Abstract

Previous research has suggested that the anomalous western North Pacific anticyclone (WNPAC) can generally persist from an El Niño mature winter to the subsequent summer, influencing southern China precipitation significantly, where southern China includes the Yangtze River valley and South China. Since the late 1970s, three extreme El Niño events have been recorded: 1982/83, 1997/98, and 2015/16. There was a sharp contrast in the change in southern China rainfall and corresponding atmospheric circulations in the decaying August between the 2015/16 extreme El Niño event and the earlier two extreme El Niño events. Enhanced rainfall in the middle and upper reaches of the Yangtze River and suppressed rainfall over South China resulted from basinwide warming in the tropical Indian Ocean induced by the extreme El Niño in August 1983 and 1998, which was consistent with previous studies. However, an anomalous western North Pacific cyclone emerged in August 2016 and then caused positive rainfall anomalies over South China and negative rainfall anomalies from the Yangtze River to the middle and lower reaches of the Yellow River. Without considering the effect of the long-term global warming trend, in August 2016 the negative SST anomalies over the western Indian Ocean and cooling in the north tropical Atlantic contributed to the anomalous western North Pacific cyclone and a rainfall anomaly pattern with opposite anomalies in South China and the Yangtze River region. Numerical experiments with the CAM5 model are conducted to confirm that cooler SST in the western Indian Ocean contributed more than cooler SST in the north tropical Atlantic to the anomalous western North Pacific cyclone and anomalous South China rainfall.

1. Introduction

El Niño–Southern Oscillation (ENSO), as the most important air–sea interaction system, influences different regions of the globe in a different manner. A previous super El Niño event in 1997/98 attracted worldwide attention and caused widespread and devastating floods over the Yangtze River valley (YRV) and northeastern China. Severe floods in the summer of 1998 afflicted over 220 million people, caused over 3000 deaths, damaged over 30 × 106 acres (1 acre ≈ 4047 m2) of farmland, and inflicted huge economic losses of over $4 billion (USD) (Yuan et al. 2017). Moreover, climate-related natural disasters concurrent with the 1997/98 El Niño event resulted in global losses ranging from $32 to $96 billion (USD) (Glantz 2001). Lessons learned from the 1997/98 El Niño forced governments, researchers, and individuals to discover ways to cope with the recurring natural hazards associated with ENSO, understand ENSO-related societal and environmental impacts, and find effective solutions. Thus, it is urgent that we uncover the mechanisms behind the occurrence, development, and decay of ENSO, quantify the potential direct and indirect effects of ENSO, and improve our corresponding predictions in a changing climate.

A wide variety of media platforms agreed that the 2015/16 El Niño was among the strongest El Niño events on record; it emerged in mid-May 2015, peaked during November 2015–January 2016, and decayed in May 2016 (L’Heureux et al. 2017). The 2015/16 El Niño had an evolution similar to that of past strong El Niño events as well as comparable accumulative intensity and peak intensity, but it differed in its unique atmosphere–ocean feedback mechanisms. L. Chen et al. (2017) found different formation mechanisms in the 2015/16 extreme El Niño and the past extreme El Niño events and suggested two distinctive ways in which extreme El Niño events can form. One is the simultaneous occurrence of an exceptionally strong, positive precursory thermocline depth anomaly and moderate (e.g., in 1982) or strong (e.g., in 1997) westerly wind events. The other is consecutive exceptionally strong westerly wind events that are concurrent with a quite weak or even negative precursory thermocline depth anomaly, which occurred in 2015 and provided a new perspective in operational forecasts. Compared to the 1982/83 and 1997/98 extreme El Niño events, the 2015/16 extreme El Niño showed remarkably above-average SST anomalies in the west-central Pacific, which may be related to the weaker westerly wind anomaly and Bjerknes feedback in the second half of 2015 (Lim et al. 2017). Paek et al. (2017) indicated that the latest 2015/16 extreme El Niño was an equal mixture of the eastern Pacific and central Pacific types of El Niño, which was related to North Pacific Oscillation forcing and the persistence of Pacific meridional mode coupling from January to October 2015. Nevertheless, the 1997/98 extreme El Niño was purely a strong eastern Pacific type, dominated by the traditional delayed oscillator mechanism. Furthermore, the latest two extreme El Niño events had different underlying dynamics and produced different impacts on the climate in the western United States in the decaying spring (Paek et al. 2017).

A large-scale anomalous western North Pacific anticyclone (WNPAC) is considered essential for conveying important impacts of El Niño to East Asia (Zhang et al. 1999; Wang et al. 2000; Wang and Zhang 2002; Wu et al. 2003; Yuan and Yang 2012; Li et al. 2017). A pronounced El Niño–related WNPAC could persist from an El Niño mature winter to the subsequent summer, even if the El Niño dissipates in spring. The WNPAC would enhance precipitation in southeastern China in the El Niño mature winter and the following spring (Zhang and Sumi 2002; Chen et al. 2014) and in the mei-yu–baiu rainband in the El Niño decaying summer (Chang et al. 2000) through modulating the western North Pacific subtropical high and further regulating moisture transport. Furthermore, the northern flank of WNPAC associated with El Niño advances northward from 27°N in June to 38°N in August, which leads to a band of increased rainfall marching northward in each month of a post–El Niño summer (Hu et al. 2017). It is noted that the WNPAC anomaly disappeared unexpectedly in August 2016, which is in contrast to the maintenance of the WNPAC in 1998 or 1983. The difference in the WNPAC evolution implies that the influences of these different extreme El Niño events on southern China rainfall anomalies in the decaying August may not be the same.

The maintenance of the WNPAC from the El Niño decaying from spring to summer is attributed to local air–sea interaction over the western North Pacific (Wang et al. 2000) and the remote forcing of basinwide warming in the tropical Indian Ocean (Yang et al. 2007; Li et al. 2008; Wu et al. 2009; Xie et al. 2009). From June to August, the contribution of western North Pacific cooling to the WNPAC gradually weakens, whereas the contribution of the tropical Indian Ocean warming is enhanced (Wu et al. 2010). The WNPAC is also associated with the north tropical Atlantic SST anomaly (Ham et al. 2013). At this time, the roles of tropical ocean SST anomalies in southern China rainfall and the corresponding atmospheric circulation anomalies are still not clear. Thus, this paper aims to elucidate the differences in rainfall in southern China and related circulation anomalies in the decaying August among extreme El Niño events and identify the relative contributions of key SST regions to them. In this paper, southern China includes the YRV and South China.

The remainder of the paper is organized as follows. Section 2 describes the data, methodology, and model applied in this study. Section 3 addresses the differences in southern China rainfall and circulation anomalies in the decaying August between the 2015/16 extreme El Niño and the earlier two extreme El Niño events. We confirm the contribution of the SST anomalies in the western Indian Ocean (WIO) and north tropical Atlantic (NTA) to southern China rainfall anomalies and western North Pacific circulation anomalies in section 4. Section 5 discusses plausible reasons for SST change in WIO in the decaying summer of the 2015/16 El Niño. A summary is given in section 6.

2. Datasets, methods, and model

The monthly mean CPC Merged Analysis of Precipitation (CMAP) used in this study is provided by NOAA/OAR/ESRL (1997) and is available at a 2.5° × 2.5° resolution from 1979 to the present (Xie and Arkin 1997). Monthly Hadley Centre Sea Ice and SST (HadISST) data are obtained from Met Office (2003) and are available on 1° × 1° resolution grids starting from January 1870 (Rayner et al. 2003). Monthly atmospheric variables are derived from the Japanese 55-year Reanalysis (JRA-55) dataset with a horizontal resolution of 1.25° × 1.25° starting in 1958 (JMA 2009; Kobayashi et al. 2015). To confirm that the main results obtained in this study are independent of a fixed dataset, the following datasets are also used for cross validation (figure not shown): monthly mean precipitation datasets on 2.5° × 2.5° resolution grids from 1979 to the present, obtained from the Global Precipitation Climatology Project (GPCP; Adler et al. 2003; NOAA/OAR/ESRL 2003a); the monthly ERA-Interim dataset at a 1° × 1° resolution from 1979 to the present, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF 2011; Dee et al. 2011); and monthly NOAA Extended Reconstruction SST, version 3b (ERSST.v3b; NOAA/OAR/ESRL 2003b; Smith et al. 2008) on 2° × 2° resolution grids from 1854 to the present. Monthly shortwave radiation, longwave radiation, sensible heat flux, and latent heat flux data provided by JRA-55 are also used in this study. Monthly oceanic current and ocean subsurface temperature are obtained from NCEP Global Ocean Data Assimilation System (GODAS) ocean reanalysis from 1980 to 2016 on a grid of 0.333° latitude × 1.0° longitude resolution (NOAA/OAR/ESRL 2004).

