1. Introduction
ENSO is the most prominent mode of climate variability in the tropical Pacific, with severe global ecological, social, and economic consequences (McPhaden et al. 2006). The theoretical explanations of ENSO can be roughly grouped into two frameworks. One suggests that “El Niño is one phase of a self-sustained, unstable, and naturally oscillatory mode of the coupled ocean-atmosphere system” (Wang et al. 2017) while the other regards “El Niño as a stable (or damped) mode triggered by or interacted with stochastic forcing or noise such as westerly wind bursts and Madden–Julian oscillation events” (Gebbie et al. 2007) and “the tropical instability waves in the eastern Pacific Ocean” (An 2008).
Bjerknes (1969) first proposed interactions between the atmosphere and the equatorial eastern Pacific Ocean, emphasizing its possible role in inducing El Niño events. In Bjerknes’ view, the east–west SST gradient is inevitably reduced through an initial positive SST anomaly in the central or eastern equatorial Pacific, resulting in weaker easterly trade winds around the equator. The weakened easterly trade winds subsequently evoke changes of the equatorial currents not only in direction but also in velocity that further facilitate the SST anomaly, establishing a positive feedback. Clearly, Bjerknes’ theory regards the positive SST anomalies as the upper stream of the cause chain of El Niño, but why and how the SSTs in the eastern equatorial Pacific change positively remains unresolved. Although the other view of ENSO can explain the diversity and unpredictability of El Niño (e.g., Landsea and Knaff 2000; Philander and Fedorov 2003) due to a consideration of external atmospheric forcing [e.g., the Madden–Julian oscillation and westerly wind bursts (Gebbie et al. 2007)] and the oceanic noise [e.g., the tropical instability waves (An 2008)], how these sporadic (5–30 days) and weak westerly winds (2 m s−1) induce the warm pool climatologically in the western equatorial Pacific to migrate along the equator to the central or eastern equatorial Pacific still remains to be discussed (McPhaden 1999).
Despite those tremendous endeavors in the theoretical model simulations and in the field research over the past decades, a widely accepted theory for its genesis and maintenance mechanisms is still vigorously debated (Thual et al. 2016). The apparent absence of a super El Niño in 2014 that was expected by many models implies that we may still not understand some fundamental aspects of the system. Presently the studies on El Niño mainly focus on the westerly wind bursts, which are considered as a potential driver by some authors because the simulated results forced with the westerly wind bursts match the observations very well (Thual et al. 2016; Lengaigne et al. 2004; Eisenman et al. 2005; Fedorov et al. 2014; Menkes et al. 2014; Levine and McPhaden 2016; Hu and Fedorov 2017; Gebbie et al. 2007). However, their simulations can only prove the synchronicity of the westerly winds and SST anomalies in the western equatorial Pacific, but cannot attest that the SST anomalies are directly and solely caused by the westerly winds because we can do an opposite model simulation showing that the SST anomalies can evoke westerly winds in the western equatorial Pacific due to a reverse of the Walker Circulation. Furthermore, Fig. 1 shows that the buoys deployed along the equatorial Pacific by NOAA since 1992 recorded wind intensity that shows no dramatic changes with direction change only during the last two major warm events (1997, 2015/16 El Niño) from prevailing easterly to anomalous westerly. Hence, the role of the westerly winds in reversing the prevailing equatorial currents from westward to eastward in such a short period remains unclear and needs to be further validated. Therefore, we hypothesize that there must exist other factors contributing to El Niño.
Observed monthly mean zonal surface winds (m s−1) during 1992–2018. Monthly mean zonal surface winds at (a) 0°, 156°E; (b) 0°, 165°E; (c) 0°, 180°; (d) 0°, 170°W; (e) 0°, 140°W; and (f) 0°, 110°W. For the zonal component, positive winds are westerly, negative winds are easterly. Discontinuous areas are data missing while transmitting from monitoring buoys to shore-based stations. Derived from NOAA/PMEL.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
The aim of this paper is devoted to a further understanding of genesis and maintenance of ENSO by proposing a previously ignored new mechanism using observational data in situ (monitoring buoys deployed by NOAA along the equatorial Pacific since 1992) and a coupled atmosphere–ocean model. The remainder of this paper are generally arranged as follows: section 2 provides observational data and fundamental analysis of the prevailing theories, further investigations and new insights are given in section 3, data and simple methods are described in section 4, simulations of ENSO are presented in section 5, and discussions and conclusions are arranged in the last section.
