The Oxford Research Encyclopedia of Climate Science will be available via subscription on April 26. Visit About to learn more, meet the editorial board, or recommend to your librarian.

Dismiss
Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CLIMATE SCIENCE (climatescience.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 24 March 2018

# Climate Dynamics of ENSO Modoki Phenomena

## Summary and Keywords

The El Niño Modoki/La Niña Modoki (ENSO Modoki) is a newly acknowledged face of ocean-atmosphere coupled variability in the tropical Pacific Ocean. The oceanic and atmospheric conditions associated with the El Niño Modoki are different from that of canonical El Niño, which is extensively studied for its dynamics and worldwide impacts. A typical El Niño event is marked by a warm anomaly of sea surface temperature (SST) in the equatorial eastern Pacific. Because of the associated changes in the surface winds and the weakening of coastal upwelling, the coasts of South America suffer from widespread fish mortality during the event. Quite opposite of this characteristic change in the ocean condition, cold SST anomalies prevail in the eastern equatorial Pacific during the El Niño Modoki events, but with the warm anomalies intensified in the central Pacific. The boreal winter condition of 2004 is a typical example of such an event, when a tripole pattern is noticed in the SST anomalies; warm central Pacific flanked by cold eastern and western regions. The SST anomalies are coupled to a double cell in anomalous Walker circulation with rising motion in the central parts and sinking motion on both sides of the basin. This is again a different feature compared to the well-known single-cell anomalous Walker circulation during El Niños. La Niña Modoki is the opposite phase of the El Niño Modoki, when a cold central Pacific is flanked by warm anomalies on both sides.

The Modoki events are seen to peak in both boreal summer and winter and hence are not seasonally phase-locked to a single seasonal cycle like El Niño/La Niña events. Because of this distinction in the seasonality, the teleconnection arising from these events will vary between the seasons as teleconnection path will vary depending on the prevailing seasonal mean conditions in the atmosphere. Moreover, the Modoki El Niño/La Niña impacts over regions such as the western coast of the United States, the Far East including Japan, Australia, and southern Africa, etc., are opposite to those of the canonical El Niño/La Niña. For example, the western coasts of the United States suffer from severe droughts during El Niño Modoki, whereas those regions are quite wet during El Niño. The influences of Modoki events are also seen in tropical cyclogenesis, stratosphere warming of the Southern Hemisphere, ocean primary productivity, river discharges, sea level variations, etc. A remarkable feature associated with Modoki events is the decadal flattening of the equatorial thermocline and weakening of zonal thermal gradient. The associated ocean-atmosphere conditions have caused frequent and persistent developments of Modoki events in recent decades.

# Introduction

The Modoki El Niño and Modoki La Niña are two phases of a new type of climate variation (ENSO Modoki) often observed in the tropical Pacific since the 1990s. The term Modoki (Japanese, meaning pseudo or quasi) was used by researchers from Japan (Ashok, Behera, Rao, Weng, & Yamagata, 2007) to describe these events, as their appearances resemble those of El Niño/La Niña events. As has been discussed in many research articles, El Niño/La Niña with their atmospheric counterpart, the Southern Oscillation, are two phases of a most dominant mode of climate variation in the world popularly known as El Niño/Southern Oscillation (ENSO). Emerging with an irregular periodicity of two to seven years, El Niño, the warm phase of the ENSO causes disorder in weather and climate across the globe. Considering those large socio-economic imprints, considerable research has been dedicated to understand and predict the phenomenon. It is noted that the phenomenon exhibits substantial variations in space and time. Specifically, studies in 1970s described the phenomenon as typically developing from the Peruvian coast and moving westward across the equatorial Pacific (cf. Rasmusson & Carpenter, 1982). In later years, however, it was noticed that ENSO SST anomalies could also start from the central and western equatorial Pacific and propagate to the eastern Pacific, which has become a notable characteristic of ENSO development post 1980s. In addition to these changes in development characteristics, the ENSO frequency and amplitude also show remarkable changes during 1990s and 2000s (e.g., Trenberth & Smith, 2006). The SST variations in these decades were more pronounced in the central Pacific compared to the conventionally observed regions of eastern Pacific and the South American coasts.

The low-frequency decadal ENSO characteristics are often linked with the mean state of the tropical Pacific climate system (e.g., McPhaden, Lee, & McClurg, 2011). While it is yet unclear how ENSO shall evolve in a changing mean state, under the stress of anthropogenic global warming, recent observations of the tropical Pacific do show a remarkable decadal change particularly visible since 1990s. Opposite to the upward trend in surface temperature observed earlier, the tropical Pacific underwent a decadal change since the beginning of 21st century. This could be related to a hiatus in the tropical warming (Kosaka & Xie, 2013), though the viewpoint of global hiatus has been revised with the analysis of more recent data (Medhaug, Stolpe, Fischer, & Knutti, 2017; Risbey & Lewandowsky, 2017). Nonetheless, the early decade of this century saw perpetually cooler eastern Pacific compared to the decades in the later part of the 20th century (Luo, Sasaki, & Masumoto, 2012). Embedded in this major change, El Niño Modoki events are frequently observed in the tropical Pacific (Ashok et al., 2007; Ashok & Yamagata, 2009; Lee & McPhaden, 2010; Newman, Shin, & Alexander, 2011; Weng, Ashok, Behera, Rao, & Yamagata, 2007; Yeh, Kug, Dewitte, Kwon, Kirtman, & Jin, 2009; Yeh, Kirtman, Kug, Park, & Latif, 2011; Yeh, Wang, Wang, & Dewitte, 2015). It is not yet clear how climate change in the 21st century is going to affect the ENSO diversity (Cai et al., 2015; Capotondi et al., 2015), but some analyses, based on global coupled model experiments with future scenarios, suggest an increase in El Niño Modoki type of variability in the later part of 21st century (e.g., Yeh et al., 2009).

Different from canonical El Niños, the El Niño Modokis (Ashok, Behera, Rao, Weng, & Yamagata, 2007) are characterized by warm central Pacific flanked by cool eastern and western Pacific. Several other terminologies are also used to describe the phenomenon, such as Trans-Niño (Trenberth & Stephaniak, 2001), Dateline El Niño (Larkin & Harrison, 2005), Central Pacific El Niño (Kao & Yu, 2009; Yeh et al., 2009), and Warm Pool El Niño (Kug & An, 2009). ENSO Modoki is seen in both boreal summer and winter, and it causes global teleconnections different from those of the canonical ENSO (Ashok et al., 2007; Cai & Cowan, 2009; Kim, Kim, & Yeh, 2012; Pradhan, Ashok, Preethi, Krishnan, & Sahai, 2011; Taschetto & England, 2009; Wang & Hendon, 2007; Weng et al., 2007, Weng, Behera, & Yamagata, 2009a; Weng, Wu, Liu, Behera, & Yamagata, 2009b). Since ENSO has dramatic societal impacts, such regional impacts need reexamination following the discovery of the ENSO Modoki phenomenon. The distinctive tripole nature of the ENSO Modoki, compared to the conventional dipole nature of ENSO, adds another dimension to the regional influences when we discuss the different teleconnections in “ENSO Modoki Influences.” The ENSO Modoki assumes both warm (when central Pacific is warmer than normal) and cold (when central Pacific is cooler than normal) phases of its behavior. The cold phase of La Niña Modoki is analogous to La Niña phase of ENSO. A typical pattern of El Niño Modoki was seen in the boreal summer of 2004, and a typical pattern of La Niña Modoki was observed in the boreal winter of 2000–2001.

# The ENSO Modoki Phenomenon

ENSOs and ENSO Modokis are recognized through the statistical analyses of the sea surface temperature (SST) variations in the tropical Pacific. The variations are often depicted as two major modes of variability in an Empirical Orthogonal Function (EOF) or a similar statistical analysis conducted on the SST anomalies (year-to-year deviation from the climatology). Particularly for the recent period from 1981, with better data quality in the era of satellite observations, the first EOF (EOF1) pattern, which explains about 45% of the variability captures the ENSO, whereas the second EOF (EOF2), which explains 12% of the SST variability, captures the ENSO Modoki. In the latter pattern, depicted by the second mode, SST anomalies of both eastern and western tropical Pacific have loadings of the same sign and are opposite to the loadings in central tropical Pacific (cf. Ashok et al., 2007). The central Pacific anomalies extend eastward toward higher latitudes in both hemispheres, like a boomerang pattern straddling the tongue of the loadings of opposite–sign in the eastern Pacific (Weng et al., 2007, 2009a).