The period of 1979–2016 is analyzed in this study. All monthly anomalies are computed as departures from a 1981–2010 monthly mean climatology. Then the long-term linear trends of all variables are removed to avoid any possible impact of global warming. Regression and correlation analyses are applied to examine the relationship between pairs of variables. To assess the statistical significance of these analyses, the two-sided Student’s t test is adopted.

Several climatic indices are used in this analysis. The variability of the SST anomaly in the equatorial eastern Pacific, equatorial central Pacific, and equatorial east-central Pacific is measured by the Niño-3 index (5°S–5°N, 90°–150°W), Niño-4 index (5°S–5°N, 160°E–150°W), and Niño-3.4 index (5°S–5°N, 120°–170°W), respectively. The eastern Pacific El Niño and central Pacific El Niño are described by the Niño-3 index and El Niño Modoki index (EMI), respectively. The EMI is defined as [SSTA]C − 0.5[SSTA]E − 0.5[SSTA]W, where [SSTA]C, [SSTA]E, and [SSTA]W represent the area-mean SST anomalies over the central (subscript C: 10°S–10°N, 165°E–140°W), eastern (subscript E: 15°S–5°N, 110°–70°W), and western (subscript W: 10°S–20°N, 125°–145°E) Pacific, respectively (Ashok et al. 2007; Weng et al. 2007, 2009; Wang and Wang 2014). The correlation between Niño-3 and EMI in winter during 1979–2014 is 0.435, significantly exceeding the 99% confidence level. Partial regression {} and partial correlation {} analysis can be used to measure the relationship between two variables without the effect of other variable.

To quantitatively estimate the contributions of different processes to SST changes in the western Indian Ocean, a mixed layer heat budget analysis has been performed in the present study, following He and Wu (2013a,b). To demonstrate the influence of underlying SST anomalies over key regions on atmospheric circulation and rainfall, we use the Community Atmosphere Model, version 5 (CAM5), the atmospheric component of the Community Earth System Model, version 1.0.6 (CESM1), with a resolution of 1.25° longitude × 0.9° latitude. The state-of-the-art computer simulations of Earth’s past, present, and future climate states can be reproduced by CESM1 as a fully coupled global climate model. CAM5 is the latest in a series of global atmospheric models and has been improved substantially in physical processes, such as the moist turbulence scheme, shallow convection scheme, cloud microphysics scheme, and so on (Neal et al. 2012). Numerical experiments using CAM5 are run for 22 years with global SST forcing. The ensemble mean outputs over the last 20 years are analyzed in this study, following Chen et al. (2014) and J. P. Chen et al. (2017).

3. Southern China rainfall anomalies and circulation systems in extreme El Niño events

a. Southern China rainfall anomalies in extreme El Niño events

To capture the delayed effects of an extreme El Niño on southern China rainfall and associated atmospheric circulation and oceanic conditions, we first examine all extreme El Niño events since 1979 in which the winter [December–February (DJF)] mean Niño-3 and Niño-3.4 indices exceed two standard deviations. There are three extreme El Niño cases (1982/83, 1997/98, and 2015/16) during 1979–2016 (Fig. 1). It is noted that the Niño-4 index exceeds 1.5 standard deviations (SD) in the winter of 2015 and the EMI reaches 0.85 SD. However, in the other two extreme El Niño cases, the Niño-4 index is less than one SD and the EMI shows negative values. This means that in the peak phase of the 2015/16 extreme El Niño, warmer SST anomalies extended more westward than in the other two cases. To exclude the influence of the central Pacific El Niño, the individual effect of the eastern Pacific El Niño is evaluated by partial regression analysis of atmospheric and oceanic variables upon the Niño-3 index during 1979–2014.

Fig. 1.

Time series of normalized (a) Niño-3 (red line) and Niño-3.4 (black line) indices and (b) Niño-4 index (blue line) and EMI (green line).

Fig. 1.

Time series of normalized (a) Niño-3 (red line) and Niño-3.4 (black line) indices and (b) Niño-4 index (blue line) and EMI (green line).

Figure 2b displays lead–lag regression coefficients of sea level pressure anomalies over the western North Pacific (10°–30°N, 120°–170°E) upon the Niño-3.4 and Niño-3 indices for December(0)–January(1)–February(1) during 1979–2014, respectively. The El Niño–developing and El Niño–decaying years are symbolized with (0) and (1) following the month, respectively. It can be seen that remarkable El Niño–induced positive anomalies of sea level pressure over the western North Pacific persist from November of the developing year to August of the decaying year. Lead–lag partial regression analysis onto Niño-3 index for winter during 1979–2014, excluding the EMI effect, captures similar features. This is consistent with previous results regarding the prolonged effect of El Niño on an anomalous high pressure system over the western North Pacific from late fall to the following summer (e.g., Zhang et al. 1996; Wang et al. 2000; Wu et al. 2003; Xie et al. 2009). In the 1982/83 and 2015/16 extreme El Niño events, anomalous western North Pacific high pressure developed rapidly in November, one month later than in the 1997/98 extreme El Niño. Then the anomalous western North Pacific high pressure persisted through the following spring until the following August, except in the 2015/16 extreme El Niño (Fig. 2a). ENSO could affect the East Asian climate via the Pacific–East Asian teleconnection (Zhang et al. 1999; Wang et al. 2000; Wang and Zhang 2002; Wu et al. 2003; Yuan and Yang 2012). In August 2016, an anomalous enhanced western North Pacific low pressure appeared. This implies that southern China rainfall anomalies in the decaying August of the 2015/16 extreme El Niño were quite different from those in the other two extreme El Niño cases, which will be confirmed by Fig. 3.

Fig. 2.

(a) Evolution of sea level pressure anomalies (hPa) over the western North Pacific (10°–30°N, 120°–170°E) from September to the following August in extreme El Niño years (2015/16, 1997/98, and 1982/83). (b) Lead–lag regression coefficients of sea level pressure anomalies (hPa) over the western North Pacific onto Niño-3.4 (red dashed line) and Niño-3 (blue dotted line) indices for December(0)–January(1)–February(1) during 1979–2014, respectively, and lead–lag partial regression coefficients of sea level pressure anomalies (hPa) over the western North Pacific onto Niño-3 (black solid line) index for December(0)–January(1)–February(1) during 1979–2014, excluding the EMI effect. The open circles, open circles with bars, and filled circles denote regression coefficients exceeding the 90%, 95%, and 99% confidence levels based on the Student’s t test, respectively.

Fig. 2.

(a) Evolution of sea level pressure anomalies (hPa) over the western North Pacific (10°–30°N, 120°–170°E) from September to the following August in extreme El Niño years (2015/16, 1997/98, and 1982/83). (b) Lead–lag regression coefficients of sea level pressure anomalies (hPa) over the western North Pacific onto Niño-3.4 (red dashed line) and Niño-3 (blue dotted line) indices for December(0)–January(1)–February(1) during 1979–2014, respectively, and lead–lag partial regression coefficients of sea level pressure anomalies (hPa) over the western North Pacific onto Niño-3 (black solid line) index for December(0)–January(1)–February(1) during 1979–2014, excluding the EMI effect. The open circles, open circles with bars, and filled circles denote regression coefficients exceeding the 90%, 95%, and 99% confidence levels based on the Student’s t test, respectively.

Fig. 3.

(a) August precipitation anomaly (mm day−1) obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficient exceeds the 90% confidence level. Precipitation anomalies (mm day−1) for August (b) 2016, (c) 1998, and (d) 1983.

Fig. 3.

(a) August precipitation anomaly (mm day−1) obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficient exceeds the 90% confidence level. Precipitation anomalies (mm day−1) for August (b) 2016, (c) 1998, and (d) 1983.