2. Observational data and fundamental analysis
An important criterion for a valid mechanism is not only its consistency with observational data but also the real existence of cause–effect relationship. Our investigation demonstrates that the agreement of the simulations and the observations is unlikely due to the role of the westerly winds as “an important potential trigger” but more likely a passive response to El Niño events. This reasoning is based on the observations that the westerly winds from the equatorial Indian Ocean (McPhaden 1999) and the easterly trade winds from the equatorial Pacific meet over the Maritime Continent and gradually decrease to zero, resulting in a strong convective zone (zero wind zone; see Figs. 2a,e). As the easterly trade winds weaken (more precisely, it is a weakening of the southeast trade winds, the possible cause of which will be analyzed later) or the westerly winds strengthen or both at the onset of El Niño, an eastward intrusion of the westerly winds is likely to occur, shifting the convective zone eastward. In contrast, the convective zone tends to migrate westward as the easterly trade winds intensify or the westerly winds weaken or both during La Niña, exhibiting a seesaw behavior (Figs. 2a,e). The magnitude of the zonal displacement of the convective zone is largely determined by the zonal dynamic contrast of the wind forcings and the variabilities of the warm pool (surface waters greater than about 29°C; McPhaden 1999). The detailed observations (Figs. 1a,b and 2e,g,h) quantitatively indicate that in the 2015 El Niño, the trade winds in the western equatorial Pacific (WEP) were observed to change from easterly to the westerly, without intensification (average 2 m s−1) but with a wider zonal extent. The records show that these reversed trade winds had a slight strengthening from 2 to ~3 m s−1 even in the most pronounced event in 1997 (Figs. 1a, 2a,c,d, and 10). Nevertheless, a weakening of the easterly trade winds is prominent from average of 6 to 2 m s−1 in each warm event in the central equatorial Pacific (Figs. 2a,e). A significant enhancement of the extended westerly trade winds or the westerly wind bursts (WWBs; both are westerly winds) in the western equatorial Pacific during the 1997 and 2015 El Niño events has not been identified with the monitoring buoys deployed by NOAA along the equator since 1992 (Figs. 1a,b and 2a,e) despite the maximum westerly winds occasionally being 7 m s−1. However, the amplitude of the westerly winds may vary if different methods of measurement are applied (Puy et al. 2016), and different studies may also quote the strength of the wind speed over different region based on the aims of the research (Harrison and Vecchi 1997). The WWBs are thought to play a role in contributing to the overall weakening of the easterly trade winds during El Niño development as it progresses toward maturity; in turn, the weakened easterly winds are likely to stimulate more occurrences of the westerly wind events (Levine et al. 2016), establishing a positive feedback. Indeed, the role of westerly winds still warrants further consideration, for example, as stated by McPhaden (1999). However, episodic wind forcing is not a sufficient condition for El Niños to occur, since such forcing is evident during non–El Niño years as well. It has also been argued that episodic wind forcing is not even a necessary condition for the development of El Niños, since many coupled ocean–atmosphere models simulate ENSO-like variability without it (McPhaden 1999; McPhaden andYu 1999).
Time vs longitude sections of (a),(e) observed surface zonal wind, (b),(f) ADCP zonal current, (c),(g) SST, and (d),(h) SST anomalies (top) from January 1996 to December 1998 and (bottom) from January 2014 to December 2016. Analyses are based on 5-day averages for between 2°S and 2°N for the TAO data. Black squares on the abscissas of the plots indicate longitudes of data availability at the start (top) and end (bottom) of the time series record. For the zonal component, positive winds are westerly, and negative winds are easterly. Positive currents are eastward, and negative currents are westward. Blank areas are data missing while transmitting from monitoring buoys to shore-based stations.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
3. Investigations and new insights
While many studies concentrate on the role of the WWBs as a potential driver in El Niño evolution, there are also other factors that may be overlooked. Recently, an eastward current along the equator during El Niños was identified (Kim et al. 2015; Kim and Cai 2014; Santoso et al. 2013). It seems to us that the underlying assumption propo6sed by Santoso et al. (2013) could be a reversed south equatorial current (SEC) associated with a positive west–east contrast of the zonal sea level pressure (SLP; Santoso et al. 2013). The eastward current is assumed to play a dominant role in shifting the eastern edge of the warm pool eastward, giving rise to El Niño (Kim et al. 2015; Santoso et al. 2013; Kim and Cai 2014; Picaut et al. 1997). This assumption, however, is questionable in light of observational records.
Our examination in terms of heat budget and origin of the equatorial countercurrent reveals that an anomalous eastward current along the equator during El Niños does exist (Figs. 2b–d,f–h and 3a–c), but it is thought to be a southward-shifted north equatorial countercurrent (NECC) usually centered at ~5°N in winter but ~8°N in summer (Chen et al. 2016) rather than a “reversed SEC”; if this eastward current along the equator were a reversed SEC, there should exist two branches of the eastward currents north of the equator at the same period, one along the normal latitudinal positions 5°–8°N and the other along the equator, obviously inconsistent with the satellite observations (only one eastward current can be seen at any time between midlatitudes and the equator). Moreover, the direction change of the zonal winds from easterly to westerly in March 1997 and from westerly to easterly in December 1997 over the WEP and central equatorial Pacific (CEP) (at 0°, 165°E) coincided with that of the equatorial current in the eastern equatorial Pacific (at 0°, 140°W) from westward to eastward and from eastward to westward, respectively (Figs. 1a,b and 4), clearly inconsistent with the feature of fluid’s enormous hysteresis if the reverse of the equatorial current were forced by the winds, suggesting that this new eastward current along the equator during El Niños is not likely to be a reversed SEC, but rather, a southward-shifted NECC. This is also supported by the notable asymmetric distributions of the warm pool at the El Niño onset with most parts shifting to south of the equator (climatological positions of the warm pool are roughly symmetric about the equator) and with a marked cooling of ~1.5°–2°C in north of the equator (at 5°N, 180°) and a warming of ~1°C in south of the equator (at 5°S, 180°) in the CEP (Fig. 5 and Figs. S4a,b in the online supplemental material). The concurrence of a cooling and a warming in the same basin (or same longitudes) is a manifestation of a southward migration of the NECC with the NEC (cool) and SEC (warm) together.
Observed daily mean zonal surface velocity (cm s−1) during 1995–2017. Daily mean zonal surface velocity at (a) 0°, 156°E during 2013–17, (b) 0°, 165°E during 1996–98, and (c) 0°, 140°W during 1995–99. For the zonal component, positive values are eastward, negative values are westward. Discontinuous areas are data missing while transmitting from monitoring buoys to shore-based stations. Derived from NOAA/PMEL.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
Five-day zonal surface winds (m s−1) in the western equatorial Pacific and 5-day zonal current velocity (cm s−1) during 1996–98: (left) 5-day zonal surface winds at 0°, 165°E during 1996–98 and (right) 5-day zonal surface current velocity at 0°, 140°W during 1996–98. The major direction change of the zonal winds from easterly to westerly in March 1997 or from westerly to easterly in December 1997 over the WEP and CEP coincided with that of the zonal current from westward to eastward or from eastward to westward in the eastern equatorial Pacific (blue and black points). Discontinuous areas are data missing while transmitting from monitoring buoys to shore-based stations. From NOAA/PMEL.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
SST in the central equatorial Pacific (at 5°N, 180° and 5°S, 180°, respectively) during 1996–98. Discontinuous areas are data missing while transmitting from monitoring buoys to shore-based stations. From NOAA/PMEL.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
Early work of Johnson et al. (2002) has revealed that the NECC shifts southward during El Niño and northward during La Niña, and subsequently being confirmed by recent study conducted by Chen et al. (2016) showing an intensification of the NECC and a migration southward during El Niño events. More significantly, this variability is found to be one month lead ahead of the Niño-3.4 index (Chen et al. 2016), manifesting a possible role of the NECC in inducing El Niño. The potential mechanism of the meridional migrations of the NECC still remains debated, however.