## Definition

Based on the unique tripolar pattern in the SST anomalies, the ENSO Modoki index (EMI) is defined as a combination of three poles (cf. Ashok et al., 2007) as expressed in the following equation.

$Display mathematics$
(1)

Each of those terms on right hand side of the equation represents the area-averaged SST anomaly over the tropical regions of central Pacific (CP: 165°E-140°W, 10°S-10°N), eastern Pacific (EP: 110°W-70°W, 15°S-5°N), and western Pacific (WP: 125°E-145°E, 10°S-20°N), respectively (Figure 1). It may be noted here that geographical domains of these three poles are somewhat different compared to the indices used to explain the canonical ENSO. Moreover, the combinations of three poles used to derive the EMI make it unique in the sense that it captures a completely different variability in the SST anomalies of the tropical Pacific. Nevertheless, the domain of Niño 3.4 index (Figure 1), sometimes used to discuss ENSO variability, on and above the Niño 3 index, shares considerable parts of the CP box domain of EMI. That is the reason why Niño 3.4 index mixes up two different types of variations in the tropical Pacific represented by ENSO and ENSO Modoki. This mix-up can be alienated by considering Niño 3 index (Figure 1) as a major indicator for ENSO variability and EMI for that of ENSO Modoki (Kao & Yu, 2009; Weng et al., 2007).

Click to view larger

Figure 1. The boxes outlining the domains of EMI, Niño4, Niño3.4, Niño3 & Niño 1&2. Those are the indices commonly used to describe the variations of ENSO Modoki and ENSO.

The differences in those two modes are further elucidated through simple statistical analyses of the principal components of the two EOF modes. For example, correlation between Niño 3 index and principal component 1 (PC1) of EOF1 is found to be very high (0.98), whereas a correlation between Niño 3 index and principal component 2 (PC2) is negligible (Ashok et al., 2007). Similarly, the correlation between EMI and PC1 is negligible, whereas the correlation between EMI and PC2 is close to 0.91. Ashok et al. (2007) concluded that the simple analysis separates the two dominant modes of the tropical Pacific and proves EMI as an essential indicator of the second mode of variability represented by ENSO Modoki. Based on a cross-correlation analysis among several other indices, Ashok et al. (2007) has pointed out that the SST anomaly of western tropical Pacific is as important as that of the eastern Pacific for the ENSO Modoki index. This viewpoint to formulate the ENSO Modoki index is quite different compared to some other indices, such as the Trans-Niño index (TNI), defined as the difference between normalized SSTA between Niño 4 (Figure 1) and Niño1+2 (Trenberth & Stepaniak, 2001). In their analysis, Ashok et al. (2007) found that the correlation between western tropical Pacific SSTA and EMI is 0.51, which is higher in magnitude than a corresponding correlation of 0.25 with TNI. Some studies have also used Niño 4 index to define the CP-El Niño (Kug, Jin, & An, 2009; Yeh et al., 2009), which is sometimes used to classify the second mode of SST variability. The Niño 4 index also shares a dominant portion of the geographical domain used to define CP pole of EMI (Figure 1). Since CP pole contributes significantly to EMI, it is natural to find that majority of the ENSO Modoki variability could be captured by Niño 4 index. Hence, these indices may capture part of the ENSO Modoki variability because of their strong similarity and association with EOF2. But, EMI by far is a better index for capturing the unique tripolar pattern of the ENSO Modoki phenomenon, statistically as well as dynamically.

Click to view larger

Figure 2. Time-series of EMI (bar) and Niño3 (green line) indices averaged for June–August. The cyan and magenta lines represent ± 1 standard deviations for EMI and Niño3 respectively.

Click to view larger

Figure 3. Time-series of EMI (bar) and Niño3 (green line) indices averaged for December–February. The cyan and magenta lines represent ±1 standard deviations for EMI and Niño3 respectively.

## Seasonality

The seasonal phase locking of its peak phase to boreal winter is a remarkable feature of ENSO. Though ENSO events develop irregularly over the years, with 2–7 year periodicity, each of them follows quite a regular seasonal evolution cycle; usually developing during boreal summer and peaking during boreal winter (Larkin & Harrison, 2002; Rasmusson & Carpenter, 1982). The seasonal synchronization is a defining characteristic of climate phenomena in understanding their underlying development mechanisms and predictability. Such seasonal phase-locking features are observed in many other climate phenomena; the Indian Ocean Dipole (IOD; Saji, Goswami, Vinayachandran, & Yamagata, 1999; Yamagata, Behera, Luo, Masson, Jury, & Rao, 2004), the subtropical Indian Ocean Dipole (Behera & Yamagata, 2001), the Atlantic Niño, and even in the recently discovered Ningaloo Niño (Feng, McPhaden, Xie, & Hafner, 2013; Kataoka, Tozuka, Behera, & Yamagata, 2013), California Niño (Yuan & Yamagata, 2014), and Dakar Niño (Oettli, Morioka, & Yamagata, 2016).

Weng et al. (2007) discussed the phase-locking features of ENSO Modoki in detail by analyzing both ENSO and ENSO Modoki indices. They have found that the El Niño events intensify in late summer to early fall, peak in winter, and decay in the following spring as has been already discussed in the literature. However, the EMI does not show such a clear phase locking to a seasonal peak compared with that of the Niño 3. In most of the strong ENSO Modoki events, such as in 1994–1995, 2002–2003, 2004–2005, and 2009–2010, the EMI amplitudes are quite similar in both summer and winter seasons. They have also found that the EMI exhibits large decadal variability particularly during first halves of the 1990s and 2000s. This is noticeable in Figure 2 and Figure 3, which show the seasonally averaged time series of EMI (in bars) for boreal summer and winter seasons. It may be noted, however, that a weaker decadal phase is observed in the summertime plot of the second half of 1980s (with three consecutive weak positive events), but it is not as clear in the wintertime plot. It is also found that the summertime correlation of the EMI with the Niño 3 is almost negligible (0.1) compared to that of the wintertime. This is quite understandable because of the strong seasonal phase locking of ENSO, the appearance of ENSO Modoki sometimes gets blended with it in winter time rather than during the summer time (Table 1).

## Typical Patterns in Boreal Summer and Winter

The characteristics of ENSO and ENSO Modoki phenomena are rooted in their associations with the annual cycle. As discussed in “Seasonality,” the ENSO Modoki events are quite distinct in both boreal summer and winter seasons. Unlike the ENSO events, which typically develop in boreal summer and mature in boreal winter, ENSO Modoki events do not always continue into boreal winter after appearing in boreal summer. Here, the years related to ENSO and ENSO Modoki events are separated based on their seasonal variability during boreal summer and winter to further distinguish the differences in the characteristics of the two phenomena. The years that exceed one standard deviation of types of the phenomena are listed in their respective categories in Table 1. The years that exceed one standard deviation for the respective season and respective time series of ENSO and ENSO Modoki are picked to capture the strong events with characteristic patterns. While most of the ENSO Modoki years are quite distinct, compared to ENSO years, some of the years of ENSO Modoki, particularly during winter season, are blended with ENSO events. These blended events appear as an extended ENSO event with central Pacific intensifications. The SST anomaly pattern shows a distinct warming (Figure 4) in the central Pacific with an extended warming in the eastern Pacific, but it is less intense compared to a conventional El Niño event (Figure 5c and Figure 6c). A typical example of this kind of blended event is the El Niño/El Niño Modoki of 2009 (Figure 4). Some studies referred this pattern of El Niño Modoki as El Niño Modoki I pattern (Wang & Wang, 2013).

Click to view larger

Figure 4. December–February SST and 850hPa wind anomalies of 2009–2010 El Niño/El Niño Modoki event.