Figure 3a shows partial regression coefficients of the August rainfall anomaly onto the normalized Niño-3 index in the preceding DJF for the period of 1979–2014. There are two obvious anomalous rainfall bands over southern China. One is the southwest–northeast-oriented positive regression from the middle reaches of the YRV to the Huang–Huai River area at the 90% confidence level according to the Student’s t test. The other is the negative regression on the coast of southeastern China. In the 1982/83 and 1997/98 extreme El Niño events, the anomalous August rainfall pattern is nearly consistent with that obtained by partial regression upon the Niño-3 index in the preceding DJF, except for decreasing rainfall in the Huang–Huai River area in the 1982/83 extreme El Niño. However, in the 2015/16 extreme El Niño, the change in August rainfall over southern China is mostly opposite, with the regression rainfall pattern associated with the preceding winter SST anomaly over the equatorial east-central Pacific. In August 2016, an increase in rainfall occurred over South China, while reduced rainfall appeared between the YRV and the middle and lower reaches of the Yellow River. Pronounced differences in August rainfall anomalies in southern China between the 2015/16 extreme El Niño and the other two cases as seen in Figs. 3b–d suggest distinctive circulation systems in August 2016, under the effect of the extreme El Niño.

b. Circulation anomalies in extreme El Niño events

To further understand the connection of southern China August rainfall anomalies and El Niño, ENSO-related SST and 850-hPa wind anomalies in August, obtained by partial regression onto the preceding DJF Niño-3 index, are displayed in Fig. 4a. There is a remarkable anomalous anticyclone over the western North Pacific and southeastern China, which is concurrent with anomalous warmer SST covering the Indian Ocean and western North Pacific. Significant westerly wind anomalies dominate the north Indian Ocean and the western North Pacific. These prominent features are consistent with the correlation with the preceding DJF Niño-3.4 SST anomalies obtained by Wu et al. (2003). The predominant changes in lower-level circulation over the western North Pacific and southern China and Indo-Pacific SST related to El Niño can be identified clearly in August 1983 and 1998, although with different magnitudes (Figs. 4c,d). In August 1998, an anomalous WNPAC is stronger and SST anomalies over the southeastern Indian Ocean and western North Pacific are warmer than in 1983. However, anomalous August SST warming over the western Indian Ocean in 1998 is weaker than in 1983. It is unexpected that some common characteristics linked to El Niño, such as the WNPAC and basinwide Indian Ocean warming, are markedly different from those in the decaying August of the 2015/16 extreme El Niño. In August 2016, enhanced cyclonic wind anomalies are observed over South China and the western North Pacific. Then the equatorial eastern Indian Ocean and western Pacific are dominated by westerly wind anomalies, while easterly wind anomalies occupy the equatorial WIO (Fig. 4b). It is noted that the 2015/16 extreme El Niño is followed by warmer SST anomalies over the southeastern Indian Ocean and cooler SST anomalies over the WIO and NTA in the decaying August, instead of the El Niño–induced basinwide Indian Ocean warming indicated by previous research (e.g., Du et al. 2009; Xie et al. 2009; Yuan et al. 2012; Xie et al. 2016). Moreover, negative SST anomalies dominate the NTA in August 2016, whereas positive SST anomalies over the north tropical Atlantic are observed by partial regression onto the Niño-3 index. Another notable difference is that an obvious anomalous anticyclone over the tropical eastern Pacific appears in the decaying August of the 2015/16 extreme El Niño but not in the other two cases. The anomalous western North Pacific cyclone is concurrent with the anomalous anticyclone over the tropical eastern Pacific in August 2016, which seems to be a response to SST anomalies over the NTA. Corresponding to cooler SST anomalies over the NTA are anomalous lower-level divergence and descending motion at 500 hPa (figure not shown). Ham et al. (2013) suggested that positive SST anomalies over the north tropical Atlantic during spring and summer could induce a low-level subtropical teleconnection with a cyclonic atmospheric flow over the eastern Pacific and an anticyclonic flow over the western Pacific. Thus, the negative SST anomalies over the WIO and NTA appear to exert combined effect on the anomalous western North Pacific cyclone in August 2016.

Fig. 4.

(a) August anomalies of SST (shading, °C) and 850-hPa winds (vectors, m s−1) obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficients exceed the 90% confidence level. Only the regression values of wind exceeding the 90% confidence level are plotted. Anomalies of SST (shading, °C) and 850-hPa winds (vectors, m s−1) for August (b) 2016, (c) 1998, and (d) 1983. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

Fig. 4.

(a) August anomalies of SST (shading, °C) and 850-hPa winds (vectors, m s−1) obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficients exceed the 90% confidence level. Only the regression values of wind exceeding the 90% confidence level are plotted. Anomalies of SST (shading, °C) and 850-hPa winds (vectors, m s−1) for August (b) 2016, (c) 1998, and (d) 1983. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

In August 2016, on the northwestern flank of the anomalous cyclone are anomalous northeasterly winds over South China, which substantially suppress water vapor transport from the South China Sea and western North Pacific to the Yangtze–Yellow River region (Fig. 5b). In the midtroposphere, an anomalous ascent is observed over the western North Pacific and South China and the anomalous descending motion extends from the East China Sea to the Yangtze–Yellow River region. These anomalous circulations play an important role in a south–north contrast in rainfall anomalies, with above-normal rainfall over South China and below-normal rainfall in the Yangtze–Yellow River region in August 2016. In contrast, in August 1983 and 1998 an anomalous anticyclone occurs over the western North Pacific and South China and then transports more moisture north of the YRV, forming an opposite rainfall pattern in which South China is dry and the middle reaches of the YRV are wet. This conforms to the results obtained by partial regression onto the Niño-3 index in the preceding winter. It can be concluded that these opposite changes in the western North Pacific circulations are responsible for the salient difference in southern China rainfall anomalies in the decaying August between the 2015/16 extreme El Niño and the other two cases.

Fig. 5.

(a) August anomalies of 500-hPa vertical velocity (shading, 10−2 Pa s−1) and vertically integrated water vapor flux [vectors, kg (m2 s)−1] from the surface to 300 hPa obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficients of SST exceed the 90% confidence level. Only the regression values of vertically integrated water vapor flux exceeding the 90% confidence level are plotted. The August anomalies of 500-hPa vertical velocity (shading, Pa s−1) and vertically integrated water vapor flux [vectors, kg (m2 s)−1] from the surface to 300 hPa for (b) 2016, (c) 1998, and (d) 1983. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

Fig. 5.

(a) August anomalies of 500-hPa vertical velocity (shading, 10−2 Pa s−1) and vertically integrated water vapor flux [vectors, kg (m2 s)−1] from the surface to 300 hPa obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Dots denote where the regression coefficients of SST exceed the 90% confidence level. Only the regression values of vertically integrated water vapor flux exceeding the 90% confidence level are plotted. The August anomalies of 500-hPa vertical velocity (shading, Pa s−1) and vertically integrated water vapor flux [vectors, kg (m2 s)−1] from the surface to 300 hPa for (b) 2016, (c) 1998, and (d) 1983. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

Corresponding to the basinwide warming in the Indian Ocean associated with El Niño, anomalous midtropospheric ascending motion and upper-level divergence are seen over the tropical Indian Ocean. The divergent wind flow from the tropical Indian Ocean to the western North Pacific, where they converge with those from South China. This contributes to the development of the WNPAC (Fig. 6a). The abovementioned circulation anomalies are very similar to those in August 1983 and 1998, but the opposite of those in August 2016 (Figs. 6b–d). During the decay phase of El Niño, basinwide warming in the Indian Ocean is often accompanied by the WNPAC. Positive SST anomalies over the tropical Indian Ocean can trigger the anomalous anticyclone over the western North Pacific as a Kelvin wave response, acting like a capacitor to prolong the delayed effect of El Niño (Xie et al. 2009; Wu et al. 2010). Alternately, positive SST anomalies over the north Indian Ocean contribute to the anomalous anticyclone over the western North Pacific by an anomalous zonal overturning circulation between the north Indian Ocean and the South China Sea–Philippine Sea (He and Wu 2014).

Fig. 6.

(a) August anomalies of velocity potential (contours, 106 m2 s−1) and corresponding divergent winds (vectors, m s−1) at 200 hPa obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Shading denotes the regression coefficients of velocity potential exceeding the 90% confidence level. The anomalies of velocity potential (shading, 106 m2 s−1) and corresponding divergent winds (vectors, m s−1) at 200 hPa for August (b) 2016, (c) 1998, and (d) 1983.

Fig. 6.

(a) August anomalies of velocity potential (contours, 106 m2 s−1) and corresponding divergent winds (vectors, m s−1) at 200 hPa obtained by partial regression onto the normalized Niño-3 index in the preceding DJF, excluding the EMI effect. Shading denotes the regression coefficients of velocity potential exceeding the 90% confidence level. The anomalies of velocity potential (shading, 106 m2 s−1) and corresponding divergent winds (vectors, m s−1) at 200 hPa for August (b) 2016, (c) 1998, and (d) 1983.

In August 2016, in response to cooler SST anomalies over the WIO, the 500-hPa vertical motion displays an anomalous descent (Fig. 5b) and the potential velocity and corresponding divergent winds at 200 hPa show anomalous convergence over the WIO (Fig. 6b), as indicated by Lindzen and Nigam (1987). Anomalous upper-level divergence is seen over the western North Pacific, with northeasterly wind anomalies from the western North Pacific to the tropical Indian Ocean (Fig. 6b). At the lower level, anomalous southwesterly wind flow from the tropical eastern Indian Ocean to the western North Pacific and South China Sea (Fig. 4b). It could be inferred that there is an anomalous southwest–northeast-oriented vertical circulation connecting the descent over the western Indian Ocean to the ascent over the western North Pacific and South China, overlain by an anomalous cyclone with a large extension. These anomalous circulations are similar to that obtained by He and Wu (2014).