In essence, an anomalous change of the NECC during El Niño was proposed by Wyrtki (1973) but largely overlooked due to unclear mechanisms at that time. The weak and episodic westerly wind forcing is considered insufficient in reversing the SECs (McPhaden 1999; McPhaden and Yu 1999), but favorable to the eastward current.
The strong support to the suggestion of a southward-shifted NECC comes from the observations in situ showing the monthly evolutions of the NECC from the normal positions 5°–8°N (blue color) in November 1996 to 5°S–5°N in June 1997 until the end of the 1997 El Niño, with the NECC in previous latitudes completely disappeared. (Fig. 6, rightmost). The synchroneity of an absence of the NECC in the normal latitudes and an emergence of a new eastward current along the equator is suggestive of a southward-shifted NECC.
Meridional position evolutions of surface currents in the central equatorial Pacific from November 1996 to June 1997. Colors in blue, red, yellow, and green represent eastward, northward, westward, and southward surface currents, respectively. Derived from http://www.oceanmotion.org/html/resources/oscar.htm.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
In addition, the investigation into the SLP in the western and eastern equatorial Pacific shows that the SLP in the eastern equatorial Pacific is rarely lower than that in the western equatorial Pacific, although the contrast of the SLP between them is largely narrowed during El Niño events (Fig. S10), indicating that “eastward equatorial current” during El Niños suggested by Santoso et al. (2013) is unlikely to be induced by a positive west–east contrast of the zonal SLP, rather, by a southward-shifted NECC.
We have examined all El Niño events during 1992–2014 in the observational records (available for both the zonal winds and zonal currents), indicating that each warm event was always preceded by a southward-shifted NECC along the equator (Figs. 2b,c,f,g), with the NECC in previous positions being totally absent. The observations show that the southward-shifted NECC persisted for a longer period from January to December 2015 with moderate intensity (30–40 cm s−1) (Fig. S7b), giving rise to a strong El Niño. However, the southward-shifted NECC stayed around the equator just for a shorter period from April to June 2014 with a relative weak intensity (20–26 cm s−1) (Fig. S7b), indicating a shortage of energy supply for further development, thus a weak or at most a moderate El Niño rather than a super one predicted by many models before. Therefore, it is very reasonable to hypothesize that the southward-shifted NECC is most likely to play a critical role for the genesis and diversity of El Niño.
Analyzing the structures and distributions of the equatorial currents and NECC is also conducive to our further understanding of the SST anomalies in the WEP. At 5°–8°N, the NECC is fed by the southward branch of the north equatorial current (NEC) (cool from higher latitudes), the northward branch of the SEC (warm from the equator) and the upwelling (cool from deep water) (Wang et al. 2016) (Fig. S1a). The warm pool in the WEP is sustained by the westward warm SEC along the equator. Nevertheless, the NECC is no longer fed by the relative warm sources after shifting to the equator, rather, by the three branches of relative cool currents, involving the southward branch of the NEC (cool), the northward branch of the SEC (becoming cool after shifting to higher latitudes) and the upwelling (cool) (Fig. S1b). The warm pool in the WEP is also replaced by the cooler pool afterward, cooling the WEP by about 2°C (Fig. S8). Furthermore, an enhancement of longwave radiative fluxes due to an eastward migration of convective clouds is considered favorable to a cooling in the WEP (Hu and Fedorov 2017).
The eastward propagation mechanism of the warm pool (actually a cooler pool) in the WEP also remains debated. Present theories can be generally categorized into three different frameworks, the most prevailing theory considering the westerly wind bursts as a direct trigger in shifting the warm pool eastward (Lengaigne et al. 2004; Fedorov et al. 2014; Menkes et al. 2014), the next regarding enhanced eastward downwelling equatorial Kelvin waves as a potential driver in generating warming phase in the eastern cold tongue (McPhaden andYu 1999; Wyrtki 1975; van Oldenborgh 2000) while another suggesting that the interplays between the ENSO-related current anomaly and the climatological current play a key role in determining the zonal propagation of SST anomalies (Santoso et al. 2013). The exact evolutions of SST anomalies (i.e., deviations from climatological norms) in different regions, for example, the negative SST anomalies in the WEP, the tiny positive SST anomalies in the CEP, and the large positive SST anomalies in the EEP, remain ambiguous.