The characteristic ENSO Modoki pattern is clearly represented in the composites of boreal summer (Figure 5a and Figure 5b). The typical tripolar structure is clearly depicted in both El Niño Modoki and La Niña Modoki anomaly patterns. The distinct opposite anomaly pattern on either side of the basin in relation to the central Pacific means that the South American coasts will experience an El Niño-like condition during the La Niña Modoki and a La Niña-like condition during El Niño Modoki. So, it will have disastrous effects if one tries to predict the coastal conditions of the American coasts based on an ENSO index from the central Pacific such as the Niño 4. On the contrary, in conventional ENSO events as anticipated from the patterns seen in Figure 5c and Figure 5d, the central Pacific index such as Niño 4 would do a good job to capture the anomalies in the coastal and near coastal regions of eastern Pacific. Therefore, it is important to comprehend both phenomena in their totality for better predictability and applicability in societal applications.

Table 1. The ENSO and ENSO Modoki Years Since 1979. Modoki years marked by an asterisk indicate ENSO and ENSO Modoki blended events with large amplitudes in central Pacific together with less than typical amplitudes of ENSO in the eastern Pacific

June–August

December–February

El Niño Modoki

1990, 1991, 1994, 2002, 2004

1990–1991, 1991–1992*, 1994–1995, 2002–2003, 2004–2005, 2009–2010*

La Niña Modoki

1983, 1984*, 1988*, 1998, 1999*, 2008

1988–1989, 1998–1999*, 1999–2000*, 2000–2001, 2007–2008*, 2008–2009, 2010–2011*, 2011–2012

El Niño

1983, 1987, 1997, 2009, 2012

1982–1983, 1986–1987, 1991–1992, 1997–1998, 2009–2010

La Niña

1984, 1985, 1988, 1999, 2007, 2009, 2010, 2013

1984–1985, 1988–1989, 1998–1999, 1999–1900, 2007–2008, 2010–2011

The SST anomalies in the mid-latitudes are quite different for El Niño and El Niño Modoki. Warm SST anomalies prevail in seas around Japan and in the Kuroshio extension regions during El Niño Modoki events, while cold SST anomalies prevail in those regions during El Niño. The differences are not so prominent in the mid-latitudes of Southern Hemisphere. On the other hand, the La Niña Modoki-associated SST anomaly patterns are quite different compared to that of La Niña in both hemispheres. The boomerang pattern of cold anomalies emanating from the central Pacific in La Niña Modoki covers much wider areas in both hemispheres (Figure 5b). The lower tropospheric winds overlaid on the SST anomalies also corroborate the distinct ocean-atmosphere conditions related to ENSO Modoki. The El Niño Modoki composite (Figure 5a) shows convergence of wind anomalies over the central Pacific on top of the warm SST anomalies. Wind anomalies diverge from the central Pacific in case of a La Niña Modoki. Compared to this central Pacific convergence/divergence, westerly/easterly anomalies prevail over the whole tropical Pacific in the composites of El Niño/La Niña (Figure 5c and Figure 5d).

Click to view larger

Figure 5. June–August composites of SST and wind anomalies for (a) El Niño Modoki, (b) La Niña Modoki, (c) El Niño, and (d) La Niña. The years for composites are picked from Table 1. The composites for El Niño Modoki and La Niña Modoki do not include the years (in orange color) that are common for El Niño and La Niña.

The SST and wind anomalies are stronger in the central Pacific during the Modokis of boreal winter season (Figure 6a and Figure 6b) compared to that of the summer season discussed earlier in this sub-section. On the other hand, the anomalies are seen extended all the way to American coasts for El Niño and La Niña, with further intensification of anomalies near the coastal regions as well as in the interior (Figure 6c and Figure 6d). Looking at the SST anomaly pattern, it is easy to conclude that Niño 3 is an appropriate measure to pick the ENSO events. As reported by previous studies, the coupling between ocean and atmosphere is stronger in El Niño/La Niña events, and that helps in the amplitude intensifications.

ENSO teleconnection also appears prominently in the Bering Sea off Alaska. This is the usual position for the winter-time Aleutian low. From low-level wind anomalies shown in the figure, it is noted that the low intensifies during El Niño, with warm SST anomalies hugging the coasts and cold SST anomalies in the interior. Opposite anomaly patterns are seen during La Niña. This teleconnected pattern is basically similar in La Niña Modoki. However, the SST and wind anomalies are different for El Niño Modoki: The anomalous low is wider, and the near shore wind anomalies are stronger compared to that of El Niño. Hence, the warm SST anomalies near the coast are wider during El Niño Modoki, extending all along the coast associated with poleward wind anomalies along the coasts. The poleward wind anomalies would favor down-welling (a process that helps to accumulate warm waters near the coast) and hence produce the warm SST anomalies. On the other hand, the wind anomalies are weaker in the interior ocean in the El Niño Modoki case; as a result, the SST anomalies are quite diffused. Actually, the boomerang pattern of the cold anomalies seen in El Niño case is not so prominent in the composite of El Niño Modoki.

Click to view larger

Figure 6. December–February composites of SST and wind anomalies for (a) El Niño Modoki, (b) La Niña Modoki, (c) El Niño, and (d) La Niña. The years for composites are picked from Table 1. The composites for El Niño Modoki and La Niña Modoki do not include the years (marked by an asterisk) that are common for El Niño and La Niña.

## Formation Mechanism

The formation mechanism of ENSO Modoki is better understood in the context of the ENSO mechanism. The coupled feedback process proposed by Bjerknes (1969) to explain the mechanism of ENSO has become a standard to define ocean-atmosphere coupled climate phenomena. In that classical hypothesis, Bjerknes suggested that the basin-scale climate anomalies associated with Southern Oscillation are closely associated with observed SST anomalies in the equatorial eastern Pacific. The Southern Oscillation is the atmospheric see-saw between Darwin and Tahiti discovered earlier by Walker (1924), and it is a consequence of vertical circulation over the equatorial Pacific—known as the Walker circulation (Figure 7). In a normal year, the Walker circulation consists of surface winds that blow from the east to the west across the tropical Pacific Ocean, and the accumulated air rises in the tropical western Pacific giving rise to atmospheric convection, clouds, and rainfall there. The rising air then feeds into the upper-level winds that blow from the west to the east, and the accumulated air in the upper troposphere sinks in the eastern tropical Pacific to complete the vertical circulation cell. Associated with the surface easterly winds of the Walker circulation, the surface and subsurface ocean produce a cooler SST and shallower thermocline (lower heat content) in the east and warmer SST and deeper thermocline (higher heat content) in the west as depicted in the schematic diagram based on the observation (Figure 7). The resultant east-west thermal gradient then helps to maintain the surface easterly winds as air blows from colder to warmer surface.

Click to view larger

Figure 7. Schematics of the observed mean ocean and atmospheric conditions of the tropical Pacific Ocean.

In an El Niño year, the easterlies at the surface weaken, giving rise to warmer SST anomaly in the equatorial eastern Pacific, which reduces the east-west SST/thermal gradient and further weakens the surface easterly winds as well as the strength of the Walker circulation. The weaker easterly winds in turn drive the ocean circulation changes that further reinforce the warm SST anomaly in the east and thereby further weaken the east-west thermal gradient in a feedback loop as depicted in the schematic diagram based on the observations (Figure 8a). An opposite situation prevails in the La Niña years, when the normal conditions get strengthened (Figure 8b). The Bjerknes hypothesis is not perfect to explain the full spectrum of ENSO variability (e.g., the turnaround of El Niño to La Niña), but essentially it explains the ocean-atmosphere feedback mechanism behind the development of either of the ENSO phases from a normal state. Several other hypotheses were proposed subsequently to explain the ENSO turnarounds (Jin, 1997; McCreary, 1983; Philander, Yamagata, & Pacanowski, 1984; Wyrtki, 1985; Yamagata, 1985) and processes for the aperiodic variation; for example, “delayed oscillator” (e.g., Battisti & Hirst, 1989; Schopf & Suarez, 1988) and “recharge/discharge oscillator” (Jin, 1997).

Click to view larger

Figure 8. Schematics of the observed anomalous ocean and atmospheric conditions associated with (a) El Niño and (b) La Niña.

