. Oceanogr., 33, 1027–1043, https://doi.org/10.1175/1520-0485(2003)033<1027:TPCTAP>2.0.CO;2.
Stellema, A., A. Sen Gupta, A. S. Taschetto, and M. Feng, 2022: Pacific Equatorial Undercurrent: Mean state, sources, and future changes across models. Front. Climate, 4, 933091, https://doi.org/10.3389/fclim.2022.933091.
Studholme, J., A. V. Fedorov, S. K. Gulev, K. Emanuel, and K. Hodges, 2022: Poleward expansion of tropical cyclone latitudes in warming climates. Nat. Geosci., 15, 14–28, https://doi.org/10.1038/s41561-021-00859-1.
Sverdrup, H. U., 1947: Wind-driven currents in a baroclinic ocean; with application to the equatorial currents of the eastern Pacific. Proc. Natl. Acad. Sci. USA, 33, 318–326, https://doi.org/10.1073/pnas.33.11.318.
Tian, B., and X. Dong, 2020: The double‐ITCZ bias in CMIP3, CMIP5, and CMIP6 models based on annual mean precipitation. Geophys. Res. Lett., 47, e2020GL087232, https://doi.org/10.1029/2020GL087232.
Tseng, Y. H., and Coauthors, 2016: North and equatorial Pacific Ocean circulation in the CORE-II hindcast simulations. Ocean Modell., 104, 143–170, https://doi.org/10.1016/j.ocemod.2016.06.003.
Voigt, A., and T. A. Shaw, 2015: Circulation response to warming shaped by radiative changes of clouds and water vapor. Nat. Geosci., 8, 102–106, https://doi.org/10.1038/ngeo2345.
Voigt, A., S. Bony, J. L. Dufresne, and B. Stevens, 2014: The radiative impact of clouds on the shift of the intertropical convergence zone. Geophys. Res. Lett., 41, 4308–4315, https://doi.org/10.1002/2014GL060354.
Wang, D., and M. A. Cane, 2011: Pacific shallow meridional overturning circulation in a warming climate. J. Climate, 24, 6424–6439, https://doi.org/10.1175/2011JCLI4100.1.
Wyrtki, K., and R. Kendall, 1967: Transports of the Pacific equatorial countercurrent. J. Geophys. Res., 72, 2073–2076, https://doi.org/10.1029/JZ072i008p02073.
Xie, S. P., and S. G. H. Philander, 1994: A coupled ocean‐atmosphere model of relevance to the ITCZ in the eastern Pacific. Tellus, 46A, 340–350, https://doi.org/10.3402/tellusa.v46i4.15484.
Xie, S. P., C. Deser, G. A. Vecchi, J. Ma, H. Teng, and A. T. Wittenberg, 2010: Global warming pattern formation: Sea surface temperature and rainfall. J. Climate, 23, 966–986, https://doi.org/10.1175/2009JCLI3329.1.
Yu, Z., and P. S. Schopf, 1997: Vertical eddy mixing in the tropical upper ocean: Its influence on zonal currents. J. Phys. Oceanogr., 27, 1447–1458, https://doi.org/10.1175/1520-0485(1997)027<1447:VEMITT>2.0.CO;2.
Yu, Z., J. P. McCreary Jr., W. S. Kessler, and K. A. Kelly, 2000: Influence of equatorial dynamics on the Pacific North Equatorial Countercurrent. J. Phys. Oceanogr., 30, 3179–3190, https://doi.org/10.1175/1520-0485(2000)030<3179:IOEDOT>2.0.CO;2.
Zhao, B., and A. Fedorov, 2020: The seesaw response of the intertropical and South Pacific convergence zones to hemispherically asymmetric thermal forcing. Climate Dyn., 54, 1639–1653, https://doi.org/10.1007/s00382-019-05076-6.