Our further investigation into the propagation mechanism of the warm pool demonstrates that the southward-shifted NECC plays a significant role not only in shifting the eastern edge of the warm pool eastward but also in elevating the mean SSTs in different ocean basins. An educated guess can be made that SST for any seawater mass horizontally moving with a speed of ~0.5 m s−1 (representative of observed mean speed of the SEC along the equator) (Bonjean and Lagerloef 2002; Admiralty 2018) in either direction along the equator, for example, from 150°E to 90°W (eastward) or from 90°W to 150°E (westward), can rise by the same values (~3.5°–4.5°C) under the same solar radiation after considering heat exchange between the atmosphere and sea. In normal condition, the mean SST can rise by about ~4.5°C as the sea waters with mean SST ~ 25.5°C in the EEP (at 0°, 90°W) travel to the WEP (at 0°, 150°E) along the equator, attaining ~30°C. The warm pool in the WEP can remain at ~30°C mainly because of a balance between the westward SEC along the equator consistently feeding the warm pool and the northward and southward branches of the SEC flowing away from the warm pool after reaching the Asia continental shelf not only in transport but also in SST (Fig. S1a). At the onset of El Niño, the previous warm pool with SST ~ 30°C in the WEP is replaced by the cooler pool with SST ~ 28°C, owing to southward migrations of the NECC (in practice, this process may take place gradually, but an abrupt shift cannot be ruled out; Fig. S8 and Figs. 1b,c). This cooler pool, similar to the westward SEC along the equator, is continuously heated by solar radiation at the same heating rate (assuming it remains constant) while moving eastward along the equator advected by the southward-shifted NECC, slightly elevating SSTs (0.5°–1.5°C) in the CEP (because of being heated for a shorter time from the WEP to the date line), largely elevating SSTs (3.5°–4.5°C) in the EEP (due to being heated for a longer time from the WEP to the EEP), and pushing down the thermocline in the eastern Pacific (EP), hence contributing to the diversity of El Niño, in contrast to the previous notion of the westerly wind bursts (Chen et al. 2015). The amplitude of SST anomalies is also closely related to the interplays between the southward-shifted NECC and the upwelling in the EP, implying that the mean SST in the EP would maintain at ~32°–33°C if the upwelling near the equator totally vanished. The mean SST in the EEP has not risen to what it could be during El Niño events mainly because of the effective suppression of the upwelling. The observations show that the equatorward upwelling in the eastern cold tongue, usually recurving along the equator, diverted toward the west at 10°–15°S (main axis) during the 1997 El Niño, but at about ~3°–4°N during the 1998 La Niña due to its meridional migrations along with other systems, to feed the westward SECs (Figs. S5a,b; www.oscar.noaa.gov/datadisplay/ or www.oceanmotion.org/html/resources/oscar.htm), suggesting its unique role in determining the amplitudes of SST anomalies. Our investigations indicate that the meridional position variations of the NECC and the upwelling are in phase with the observed SST anomalies in the Niño-3.4 region (Figs. 7a–c), with correlation coefficient r = 0.94 and 0.91, respectively. It is, therefore, reasonable to hypothesize that the southward-shifted NECC plays a critical role in cooling the WEP, extending the eastern edge of the warm pool eastward and elevating the SSTs in the CEP and EEP at the onset of El Niño, with the interactions of the eastward warm pool and the upwelling in the eastern cold tongue ascertaining the amplitudes of SST anomalies.
Comparisons of observed SST anomalies in the Niño-3.4 region and meridional position variations of the NECC and the upwelling from 1997 to 2018. (a) SST anomalies in the Niño-3.4 region. Red shadings are warm phases, and blue shadings are cold phases. (b) Meridional position variations of the NECC (center axis along 160°W). The red side bar represents the tendency of the warm heat flux. (c) Meridional position variations of the upwelling (main axis along 120°W). The blue side bar denotes the tendency of the cold heat flux. (Data available at www.esrl.noaa.gov/ and www.oceanmotion.org/html/resources/oscar.htm).
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
4. Data and methods
To demonstrate the spatiotemporal patterns of the El Niño events, high-resolution satellite observations are employed. The SST data used in this study are the Hadley Centre Global Sea Ice and Sea Surface Temperature (HadISST) version 1.1 from 1961 to 2010 with a resolution of 1° × 1° (http://www.metoffice.gov.uk/hadobs/hadisst/data/download.html). The monthly outgoing longwave radiation (OLR) data are obtained from the National Oceanic and Atmospheric Administration (NOAA) polar-orbiting satellites (Liebmann and Smith 1996). OLR from the model and observations have been used as a proxy for deep tropical atmospheric convection. Observed anomalies are calculated over the 1981–2001 common period. For the energy flux analysis, we use surface and radiative fluxes from the ERA-Interim reanalysis (http://www.ecmwf.int/en/research/climate-reanalysis/era-interim). The warm-water volume, which is defined as the integral of water above the 20°C isotherm over the equatorial Pacific (140°E–80°W, 5°S–5°N), is derived from the potential temperature and salinity datasets from the NCEP Global Ocean Data Assimilation System (GODAS) reanalysis from 1980 to 2014 (http://www.esrl.noaa.gov/psd/data/gridded/data.godas.html).
The subsequent simulations are forced with the heat flux (HF) Q by establishing a relationship between the heat flux Q in the vicinity of the equator and the ranges of the NECC and the upwelling (UPG) from the equator based on an empirical formula (a simplified scheme of coupled HF–NECC and HF–UPG relationship). The heat flux QN basically remains with a linear change as 0° < φN < 4° whereas it exhibits nonlinearity as 4° ≤ φN ≤ 6.5°. The heat flux Qu maintains a linear relationship with the meridional position of the upwelling. For the NECC, the closer to the equator, the warmer heat flux QN. For the upwelling, the closer to 3°N, the cooler heat flux Qu. The amplitude of SST anomalies is determined by the interactions of the QN and Qu. For more details of the model description, see the online supplemental material.