Similar to the canonical ENSO, ENSO Modoki events are characterized by interactions among surface wind, SST, and subsurface heat anomalies. But those anomalies essentially remain confined to the central Pacific throughout the event cycle (Ashok et al., 2007; Kao & Yu, 2009; Kug et al., 2009) unlike the basin wide thermocline and surface wind variations associated with ENSO. It is noted that the thermocline feedback is more effective in eastern Pacific associated with El Niño/La Niña, whereas the zonal advection near the dateline is most effective for the development of Modoki events in central Pacific (e.g., Capotondi et al., 2015). Ashok et al. (2007) explained the formation mechanism by correlating EMI with the satellite-derived sea surface height (SSH) anomalies (which represent the subsurface heat content anomalies). In addition, they have regressed the EMI with surface wind anomalies to demonstrate the ocean-atmosphere coupling process. From the lead/lag correlation and regression it is found that the warm signal is excited by westerly wind anomalies in the western Pacific one to two seasons before the peak phase. This helps the El Niño Modoki evolution by transporting the warm water from the off-equatorial regions to the equator through the down-welling equatorial Kelvin waves that subsequently deepen the thermocline in the central Pacific, as depicted in the schematic diagram based on observations (Figure 9). The off-equatorial influence is also reported from north Pacific. Yeh et al. (2015) suggested that the anomalous SST of north Pacific and the associated atmospheric variability are more closely associated with the occurrence of the El Niño Modoki after 1990.

In addition to Ashok et al. (2007), several recent studies based on model experiments have suggested the importance of westerly wind burst (WWB) to induce the El Niño Modoki. Chen et al. (2015) suggested that the asymmetry, irregularity, and extremes of El Niño and El Niño flavors result from westerly wind bursts in the equatorial Pacific. Based on a heat budget analysis of WWB-forced model experiments, they have found that the zonal and meridional advections are most important in the development of El Niño Modokis. This is unlike El Niño, where the vertical advection plays an important role. In a similar study, based on coupled general circulation model experiments, Fedorov, Hu, Lengaigne, and Guilyardi (2015) found that a recharge state (in which the initial ocean heat content of the system is higher than the model climatology) can develop into either an El Niño Modoki (following a weak wind burst) or El Niño (following a stronger wind burst). On the other hand, started from a neutral state (with a near normal state of initial heat content), the model produces typically an El Niño Modoki event after a strong wind burst. In another study, Lee, Yeh, and Jo (2017) also found that the atmospheric weather noise may play a more important role than the climatological mean state in the increased frequency of El Niño Modoki occurrences since the 1990s.

After the initiation of ENSO Modoki events, the relationship among SST, wind, and SSH anomalies becomes stronger (see Figure 6 of Ashok et al., 2007) in subsequent months until those events reach their peak levels (either June-August or December-February). Easterly wind anomalies prevail in the eastern Pacific to the east of the warm SST anomalies, giving rise to the double Walker circulation cells (Figure 9), unlike the customary single Walker circulation cell associated with El Niño. The easterly wind anomalies in the east prevent further spreading of warm ocean waters to the east as usually observed during El Niño. Therefore, cold SST anomalies prevail in the coastal regions of South America and eastern Pacific during El Niño Modoki. After the peak phase, anomalous easterlies in the eastern Pacific are strengthened, and the equatorial upwelling is also strengthened. The associated down-welling Rossby waves propagate west. Together with the weakening of westerlies in the western Pacific, the down-welling Rossby waves smear out the cold anomaly in the western Pacific and eventually terminate the Modoki event. The mechanism for La Niña Modoki is essentially opposite of that explained here for El Niño Modoki.

Click to view larger

Figure 9. Schematics of the observed anomalous ocean and atmospheric conditions associated with (a) El Niño Modoki and (b) La Niña Modoki.

The ENSO Modoki is expected to have a higher mode of ocean-atmosphere coupling, compared to the gravest mode associated with ENSO, due to the double cell structure in the vertical atmospheric circulation and the tripole structure in SST anomalies associated with the phenomenon. Since the coupled SST and wind anomalies do not propagate beyond the central Pacific, the extent of the coupling with subsurface is less deep in the case of the ENSO Modoki. In an El Niño event, on the other hand, the extent of the coupling in the subsurface is deeper, and as a result there is a great amount of poleward heat discharge from the equator during the event. The associated dynamical feedback helps for a quick transition to a La Niña from El Niño as per the recharge oscillator paradigm (Jin, 1997). In contrast, owing to shallower coupling in the subsurface, during El Niño Modoki, the discharge process is much weaker than a canonical El Niño. In addition, the SST anomaly during El Niño Modoki is thermally damped via evaporative cooling (Kug et al., 2009). Therefore, there is less likelihood of observing a La Niña Modoki event following an El Niño Modoki event. Moreover, because a La Niña Modoki usually does not succeed an El Niño Modoki, as generally a La Niña succeeds a canonical El Niño event, it is believed that El Niño Modoki occurrences would contribute to the warming of the mean state of the tropical Pacific according to the tendency found in model results (Kug et al., 2009). Indeed, based on several coupled model projections, Yeh et al. (2009) suggested that the occurrence of El Niño Modoki is going to increase under a global warming scenario.

# ENSO Modoki Influences

The importance of studying the difference between El Niño Modoki and the canonical El Niño lies in their unique influences on the surrounding climate. Considering the differences in the atmospheric circulation pattern, it is expected that the impacts arising from both phenomena will be very different. Particularly, the eastern Pacific experiences a completely different response owing to the presence of a double Walker-circulation cell of ENSO Modoki. The associated variations in atmospheric convections are located in the central Pacific. Therefore, the direct and indirect (teleconnection) impacts associated with the atmospheric convections are different between the events over space and time. Particularly, the impacts of the teleconnection arising from the atmospheric diabatic heating in the central Pacific are different between the boreal summer (June-August) and winter (December-February) seasons. Hence, in the following two sub-sections, “Boreal Summer Influences” and “Boreal Winter Influences,” we have discussed the impacts separately for those two seasons. It may be noted that the influences discussed here are mostly based on El Niño Modoki events with an understanding that the influences arising from La Niña Modoki events are basically opposite of what is seen during El Niño Modoki events.

## Boreal Summer Influences

The boomerang pattern in SST anomalies, extending from central Pacific to the coasts of America, associated with El Niño Modoki (Figure 6) causes warmer than normal surface temperature in the western states, while cooler than normal temperature prevails in the central and eastern states of the United States during boreal summer. The persistent summer drought in the western United States is caused not only by below normal rainfall (Figure 10a) but also by higher than normal surface temperature associated with El Niño Modoki summers (Weng et al., 2007). In contrast to El Niño Modoki years, temperature in most areas of the USA, except for the southeastern and northwestern states, is basically cooler than normal during El Niño years (Weng et al., 2007). Therefore, the combined effects of rainfall and temperature anomalies during the El Niño Modoki-related summers exacerbate the drought conditions in the western USA. Opposite conditions are expected during La Niña Modoki. With warmer anomalies in eastern and western tropical Pacific, the influence on the American side is opposite to that of El Niño Modoki and La Niña, with unexpected spells of rainfall.

The atmospheric teleconnection from the central Pacific diabatic heating during El Niño Modoki induces significant cooling in the northeastern parts and near-neutral conditions elsewhere in tropical Atlantic (Amaya & Foltz, 2014). This is in contrast to the generally observed warming subsequent to canonical El Niño events. It is also noted that the difference in SST response stems primarily from Pacific/North American (PNA) teleconnection pattern. A much stronger PNA and a stronger atmospheric Kelvin wave response are evident during canonical events compared to Modoki events. The stronger PNA pattern and associated Kelvin waves during canonical El Niño events generate anomalously weak surface winds in the tropical North Atlantic. This results in anomalously weak evaporative cooling and warmer SSTs there (Amaya & Foltz, 2014). A weaker teleconnection results in weaker SST anomalies during Modoki events. Hence, the lesser warming due to El Niño Modoki is suggested to have lesser impacts on Atlantic hurricane activities compared to El Niño (Larson, Lee, Wang, Chung, & Enfield, 2012). Nevertheless, Kim, Webster, and Curry (2009) suggested that warmer anomalies in the central Pacific, as seen in El Niño Modoki years, have potential for above normal landfall frequencies of hurricanes along the Gulf of Mexico coast and Central America compared to conventional El Niño with a warmer eastern Pacific. Differences are associated with the modulation of vertical wind shear in the main hurricane development region forced by differential teleconnection patterns emanating from the Pacific.