4. The contribution of western Indian Ocean and north tropical Atlantic SST anomalies

To confirm the influence of the WIO SST anomalies on the western North Pacific circulations and southern China rainfall, we conduct a sensitivity experiments with SST anomalies specified in the WIO. A control run is carried out with the climatological mean seasonal cycle of SST forcing in the global ocean. The WIO SST-forced run is the same as the control run except that the negative SST anomalies over the WIO (20°S–20°N, 40°–90°E) from June to August are imposed in the climatological mean seasonal cycle of SST (Figs. 7a–c).

Fig. 7.

SST anomaly (°C) distribution in (a) June, (b) July, and (c) August for the WIO SST experiment. (d) Composite differences of precipitation (shading, mm day−1) and 850-hPa winds (vectors, m s−1) in August between the WIO SST experiment and the CAM5 control run. The red letter C represents the anomalous cyclone.

Fig. 7.

SST anomaly (°C) distribution in (a) June, (b) July, and (c) August for the WIO SST experiment. (d) Composite differences of precipitation (shading, mm day−1) and 850-hPa winds (vectors, m s−1) in August between the WIO SST experiment and the CAM5 control run. The red letter C represents the anomalous cyclone.

To highlight how the atmosphere responds to the specified SST forcing, the differences in August rainfall and 850-hPa winds between the SST-forced run and the control run are shown in Fig. 7d. In the WIO SST-forced run, the anticyclonic wind anomalies dominate the WIO, with southeasterly anomalies south of equator and westerly anomalies north of equator. Meanwhile, an anomalous cyclone covers South China, the South China Sea, and the western North Pacific, where an increase in rainfall is seen over South China, coincident with the observation in August 2016. Anomalous easterly winds prevail over South China, the northern Bay of Bengal, and northern India. Then they suppress warm moisture transport from the Indian Ocean and South China Sea to the Yangtze River and induce below-normal rainfall in the middle and lower reaches of the YRV. The center of the negative rainfall anomaly appears in the Yangtze River estuary, which agrees with the observation in August 2016 (Fig. 7d). These changes in circulation and rainfall feature a Kelvin wave response to cooler SST over the WIO as well as the observation.

In addition, cooler SST anomalies over the WIO persist from June to August, but the western North Pacific cyclone emerges in August 2016 and is related to the change in the tropical eastern Indian Ocean SST. In June and July 2016, tropical eastern Indian Ocean SST anomalies are obviously increasing, accompanied by enhanced convection (figure not shown). J. P. Chen et al. (2017) indicated that warmer SST anomalies over the tropical eastern Indian Ocean could induce anomalous descending motion over South China and the western North Pacific. Thus, SST forcing of warmer SST anomalies over the tropical eastern Indian Ocean is conducive to the maintenance of the WNPAC, which has been demonstrated by Chen et al. (2018). However, warmer SST anomalies weaken and anomalous convection is suppressed over the tropical eastern Indian Ocean in August 2016. Therefore, the role of the cooler SST anomalies over the WIO become prominent in the western North Pacific cyclone in August 2016.

The observed August feature shows that the WIO and NTA SST anomalies act in concert in 2016 but not in 1983 and 1998. To quantify the relative contributions of the WIO and NTA SST anomalies to the western North Pacific circulations and southern China rainfall, we conduct another sensitivity experiment with SST anomalies specified in the NTA (Figs. 8a–c). In the NTA SST-forced run, the response of lower-level circulation to negative SST anomalies over the NTA is characterized by a subtropical teleconnection with an anticyclone over the eastern Pacific and a cyclone over the western Pacific. Since the location of the anomalous cyclone is farther east than the observation in August 2016, the increase in rainfall is confined to the southeastern coast of China (Fig. 8d). In contrast, cyclonic winds over southeastern China, the South China Sea, and the western North Pacific are slight weaker and more southeastward than those seen in the WIO SST-forced run. South China (18°–25°N, 108°–120°E) rainfall anomalies in response to the negative WIO SST anomalies are 0.75 mm day−1, which is about 9 times larger than the contribution of the negative NTA SST anomalies. Nevertheless, In the NTA SST-forced run, rainfall anomalies in the middle reaches of the YRV (28°–33°N, 103°–110°E) are −0.94 mm day−1, which is twice as large as the contribution of negative WTO SST anomalies. These numerical experiments demonstrate that the combined effects of cooler SST anomalies over the WIO and NTA contribute predominantly to the western North Pacific cyclone and southern China rainfall anomalies in August 2016. As to the formation of the negative SST anomalies over the NTA, it is still unclear and remains an interesting topic for future study.

Fig. 8.

SST anomaly (°C) distribution in (a) June, (b) July, and (c) August for the NTA SST experiment. (d) Composite differences of precipitation (shading, mm day−1) and 850-hPa winds (vectors, m s−1) in August between the NTA SST experiment and the CAM5 control run. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

Fig. 8.

SST anomaly (°C) distribution in (a) June, (b) July, and (c) August for the NTA SST experiment. (d) Composite differences of precipitation (shading, mm day−1) and 850-hPa winds (vectors, m s−1) in August between the NTA SST experiment and the CAM5 control run. The red letters A and C represent the anomalous anticyclone and cyclone, respectively.

It should be noted that this study does not consider the contributions of intraseasonal oscillations (Chen et al. 2015) and tropical cyclones (Chen et al. 2012; Wang and Wang 2013) to southern China rainfall anomalies in the decaying August of the 2015/16 extreme El Niño. In August 2016, intraseasonal oscillations with strong signals on both 10–30-day and 30–60-day time scales were extremely active (Yuan et al. 2017; Zhan et al. 2017) and tropical cyclone activity over the western North Pacific was enhanced (Zhan et al. 2017), which was associated with an anomalous cyclonic circulation over the western North Pacific. Zhan et al. (2017) suggested that the enhanced genesis of tropical cyclones during July–August 2016 did not seem to be related to the active phases of the intraseasonal oscillation. The relative contributions of intraseasonal oscillations and tropical cyclones to rainfall anomalies in South China in August 2016 need to be further investigated in future work.

5. Discussion of reasons for SST change in the western Indian Ocean in the decaying summer of the 2015/16 El Niño

The present analysis shows that Indian Ocean SST anomalies in the decaying August of the 2015/16 extreme El Niño are very different from those in the other two extreme El Niño events on record (i.e., the 1982/83 and 1997/98 events). A basinwide warming mode in the Indian Ocean cannot be identified in the decaying August of the 2015/16 extreme El Niño, which deviates from the statistical relationship between ENSO and the Indian Ocean SST. To understand the role of different terms in the SST changes, we show relative contributions of anomalies of surface heat flux, ocean horizontal advection, and vertical advection in Fig. 9a. For the warming of the WIO (20°S–8°N, 40°–80°E) in August 1983 and 1998, the net surface heat flux has the largest contribution in August 1983 and 1998. In terms of surface heat flux anomalies (Fig. 9b), it can be seen that surface latent heat flux has a dominant positive contribution, which is related to the reduced surface winds when anomalous northwesterly winds are opposite the mean southeasterly winds. The shortwave radiation contribution is negative and acts as a damping effect by cloud–radiation feedback, which is consistent with Wu and Yeh (2010). The contributions of the horizontal advection and vertical advection are small. The dominant role of surface heat flux anomalies in SST changes in most of the tropical Indian Ocean has been indicated by Wu et al. (2008) and Wu and Yeh (2010). However, in August 2016, the oceanic dynamic has a larger contribution to the WIO SST change than the net surface heat flux. The ocean horizontal advection has a large positive effect, which is associated with equatorward anomalous oceanic currents leading to anomalous cold meridional advection. The negative vertical advection contribution to the SST changes averaged over the WIO is small. The ocean upwelling has a large positive contribution confined to the east coast of African and near the equator (figure not shown).

Fig. 9.

(a) Relative contributions (%) of anomalies of net surface heat flux (black bar), horizontal advection (red bar), and vertical advection (blue bar) averaged over the WIO (20°S–8°N, 40°–80°E) in August 2016, 1998, and 1983 to SST change. (b) Terms of surface heat flux anomalies (°C month−1) averaged over the WIO in August 2016 (red line with filled circles), 1998 (blue line with open circles), and 1983 (black line). From left to right, the terms are shortwave radiation flux, penetrative shortwave radiation flux, longwave radiation flux, sensible heat flux, and latent heat flux, respectively. The convention for surface heat fluxes is positive values for downward heat fluxes.

Fig. 9.

(a) Relative contributions (%) of anomalies of net surface heat flux (black bar), horizontal advection (red bar), and vertical advection (blue bar) averaged over the WIO (20°S–8°N, 40°–80°E) in August 2016, 1998, and 1983 to SST change. (b) Terms of surface heat flux anomalies (°C month−1) averaged over the WIO in August 2016 (red line with filled circles), 1998 (blue line with open circles), and 1983 (black line). From left to right, the terms are shortwave radiation flux, penetrative shortwave radiation flux, longwave radiation flux, sensible heat flux, and latent heat flux, respectively. The convention for surface heat fluxes is positive values for downward heat fluxes.

Plausible causes for the differences in Indian Ocean SST in the decaying August among extreme El Niño events are discussed based on previous relevant research and the evidence seen in the correlation or regression fields in this study. One factor that may have contributed to the disappearance of the basinwide warming in the Indian Ocean in the decaying August of the 2015/16 extreme El Niño is the westward extension of warmer SST anomalies over the equatorial east-central Pacific in the peak phase. The SST anomaly in the Niño-4 region in the winter of the 2015/16 extreme El Niño was larger than 1°C and become the maximum of 1979–2016. The 2015/16 extreme El Niño is considered a mixture of the eastern Pacific and central Pacific types (Lim et al. 2017; Paek et al. 2017). Moreover, there is a significant negative relationship between equatorial central Pacific SST in the preceding winter and tropical western Indian Ocean SST in August. The negative relationship appears in June (figure not shown) and expands northward in August, as shown in the partial regressions of SST and 850-hPa winds in August onto the normalized EMI in the preceding DJF during 1979–2014, after removing the effect of the Niño-3 index (Fig. 10b). The lower-level wind anomalies feature an anticyclone with southeasterly anomalies over the southwestern Indian Ocean and southwesterly anomalies over the northwestern Indian Ocean, which enhances the prevailing southeasterly trades south of the equator and the prevailing southwesterly trades north of the equator, and then helps to cool the WIO SST through wind–evaporation effect. Thus, negative latent heat flux has a large contribution in the tropical central-western Indian Ocean (figure not shown). The dominant role of surface heat flux anomalies in SST changes in most of the tropical Indian Ocean has been emphasized by Wu et al. (2008) and Wu and Yeh (2010). The anomalous cyclonic winds emerge off the east coast of Africa where the mean thermocline is shallow, and may induce Ekman upwelling in favor of the SST cooling. Meanwhile, the cyclonic winds related to the equatorial central Pacific SST in the preceding winter occur over South China and the South China Sea in August.

Fig. 10.

(a) August climatology of SST (shading, °C) and 850-hPa winds (vectors, m s−1). (b) Partial regressions of SST (shading) and 850-hPa winds (vectors) in August with the normalized EMI in the preceding DJF during 1979–2014, excluding the Niño-3 SST effect. (c) Composite August SST (shading, °C) and 850-hPa wind anomalies (vectors, m s−1) for the three negative WIO SST cases (1992, 1993, and 2002). The dots denote partial regression coefficients and composite anomalies of SST above the 90% confidence level. Partial regression coefficients and composite anomalies of the winds exceeding the 90% confidence level are shown with thick vectors. The red letter C indicates the cyclonic circulation.

Fig. 10.

(a) August climatology of SST (shading, °C) and 850-hPa winds (vectors, m s−1). (b) Partial regressions of SST (shading) and 850-hPa winds (vectors) in August with the normalized EMI in the preceding DJF during 1979–2014, excluding the Niño-3 SST effect. (c) Composite August SST (shading, °C) and 850-hPa wind anomalies (vectors, m s−1) for the three negative WIO SST cases (1992, 1993, and 2002). The dots denote partial regression coefficients and composite anomalies of SST above the 90% confidence level. Partial regression coefficients and composite anomalies of the winds exceeding the 90% confidence level are shown with thick vectors. The red letter C indicates the cyclonic circulation.

To further confirm the relationship between EMI and WIO SST and its influence on circulation anomalies over South China and the western North Pacific, a conditional composite analysis is performed. There are three years (1992, 1993, and 2002) in which the normalized EMI in the preceding DJF is more than 0.5 SD and the August SST anomalies averaged over the WIO (20°S–8°N, 40°–80°E) is less than −0.5 SD, while the magnitude of the August SST anomalies averaged over the South China Sea (0°–20°N, 105°–120°E) is less than 0.5 SD. Figure 10c shows composite anomalies of SST and 850-hPa winds for the selected three events, excluding the local SST effect over the South China Sea. Negative SST anomalies are observed in the western Indian Ocean, which are larger compared with those in Fig. 10b. There are obvious southwesterly anomalies over the northwestern Indian Ocean and cyclonic circulation over the South China Sea and South China, which is similar to the partial regression pattern with EMI (Fig. 10b). These similarities illustrate anomalous cyclonic circulation over the South China Sea and South China that acts like a response to the negative WIO SST anomalies in August, associated with the positive equatorial central Pacific SST anomalies in the preceding DJF.

This result is consistent with the previous suggestion of Yuan et al. (2012) that a basinwide warming mode in the Indian Ocean cannot be identified and the anomalous WNPAC disappears in the decaying summer during the central Pacific El Niño. Note that the region of negative SST anomaly over the western Indian Ocean in August 2016 is larger than the partial correlation pattern of SST with the EMI in the preceding DJF. This implies that another factor is also influencing the change in the WIO SST.

Another factor discussed by Ren et al. (2016) is the seasonal timing of the El Niño decay phase. Significantly warmer SST anomalies over the tropical Indian Ocean in spring can persist into summer only in a later-decaying El Niño, not in a normally decaying El Niño. In the 2015/16 extreme El Niño, the warmer SST anomalies in the Niño-3 region decay rapidly from winter to spring and begin to lessen by 0.5°C after April (Fig. 11a). In the late spring, the difference in the Niño-3 SST anomaly between the 2015/16 extreme El Niño and the earlier two extreme events reaches a maximum at 1°C. According to the definition of Ren et al. (2016) that the amplitude of the Niño-3 index in the decaying April is larger than the average amplitude of all El Niño events, extreme El Niño events are categorized as the later-decaying type, except for the 2015/16 extreme El Niño. The long persistence of El Niño–induced tropical Indian Ocean warming from winter to the following summer is attributed to local air–sea interaction within the tropical Indian Ocean (Du et al. 2009). During an El Niño event, the anomalous descending branch of the Walker circulation that lies over the western equatorial Pacific and Indonesian archipelago (Julian and Chervin 1978; Oort and Yienger 1996) initiates basinwide warming in the tropical Indian Ocean through cloud–radiation and wind–evaporation mechanisms. The anticyclonic wind-curl anomalies in the southeastern tropical Indian Ocean act to force a downwelling and westward-propagating oceanic Rossby wave over the south Indian Ocean (Huang and Kinter 2002; Masumoto and Meyers 1998; Perigaud and Delecluse 1993; Yu et al. 2005). Then it deepens the thermocline and contributes to the SST warming in the southwestern Indian Ocean, where the mean thermocline is shallow in spring (Xie et al. 2002). Wu and Yeh (2010) performed a mixed layer heat budget analysis to quantify the roles of different progresses in relation to the southwestern Indian Ocean warming and emphasized that surface heat flux is more important than oceanic processes. The warming of the southwestern Indian Ocean contributes to the establishment of the southward SST gradient and thus induces an asymmetric wind anomaly pattern with northeasterly (northwesterly) anomalies north (south) of the equator in spring, which persists through May–June (Wu et al. 2008; Du et al. 2009). The development and maintenance of the asymmetric atmospheric pattern are contributed by the positive wind–evaporation–SST feedback (Wu et al. 2008). In July–August following El Niño, the northeasterly anomalies retreat into the Bay of Bengal and South China Sea and sustain the warming in the north Indian Ocean by reducing the wind speed and surface evaporation, since the mean winds switch from easterly to westerly (Du et al. 2009). However, in normally decaying El Niño years the anomalous Walker circulation corresponding to the anomalous descent over the tropical Indo-Pacific region diminishes and the oceanic Rossby wave in the southwestern Indian Ocean weakens substantially in summer (Ren et al. 2016).

Fig. 11.

Evolution of SST anomalies (°C) in the (a) Niño-3, (b) Niño-3.4, and (c) Niño-4 regions from September to the following August in extreme El Niño years (2015/16, 1997/98, and 1982/83). The dotted and dashed lines denote the SST anomalies at 0° and 0.5°C, respectively.

Fig. 11.

Evolution of SST anomalies (°C) in the (a) Niño-3, (b) Niño-3.4, and (c) Niño-4 regions from September to the following August in extreme El Niño years (2015/16, 1997/98, and 1982/83). The dotted and dashed lines denote the SST anomalies at 0° and 0.5°C, respectively.

6. Summary

In this study, remarkable differences in rainfall in southern China in August and the closely related western North Pacific circulations among 1983, 1998, and 2016—namely, in the decaying years of three extreme El Niño events (1982/83, 1997/98, and 2015/16)—are examined through statistical analyses and a numerical model. The above-normal rainfall in the middle reaches of the YRV and the lower reaches of the Yellow River and the below-normal rainfall in South China in August are associated with El Niño. The anomalous rainfall patterns over southern China in August 1983 and 1998 are quite similar to the partial regressed pattern of rainfall with respect to the Niño-3 index in the preceding winter. However, below-normal rainfall in the middle and lower reaches of the YRV and above-normal rainfall in South China are observed in August 2016, although equatorial east-central Pacific SST anomalies reach the intensity of an extreme El Niño event in the preceding winter. The salient differences in southern China rainfall anomalies between August 2016 and the other two El Niño years can be explained by the anomalous western North Pacific circulations, which are attributed to SST anomalies over the WIO and NTA associated with the 2015/16 extreme El Niño.

Consistent with the previously documented relationship between the western North Pacific circulation and ENSO, the WNPAC persists from November to the succeeding August in the first two extreme El Niño events (1982/83 and 1997/98). In the 2015/16 extreme El Niño, the evolution of the WNPAC from November to the succeeding July is similar to the first two extreme El Niño events, but in the succeeding August an obviously enhanced cyclone dominates the western North Pacific. In August 1983 and 1998, the WNPAC is attributed to basinwide warming SST in the Indian Ocean (Fig. 12a), which is consistent with previous research. In August 2016, cooler SST anomalies over the WIO contribute to the anomalous western North Pacific cyclone and a south–north contrast of rainfall anomalies in southern China. Negative SST anomalies in the WIO could trigger local anomalous upper-level convergence and descent in situ and then enhance anomalous upper-level divergence and ascent over the western North Pacific (WNP) and South China. Furthermore, an anomalous lower-level cyclone is generated and leads to above-normal rainfall from the WNP to South China and below-normal rainfall in the middle and lower reaches of the YRV (Fig. 12b). The heat budget analysis shows that in August of 1983 and 1998 the surface heat fluxes play a dominant role in SST change in the western Indian Ocean, but in August 2016 the SST cooling is primarily due to oceanic dynamics. In addition to the role of the negative WIO SST anomalies, the negative NTA SST anomalies can trigger a subtropical teleconnection with the anomalous eastern Pacific anticyclone and the anomalous western Pacific cyclone and then result in the above-normal rainfall in South China and the below-normal rainfall in the middle and lower reaches of the YRV (Fig. 12b). The contribution of the negative WIO SST anomalies to the anomalous western Pacific cyclone and the above-normal rainfall in South China is larger than that of the cooling in the NTA. However, the negative NTA SST anomalies play a more important role in below-normal rainfall in middle reaches of the YRV.

Fig. 12.

Schematic diagrams summarizing the influence of (a) tropical Indian Ocean and (b) WIO and NTA SST anomalies on southern China rainfall anomalies and western North Pacific circulation anomalies in the decaying August of the extreme El Niño events (1982/83, 1997/98, and 2015/16). Solid red (blue) ellipses with arrows denote anomalous cyclone (anticyclone) associated with tropical Indian Ocean or WIO SST anomalies. Dashed red (blue) ellipses with arrows denote anomalous cyclone (anticyclone) related to NTA SST anomalies. Dark red (blue) shading indicates warmer (cooler) SST anomalies. Light blue (red) shading indicates positive (negative) rainfall anomalies.

Fig. 12.

Schematic diagrams summarizing the influence of (a) tropical Indian Ocean and (b) WIO and NTA SST anomalies on southern China rainfall anomalies and western North Pacific circulation anomalies in the decaying August of the extreme El Niño events (1982/83, 1997/98, and 2015/16). Solid red (blue) ellipses with arrows denote anomalous cyclone (anticyclone) associated with tropical Indian Ocean or WIO SST anomalies. Dashed red (blue) ellipses with arrows denote anomalous cyclone (anticyclone) related to NTA SST anomalies. Dark red (blue) shading indicates warmer (cooler) SST anomalies. Light blue (red) shading indicates positive (negative) rainfall anomalies.

A basinwide warming mode in the Indian Ocean cannot be identified in the decaying August of the 2015/16 extreme El Niño as it can during the 1982/83 and 1997/98 extreme El Niño events, which may be attributed to two factors. One is the different location of the warmer SST anomalies in the peak and decay phases in the 2015/16 extreme El Niño and the earlier two extreme events. The 2015/16 extreme El Niño, with a maximum SST anomaly in the Niño-3.4 region and a westward extension of warmer SST anomalies to the west of the date line in winter, is characterized by a mixed type of eastern Pacific El Niño and central Pacific El Niño, whereas the 1982/83 and 1997/98 extreme El Niño events are the pure eastern Pacific type (Lim et al. 2017; Paek et al. 2017). In August, the anticyclonic wind anomalies in the WIO with southeasterly anomalies south of the equator and southwesterly anomalies north of the equator, associated with equatorial central Pacific SST anomalies in the preceding winter, is conducive to SST cooling through enhancing the mean summer monsoon. The other factor is the different seasonal timing of the El Niño decay phase in the 2015/16 extreme El Niño and the earlier two extreme events. In the decaying spring, warmer SST anomalies over the equatorial Pacific during the 2015/16 extreme El Niño decay from the eastern Pacific, while warming starts to weaken from the equatorial central Pacific during the 1982/83 and 1997/98 extreme El Niño events (Fig. 11). Thus, the warming in the equatorial eastern Pacific decays faster during the 2015/16 extreme El Niño than in the earlier two events. The 1982/83 and 1997/98 extreme El Niño events are considered later-decaying El Niño events, during which the significantly warmer SST anomalies in the tropical Indian Ocean in spring can persist into summer, but not during normally decaying El Niño (Ren et al. 2016).

Acknowledgments

This work was jointly supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant XDA11010403), the CAS/SAFEA International Partnership Program for Creative Research Teams, the National Natural Science Foundation of China (Grants 41506004, 41422601, 41376025, 41675062, and 41530530), the Guangdong Natural Science Foundation (Grant 2016A030310113), the project of the Guy Carpenter Asia–Pacific Climate Impact Centre (Grant 9360126), and the Independent Research Project Program of State Key Laboratory of Tropical Oceanography (LTOZZ1702). Author C. Wang acknowledges the support of the National Natural Science Foundation of China (41731173), the Leading Talents of Guangdong Province Program, the Pioneer Hundred Talents Program of the Chinese Academy of Sciences, and the National Program on Global Change and Air–Sea Interaction (GASI-IPOVAI-04). The authors gratefully acknowledge the use of the High-Performance Computing Center (HPCC) for all numeric simulations and data analysis at the South China Sea Institute of Oceanology, Chinese Academy of Sciences. The authors declare no competing interests.

REFERENCES

REFERENCES
Adler
,
R. F.
, and Coauthors
,
2003
:
The Version-2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979–Present)
.
J. Hydrometeor.
,
4
,
1147
1167
, https://doi.org/10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Ashok
,
K.
,
S. K.
Behera
,
S. A.
Rao
,
H.
Weng
, and
T.
Yamagata
,
2007
:
El Niño Modoki and its possible teleconnection
.
J. Geophys. Res.
,
112
,
C11007
, https://doi.org/10.1029/2006JC003798.
Chang
,
C. P.
,
Y. S.
Zhang
, and
T.
Li
,
2000
:
Interannual and interdecadal variations of the East Asian summer monsoon and tropical Pacific SSTs. Part I: Roles of the subtropical ridge
.
J. Climate
,
13
,
4310
4325
, https://doi.org/10.1175/1520-0442(2000)013<4310:IAIVOT>2.0.CO;2.
Chen
,
J. P.
,
R. G.
Wu
, and
Z. P.
Wen
,
2012
:
Contribution of South China Sea tropical cyclones to an increase in southern China summer rainfall around 1993
.
Adv. Atmos. Sci.
,
29
,
585
598
, https://doi.org/10.1007/s00376-011-1181-6.
Chen
,
J. P.
,
Z. P.
Wen
,
R. G.
Wu
,
Z. S.
Chen
, and
P.
Zhao
,
2014
:
Interdecadal changes in the relationship between SC winter-spring precipitation and ENSO
.
Climate Dyn.
,
43
,
1327
1338
, https://doi.org/10.1007/s00382-013-1947-x.
Chen
,
J. P.
,
Z. P.
Wen
,
R. G.
Wu
,
Z. S.
Chen
, and
P.
Zhao
,
2015
:
Influences of northward propagating 25–90-day and quasi-biweekly oscillations on eastern China summer rainfall
.
Climate Dyn.
,
45
,
105
124
, https://doi.org/10.1007/s00382-014-2334-y.
Chen
,
J. P.
,
Z. P.
Wen
,
R. G.
Wu
,
X.
Wang
,
C.
He
, and
Z. S.
Chen
,
2017
:
An interdecadal change in the intensity of interannual variability in summer rainfall over southern China around early 1990s
.
Climate Dyn.
,
48
,
191
207
, https://doi.org/10.1007/s00382-016-3069-8.
Chen
,
L.
,
T.
Li
,
B.
Wang
, and
L.
Wang
,
2017
:
Formation mechanism for 2015/16 super El Niño
.
Sci. Rep.
,
7
,
2975
, https://doi.org/10.1038/s41598-017-02926-3.
Chen
,
Z. S.
,
Y.
Du
,
Z. P.
Wen
,
R. G.
Wu
, and
C. Z.
Wang
,
2018
:
Indo-Pacific climate during the decaying phase of the 2015/16 El Niño: Role of southeast tropical Indian Ocean warming
.
Climate Dyn.
,
50
,
4707
4719
, https://doi.org/10.1007/s00382-017-3899-z.
Dee
,
D. P.
, and Coauthors
,
2011
:
The ERA-Interim reanalysis: Configuration and performance of the data assimilation system
.
Quart. J. Roy. Meteor. Soc.
,
137
,
553
597
, https://doi.org/10.1002/qj.828.
Du
,
Y.
,
S. P.
Xie
,
G.
Huang
, and
K. M.
Hu
,
2009
:
Role of air–sea interaction in the long persistence of El Niño–induced north Indian Ocean warming
.
J. Climate
,
22
,
2023
2038
, https://doi.org/10.1175/2008JCLI2590.1.
ECMWF
,
2011
: ERA-Interim project, version 2. European Centre for Medium-Range Weather Forecasts, accessed 4 April 2017, http://apps.ecmwf.int/datasets/data/interim-full-moda/levtype=sfc/.
Glantz
,
M. H.
,
2001
: Lessons learned from the 1997–98 El Niño: Once burned, twice shy? United Nations University Press, 294 pp.
Ham
,
Y. G.
,
J. S.
Kug
,
J. Y.
Park
, and
F. F.
Jin
,
2013
:
Sea surface temperature in the north tropical Atlantic as a trigger for El Niño/southern oscillation events
.
Nat. Geosci.
,
6
,
112
116
, https://doi.org/10.1038/ngeo1686.
He
,
Z. Q.
, and
R. G.
Wu
,
2013a
:
Coupled seasonal variability in the South China Sea
.
J. Oceanogr.
,
69
,
57
69
, https://doi.org/10.1007/s10872-012-0157-1.
He
,
Z. Q.
, and
R. G.
Wu
,
2013b
:
Seasonality of interannual atmosphere–ocean interaction in the South China Sea
.
J. Oceanogr.
,
69
,
699
712
, https://doi.org/10.1007/s10872-013-0201-9.
He
,
Z. Q.
, and
R. G.
Wu
,
2014
:
Indo-Pacific remote forcing in summer rainfall variability over the South China Sea
.
Climate Dyn.
,
42
,
2323
2337
, https://doi.org/10.1007/s00382-014-2123-7.
Hu
,
K.
,
S.
Xie
, and
G.
Huang
,
2017
:
Orographically anchored El Niño effect on summer rainfall in central China
.
J. Climate
,
30
,
10 037
10 045
, https://doi.org/10.1175/JCLI-D-17-0312.1.
Huang
,
B. H.
, and
J. L.
Kinter
,
2002
:
Interannual variability in the tropical Indian Ocean
.
J. Geophys. Res.
,
107
,
3199
, https://doi.org/10.1029/2001JC001278.
JMA
,
2009
: The Japanese 55-year Reanalysis (JRA55) project. JMA/Climate Prediction Division/Global Environment and Marine Department, accessed 31 March 2017, http://jra.kishou.go.jp/JRA-55/index_en.html.
Julian
,
P. R.
, and
R. M.
Chervin
,
1978
:
Study of southern oscillation and Walker circulation phenomenon
.
Mon. Wea. Rev.
,
106
,
1433
1451
, https://doi.org/10.1175/1520-0493(1978)106<1433:ASOTSO>2.0.CO;2.
Kobayashi
,
S.
, and Coauthors
,
2015
:
The JRA-55 reanalysis: General specifications and basic characteristics
.
J. Meteor. Soc. Japan
,
93
,
5
48
, https://doi.org/10.2151/jmsj.2015-001.
L’Heureux
,
M.
, and Coauthors
,
2017
:
Observing and predicting the 2015/16 El Niño
.
Bull. Amer. Meteor. Soc.
,
98
,
1363
1382
, https://doi.org/10.1175/BAMS-D-16-0009.1.
Li
,
G.
,
J. P.
Chen
,
X.
Wang
,
Y. K.
Tan
, and
X. H.
Jiang
,
2017
:
Modulation of Pacific Decadal Oscillation on the relationship of El Niño with southern China rainfall during early boreal winter
.
Atmos. Sci. Lett.
,
18
,
336
341
, https://doi.org/10.1002/asl.761.
Li
,
S. L.
,
J.
Lu
,
G.
Huang
, and
K. M.
Hu
,
2008
:
Tropical Indian Ocean basin warming and East Asian summer monsoon: A multiple AGCM study
.
J. Climate
,
21
,
6080
6088
, https://doi.org/10.1175/2008JCLI2433.1.
Lim
,
Y. K.
,
R. M.
Kovach
,
S.
Pawson
, and
G.
Vernieres
,
2017
:
The 2015/16 El Niño event in context of the MERRA-2 reanalysis: A comparison of the tropical Pacific with 1982/83 and 1997/98
.
J. Climate
,
30
,
4819
4842
, https://doi.org/10.1175/JCLI-D-16-0800.1.
Lindzen
,
R. S.
, and
N.
Nigam
,
1987
:
On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics
.
J. Atmos. Sci.
,
44
,
2418
2436
, https://doi.org/10.1175/1520-0469(1987)044<2418:OTROSS>2.0.CO;2.
Masumoto
,
Y.
, and
G.
Meyers
,
1998
:
Forced Rossby waves in the southern tropical Indian Ocean
.
J. Geophys. Res.
,
103
,
27 589
27 602
, https://doi.org/10.1029/98JC02546.
Met Office
,
2003
: Hadley Centre Sea Ice and SST (HadISST), version 1. Met Office Hadley Centre for Climate Change, accessed 23 December 2016, https://www.metoffice.gov.uk/hadobs/hadisst/data/download.html.
Neal
,
R. B.
, and Coauthors
,
2012
: Description of the NCAR Community Atmosphere Model (CAM 5.0). NCAR Tech. Note NCAR/TN-486+STR, 274 pp., www.cesm.ucar.edu/models/cesm1.0/cam/docs/description/cam5_desc.pdf.
NOAA/OAR/ESRL
,
1997
: CPC Merged Analysis of Precipitation excluding NCEP Reanalysis, continuing from January 1979 (updated monthly). NOAA/OAR/ESRL Physical Sciences Division (PSD), accessed 24 February 2017, https://www.esrl.noaa.gov/psd/data/gridded/data.cmap.html#plot.
NOAA/OAR/ESRL
,
2003a
: Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979-Present), version 2.3. NOAA/OAR/ESRL Physical Sciences Division (PSD), accessed 27 June 2017, https://www.esrl.noaa.gov/psd/data/gridded/data.gpcp.html.
NOAA/OAR/ESRL
,
2003b
: NOAA Extended Reconstructed Sea Surface Temperature, version 3b. NOAA/OAR/ESRL Physical Sciences Division (PSD), accessed 22 March 2017, https://www.esrl.noaa.gov/psd/data/gridded/data.noaa.ersst.v3.html.
NOAA/OAR/ESRL
,
2004
: NCEP Global Ocean Data Assimilation System (GODAS). NOAA/OAR/ESRL Physical Sciences Division (PSD), accessed 6 February 2018, https://www.esrl.noaa.gov/psd/data/gridded/data.godas.html#detail.
Oort
,
A. H.
, and
J. J.
Yienger
,
1996
:
Observed interannual variability in the Hadley circulation and its connection to ENSO
.
J. Climate
,
9
,
2751
2767
, https://doi.org/10.1175/1520-0442(1996)009<2751:OIVITH>2.0.CO;2.
Paek
,
H.
,
J. Y.
Yu
, and
C.
Qian
,
2017
:
Why were the 2015/2016 and 1997/1998 extreme El Niños different?
Geophys. Res. Lett.
,
44
,
1848
1856
, https://doi.org/10.1002/2016GL071515.
Perigaud
,
C.
, and
P.
Delecluse
,
1993
:
Interannual sea level variations in the tropical Indian Ocean from Geosat and shallow water simulations
.
J. Phys. Oceanogr.
,
23
,
1916
1934
, https://doi.org/10.1175/1520-0485(1993)023<1916:ISLVIT>2.0.CO;2.
Rayner
,
N. A.
,
D. E.
Parker
,
E. B.
Horton
,
C. K.
Folland
,
L. V.
Alexander
,
D. P.
Rowell
,
E. C.
Kent
, and
A.
Kaplan
,
2003
:
A global analysis of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century
.
J. Geophys. Res.
,
108
,
4407
, https://doi.org/10.1029/2002JD002670.
Ren
,
R.
,
S.
Sun
,
Y.
Yang
, and
Q.
Li
,
2016
:
Summer SST anomalies in the Indian Ocean and the seasonal timing of ENSO decay phase
.
Climate Dyn.
,
47
,
1827
1844
, https://doi.org/10.1007/s00382-015-2935-0.
Smith
,
T.
,
R.
Reynolds
,
T.
Peterson
, and
J.
Lawrimore
,
2008
:
Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006)
.
J. Climate
,
21
,
2283
2296
, https://doi.org/10.1175/2007JCLI2100.1.
Wang
,
B.
, and
Q.
Zhang
,
2002
:
Pacific–East Asian teleconnection. Part II: How the Philippine Sea anomalous anticyclone is established during El Niño development?
J. Climate
,
15
,
3252
3265
, https://doi.org/10.1175/1520-0442(2002)015<3252:PEATPI>2.0.CO;2.
Wang
,
B.
,
R. G.
Wu
, and
X. H.
Fu
,
2000
:
Pacific–East Asian teleconnection: How does ENSO affect East Asian climate?
J. Climate
,
13
,
1517
1536
, https://doi.org/10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.
Wang
,
C. Z.
, and
X.
Wang
,
2013
:
Classifying El Niño Modoki I and II by different impacts on rainfall in southern China and typhoon tracks
.
J. Climate
,
26
,
1322
1338
, https://doi.org/10.1175/JCLI-D-12-00107.1.
Wang
,
X.
, and
C. Z.
Wang
,
2014
:
Different impacts of various El Niño events on the Indian Ocean Dipole
.
Climate Dyn.
,
42
,
991
1005
, https://doi.org/10.1007/s00382-013-1711-2.
Weng
,
H.
,
K.
Ashok
,
S. K.
Behera
, and
S. A.
Rao
,
2007
:
Impacts of recent El Niño Modoki on dry/wet conditions in the Pacific Rim during boreal summer
.
Climate Dyn.
,
29
,
113
129
, https://doi.org/10.1007/s00382-007-0234-0.
Weng
,
H.
,
S. K.
Behera
, and
T.
Yamagata
,
2009
:
Anomalous winter climate conditions in the Pacific Rim during recent El Niño Modoki and El Niño events
.
Climate Dyn.
,
32
,
663
674
, https://doi.org/10.1007/s00382-008-0394-6.
Wu
,
B.
,
T. J.
Zhou
, and
T.
Li
,
2009
:
Seasonally evolving dominant interannual variability modes of East Asian climate
.
J. Climate
,
22
,
2992
3005
, https://doi.org/10.1175/2008JCLI2710.1.
Wu
,
B.
,
T.
Li
, and
T. J.
Zhou
,
2010
:
Relative contributions of the Indian Ocean and local SST anomalies to the maintenance of the western North Pacific anomalous anticyclone during the El Niño decaying summer
.
J. Climate
,
23
,
2974
2986
, https://doi.org/10.1175/2010JCLI3300.1.
Wu
,
R. G.
, and
S. W.
Yeh
,
2010
:
A further study of the tropical Indian Ocean asymmetric mode in boreal spring
.
J. Geophys. Res.
,
115
,
D08101
, https://doi.org/10.1029/2009JD012999.
Wu
,
R. G.
,
Z. Z.
Hu
, and
B. P.
Kirtman
,
2003
:
Evolution of ENSO-related rainfall anomalies in East Asia
.
J. Climate
,
16
,
3742
3758
, https://doi.org/10.1175/1520-0442(2003)016<3742:EOERAI>2.0.CO;2.
Wu
,
R. G.
,
B. P.
Kirtman
, and
V.
Krishnamurthy
,
2008
:
An asymmetric mode of tropical Indian Ocean rainfall variability in boreal spring
.
J. Geophys. Res.
,
113
,
D05104
, https://doi.org/10.1029/2007JD009316.
Xie
,
P. P.
, and
P. A.
Arkin
,
1997
:
Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs
.
Bull. Amer. Meteor. Soc.
,
78
,
2539
2558
, https://doi.org/10.1175/1520-0477(1997)078<2539:GPAYMA>2.0.CO;2.
Xie
,
S. P.
,
H.
Annamalai
,
F. A.
Schott
, and
J. P.
McCreary
,
2002
:
Structure and mechanisms of south Indian Ocean climate variability
.
J. Climate
,
15
,
864
878
, https://doi.org/10.1175/1520-0442(2002)015<0864:SAMOSI>2.0.CO;2.
Xie
,
S. P.
,
K. M.
Hu
,
J.
Hafner
,
H.
Tokinaga
,
Y.
Du
,
G.
Huang
, and
T.
Sampe
,
2009
:
Indian Ocean capacitor effect on Indo-western Pacific climate during the summer following El Niño
.
J. Climate
,
22
,
730
747
, https://doi.org/10.1175/2008JCLI2544.1.
Xie
,
S. P.
,
Y.
Kosaka
,
Y.
Du
,
K. M.
Hu
,
J. S.
Chowdary
, and
G.
Huang
,
2016
:
Indo-Western Pacific Ocean capacitor and coherent climate anomalies in post-ENSO summer: A review
.
Adv. Atmos. Sci.
,
33
,
411
432
, https://doi.org/10.1007/s00376-015-5192-6.
Yang
,
J. L.
,
Q. Y.
Liu
,
S. P.
Xie
,
Z. Y.
Liu
, and
L. X.
Wu
,
2007
:
Impact of the Indian Ocean SST basin mode on the Asian summer monsoon
.
Geophys. Res. Lett.
,
34
,
L02708
, https://doi.org/10.1029/2006GL028571.
Yu
,
W. D.
,
B. Q.
Xiang
,
L.
Liu
, and
N.
Liu
,
2005
:
Understanding the origins of interannual thermocline variations in the tropical Indian Ocean
.
Geophys. Res. Lett.
,
32
,
L24706
, https://doi.org/10.1029/2005GL024327.
Yuan
,
Y.
, and
S.
Yang
,
2012
:
Impacts of different types of El Niño on the East Asian climate: Focus on ENSO cycles
.
J. Climate
,
25
,
7702
7722
, https://doi.org/10.1175/JCLI-D-11-00576.1.
Yuan
,
Y.
,
S.
Yang
, and
Z.
Zhang
,
2012
:
Different evolutions of the Philippine Sea anticyclone between the eastern and central Pacific El Niño: Possible effects of Indian Ocean SST
.
J. Climate
,
25
,
7867
7883
, https://doi.org/10.1175/JCLI-D-12-00004.1.
Yuan
,
Y.
,
H.
Gao
,
W. J.
Li
,
Y. J.
Liu
,
L. J.
Chen
,
B.
Zhou
, and
Y. H.
Ding
,
2017
:
The 2016 summer floods in China and associated physical mechanisms: A comparison with 1998
.
J. Meteor. Res.
,
31
,
261
277
, https://doi.org/10.1007/s13351-017-6192-5.
Zhan
,
R.
,
Y.
Wang
, and
Q.
Liu
,
2017
:
Salient differences in tropical cyclone activity over the western North Pacific between 1998 and 2016
.
J. Climate
,
30
,
9979
9997
, https://doi.org/10.1175/JCLI-D-17-0263.1.
Zhang
,
R. H.
, and
A.
Sumi
,
2002
:
Moisture circulation over East Asia during El Niño episode
.
J. Meteor. Soc. Japan
,
80
,
213
227
, https://doi.org/10.2151/jmsj.80.213.
Zhang
,
R. H.
,
A.
Sumi
, and
M.
Kimoto
,
1996
:
Impact of El Niño on the East Asian monsoon: A diagnostic study of the ‘86/87 and ’91/92 events
.
J. Meteor. Soc. Japan
,
74
,
49
62
, https://doi.org/10.2151/jmsj1965.74.1_49.
Zhang
,
R. H.
,
A.
Sumi
, and
M.
Kimoto
,
1999
:
A diagnostic study of the impact of El Niño
.
Adv. Atmos. Sci.
,
16
,
229
241
, https://doi.org/10.1007/BF02973084.

Footnotes

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).