5. Simulating ENSO with meridional displacements of the NECC
To further elucidate and validate our hypothesis, an updated model (HadOPA) which has achieved great success in previous El Niño simulations is employed. This model is used to simulate the El Niño and La Niña by adding a meridional migration of the South Pacific subtropical high (SPSH) in the EP as a perturbation to an intermediate ocean–atmosphere coupled model (Lengaigne et al. 2006) (see supplemental material). When the SPSH shifts southward with meridional position anomalies of −4° at a speed of 0.25° day−1 (continuous moving excluding instantaneous speed), the main axis of the upwelling subsequently shifts from 0° to ~2°S (Δφu = −2°) and the center axis of the NECC migrates from ~6.5° to ~4°N (ΔφN = −2.5°) in response to changes of the SPSH. However, the interannual oscillations with SST anomalies retain little change due to its nonlinear effects (Figs. 8a,b) but start to surge and become highly irregular as the SPSH continuously moves southward at a speed of 0.35° day−1 in early spring. When further perturbation is imposed in late spring, the model produces a broad continuum of El Niño events subsequently in position ranging from the date line to the EEP. A strong El Niño occurs in winter in the EEP as the SPSH displaces further southward with meridional position anomalies of −8° at a speed of 0.38° day−1. The NECC is observed to migrate to the equator (ΔφN = −6.5°, the strongest) and a large southward migration of the upwelling (Δφu = −10°, relatively weak) superimposed on the seasonal cycle is also noted (Figs. 8e,f). However, a relative weak but more frequent El Niño appears around the date line in summer when the SPSH shifts southward with meridional position anomalies of −6° at a speed of 0.31° day−1 (the NECC and the upwelling are observed to shift southward with ΔφN = −5.5°, relatively strong; Δφu = −5°, relatively strong) (Figs. 8c,d). We run this model by shifting the meridional position of the SPSH southward further at a speed of 0.39° day−1 and meridional position anomalies of −10° and −12°, respectively, (the upwelling is subsequently recorded to shift southward with Δφu = −11°, very weak, and Δφu = −13°, the weakest), to simulate the El Niño episodes, respectively. As expected, the warm events quickly develop into extreme EP El Niño events with SST anomalies in Niño-3 exceeding 3.8°C (Figs. 8e,f) and 4.2°C (Figs. 8g,h), respectively, consistent with the observations.
(left) Simulated sea surface temperature and (right) corresponding sea surface temperature anomalies in the tropical Pacific (°C). (a),(b) SST and SST anomalies as meridional position anomaly of the SPSH = 4°, VSH = 0.25° day−1, ΔφN = −2.5°, and Δφu = −2° (almost in normal condition). (c),(d) SST and SST anomalies as meridional position anomaly of the SPSH = −6°, VSH = 0.31° day−1, ΔφN = −5.5°, and Δφu = −5° (date line El Niño). (e),(f) SST and SST anomalies as meridional position anomaly of the SPSH = −8°, VSH = 0.38° day−1, ΔφN = −6.5°, and Δφu = −11° (strong El Niño). (g),(h) SST and SST anomalies as meridional position anomaly of the SPSH = −12°, VSH = 0.39° day−1, ΔφN = −6.5°, and Δφu = −13° (extreme El Niño). (i),(j) SST and SST anomalies as meridional position anomaly of the SPSH = 2°, VSH = 0.25° day−1, ΔφN = 0°, and Δφu = +3° (La Niña). All simulated SST and SST anomalies are consistent with observations throughout observed periods which are significant at the p = 0.1 level based on a two-sided t test.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
Similar model run has been executed to simulate the La Niña episodes by shifting the SPSH northward. As anticipated, the warm phase in the eastern tropical Pacific subsequently evolves into a cold phase in late summer next year with 2°–3°C cooling of SST anomalies in Niño-3 region as the SPSH retreats back to its climatological position but slightly north, immediately followed by a northward meridional migration of the NECC and the upwelling with ΔφN = −0°, φN = −6.5°N, the weakest, and Δφu = +3°, φu = 3°N, the strongest) (Figs. 8i,j), reasonably consistent with the observed records. The disproportional alterations of the relative position in latitude between the SPSH and the NECC and the upwelling are likely to lead to flavors of El Niño or La Niña, implying that the strongest NECC and the weakest upwelling tends to produce a strong El Niño and a moderate NECC, a moderate upwelling leads to a weak El Niño, and the weakest NECC and the strongest/moderate upwelling generate a strong/moderate La Niña. Actually, there are more situations than described here that can result in a weak event near the date line, for instance, a strong NECC plus a strong upwelling, a weak NECC plus a weak upwelling, etc., as long as the NECC and the upwelling are equally strong or the former is slightly stronger. This explains why a date line El Niño is more frequent. Note that for simplicity, the intensity of the upwelling measured in the vicinity of the equator is assumed to linearly change with its own meridional position in spite of its possible nonlinearity. The observations quantitatively indicate that the easterly trade winds in the central equatorial Pacific resumed from the weakest (2–3 m s−1) during the 1997 El Niño to its climatological norm (6 m s−1) during the 1998 La Niña without intensification (Figs. 1a–c, 2a, and 10), with a strengthening and a more north latitude of the upwelling only (Fig. S5b, www.oscar.noaa.gov/datadisplay/, or www.oceanmotion.org/html/resources/oscar.htm), suggesting a significant role played by the upwelling feedbacks in inducing a La Niña, with the zonal winds in the equatorial Pacific playing a minor role.
To further evaluate the response of SST in the equatorial Pacific to the proposed mechanism, a sensitivity experiment has been performed. A 10-member ensemble integration with slightly perturbed initial conditions has been carried out using a prescribed coupled model. Each member of this ensemble covered one full year of integration from August of the start year 1997. The SST anomalies in Niño-3 and Niño-4 change quickly once a modification is made in meridional positions of the NECC and the upwelling, suggesting that SSTs in the equatorial Pacific are very sensitive to changes of meridional positions of the NECC and the upwelling. Figure 9 shows the ensemble mean and each individual member of the SST anomalies in Niño-3 and Niño-4. The maximum SST anomalies in Niño-3, obviously lagged that in Niño-4 for about 1.5–2 months due to a gradual eastward advection of the warm pool, appear in December with amplitude up to 3.8°C, consistent with the observations. From late spring of the next year, the entire equatorial Pacific starts to cool down with strong negative SST anomalies in the CEP and EEP after August, indicating a substantial swing into a cold phase. The pronounced feature is that the ensemble in Niño-3 demonstrates a relatively narrow ensemble spread with a standard deviation of SST anomalies which remain below 0.5°C. All ensemble members in Niño-3 illustrate maximum SST anomalies during December–January followed by a cooling (~0.6°–0.8°C month−1).