On the Southern Hemisphere, the Central South American coasts experience rainfall anomalies in ENSO Modoki years that are opposite of what is seen in a canonical ENSO (Figure 10). For example, during El Niño years, these otherwise dry coastal regions get abundant rainfall followed by widespread flower blooming in coastal deserts. However, during the El Niño Modoki years, these areas remain dry as a result of cooler than normal coastal waters. In another study, it is found that El Niño Modoki events are associated with a significant rainfall decrease over the northern, central, and western Colombia during both boreal summer and winter seasons like El Niño (Córdoba-Machadoa et al., 2015). The southwestern region of Colombia exhibits an opposite behavior between canonical El Niño and El Niño Modoki, showing different seasonal precipitation response to those two phenomena.

Click to view larger

Figure 10. Boreal summer partial correlation of precipitation anomalies (shading) and partial regressions of the 500 hPa geopotential height (blue contours; unit: m) and 850 hPa wind (red streamlines), with (a) EMI and (b) Niño3. The precipitation correlation coefficients that are not significant at the 90% level are omitted (after Weng, Ashok, Behera, Rao, & Yamagata, 2007).

In the western North Pacific, El Niño Modoki is associated with a positive Pacific-Japan pattern: enhanced western-north Pacific summer monsoon, but weakened East Asian summer monsoon. This dipole in the response causes droughts in most parts of Japan and the central eastern China, while southern China is flooded (Weng et al., 2007; Weng et al., 2009b; Wang & Wang, 2013). The tropical storm activities near Japan and the southeastern United States are expected to be higher during El Niño Modoki events. A typical year was 2004, when many tropical cyclones had landfall over Japan as well as the Korean peninsula. It is also noted that the concurrent occurrence of El Niño Modoki and positive IOD favors a genesis of more number of tropical cyclones over northwest Pacific (Pradhan et al., 2011). Though the correlation between the IOD and ENSO Modoki is weak as noted in Ashok et al. (2007), there are years like 1994 when El Niño Modoki and positive IOD have occurred concurrently.

Like the northwest Pacific, negative rainfall anomalies are seen in several parts of Australia during El Niño Modoki. While canonical El Niños are associated with a significant reduction in rainfall over northeastern Australia, Modoki events appear to drive a large-scale decrease in rainfall over southeastern Australia. This results from a basin-wide low in the mid-latitude with an enhanced Australian high and the eastern South Pacific subtropical high. This atmospheric circulation pattern favors a dry periphery (Weng et al., 2007) around the wet central Pacific, which also displaces the South Pacific Convergence Zone (SPCZ). The rainfall anomaly distribution in Australia and its adjacent region is very sensitive to the displacement of the SPCZ. Hence, as a result of such a displacement, the dryness in eastern Australia and northern New Zealand is expected to be severe in El Niño Modoki compared to El Niño. This otherwise anticipated severe condition may appear moderate in some of the Modoki years, when unusually frequent extratropical storms affect the rainfall during boreal summer, since the rainfall anomaly there is a result of the tropical–extratropical interaction. It is also noted that rainfall variations are more sensitive to the Modoki SST anomaly pattern than the conventional El Niño anomalies during March–May season as discussed in the following sub-section, “Boreal Winter Influences.”

## Boreal Winter Influences

The ENSO Modoki influences in boreal winter months vary from tropical to extratropical regions. Its influence on the extratropical winter hemisphere is attenuated by the influences arising for extratropical storms and other transient disturbances. Moreover, some of the Modoki events are blended with El Niño events in boreal winter. A typical example is the winter months from December 2009 to February 2010, when large parts of the world witnessed extreme conditions. In a study with observational data and atmospheric general circulation model experiments, Ratnam, Behera, Masumoto, Takahashi, and Yamagata (2011) showed that the atmospheric teleconnection arising from the heating associated with the El Niño Modoki in the central tropical Pacific accounted for most of the anomalous conditions observed over southern parts of North America, Europe, and several countries in the Southern Hemisphere.

Based on the reanalysis data for 1979–2005 period, Weng et al. (2009a) observed that more moisture is transported from the tropics to higher latitudes during El Niño Modoki owing to poleward displacement of the wet boomerang arms from the central Pacific. They have also noted stronger interactions between tropical and extratropical systems owing to the discontinuities at outer edges (toward American coastlines) of the boomerang. The PNA pattern and related climate anomalies in North America are different for two phenomena. In the western United States, a dipole response with dry north and wet south is seen during El Niño Modoki while it is wet throughout the western the United States during El Niño (Figure 11). The moisture to the southwestern United States is transported from the northwardly shifted ITCZ during El Niño Modoki, while it is carried by the storms traveling along the polar front jet shifted to south during El Niño.

Weng et al. (2009a) further noted that the location of the warm arm of the north Pacific boomerang during El Niño Modoki is closely related to the path of moisture transport from the tropics to the subtropics. Therefore, the boomerang arm could explain the so-called atmospheric river (Newell, 1992; Ralph, Neiman, & Wick, 2004; Zhu & Newell, 1994) extending from north Pacific to the West Coast of the United States. For example, an atmospheric river seen in early January of 2005 dumped a meter-high measure of rain onto southern California just in a few days, causing widespread floods and massive mudslides (Kerr, 2006). This was an El Niño Modoki winter. This is less likely to occur during an El Niño winter, because the anomalous southwesterlies associated with El Niño event carry the moisture from the eastern tropical Pacific to the Caribbean, the Gulf of Mexico, and the southern and southeastern United States rather than to the southwestern states, as in El Niño Modoki event. On the South American side, under the direct influence of the double Walker cells, the Amazon basin and northern Brazil suffer from drier conditions. Hence, 90% of the extremely low discharge events of the Paranaıba River during the peak streamflow season of December-February are associated with the El Niño Modoki (Sahu et al., 2014).

Click to view larger

Figure 11. Partial regression patterns of anomalous precipitation (shading) and surface wind (stream) with (a) EMI and (b) Niño3 (after Weng, Behera, & Yamagata, 2009a) during boreal winter.

In a recent study, Zhang, Wang, Xiang, Qi, and He (2015) examined the differences in the teleconnections arising from La Niña and La Niña Modoki. They found that, compared to La Niña, the La Niña Modoki has an opposite effect on North Atlantic Oscillation (NAO). The Western European region experiences a cooler and drier than normal winter in response to a negative NAO-like condition during La Niña. However, during Modoki La Niña, a positive NAO-like climate anomaly is observed with a stronger Atlantic jet and a warmer and wetter than normal winter over Western Europe. Their modeling experiments indicated that the contrasting atmospheric anomalies are mainly owing to the differences in SST cooling patterns in tropical central Pacific associated with the two types of La Niña.

On the western side of the Pacific, the ENSO Modoki influence is limited to lower latitudes, including southeastern China, Taiwan, and southern Japan. The climate in northern China is not directly influenced by the ENSO Modoki (Weng et al., 2009a, 2009b). In the Southern Hemisphere, El Niño Modoki events appear to drive a large-scale decrease in rainfall over northwestern and northern Australia. During canonical El Niño events, a significant reduction in rainfall is usually noticed over northeastern and southeastern Australia (Figure 11). In addition, Tashetto and England (2009) found that the rainfall variations during the March-May season are more sensitive to the Modoki SST anomaly compared to the El Niño anomaly. Based on an SVD analysis performed on tropical Pacific SST and rainfall over Australia for the period of 1979–2006, they suggested that the leading mode in this season has a Modoki SST pattern associated with rainfall reduction over eastern Australia. On the other hand, the El Niño influence is more dominant in other seasons, particularly during September-November. Besides the Modoki, the negative rainfall anomalies over southeastern Australia are also related to the IOD (Ashok et al., 2003).

The rainfall variations in southern Africa are traditionally linked with the ENSO variability. Particularly, the austral summer (December-February) rainfall in southern Africa is generally adversely affected by El Niño. Based on the observed data, Ratnam, Behera, Masumoto, and Yamagata (2014) have reported that southern Africa receives significantly below normal rainfall during El Niño events compared to El Niño Modoki events. Though southern Africa as a whole receives below normal rainfall, the reduction in rainfall is not so significant during Modoki event. Those differences in the spatial distribution of rainfall anomalies between the two types of El Niños are rooted in the differences in SST anomaly-linked atmospheric teleconnections. The teleconnections arising from the diabatic heating in the central Pacific during Modoki events do not produce a strong enough surface divergence to significantly reduce the rainfall over southern Africa. On the other hand, the Matsuno-Gill response to atmospheric heating is intense during El Niño events, and that causes stronger teleconnection to southern Africa, anomalous sinking motion because of the stationary wave propagation over southern Africa, and severe drought there (as was the case during the summer, 2015–2016). In addition, the low-level (850 hPa) Matsuno-Gill response to anomalously high precipitation over the Pacific during El Niño events results in an anomalous anticyclone extending from the equatorial south Indian Ocean to subtropical South Indian Ocean. These anomalous anticyclonic winds weaken the subtropical easterlies and transport the tropical moisture out of southern Africa landmass (Ratnam et al., 2014).

The atmospheric teleconnection associated with the El Niño Modoki influences the decadal Antarctic temperature. Ding et al. (2014) have suggested that recent warming in continental West Antarctica is linked to sea surface temperature changes in the tropical central Pacific associated with El Niño Modoki over the past 30 years. An atmospheric Rossby wave response to heating in the central Pacific influences atmospheric circulation over the Amundsen Sea, causing increased advection of warm air to the Antarctic continent. Since no significant trend is observed in the Southern Oscillation Index or other indices of the main ENSO region (that is, the eastern tropical Pacific), they conclude that the ENSO has at most a weak contribution to the warming.

In a similar study, Zubiaurre and Calvo (2012) demonstrated an ENSO Modoki teleconnection to stratosphere. Using the chemistry-climate Whole Atmosphere Community Climate Model, they showed a significant warming in the polar stratosphere of the Southern Hemisphere associated with El Niño Modoki during boreal winter months. The signal propagates downward in early spring. A significant anomalous cooling is present in the model simulations in La Niña Modoki events. In the Northern Hemisphere stratosphere, on the contrary, the anomalous warming is associated with El Niños during boreal winter. The signal is not statistically significant during El Niño Modoki events. These differences are associated with differences in tropical convection and tropospheric teleconnections between events such as the intensification of the Pacific-South American teleconnection pattern and the weakening of the Aleutian Low (Figure 6a and Figure 6c).

# Oceanic Connections

ENSO Modoki is also associated with anomalous oceanic conditions. Behera and Yamagata (2010) have shown that a decadal signal in observed sea level of the tropical Pacific is associated with the decadal variability of ENSO Modoki. The sea level pattern associated with Modoki El Niño is different from that of typical El Niño. Like SST pattern, a tripolar pattern is seen in the sea level anomalies during Modoki events compared to a dipole pattern in conventional ENSO events (Figure 12).

Click to view larger

Figure 12. The boreal winter sea-level anomalies associated with (a) 1997 El Niño and (b) 2002 El Niño Modoki.

During El Niño Modoki, higher than normal sea level prevails in the central Pacific, flanked by lower than normal sea level on either side (Figure 12). On a decadal scale, such a pattern dominated the basin during the early part of this century. The abnormal condition is evidently aided by frequent occurrences of El Niño Modoki events and associated wind convergence to the dateline during 2000–2004 (Behera & Yamagata, 2010). The sea level rise in the central Pacific succeeded a phase of lower than normal sea level associated with La Niña Modoki events toward the end of the last century. The influence can even be seen in remote regions, such as the coasts of California and Mauritius, through atmospheric teleconnections. These variations in the world oceans are in addition to the variations that arise from ENSO (e.g., Cazenave, Minh, & Gennero, 2004; Chambers, Mehlhaff, Urban, Fujii, & Nerem, 2002; Church, White, & Hunter, 2006; Llovel, Cazenave, Rogel, & Berge‐Nguyen, 2009) and IOD. Therefore, it is logical to state that the regional distributions of sea level changes on shorter time scales could be mostly associated with modes of climate variations, particularly variations in regional wind distributions. Accumulation of residuals from such climate events could in fact explain some of the decadal variations that are of immediate concern for some of the island nations.

Like SST and sea level, sea surface salinity varies between El Niño and El Niño Modoki. A recent study points out reduced eastward displacements of the eastern edge of the low-salinity warm pool waters during Modoki events (∼15°) compared to that of the canonical El Niño events (∼30° longitude). The equatorial sea surface freshening in the South Pacific convergence zone is also reduced during Modoki events (Singh, Delcroix, & Cravatte, 2011). Based on a qualitative analysis, they suggest that the changes in zonal currents and precipitation can account for the observed differences in the anomaly patterns of sea surface salinity. The disparity between two types of El Niños is also palpable in their influences on the ocean primary productivity. New production (NP) and total primary production (PP) of chlorophyll-a (Chl-a) in the equatorial Pacific has a better association with El Niño Modoki. During Modoki El Niños, NP, Chl-a, and PP in the central basin are depressed relative to El Niños, and lower values of Chl-a and PP coincide spatially with higher SST anomaly in the central Pacific (Turk, Meinen, Antoine, McPhaden, & Lewis, 2011).

Modoki events are also seen to influence the coastal erosion. Barnard, Allan, Hansen, Kaminsky, Ruggiero, and Doria (2011) discussed the coastal erosion during the 2009–2010 El Niño/El Niño Modoki. They found that the shoreline retreat in that year exceeded the mean by 36%, and absolute shoreline positions were pushed close to or beyond recently recorded shoreline along California beaches. This winter shoreline retreat in the region exceeded the one observed during the El Niño of 1997–1998, which was the strongest El Niño event in recent decades. Observed extreme shoreline retreat in the southern ends of the region is linked to wave direction anomalies. Barnard et al. (2011) also noted elevated wave energy levels that persisted through spring and summer of 2010.

# Conclusion

The ENSO Modoki is a unique ocean-atmosphere coupled phenomenon of the tropical Pacific. The phenomenon only shares a part of its nomenclature to the ENSO. In its dynamics and impacts on land and ocean, ENSO Modoki is distinctly different from its cousin ENSO in the domain of the tropical Pacific. Since the structures and development processes of the two phenomena are different (e.g., the EOF1 and EOF2 of SST anomalies), it is possible that they are not dependent on the residuals of each other to arbitrarily trigger one or the other. Different physical processes and initial conditions lead ENSO Modoki events to amplify in the central Pacific, sometimes peaking twice in a year during boreal summer and winter. Most of the ENSO Modoki events are not a part of ENSO evolution. The central SST warming/cooling and associated zonal wind convergence/divergence on either side of the warming/cooling sometimes sustain the Modoki El Niño/La Niña for quite a long time, and hence the coupled mode appears almost as a standing mode in the central Pacific, without evolving into a full-fledged El Niño/La Niña event.

The direct impact of ENSO Modoki in tropics is due to anomalous twin-cells of Walker circulation. In addition, its impacts are transmitted to higher latitudes through atmospheric teleconnection, as in the case of ENSO. Nevertheless, the teleconnections and the direct influences are quite distinct between the two phenomena. For example, the West Coast of the United States enjoys good rainfall during an El Niño event but suffers from a drought in an El Niño Modoki event. Similarly, the Amazon basin and the adjacent regions of northern Brazil suffer from drier conditions during El Niño Modoki giving rise to some of the extremely low river discharges.