Simulated time series for ensemble members (color thin lines) and ensemble mean (black thick line) of (bottom) Niño-3 and (top) Niño-4 SST anomalies for perturbed experiments. Each member from A to J, respectively, corresponds to an index from 11.5 to 16, respectively, with an interval 0.5 (produced by computing Q parameters). All simulated time series are consistent with the observations throughout observed periods at the p = 0.1 (two-tailed) confidence level. Shading indicates the corresponding standard deviation interval.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
6. Discussions and conclusions
Observational evidence reaffirms that the driver for the southward migrations of the NECC, the trade winds, and the upwelling is rooted in the SPSH (Trenberth 1976; Rollenbeck et al. 2015; Linacre and Geerts 1998). Long-term observations (1949–2012) show (see Fig. S11) that each El Niño event corresponds to an anomalous southward shift of the SPSH (Ancapichun and Garcés-Vargas 2015). In the South Pacific, the southern subsiding branch of the Hadley cells determines the presence of a quasi-permanent belt of high surface pressure around 30°S (Held and Hou 1980), whose development over the South Pacific Ocean is known as the SPSH. At the northern edge of the SPSH, the air masses flow westward, producing the belt of the southeast trade winds (tropical easterlies). South of the SPSH, a westerly wind belt, known as the southern westerly winds (westerlies), is developed in the midlatitudes, peaking around 50°S (Varma et al. 2012). Between the eastern edge of the SPSH and the west coast of South America, strong southerly alongshore winds exist, generating intense wind-driven upwelling along the southeastern Pacific, in turn, cooling the eastern tropical Pacific. Therefore, the SPSH is considered a predominant factor in modulating the climate change in the South Pacific basin and adjacent regions, particularly South America. (Flores-Aqueveque et al. 2020). Significant changes of the SPSH in position and intensity are able to induce an alteration of the prevailing winds, ocean circulations, upwelling, and precipitation patterns in the South Pacific.
When the SPSH is occasionally forced to anomalously move southward by some external forcings (the cause is not very clear and further studies are needed), the southeast trade winds near the equator must weaken, inevitably resulting in a southward incursion of the northeast trade winds and thus, a southward shift of the trade wind system. As the trade wind system migrates southward, so do the trade-wind-induced equatorial currents, including the NECC located between the SEC and NEC. Once the northeast trade winds cross the boundary (5°–8°N) of the southeast and northeast trade winds (the northeast and southeast trade winds are symmetric about 5°–8°N, rather than the geographic equator, see Figs. S1a–S1c), the northeast trade winds would become northwest under the combined effects of the Coriolis force and other favorable flow fields, such as a reverse of the Walker Circulation, leading to a broader and stronger southward-shifted NECC, consistent with the observations showing that the prevailing winds over the WEP and CEP during the 1997 and 2015 El Niños were northwest [derived from the equal zonal (west) and meridional (north) components of the winds; see Figs. S3a, S3b, and S3e–S3f]. A limited northern boundary (~5°N) and a much broader southern boundary (~15°S) of the westerly winds over the WEP and CEP (Fedorov et al. 2014) are in conformity with the characteristics of the veered northeast trade winds, suggesting that the abovementioned westerly winds observed during El Niño events may partly consist of the veered northeast trade winds.
The observations indicate that the SPSH in 1997 was observed to shift from 17°S in March to 27°S in May, to 40°S in June, all at about 77°W (Linacre and Geerts 1998), with the zonal current intensifying at 0°, 160°E from +25 cm−1 (eastward) in May to +70 cm−1 in August, to +90 cm−1 in November (Figs. 3b,c and Fig. S9). Clearly, the strengthening of the zonal current (2°–2°S) was preceded by the southward migrations of the SPSH. Figure 10 shows that the westerly winds basically remained unchanged in mean intensity from March to October 1997, but an increase of the zonal current (eastward) at 0°, 165°E was evident, from near zero to +90 cm−1 (Fig. 3b), implying that the NECC was gradually approaching the equator (a narrow band of the zonal zero-speed current exists between the NECC and SEC), with the westerly winds playing a secondary role. The superposition of the El Niño–related southward shifts of the SPSH onto the seasonal cycles makes the average speed of the SPSH moving nearly 2–3 times faster than usual, serving as an alternative precursor for the initial development of the event. This offers the scientists new insights into monitoring and prediction of the El Niño onset.
Daily zonal surface winds (m s−1) in the western equatorial Pacific (at 0°, 165°E) during 1996–98. Discontinuous areas are data missing while transmitting from monitoring buoys to shore-based stations. From NOAA/PMEL.
Citation: Journal of Physical Oceanography 51, 7; 10.1175/JPO-D-19-0323.1
A key question arises whether the southward migration of the SPSH is a result of the eastward warm pool or is driving the warm pool eastward. Traditionally, the southward migration of the SPSH is thought to be forced by the eastward warm pool along the equator (McPhaden 1999). In contrast, our investigation demonstrates that the southward migration of the SPSH is not a passive response to the eastward warm pool, but is driving the warm pool eastward, giving rise to El Niño. This reasoning is based on the fact that the North Pacific subtropical high (NPSH) has never been found to be affected significantly by the eastward warm pool (manifested by little anomalous change of the upwelling off the California coast during El Niños; see Figs. S4a,b). The asymmetric response of the SPSH and NPSH to the warm pool is indicative of an active role of the SPSH, consistent with our model simulations. This is further supported by the enhancement of the eastward current along the equator always lagging the southward shifts of the SPSH during El Niños.
Phase correlation analysis of the meridional migrations of the SPSH and the NECC illustrates that the maximum correlation occurs around −1.8 months during boreal spring and summertime (see Fig. S12), implying that the meridional displacement of the NECC lags the SPSH by about 1.5–2 months, further suggesting the dominant role of the SPSH in inducing the SST anomalies in the CEP and EEP.
This finding may be profound because it exhibits an alternative explanation of El Niño/La Niña. This being said, we do not want to downplay other mechanisms that may also play a role at the onset of El Niño/La Niña. The apparent lack of real-time forecasting and long-term predictability of El Niño (Tippett et al. 2012; Wang et al. 2010; IRI 2021) implies that we have some way to go in fully understanding the real physical mechanisms of the El Niño/La Niña phenomenon. It is believed that our new findings can better shed light on the coupled effects of the NECC and the upwelling in the genesis of El Niño/La Niña and may lead to more accurate predictions for a longer period in the future.
Acknowledgments
Both authors contributed equally to this work and should be considered co-first authors. Y.J. Zou designed this work, prepared the manuscript and figures, and interpreted results. X.Y. Xi was responsible for laboratory efforts and contributed to the computer programming and the model simulating. The authors declare no competing financial interests.
REFERENCES
Admiralty, 2018: Ocean Passages for the World (NP136). 1st ed. UK Hydrographic Office, 224–225.
An, S. I., 2008: Interannual variations of the tropical ocean instability wave and ENSO. J. Climate, 21, 3680–3686, https://doi.org/10.1175/2008JCLI1701.1.
Ancapichun, S., J. Garcés-Vargas, 2015: Variability of the southeast pacific subtropical anticyclone and its impact on sea surface temperature off north-central Chile. Cienc. Mar., 41 (1), 1–20, https://doi.org/10.7773/CM.V41I1.2338.
Bjerknes, J., 1969: Atmospheric teleconnections from the equatorial Pacific. Mon. Wea. Rev., 97, 163–172, https://doi.org/10.1175/1520-0493(1969)097<0163:ATFTEP>2.3.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.
Chen, D. K., and Coauthors, 2015: Strong influence of westerly wind bursts on El Niño diversity. Nat. Geosci., 8, 339–345, https://doi.org/10.1038/ngeo2399.
Chen, X., B. Qiu, Y. Du, S. Chen, and Y. Qi, 2016: Interannual and interdecadal variability of the North Equatorial Countercurrent in the western Pacific. J. Geophys. Res. Oceans, 121, 7743–7758, https://doi.org/10.1002/2016JC012190.
Eisenman, I., L. Yu, and E. Tziperman, 2005: Westerly wind bursts: ENSO’s tail rather than the dog. J. Climate, 18, 5224–5238, https://doi.org/10.1175/JCLI3588.1.
Fedorov, A. V., S. Hu, M. Lengaigne, and E. Guilyardi, 2014: The impact of westerly wind bursts and ocean initial state on the development and diversity of El Niño events. Climate Dyn., 44, 1381–1401, https://doi.org/10.1007/s00382-014-2126-4.
Flores-Aqueveque, V., M. Rojas, C. Aguirre, P. A. Arias, and C. González, 2020: South Pacific subtropical high from the late Holocene to the end of the 21st century: Insights from climate proxies and general circulation models. Climate Past, 16, 79–99, https://doi.org/10.5194/cp-16-79-2020.
Gebbie, G., I. Eisenman, A. Wittenberg, and E. Tziperman, 2007: Modulation of westerly wind bursts by sea surface temperature: As emistochastic feedback for ENSO. J. Atmos. Sci., 64, 3281–3295, https://doi.org/10.1175/JAS4029.1.
Harrison, D. E., and G. A. Vecchi, 1997: Westerly wind events in the tropical Pacific, 1986–95. J. Climate, 10, 3131–3156, https://doi.org/10.1175/1520-0442(1997)010<3131:WWEITT>2.0.CO;2.
Held, I. M., and A. Y. Hou, 1980: Nonlinear axially symmetric circulations in a nearly inviscid atmosphere. J. Atmos. Sci., 37, 515–533, https://doi.org/10.1175/1520-0469(1980)037<0515:NASCIA>2.0.CO;2.
Hu, S., and A. V. Fedorov, 2017: The extreme El Niño of 2015–2016: The role of westerly and easterly wind bursts, and preconditioning by the failed 2014 event. Climate Dyn., 52, 7339–7357, https://doi.org/10.1007/S00382-017-3531-2.
IRI, 2021: IRI ENSO forecast. International Research Institute for Climate and Society, Columbia University, http://iri.columbia.edu/our-expertise/climate/forecasts/enso/current.
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.
Kim, W., and W. Cai, 2014: The importance of the eastward zonal current for generating extreme El Niño. Climate Dyn., 42, 3005–3014, https://doi.org/10.1007/s00382-013-1792-y.
Kim, W., W. Cai, and J. S. Kug, 2015: Migration of atmospheric convection coupled with ocean currents pushes El Niño to extremes. Geophys. Res. Lett., 42, 3583–3590, https://doi.org/10.1002/2015GL063886.
Landsea, C. W., and J. A. Knaff, 2000: How much skill was there in forecasting the very strong 1997-98 El Niño? Bull. Amer. Meteor. Soc., 81, 2107–2119, https://doi.org/10.1175/1520-0477(2000)081<2107:HMSWTI>2.3.CO;2.
Lengaigne, M., and Coauthors, 2004: Triggering of El Niño by westerly wind events in a coupled general circulation model. Climate Dyn., 23, 601–620, https://doi.org/10.1007/s00382-004-0457-2.
Lengaigne, M., J. P. Boulanger, C. Menkes, and H. Spencer, 2006: Influence of the seasonal cycle on the termination of El Niño events in a coupled general circulation model. J. Climate, 19, 1850–1868, https://doi.org/10.1175/JCLI3706.1.
Levine, A., and M. J. McPhaden, 2016: How the July 2014 easterly wind burst gave the 2015–2016 El Niño a head start. Geophys. Res. Lett., 43, 6503–6510, https://doi.org/10.1002/2016GL069204.
Levine, A., F. F. Jin, and M. J. McPhaden, 2016: Extreme noise-extreme El Nino: How state-dependent noise forcing creates El Niño–La Niña asymmetry. J. Climate, 29, 5483–5499, https://doi.org/10.1175/JCLI-D-16-0091.1.
Liebmann, B., and C. A. Smith, 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77, 1275–1277, https://doi.org/10.1175/1520-0477-77.6.1274.
Linacre, E., and B. Geerts, 1998: The South Pacific high, ocean currents, and El Niño. Climates and Weather Explained, Routledge, 348–382, http://www-das.uwyo.edu/~geerts/cwx/notes/chap11/fujita.html.
McPhaden, M. J., 1999: Climate oscillations: Genesis and evolution of the 1997–98 El Niño. Science, 283, 950–954, https://doi.org/10.1126/science.283.5404.950.
McPhaden, M. J., and X. Yu, 1999: Equatorial waves and the 1997–1998 El Nino. Geophys. Res. Lett., 26, 2961–2964, https://doi.org/10.1029/1999GL004901.
McPhaden, M. J., S. E. Zebiak, and M. H. Glantz, 2006: ENSO as an integrating concept in 360 Earth. Science, 314, 1740–1745, https://doi.org/10.1126/science.1132588.
Menkes, C. E., and Coauthors, 2014: About the role of westerly wind events in the possible development of an El Niño in 2014. Geophys. Res. Lett., 41, 6476–6483, https://doi.org/10.1002/2014GL061186.
Philander, S. G., and A. Fedorov, 2003: Is El Niño sporadic or cyclic? Annu. Rev. Earth Planet. Sci., 31, 579–594, https://doi.org/10.1146/annurev.earth.31.100901.141255.
Picaut, J., F. Masia, and Y. D. Penhoat, 1997: An advective-reflective conceptual model for the oscillatory nature of the ENSO. Science, 277, 663–666, https://doi.org/10.1126/science.277.5326.663.
Puy, M., J. Vialard, M. Lengaigne, and E. Guilyardi, 2016: Modulation of equatorial Pacific westerly/easterly wind events by the Madden-Julian oscillation and convectively-coupled Rossby waves. Climate Dyn., 46, 2155–2178, https://doi.org/10.1007/s00382-015-2695-x.
Rollenbeck, R., and Coauthors, 2015: Climatic cycles and gradients of the El Niño core region in North Peru. Adv. Meteor., 2015, 750181, https://doi.org/10.1155/2015/750181.
Santoso, A., and Coauthors, 2013: Late-twentieth-century emergence of the El Nino propagation asymmetry and future projections. Nature, 504, 126–130, https://doi.org/10.1038/nature12683.
Thual, S., A. J. Majda, N. Chen, and S. N. Stechmann, 2016: Simple stochastic model for El Niño with westerly wind bursts. Proc. Natl. Acad. Sci. USA, 113, 10 245–10 250, https://doi.org/10.1073/pnas.1612002113.
Tippett, M. K., A. G. Barnston, and S. Li, 2012: Performance of recent multimodel ENSO forecasts. J. Appl. Meteor. Climatol., 51, 637–654, https://doi.org/10.1175/JAMC-D-11-093.1.
Trenberth, K. E., 1976: Spatial and temporal variations in the southern oscillation. Quart. J. Roy. Meteor. Soc., 102, 639–653, https://doi.org/10.1002/qj.49710243310.
van Oldenborgh, G. J. V., 2000: What caused the onset of the 1997–98 El Niño? Mon. Wea. Rev., 128, 2601–2607, https://doi.org/10.1175/1520-0493(2000)128<2601:WCTOOT>2.0.CO;2.
Varma, V., and Coauthors, 2012: Holocene evolution of the Southern Hemisphere westerly winds in transient simulations with global climate models. Climate Past, 8, 391–402, https://doi.org/10.5194/cp-8-391-2012.
Wang, C., C. Deser, J. Y. Yu, P. DiNezio, and A. Clement, 2017: El Niño and Southern Oscillation (ENSO): A review. Coral Reefs of the Eastern Tropical Pacific, P. Glynn, D. Manzello, and I. Enochs, Eds., Coral Reefs of the World, Vol 8, Springer, 85–106.
Wang, F., Y. L. Li, and J. N. Wang, 2016: Intraseasonal variability of the surface zonal currents in the western tropical Pacific Ocean: Characteristics and mechanisms. J. Phys. Oceanogr., 46, 3639–3660, https://doi.org/10.1175/JPO-D-16-0033.1.
Wang, W., M. Chen, and A. Kumar, 2010: An assessment of the CFS real-time seasonal forecasts. Wea. Forecasting, 25, 950–969, https://doi.org/10.1175/2010WAF2222345.1.
Wyrtki, K., 1973: Teleconnections in the equatorial Pacific Ocean. Science, 180, 66–68, https://doi.org/10.1126/science.180.4081.66.
Wyrtki, K., 1975: El Niño—The dynamic response of the equatorial Pacific Ocean to atmospheric forcing. J. Phys. Oceanogr., 5, 572–584, https://doi.org/10.1175/1520-0485(1975)005<0572:ENTDRO>2.0.CO;2.