The background mean state is associated with the evolution of both phenomena. A warmer tropical Pacific and a flatter thermocline seem to favor frequent El Niño Modoki compared to El Niño. This has been the case in the recent decades. It remains to be seen how the anthropogenic forcing might drive a change in the dynamics of the basin and hence might contribute to the frequency and amplitude modulations of the two phenomena. In turn, the decadal modulations of El Niño Modoki events, with decades of warm events, are expected to contribute to the tropical warming by reducing the discharge of tropical heat to high latitudes. This could be verified once enough observational data is accumulated in the coming decades.

## References

Amaya, D. J., & Foltz, G. R. (2014). Impacts of canonical and Modoki El Niño on tropical Atlantic SST. Journal of Geophysical Research: Oceans, 119, 777–789.Find this resource:

Ashok, K., Behera, S. K., Rao, S. A., Weng, H., & Yamagata, T. (2007). El Niño Modoki and its possible teleconnection. Journal of Geophysical Research, 112, C11007.Find this resource:

Ashok, K., Guan, Z., & Yamagata, T. (2003). Influence of the Indian Ocean Dipole on the Australian winter rainfall. Geophysical Research Letters, 30.Find this resource:

Ashok, K., & Yamagata, T. (2009). Climate change: The El Niño with a difference. Nature, 461, 481–484.Find this resource:

Barnard, P. L., Allan, J., Hansen, J. E., Kaminsky, G. M., Ruggiero, P., & Doria, A. (2011). The impact of the 2009–10 El Niño Modoki on U.S. West Coast beaches. Geophysical Research Letters, 38, L13604.Find this resource:

Battisti, D. S., & Hirst, A. C. (1989). Interannual variability in a tropical atmosphere-ocean model: Influence of the basic state, ocean geometry and nonlinearity. Journal of Atmospheric Sciences, 46, 1687–1712.Find this resource:

Behera, S., & Yamagata, T. (2010). Imprint of the El Niño Modoki on decadal sea level changes. Geophysical Research Letters, 37, L23702.Find this resource:

Behera, S. K., & Yamagata, T. (2001). Subtropical SST dipole events in the southern Indian Ocean. Geophysical Research Letters, 28(2), 327–330.Find this resource:

Bjerknes, J. (1969). Atmospheric teleconnections from the equatorial Pacific. Monthly Weather Review, 97(3), 163–172.Find this resource:

Cai, W., & Cowan, T. (2009). La Niña Modoki impacts Australia autumn rainfall variability. Geophysical Research Letters, 36, L12805.Find this resource:

Cai, W., Santoso, A., Wang, G., Yeh, S. W., An, S. I., Cobb, K. M., . . . Wu, L. (2015). ENSO and greenhouse warming. Nature Climate Change, 5(9), 849–859.Find this resource:

Capotondi, A., Wittenberg, A. T., Newman, M., Di Lorenzo, E., Yu, J. Y., Braconnot, P., . . . Yeh, S-W. (2015). Understanding ENSO diversity. Bulletin of the American Meteorological Society, 96, 921–938.Find this resource:

Cazenave, A., Minh, K. D., & Gennero, M. C. (2004). Present-day sea level rise: From satellite and in-situ observations to physical causes. In C. Hwang, C. Shum, & J. Li (Eds.), Satellite altimetry for geodesy, geophysics and oceanography. International Association of Geodesy Symposium (Vol. 126, pp. 23–31). New York: Springer.Find this resource:

Chambers, D. P., Mehlhaff, C. A., Urban, T. J., Fujii, D., & Nerem, R. S. (2002). Low-frequency variations in global mean sea level: 1950–2000. Journal of Geophysical Research, 107, (C4), 3026.Find this resource:

Chen, D., Lian, T., Fu, C., Cane, M. A., Tang, Y., Murtugudde, R., . . ., Zhou, L. (2015). Strong influence of westerly wind bursts on El Nino diversity. Nature Geoscience, 8(5), 339–345Find this resource:

Church, J. A., White, N. J., & Hunter, J. R. (2006). Sea-level rise at tropical Pacific and Indian Ocean islands. Global and Planetary Change, 53, 155–168.Find this resource:

Córdoba-Machadoa, S., Palomino-Lemusa, R., Raquel Gámiz-Fortisa, S., Castro-Díeza, Y., & Esteban-Parraa, M. J. (2015). Assessing the impact of El Niño Modoki on seasonal precipitation in Colombia. Global and Planetary Change, 124, 41–61.Find this resource:

Ding, Q., Wallace, J. M., Battisti, D. S., Steig, E. J., Gallant, J. J., Kim, H. J., . . . Geng, L. (2014). Tropical forcing of the recent rapid Arctic warming in northeastern Canada and Greenland. Nature, 509, 209–212.Find this resource:

Fedorov, A. V., Hu, S., Lengaigne, M., & Guilyardi, E. (2015). The impact of westerly wind bursts and ocean initial state on the development and diversity of El Nino events. Climate Dynamics, 44, 1381–1401.Find this resource:

Feng, M., McPhaden, M. J., Xie, S., & Hafner, J. (2013). La Niña forces unprecedented Leeuwin Current warming in 2011. Scientific Reports, 3, 1277.Find this resource:

Jin, F.-F. (1997). An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. Journal of Atmospheric Sciences, 54, 811–829.Find this resource:

Kao, H.-Y., & Yu, J.-Y. (2009). Contrasting eastern-Pacific and central-Pacific types of ENSO. Journal of Climate, 22, 615–632.Find this resource:

Kataoka, T., Tozuka, T. S., Behera, K., & Yamagata, T. (2013). On the Ningaloo Niño/Niña. Climate Dynamics, 43(5–6), 1463–1482.Find this resource:

Kerr, R. A. (2006). Rivers in the sky are flooding the world with tropical waters. Science, 313(5786), 435.Find this resource:

Kim, H.-M., Webster, P. J., & Curry, J. A. (2009). Impact of shifting patterns of Pacific Ocean warming on North Atlantic tropical cyclones. Science, 325, 77–80.Find this resource:

Kim, J.-S., Kim, K.-Y., & Yeh, S.-W. (2012). Statistical evidence for the natural variation of the central Pacific El Niño. Journal of Geophysical Research, 117.Find this resource:

Kosaka, Y., & Xie, S.-P. (2013). Recent global-warming hiatus tied to equatorial Pacific surface cooling. Nature, 501, 403–407.Find this resource:

Kug, J.-S., Jin, F.-F., & An, S.-I. (2009). Two types of El Niño events: Cold tongue El Niño and warm pool El Niño. Journal of Climate, 22, 1499–1515.Find this resource:

Larkin, N. K., & Harrison, D. E. (2002). ENSO warm (El Niño) and cold (La Niña) event life cycles: Ocean surface anomaly patterns, their symmetries, asymmetries, and implications. Journal of Climate, 15, 1118–1140.Find this resource:

Larkin, N. K., & Harrison, D. E. (2005). On the definition of El Niño and associated seasonal average U.S. weather anomalies. Geophysical Research Letters, 32.Find this resource:

Larson, S., Lee, S.-K., Wang, C., Chung, E.-S., & Enfield, D. (2012). Impacts of non-canonical El Niño patterns on Atlantic hurricane activity. Geophysical Research Letters, 39. Find this resource:

Lee, J.-W., Yeh, S.-W., & Jo, H.-S. (2017). Weather noise leading to El Nino diversity in an ocean general circulation model. Climate Dynamics. Advance online publication.Find this resource:

Lee, T., & McPhaden, M. J. (2010). Increasing intensity of El Nino in the central-equatorial Pacific. Geophysical Research Letters, 37.Find this resource:

Llovel, W., Cazenave, A., Rogel, P., & Berge-Nguyen, M. (2009). 2-D reconstruction of past sea level (1950–2003) using tide gauge records and spatial patterns from a general ocean circulation model. Climate of the Past Discussions, 5, 1109–1132.Find this resource:

Luo, J.-J., Sasaki, W., & Masumoto, Y. (2012). Indian Ocean warming modulates Pacific climate changes, Proceedings of the National Academy of Sciences USA, 109(18), 701–718.Find this resource:

McCreary, J. P. (1983). A model of tropical ocean-atmosphere interaction. Monthly Weather Review, 111(2), 370–387.Find this resource:

McPhaden, M. J., Lee, T., & McClurg, D. (2011). El Niño and its relationship to changing background conditions in the tropical Pacific Ocean. Geophysical Research Letters, 38, L15709.Find this resource:

Medhaug, I., Stolpe, M. B., Fischer, E. M., & Knutti, R. (2017). Reconciling controversies about the “global warming hiatus.” Nature, 545, 41–47. Find this resource:

Newell, R. E., Newell, N. E., Zhu, Y., & Scott, C. (1992). Tropospheric rivers? A pilot study. Geophysical Research Letters, 19(24), 2401–2404.Find this resource:

Newman, M., Shin, S.-I., & Alexander, M. A. (2011). Natural variation in ENSO flavors. Geophysical Research Letters, 38, L14705.Find this resource:

Oettli, P., Morioka, Y., & Yamagata, T. (2016). A regional climate mode discovered in the North Atlantic: Dakar Niño/Niña. Scientific Reports, 6.Find this resource:

Philander, S. G. H., Yamagata, T., & Pacanowski, R. C. (1984). Unstable air-sea interactions in the tropics, Journal of Atmospheric Sciences, 41, 604–613.Find this resource:

Pradhan, K. P., Ashok, K., Preethi, B., Krishnan, R., & Sahai, A. K. (2011). Modoki, Indian Ocean Dipole, and western North Pacific typhoons: Possible implications for extreme events. Journal of Geophysical Research: Atmospheres, 116, D18108. Find this resource:

Ralph, F. M., Neiman, P. J., & Wick, G. A. (2004). Satellite and CALJET aircraft observations of atmospheric rivers over the eastern North Pacific Ocean during the winter of 1997/98. Monthly Weather Review, 132, 1721–1745.Find this resource:

Rasmusson, E. M., & Carpenter, T. H. (1982). Variation in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation/El Niño. Monthly Weather Review, 110, 354–384.Find this resource:

Ratnam, J. V., Behera, S. K., Masumoto, Y., Takahashi, K., & Yamagata, T. (2011). Anomalous climatic conditions associated with the El Niño Modoki during boreal winter of 2009. Climate Dynamics, 39(1–2), 227–238. Find this resource:

Ratnam, J. V., Behera, S. K., Masumoto, Y., & Yamagata, T. (2014). Remote effects of El Niño and Modoki events on the austral summer precipitation of Southern Africa. Journal of Climate, 27, 3802–3815.Find this resource:

Risbey, J. S., & Lewandowsky, S. (2017). Climate science: The “pause“ unpacked. Nature, 545, 37–39. Find this resource:

Sahu, N., Behera, S., Ratnam, J. V., Da Silva, R. V., Parhi, P., Duan, W., . . ,. Yamagata, T. (2014). El Niño Modoki connection to extremely low streamflow of the Paranaíba River in Brazil. Climate Dynamics, 42(5–6), 1509–1516.Find this resource:

Saji, N. H., Goswami, B. N., Vinayachandran, P. N., & Yamagata, T. (1999). A dipole mode in the tropical Indian Ocean. Nature, 401, 360–363.Find this resource:

Schopf, P. S., & Suarez, M. J. (1988). Vacillations in a coupled ocean-atmosphere model. Journal of Atmospheric Sciences, 45, 549–566.Find this resource:

Singh, A., Delcroix, T., & Cravatte, S. (2011). Contrasting the flavors of El Niño-Southern oscillation using sea surface salinity observations. Journal of Geophysical Research, 116.Find this resource:

Taschetto, A. S., & England, M. H. (2009). El Niño Modoki impacts on Australian rainfall. Journal of Climate, 22, 3167–3174.Find this resource:

Trenberth, K. E., & Smith, L. (2006). The vertical structure of temperature in the tropics: Different flavors of El Niño. Journal of Climate, 19, 4956–4973.Find this resource:

Trenberth, K. E., & Stepaniak, D. P. (2001). Indices of El Niño evolution. Journal of Climate, 14, 1697–1701.Find this resource:

Turk, D., Meinen, C. S., Antoine, D., McPhaden, M. J., & Lewis, M. R. (2011). Implications of changing El Niño patterns for biological dynamics in the equatorial Pacific Ocean. Geophysical Research Letters, 38, L23603.Find this resource:

Walker, G. T. (1924). Correlations in seasonal variations of weather. I. A further study of world weather. Memoirs Indian Meteorological Department, 24(4), 275–332.Find this resource:

Walker, N. D. (1990). Links between South African summer rainfall and temperature variability of the Agulhas and Benguela current systems. Journal of Geophysical Research, 95, 3297–3319.Find this resource:

Wang, C., & Wang, X. (2013). Classifying El Niño Modoki I and II by different impacts on rainfall in southern China and typhoon tracks. Journal of Climate, 26, 1322–1338.Find this resource:

Wang, G., & Hendon, H. H. (2007). Sensitivity of Australian rainfall to inter–El Niño variations. Journal of Climate, 20, 4211–4226.Find this resource:

Weng, H., Ashok, K., Behera, S. K., Rao, S. A., & Yamagata, T. (2007). Impacts of recent El Niño Modoki on dry/wet conditions in the Pacific Rim during boreal summer. Climate Dynamics, 29, 113–129.Find this resource:

Weng, H., Behera, S. K., & Yamagata, T. (2009a). Anomalous winter climate conditions in the Pacific Rim during recent El Niño Modoki and El Niño events. Climate Dynamics, 32, 663–674.Find this resource:

Weng, H., Wu, G., Liu, Y., Behera, S. K., & Yamagata, T. (2009b). Anomalous summer climate in China influenced by the tropical Indo-Pacific Oceans. Climate Dynamics, 36(3–4), 769–782. Find this resource:

Wyrtki, K. (1985). Water displacements in the Pacific and genesis of El Niño cycles. Journal of Geophysical Research, 90, 7129–7132.Find this resource:

Yamagata, T. (1985). Stability of a simple air-sea coupled model in the tropics. In J. C. J. Nihoul (Ed.), Coupled ocean-atmosphere models (pp. 637–657). Elsevier Oceanography Series. Amsterdam: Elsevier.Find this resource:

Yamagata, T., Behera, S. K., Luo, J-J., Masson, S., Jury, M. R., & Rao, S. A. (2004). The coupled ocean- atmosphere variability in the tropical Indian Ocean. In C. Wang, S. P. Xie, & J. A. Carton (Eds.), Earth’s climate: The ocean-atmosphere interaction (pp. 189–211). Washington, DC: American Geophysical Union.Find this resource:

Yamagata, T., Morioka, Y., & Behera, S. (2015). Old and new faces of climate variations. In S. Behera & T. Yamagata (Eds.), Indo-Pacific Climate Variability and Predictability (pp. 1–23). World Scientific Series on Asia-Pacific Weather and Climate, Vol. 7. Hackensack, NJ: World Scientific.Find this resource:

Yeh, S.-W., Kug, J., Dewitte, B., Kwon, M.-H., Kirtman, P., & Jin, F. F. (2009). El Niño in a changing climate. Nature, 461, 511–514.Find this resource:

Yeh, S.-W., Kirtman, B. P., Kug, J.-S., Park, W., & Latif, M. (2011). Natural variability of the central Pacific El Nino event on multi-centennial timescales. Geophysical Research Letters, 38, L02704.Find this resource:

Yeh, S.-W., Wang, X., Wang, C., & Dewitte, B. (2015). On the relationship between the North Pacific climate variability and Central Pacific El Nino. Journal of Climate, 28, 663–677.Find this resource:

Yuan, C., & Yamagata, T. (2014). California Niño/Niña. Scientific Reports, 4, 4801.Find this resource:

Zhang, W., Wang, L., Xiang, B., Qi, L., & He, J. (2015). Impacts of two types of La Niña on the NAO during boreal winter. Climate Dynamics, 44(5–6). Find this resource:

Zhu, Y., & Newell, R. E. (1994). Atmospheric rivers and bombs. Geophysical Research Letters, 21(18), 1999–2002.Find this resource:

Zubiaurre, I., & Calvo, N. (2012). The El Niño–Southern Oscillation (ENSO) Modoki signal in the stratosphere. Journal of Geophysical Reseach, 117(D4).Find this resource: