1. Introduction
El Niño–Southern Oscillation (ENSO) has a pronounced impact on South China Sea (SCS) climate (e.g., Huang and Wu 1989; Liu et al. 2004). Sea surface temperature (SST) anomalies over the SCS usually peak twice, during the mature phase and the following summer of ENSO (Wang et al. 2006). The first warming is mainly attributed to shortwave radiation and latent heat flux anomalies, resulting from the direct effect of the atmospheric bridge from the Pacific (Klein et al. 1999; Wang et al. 2006). The second warming is related to the air–sea interaction over the tropical Indian Ocean that is a delayed response to ENSO (Wu et al. 2008; Xie et al. 2009, 2010; Du et al. 2009, 2013). The ocean Rossby wave–triggered SST warming/deep convection in the southwestern tropical Indian Ocean induces an asymmetric winds pattern during the spring of the ENSO decay year, with anomalous northeasterlies to the north and anomalous northwesterlies to the south of the equator. The northeasterly anomalies persist into the summer and reduce the mean southwest monsoon, giving rise to the second warming over the north Indian Ocean (NIO) (Du et al. 2009, 2013; Schott et al. 2009). The NIO warming excites a baroclinic Kelvin wave propagating into the equatorial western Pacific in the troposphere, which causes Ekman divergence over the northwestern Pacific (NWP), suppresses atmospheric convection, and triggers an anomalous anticyclonic circulation. The easterlies on the southwest flank of the anticyclonic circulation in turn reduce the southwest monsoon over the SCS and induce SST warming during the summer (Xie et al. 2009).
Recent studies indicate that the Indo–western Pacific climate has experienced interdecadal modulations during the past decades (Tokinaga et al. 2012a,b). The Indian Ocean capacitor effect or Indian Ocean basinwide warming (Yang et al. 2007; Xie et al. 2009; Du et al. 2009), which is key to the post-ENSO summer impacts, undergoes a dramatic change during the 1976/77 regime shift (Xie et al. 2010). After 1976/77, the Indian Ocean warming persists longer, resulting in a stronger anticyclonic circulation and air–sea feedback in the post–El Niño summer over the NWP (Huang et al. 2010). Chowdary et al. (2012) extended the analysis and showed that the persistence of ENSO-induced NIO warming in different epochs during the last 138 yr contributes to the interdecadal variability of the Indian Ocean capacitor effect on the NWP climate change. They also found that the patterns of precipitation over the Indo-western Pacific vary from decade to decade, which is crucial to the lives of the people in surrounding area.
In this study, we investigate the relationship between the SCS SST and ENSO during the past 138 years and discuss the interdecadal modulations of ENSO effect on the rainfall variability around the SCS. We use observational data obtained from frequent ship lanes over the SCS. The data start from the late nineteenth century, enabling us to investigate the long-term variability of ENSO teleconnections to the SCS. We examine the interannual variability over the SCS associated with ENSO for different epochs. The results indicate significant interdecadal modulations of ENSO effects on the SCS, which are different from the response of the NIO. Both the Indian Ocean capacitor effect and Pacific Japan/East Asian–Pacific (PJ/EAP) atmospheric teleconnection patterns (Nitta 1987; Huang and Yan 1999; Kosaka and Nakamura 2006; Kosaka et al. 2013) are important to sustain the SCS SST warming through the summer even fall following El Niño. Besides, the rainfall anomalies around the SCS display regional dependent long-term changes, which is a part of interdecadal modulation of ENSO.
The paper is organized as follows. The following section introduces the datasets used in the study and briefly describes the methods. Section 3 distinguishes the variability of the relationship between ENSO and the SCS into different epochs. Section 4 discusses rainfall changes around the SCS for the past decades. Section 5 discusses how ENSO influences the SCS in different epochs and why different responses exist between the NIO and SCS. Section 6 gives the major conclusions.
2. Data and methods
The frequent ship lane data over the SCS, including SST, surface wind, cloudiness, sea level pressure (SLP), and latent heat flux, are extracted from the International Comprehensive Ocean–Atmosphere Dataset (ICOADS) 2.5 (Woodruff et al. 2011) for the period of 1870–2007 (the Suez Canal opened in 1869). The grid points with 2° resolution that contain monthly SSTs for more than 80% of the total months during the whole period are determined for the ship route over the SCS (Fig. 1a).Other atmospheric data (surface wind, cloudiness, SLP, and latent heat flux) are extracted from the same SST ship route. There are significant data gaps in the 1870s and 1940s along the ship lanes (Fig. 1b). We have filled the gaps with Extended Reconstructed SST (ERSST) values (Fig. 3). Nevertheless, to minimize the influence of data discontinuity on our SST analysis, the 1940s has not been considered when we divide the period of 1870–2007 into different epochs. In addition, the ship-observed cloudiness data may be of doubtful quality because of systematic biases (Norris 1999). However, Tokinaga et al. (2012b) demonstrated the reliability of cloudiness by comparing the ICOADS cloudiness and GPCP rainfall. In addition, the cloudiness from ICOADS also exhibited physical consistence with additional climate parameters in its long-term changes (Deser et al. 2010).
(a) The percentage of monthly SSTs in the total months from 1870–2007 (shaded) and the selected ship route over the SCS (blue line). (b) The number of observations in a grid.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
To validate the ICOADS ship track data, we also use objectively-analyzed SST products: the Hadley Center Sea Ice and Sea Surface Temperature dataset (HadISST) on a 1° latitude–longitude grid (Rayner et al. 2003), the National Oceanic and Atmospheric Administration (NOAA)–National Climatic Data Center (NCDC) ERSST version 2 (Smith and Reynolds 2004) on a 2° latitude–longitude grid, the extended Kaplan SST dataset (Kaplan et al. 1998) on a 5° latitude–longitude grid, the Centennial in Situ Observation-Based Estimates of SSTs (COBE-SST; Ishii et al. 2005) on a 1° latitude–longitude grid, and the NOAA Optimum Interpolation SST (OISST) version 2 (Reynolds et al. 2002) on a 1° latitude–longitude grid. The available periods in ERSST, HadISST, and Kaplan SST are from 1870 to 2007 except for the COBE-SST, which is from 1891 to 2007. The monthly values of these SST products are extracted along similar routes of ship data. Correlation analysis is conducted to check the consistency between SST products and ship route SST. Moreover, the effective degrees of freedom (Bretherton et al. 1999) of ship route SST are calculated based on the equation 
In addition to the ship measured data, three rainfall datasets are used in this study. The University of Delaware precipitation version 2 (Willmott and Matsuura 2001) with a resolution of 1° × 1° is obtained from the NOAA Earth System Research Laboratory (ESRL), Physical Sciences Division (PSD). The Philippine stations’ precipitation is from the Monthly Bulletins of Philippine Weather Bureau (Kubota and Chan 2009). We select 4 out of 37 stations in this study. The Global Precipitation Climatology Project (GPCP) monthly precipitation dataset version 2.2 for 1979–2010 combines observations and satellite precipitation data with a resolution of 2.5° × 2.5° (Adler et al. 2003).
We use Niño-3.4 SST and Southern Oscillation indices to study the long-term modulation of ENSO with regard to the SCS. The Niño-3.4 index (Trenberth et al. 2002) is extracted from the ICOADS 2.5 by averaging the SST anomalies in the area of 5°S–5°N, 170°–120°W. Hereafter, the developing phase of ENSO is denoted as year 0 and the decay phase as year 1. The Southern Oscillation index (SOI) from the Climatic Research Unit is used (Allan et al. 1991).
We detect the ENSO-induced atmospheric circulation anomalies over the SCS for the past 138 years based on the Twentieth Century Reanalysis (20CR; Compo et al. 2011), to be compared with the results of the ship data for the impacts of ENSO on the SCS in different epochs. The 20CR, using a state-of-the-art data assimilation system and surface pressure observations, is obtained from NOAA/ESRL/PSD.
We use the season-reliant EOF (S-EOF) analysis (Wang and An 2005; Chowdary et al. 2012) to investigate the evolution of SST anomalies associated with ENSO. The aim is to get main SST modes in different epochs. Rather than examining single seasonal mean SST anomalies for each year or for all seasons (Wang and An 2005) by a regular EOF, the S-EOF detect the SST anomalies in a sequence beginning from the January(0) to the following December(1) in this study. The derived spatial pattern for each S-EOF mode will include 12 sequential patterns indicating seasonal evolution of the SST anomalies; the 12 patterns share the same yearly value in their corresponding principal components. Before we discuss the differences of ENSO-related atmospheric anomalies among these epochs, the latent heat flux (LHF) is decomposed as the atmospheric forcing (AtF-L) and oceanic response. The former is mainly influenced by atmospheric changes such as wind speed and air–sea temperature difference, and the latter is considered as Newtonian cooling due to sea surface evaporation (de Szoeke et al. 2007; Du et al. 2009; Wu et al. 2012; Chowdary et al. 2012). The AtF-L, generally, can be regarded as an active forcing term. Hereafter a positive AtF-L denotes the ocean gaining heat to warm the SST. The calculation of AtF-L relies heavily on the amount of instrumentally observed wind data. Considering the quality of wind observations of ICOADS, we have presented results of AtF-L only during epochs 3 and 4, since the number of instrumentally observed winds drastically increased after the 1960s (Worley et al. 2005; Tokinaga and Xie 2011).
3. Interdecadal ENSO–SCS SST relationship
a. Ship route SST over the SCS
The ship route in the SCS we selected extends from southwest of Taiwan Island toward the Strait of Malacca in the southwest (Fig. 1). The monthly SST time series averaged from 0° to 22°N show a significant warming trend for the past 138 years (figure not shown), confirming the SST rise in the tropical oceans as part of global warming (Alory et al. 2007; Du and Xie 2008; Chowdary et al. 2012). The annual variance of detrended ICOADS ship data and other SST products indicates high consistency with each other (Figs. 2a and 3). Although objectively analyzed SST products are weaker in amplitude, probably due to spatial interpolations such as optimal interpolation, reduced spatial optimal interpolation, and smoother methods (Kaplan et al. 1998), they capture the main characteristics of the ship route SST. Large SST variance appears in the central and southern SCS, with the former in the summer [June–August (JJA)] and the later in the winter (November–February) (Fig. 2b). The monsoon circulations influence the SST seasonality. In the summer, the offshore wind jet, which induced by the southwest summer monsoon and coastal mountain range, favors the upwelling off the southern Vietnam coast and advects cold coastal water into the open ocean (Xie et al. 2003), forming a cold filament east of Vietnam around 10°N. The coastal wind and cold filament vary with ENSO, forming an SST variance maximum in the summer. In the winter, the monsoon transits to the northeast and drives cyclonic ocean circulation over the basin. The southward flow in the west boundary advects cold water from the north to the south, forming a cold tongue in SST climatology (Liu et al. 2004; Fig. 2b). The winter cold tongue also varies with ENSO (Liu et al. 2004). ICOADS also exhibits a high consistency with other SST products for both the Niño-3.4 and SCS SST indices with significant positive correlations (Fig. 3).
(a) Standard deviation (°C) of ICOADS SST, Kaplan SST, ERSST, HadISST, and COBE SST. (b) The standard deviation (shaded, °C) and climatology (contour, °C) of ICOADS SST as a function of calendar month.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
SST correlation between ICOADS and other SST products calculated in the SCS (x axis) and Niño-3.4 (y axis) regions. The correlation coefficients all exceed the 99% significance level, which denotes coherence between ship data and other SST products.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
b. SCS SST anomalies during ENSO
SST anomalies appear over the SCS in interannual time scale during ENSO period (Wang et al. 2006).We first evaluate the ENSO-induced variability in the past 138 yr to figure out if it shows any long-term changes. Figure 4a shows the correlation in 21-yr sliding windows between the SCS SST anomalies averaged from 3° to 16°N along the ship route and the November–January [NDJ(0/1)] Niño-3.4 index. The SST anomalies are detrended and filtered with 4–84 months bandpass before the correlation analysis to highlight the interannual variability associated with ENSO. Since the ICOADS data are inhomogeneous during the analysis period (Kent et al. 2007), the confidence levels of the correlation coefficients are calculated according to the effective degrees of freedom of the bandpass filtered data (figure not shown). Note that the data coverage in the early twentieth century is slightly lower than in the modern era. To illustrate the rationality of our analysis, we have estimated the measurement and sampling uncertainties of SCS SST by using HadSST3 (downloaded from http://www.metoffice.gov.uk/hadobs/hadsst3/data/download.html). The mean measurement and sampling uncertainties over the SCS during epochs 1, 2, 3, and 4 are about 0.3°, 0.2°, 0.2°, and 0.2°C, respectively (figure not shown). The values are much less than the interannual standard deviation of SCS SST (about 0.6°C averaged from 3° to 16°N), indicating that the SST uncertainty is small enough to discuss the interannual SST variations in the SCS even before the 1960s. We performed the same correlation analysis using the HadSST3 product and obtained a similar pattern of SST anomalies (figure not shown). Such a similarity between the HadSST3 and our ICOADS ship track dataset supports our results in the SCS.
Correlation (shaded) in 21-yr sliding windows between the SCS SST anomalies and (a) NDJ(0/1) Niño-3.4 and (b) DJF(0/1) SOI index. Green lines denote four epochs of decadal variability of the SCS SST anomalies associated with ENSO. Only correlation coefficients with a significance level exceeding 95% are shown, and white contours represent correlations of ±0.7.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
Based on the seasonal evolution and duration of SST warming from visual inspection, we divide the 138 yr into four periods. They are 1892–1915, 1930–40, 1960–83, and 1984–2007. The reason for selecting epoch 1 is that the large correlations (above 0.7) persist from January(1) to May(1) (Fig. 4a) and there is no significant second peak (Fig. 4b). From 1916 to 1929, the correlations peak at March(1) to April(1) (Fig. 4), which is quite different from other periods. So we ruled this period out from our selection of the four epochs. There is a significant single peak in the period from 1930 to 1940, which is the main reason to select epoch 2 (Fig. 4a). Since the double-peak structure is dominant after the 1960s, the division of epochs 3 and 4 is based on the time of the second peak. In epochs 3 and 4, the second peak occurs in May–August [MJJA(1)] and June–September [JJAS(1)], respectively. The correlation is stronger at the second than the first peak. Comparative analysis is conducted using DJF(0/1) SOI and we obtained similar results (Fig. 4b).
Figure 5 shows the correlations of SST and surface winds anomalies with the NDJ(0/1) Niño-3.4 index as a function of latitude and calendar month along the ship route in each of these epochs. The SST patterns in the four epochs are similar to that of Fig. 4. In epoch 1, the SST warming occurs south of the Vietnam area in November–January [NDJ(0/1)] and then persists to March–May [MAM(1)] over the central SCS. In epoch 2, significant correlations are found from November(0) to May(1), mainly in the central and southern SCS. The SST warming in epochs 3 and 4 occurs in the winter and the following summer, centered at the southern and central SCS, respectively. The second warming has a longer duration in the epoch 4 than 3, persisting until September(1).
Correlation of the SCS SST anomalies (shaded) and surface wind (vectors) with the NDJ(0/1) Niño-3.4 index as a function of calendar month and latitude in (a) epoch 1 (1892–1915), (b) epoch 2 (1930–40), (c) epoch 3 (1960–83), and (d) epoch 4 (1984–2007). Black contours indicate confidence level of 95%.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
Wind correlations associated with the winter warming are strong and common to all epochs, with southerlies over the southern SCS. During the decay year of El Niño, wind anomalies vary in timing and magnitude (Fig. 5). In epoch 1, southerly wind anomalies are seen around MAM(1) over the central SCS. In epoch 2, wind signals dissipate quickly after March(1). In epochs 3 and 4, northeasterly anomalies prevail in the summer following El Niño over the central SCS. The wind anomalies are more pronounced in epoch 4 and persist longer through JJA(1) than in other epochs. The anomalous winds oppose the climatological winds during both the winter and summer during and after El Niño (Xie et al. 2003; Liu et al. 2004). The wind anomalies play an important role in inducing SST warming by modulating the latent heat flux (Du et al. 2009; Chowdary et al. 2012; Wu et al. 2014).
We highlight the changes in anomaly patterns by using S-EOF analysis (Fig. 6). The fractional variance explained by these first SST S-EOF modes in the four epochs is 24.4%, 32.4%, 34.3%, and 22%, respectively. The principal components of the first S-EOF (PC1s) during all four epochs are highly correlated with the NDJ(0/1) Niño-3.4 SST index, indicating a strong link between the SCS SST and ENSO. Epochs 1 and 2 display a single-peak structure while epochs 3 and 4 show a double-peak structure, similar to the correlation analysis in Fig. 5. The regressions of surface wind anomalies upon the corresponding PC1s in different epochs also show similar patterns to the correlation in Fig. 5. Note that wind anomalies in S-EOF analysis are much coherent than in the correlation analysis (Fig. 5a) for epoch 1 (1892–1915), with an indication of northeasterly wind anomalies in the post–El Niño summer. These results indicate that our selection of epochs is reasonable.
S-EOFs of the SCS SST anomalies from April(0) (ENSO developing phase) to October(1) (ENSO decaying phase) for epochs (a) 1, (b) 2, (c) 3, and (d) 4. Regression of surface wind (vectors, m s−1) upon the principal component corresponding to the spatial patterns (shaded) of the four epochs. Percentage of variance explained by S-EOF is shown at the lower right of each panel.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
c. Atmospheric anomalies associated with ENSO
ENSO can affect the tropical Indian Ocean and the NWP through changes in atmospheric circulation (Weisberg and Wang 1997; Wang et al. 1999, 2000, 2003; Tokinaga and Tanimoto 2004; Yang et al. 2007; Wu et al. 2008; Xie et al. 2009; Du et al. 2009). Many studies demonstrate that heat flux changes associated with atmospheric anomalies contribute to the SST warming during ENSO period (Wang et al. 2006; Xie et al. 2009; Du et al. 2009; Chowdary et al. 2012). Using atmospheric variables of ship reports, we investigate the relationships between SST anomalies and ENSO in all the four epochs in this section. Figure 7 shows the regression coefficients of SST, AtF-L, surface winds, and cloudiness along the SCS ship track upon the NDJ(0/1) Niño-3.4 index in all the epochs. Positive (negative) values of AtF-L mean the ocean is gaining (losing) heat.
Regression coefficients of SST (white contours, °C), AtF-L (black contours, W m2), surface winds (vectors, m s−1), and clouds (shaded, oktas) along the SCS ship track upon the NDJ(0/1) Niño-3.4 index as a function of calendar month and latitude in epochs (a) 1, (b) 2, (c) 3, and (d) 4. White contours for SST regressions are shown at values of ±0.1, ±0.2, ±0.3, ±0.4, ±0.5, and ±0.6; black contours for AtF-L regressions are shown at values of ±(5, 10, 15). (Note: The AtF-L is not shown in (a) and (b) (epochs 1 and 2) due to missing data; the cloud result for epoch 1 is from 1901 to 1915.)
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
In epoch 1, SST warming is observed in DJF(0/1). A large contribution comes from the decreased cloudiness (Fig. 7a; Table 1). After May(1), the increased cloudiness reduces shortwave radiation and cools the SCS. Epoch 2 is similar to epoch 1 (Fig. 7b; Table 1), but SST anomalies persist longer, until JJA(1). It seems that the local air–sea interaction favors the longer warming in this epoch. We will discuss this in the next section.
The contribution to the SCS warming during different epochs (SW = shortwave radiation; DJFMAM = December–May).
In epochs 3 and 4, both the cloudiness and the AtF-L anomalies contribute to the first warming in DJF(0/1) (Table 1). For the second warming, the AtF-L anomalies are the most important driver while the negative cloudiness anomalies dissipate, consistent with previous studies (Liu et al. 2004; Wang et al. 2006; Du et al. 2009; Table 1). The second warming persists longer in the recent epoch (Figs. 5d, 6d), due to the anomalous easterlies and positive AtF-L anomalies during JAS(1) (Fig. 7d; Table 1). In addition, stronger signals of cloudiness are also seen in epoch 4 (Fig. 7d; Table 1).
The anomalous winds also drive changes in ocean dynamic processes. In winter, the southerly wind anomalies in the SCS are part of the anomalous anticyclone circulation over the western Pacific (Alexander et al. 2002; Lau and Nath 2000), reducing the southward oceanic cold advection and favoring SST warming (Liu et al. 2004). In the summer following El Niño, the anomalous easterly winds weaken the upwelling along the east coast of Vietnam (Xie et al. 2003).
Wang et al. (2006) also showed the contribution of oceanic advections in inducing the SST warming in addition to surface heat flux during the ENSO event. We have examined the variability of oceanic advections by analyzing the Simple Oceanic Data Assimilation (SODA) version 2.2.4 (Giese and Ray 2011) for each epoch. Our oceanic heat budget analysis suggests that the oceanic advections contribute to the second warming during epochs 3 and 4 (figure not shown), consistent with the study of Wang et al. (2006).
4. Rainfall changes in surrounding regions
Pronounced SST and atmospheric anomalies are observed over the SCS in the previous sections, which can be crucial to the rainfall variability around this region. In this section, we evaluate how rainfall response to the decadal teleconnections of ENSO by using observed rainfall datasets. Figures 8 and 9 show the regression of rainfall upon the DJF(0/1) SOI index in 21-yr sliding windows. Since a positive (negative) SOI denotes a La Niña (an El Niño) situation, the SOI was multiplied by −1 before the correlation analysis to denote rainfall variability during El Niño event. The regions in southeastern China and Malaysia are selected to evaluate the impacts of ENSO on areas north and south of the SCS, respectively. Four stations in the Philippines use rain gauge data (Fig. 8a). In climatology, the rainy season is winter over Malaysia, the Philippines, and adjacent regions while the rainy season is summer over the eastern Indo-China Peninsula, south coast of China, and northwest Philippines (Fig. 8a). In the El Niño winter, rainfall decreases over Malaysia and the two southern Philippine stations (Figs. 8c, 9c,d), associated with the reduction of deep convection over the broad Maritime Continent. The El Niño–induced decrease of rainfall strengthens after the 1980s, from the winter [DJF(0/1)] and to the spring [MAM(1)]. Meanwhile, significant positive precipitation anomalies are found over southeastern China due to the anomalous southerly winds bringing more water vapor to the continent (Figs. 7c,d).These anomalies in southeastern China persist through December–April [DJFMA(0/1)] during the recent two epochs (Fig. 8b). The two stations in the northern Philippines are influenced by the anticyclonic atmospheric circulation over the northwest Pacific (Fig. 6 of Xie et al. 2009), with the summer rainfall anomalies strengthening after the 1980s (Figs. 9a,b). The El Niño–induced rainfall anomalies in the two stations of the southern Philippines take place in the winter (Figs. 9c,d). There are also significant precipitation anomalies around fall, that is September–November [SON(0)], after the mid-1970s (Figs. 9b,c), probably due to frequent tropical cyclone activities (Kubota and Wang 2009). Note that negative rainfall anomalies patterns during JAS(0) persist throughout the entire analysis period over northern Malaysia (Fig. 8c), which is due to high pressure anomalies controlling Malaysia in all of the four epochs in this season (figure not shown).Overall, the rainfall variability in the neighboring regions of the SCS indicates significant interdecadal changes in the response to ENSO.
(a) GPCP precipitation averaged for JJA (shaded, mm day−1) and DJF (contours, mm day−1) 1979–2009. The 21-yr running regression (shaded, mm day−1) of University of Delaware land rainfall anomalies in (b) southeastern China and (c) northern Malaysia upon the DJF(0/1) SOI index. Only rainfall anomalies with significant level exceeding 90% are shown. The red boxes in (a) show the selected two-regions and the red solid circles denote the selected stations in the Philippines (Fig. 9). (Note: the sign of SOI is flipped to denote rainfall variability during El Niño event).
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
The 21-yr running regression (shaded, mm day−1) of Philippine rainfall anomalies in (a)–(d) four stations for the DJF(0/1) SOI index from 1961 to 1996. Only rainfall anomalies exceeding a 90% significance level are shown. The locations of the four stations are shown with red solid circles in Fig. 8a. (Note: the sign of SOI is flipped to denote rainfall variability during El Niño event).
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
5. Discussion
a. Amplitude of ENSO and teleconnection to the SCS
The above analyses show remarkable interdecadal modulations of the impacts of ENSO on the SCS climate variability. Previous studies suggested that the strength of ENSO is an important factor for tropical climate variation through the teleconnection (e.g., Deser et al. 2010; Chowdary et al. 2012). Figure 10 shows that ENSO intensified after the 1970s when the SCS response displayed a double-peak SST structure. Enhanced ENSO activity strengthens the ENSO influence on the SCS SST. A similar relationship exists for precipitation (Figs. 8b,c). Some earlier studies suggested that the Pacific decadal oscillation (PDO) and interdecadal Pacific oscillation (IPO) display some coherence with ENSO teleconnections to the Indo–western Pacific after the 1950s. However, the relationships between the two variations and the Indo–western Pacific are very weak prior to the 1950s (Chowdary et al. 2012).
The 21-yr running standard deviation of NDJ(0/1) Niño-3.4 index (red line) and DJF(0/1) SOI index (blue line).
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
b. The role of the southeast tropical Indian Ocean
In the NIO, the SST response to ENSO peaks twice, much as in the SCS (Du et al. 2009). This relationship with ENSO experiences significant interdecadal changes (Xie et al. 2010; Chowdary et al. 2012). Because of the thermocline shoaling over the southwest tropical Indian Ocean, the NIO warming is more significant after the mid-1970s, resulting in significant atmospheric anomalies over the northwest Pacific (Xie et al. 2010; their Fig. 14). Compared to the single-peak structure over the NIO from 1950 to 1976 in Chowdary et al. (2012), however, the SCS SST response to ENSO shows a double-peak structure in epoch 3 (1960–83) (Figs. 4–7). Here, we examine the contribution of tropical Indian Ocean SST to the JJA(1) SCS warming during epochs 3 and 4. Figure 11 shows the regression of SST, SLP, surface wind, and precipitation anomalies upon the JJA(1) SCS SST (ICOADS SST averaged between 3° and 16°N). In the most recent epoch, a hemispherically asymmetric wind pattern develops in MAM(1) and the NIO warming persists to JJA(1) (Xie et al. 2009; Du et al. 2009). Meanwhile, the anticyclone over NWP in JJA(1) is stronger, reflecting the stronger influences of El Niño on Indo–western Pacific climate (Xie et al. 2010; there Fig. 14). The patterns are weak or different in epoch 3, except that the SST anomalies over the southeastern IO have not changed much between the recent two epochs. The SST signals are even larger in the period of 1960–83. Thus, enhanced convection associated with positive precipitation anomalies over the southeastern IO contributes to easterly wind anomalies over the SCS during JJA(1) in epoch 3 (Fig. 11c), which is important for the second summer warming over the region.
Regression coefficients of SLP (green contours, hPa), precipitation (blue contours; kg m−2 s−1), ERSST (shaded; °C), and surface wind (vectors, m s−1) upon the JJA(1) SCS SST in (a),(b) March(1)–May(1) and (c),(d) June(1)–August(1) for (left) 1960–83 and (right) 1984–2007.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
c. The influence of the Pacific–Japan/East Asian–Pacific (PJ/EAP) pattern
The summer SST warming over the SCS persists 1–2 months longer during epoch 4 than that during epoch 3. We suggest that the PJ/EAP pattern, an important mode of atmospheric variability during the summer over the NWP (Nitta 1987; Huang and Li 1987; Huang and Yan 1999; Kosaka and Nakamura 2006; Qu and Huang 2012; Kosaka et al. 2013), acts to strengthen the ENSO teleconnection to the SCS in the recent epoch. The PJ/EAP pattern index defined as the SLP difference between Yokohama, Japan, and Hengchun, Taiwan, in JJA (Huang and Yan 1999; Kubota et al. 2015). Figure 12 shows that the PJ/EAP pattern intensified in the most recent epoch, consistent with the elevated amplitude of ENSO. Epoch 2 is similar, but with a weaker amplitude. The SCS summer SST is highly correlated with the PJ/EAP index in epochs 2 and 4 (Fig. 12b), suggesting a contribution of PJ/EAP to the persistence of SST warming in these periods. Moreover, PJ/EAP is stronger and persists longer in epoch 4 (Fig. 13), as does the SST warming. This is due to the high pressure anomalies associated with the PJ/EAP pattern that change the shortwave radiation and surface winds to favor SST warming over the SCS (Fig. 7).
(a) The 21-yr running correlation of the NDJ (0/1) Niño-3.4 index and JJA(1) PJ/EAP index (red line) and 21-yr running STD of JJA (1) PJ/EAP index (blue line). (b) The 21-yr running correlation of SST anomalies (shaded) along the SCS ship track with the JJA(1) PJ/EAP index. Black dashed contours indicate a confidence level of 95%.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
Schematic diagram of the mechanism of the SCS second warming in the summer following El Niño during the more recent two epochs. The Indian Ocean basinwide warming (IOB; area 2) induced by ENSO (area 1) excites stronger anticyclonic circulation over NWP, accompanied by the PJ/EAP pattern (area 3) to warm the SCS during the period of 1984–2007. The southeastern tropical Indian Ocean (SETIO) warming (area 2) associated with ENSO (area 1) sustains easterly wind anomalies over the SCS to warm the SST during the period of 1960–83.
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
Figure 14 shows regression coefficients of SLP, latent heat flux, and surface winds of the twentieth-century reanalysis datasets for the NDJ(0/1) Niño-3.4 index for June(1)–July(1) [JJ(1)] and August(1)–September(1) [AS(1)] in epochs 3 and 4. The meridional dipole of PJ/EAP in SLP occurs in JJ(1) over the NWP and SCS in epoch 4 but not in epoch 3. In the recent epoch, an anticyclonic circulation, accompanied by positive SLP and latent heat flux anomalies, occupies the SCS. The anomalous circulations persist through AS(1), reflecting a stronger link between the PJ/EAP pattern and ENSO.
Regression coefficients of SLP (shaded, Pa), latent heat flux (white contours, W m2), and surface wind (vectors, m s−1) upon the NDJ(0/1) Niño-3.4 index in (a),(c) June(1)–July(1) and (b),(d) August(1)–September(1) of epochs (left) 3 and (right) 4. White contours for latent heat flux regressions are shown at values of ±(3, 6, 9).
Citation: Journal of Climate 28, 18; 10.1175/JCLI-D-15-0057.1
6. Summary
We have investigated the slow modulations of ENSO effects on climate variability over the SCS. The ENSO-induced SCS SST warming varies from decade to decade. Using frequent ship track observations, four epochs are selected based on the warming features and their relationship with ENSO: 1892–1915 (epoch 1), 1930–40 (epoch 2), 1960–83 (epoch 3), and 1984–2007 (epoch 4).
A double-peak SST warming is found in epochs 3 and 4 and a single-peak SST warming in epochs 1 and 2. In epochs 3 and 4, the first SST warming occurs in the mature phase of ENSO [DJF(0/1)] whereas the second SST warming occurs in JJA(1) and JJAS(1), respectively. The SST warming in the recent epoch is more significant and persists longer. For the SST warming in the winter, the cloudiness decreases first followed by southerly wind anomalies. Both changes cause SST warming by increasing downward shortwave radiation and reducing the ocean surface latent heat release, respectively. For the second warming in the following summer in epochs 3 and 4, the positive AtF-L anomalies due to the easterly wind anomalies are the main driver. Consistently, precipitation around the SCS shows similar interdecadal changes in the relationships to ENSO.
In epoch 3 (from 1960 to 1983), the second SST warming is present in the SCS but absent in the NIO (Xie et al. 2010; Chowdary et al. 2012). Further investigation suggests that the SST warming and enhanced convection over the southeastern tropical Indian Ocean sustain the SCS warming. In epoch 4, from 1984 to 2007, besides the influence of the capacitor effect from the tropical Indian Ocean, the PJ/EAP pattern helps prolong the second SCS SST warming to the early fall season, 1–2 months longer than that in epoch 3. A similar PJ/EAP effect on the SCS SST is found in epoch 2. The persistence of positive SLP anomalies over the NWP and the easterly winds anomalies of the PJ/EAP pattern helps sustain the SCS SST warming through JJAS(1) in the recent epoch. The result supports the idea that the interaction between the PJ/EAP pattern and NIO–SCS SST results in positive feedback, which helps sustain the coupled anomalies during the summer (Kosaka et al. 2013).
Acknowledgments
We thank J. S. Chowdary of the Indian Institute of Tropical Meteorology, Y. Kosaka of the University of Tokyo, and G. Li of LTO/SCSIO for their valuable and constructive comments, which helped to improve the manuscript. We acknowledge the Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory (http://rda.ucar.edu/datasets/ds540.0) for providing ICOADS version 2.5 and the Met Office Hadley Centre (http://www.metoffice.gov.uk) for providing HadISST. The ERSST, Kaplan SST, OISST, COBE-SST, the University of Delaware precipitation, GPCP precipitation, and the 20th Century Reanalysis datasets were obtained from NOAA/ESRL (http://www.esrl.noaa.gov/psd/data). The Philippine stations precipitation and PJ/EAP pattern index were provided by H. Kubota. This work is supported by the Ministry of Science and Technology (2012CB955603), the Chinese Academy of Sciences (XDA11010103), the National Science Foundation of China (41176024), and the CAS/SAFEA International Partnership Program for Creative Research Teams.
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, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Alexander, M. A., I. Bladé, M. Newman, J. R. Lanzante, N.-C. Lau, and J. D. Scott, 2002: The atmospheric bridge: The influence of ENSO teleconnections on air–sea interaction over the global oceans. J. Climate, 15, 2205–2231, doi:10.1175/1520-0442(2002)015<2205:TABTIO>2.0.CO;2.
Allan, R. J., N. Nicholls, P. D. Jones, and I. J. Butterworth, 1991: A further extension of the Tahiti–Darwin SOI, early SOI results and Darwin pressure. J. Climate, 4, 743–749, doi:10.1175/1520-0442(1991)004<0743:AFEOTT>2.0.CO;2.
Alory, G., S. Wijffels, and G. Meyers, 2007: Observed temperature trends in the Indian Ocean over 1960–1999 and associated mechanisms. Geophys. Res. Lett., 34, L02606, doi:10.1029/2006GL028044.
Bretherton, C. S., M. Widmann, V. P. Dymnikov, J. M. Wallace, and I. Bladé, 1999: The effective number of spatial degrees of freedom of a time-varying field. J. Climate, 12, 1990–2009, doi:10.1175/1520-0442(1999)012<1990:TENOSD>2.0.CO;2.
Chowdary, J. S., S.-P. Xie, H. Tokinaga, Y. M. Okumura, H. Kubota, N. Johnson, and X.-T. Zheng, 2012: Interdecadal variations in ENSO teleconnection to the Indo–western Pacific for 1870–2007. J. Climate, 25, 1722–1744, doi:10.1175/JCLI-D-11-00070.1.
Compo, G. P., and Coauthors, 2011: The Twentieth Century Reanalysis project. Quart. J. Roy. Meteor. Soc., 137, 1–28, doi:10.1002/qj.776.
Deser, C., A. S. Phillips, and M. A. Alexander, 2010: Twentieth century tropical sea surface temperature trends revisited. Geophys. Res. Lett., 37, L10701, doi:10.1029/2010GL043321.
de Szoeke, S. P., S.-P. Xie, T. Miyama, K. J. Richards, and R. J. O. Small, 2007: What maintains the SST front north of the eastern Pacific equatorial cold tongue? J. Climate, 20, 2500–2514, doi:10.1175/JCLI4173.1.
Du, Y., and S.-P. Xie, 2008: Role of atmospheric adjustments in the tropical Indian Ocean warming during the 20th century in climate models. Geophys. Res. Lett., 35, L08712, doi:10.1029/2008GL033631.
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, doi:10.1175/2008JCLI2590.1.
Du, Y., S.-P. Xie, Y.-L. Yang, X.-T. Zheng, L. Liu, and G. Huang, 2013: Indian Ocean variability in the CMIP5 multimodel ensemble: The basin mode. J. Climate, 26, 7240–7266, doi:10.1175/JCLI-D-12-00678.1.
Giese, B. S., and S. Ray, 2011: El Niño variability in simple ocean data assimilation (SODA), 1871–2008. J. Geophys. Res., 116, C02024, doi:10.1029/2010JC006695.
Huang, G., and Z. Yan, 1999: The East Asian summer monsoon circulation anomaly index and its interannual variations. Chin. Sci. Bull., 44, 1325–1329, doi:10.1007/BF02885855.
Huang, G., K. Hu, and S.-P. Xie, 2010: Strengthening of tropical Indian Ocean teleconnection to the northwest Pacific since the mid-1970s: An atmospheric GCM study. J. Climate, 23, 5294–5304, doi:10.1175/2010JCLI3577.1.
Huang, R., and W. Li, 1987: Influence of heat source anomaly over the western tropical Pacific on the subtropical high over East Asia. Proc. Int. Conf. on the General Circulation of East Asia, Chengdu, China, 40–51.
Huang, R., and Y. Wu, 1989: The influence of ENSO on the summer climate change in China and its mechanism. Adv. Atmos. Sci., 6, 21–32, doi:10.1007/BF02656915.
Ishii, M., A. Shouji, S. Sugimoto, and T. Matsumoto, 2005: Objective analyses of sea surface temperature and marine meteorological variables for the 20th century using ICOADS and the Kobe collection. Int. J. Climatol., 25, 865–879, doi:10.1002/joc.1169.
Kaplan, A., M. A. Cane, Y. Kushnir, A. C. Clement, M. B. Blumenthal, and B. Rajagopalan, 1998: Analyses of global sea surface temperature 1856–1991. J. Geophys. Res., 103, 18 567–18 589, doi:10.1029/97JC01736.
Kent, E. C., S. D. Woodruff, and D. I. Berry, 2007: Metadata from WMO Publication No. 47 and an assessment of voluntary observing ship observation heights in ICOADS. J. Atmos. Oceanic Technol., 24, 214–234, doi:10.1175/JTECH1949.1.
Klein, S. A., B. J. Soden, and N.-C. Lau, 1999: Remote sea surface temperature variations during ENSO: Evidence for a tropical atmospheric bridge. J. Climate, 12, 917–932, doi:10.1175/1520-0442(1999)012<0917:RSSTVD>2.0.CO;2.
Kosaka, Y., and H. Nakamura, 2006: Structure and dynamics of the summertime Pacific–Japan teleconnection pattern. Quart. J. Roy. Meteor. Soc., 132, 2009–2030, doi:10.1256/qj.05.204.
Kosaka, Y., S.-P. Xie, N.-C. Lau, and G. A. Vecchi, 2013: Origin of seasonal predictability for summer climate over the northwestern Pacific. Proc. Natl. Acad. Sci. USA, 110, 7574–7579, doi:10.1073/pnas.1215582110.
Kubota, H., and J. C. L. Chan, 2009: Interdecadal variability of tropical cyclone landfall in the Philippines from 1902 to 2005. Geophys. Res. Lett., 36, L12802, doi:10.1029/2009GL038108.
Kubota, H., and B. Wang, 2009: How much do tropical cyclones affect seasonal and interannual rainfall variability over the western North Pacific? J. Climate, 22, 5495–5510, doi:10.1175/2009JCLI2646.1.
Kubota, H., Y. Kosaka, and S.-P. Xie, 2015: A 116-year long index of the Pacific-Japan pattern and its interdecadal variability. Int. J. Climatol., in press.
Lau, N.-C., and M. J. Nath, 2000: Impact of ENSO on the variability of the Asian–Australian monsoons as simulated in GCM experiments. J. Climate, 13, 4287–4309, doi:10.1175/1520-0442(2000)013<4287:IOEOTV>2.0.CO;2.
Liu, Q., X. Jiang, S.-P. Xie, and W. T. Liu, 2004: A gap in the Indo-Pacific warm pool over the South China Sea in boreal winter: Seasonal development and interannual variability. J. Geophys. Res., 109, C07012, doi:10.1029/2003JC002179.
Nitta, T., 1987: Convective activities in the tropical western Pacific and their impacts on the Northern Hemisphere summer circulation. J. Meteor. Soc. Japan, 65, 373–390.
Norris, J. R., 1999: On trends and possible artifacts in global ocean cloud cover between 1952 and 1995. J. Climate, 12, 1864–1870, doi:10.1175/1520-0442(1999)012<1864:OTAPAI>2.0.CO;2.
Qu, X., and G. Huang, 2012: Impacts of tropical Indian Ocean SST on the meridional displacement of East Asian jet in boreal summer. Int. J. Climatol., 32, 2073–2080, doi:10.1002/joc.2378.
Rayner, N. A., and Coauthors, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, doi:10.1029/2002JD002670.
Reynolds, R. W., N. A. Rayner, T. M. Smith, D. C. Stokes, and W. Wang, 2002: An improved in situ and satellite SST analysis for climate. J. Climate, 15, 1609–1625, doi:10.1175/1520-0442(2002)015<1609:AIISAS>2.0.CO;2.
Schott, F. A., S.-P. Xie, and J. P. McCreary Jr., 2009: Indian Ocean circulation and climate variability. Rev. Geophys., 47, RG1002, doi:10.1029/2007RG000245.
Smith, T. M., and R. W. Reynolds, 2004: Improved extended reconstruction of SST (1854–1997). J. Climate, 17, 2466–2477, doi:10.1175/1520-0442(2004)017<2466:IEROS>2.0.CO;2.
Tokinaga, H., and Y. Tanimoto, 2004: Seasonal transition of SST anomalies in the tropical Indian Ocean during El Niño and Indian Ocean dipole years. J. Meteor. Soc. Japan, 82, 1007–1018, doi:10.2151/jmsj.2004.1007.
Tokinaga, H., and S.-P. Xie, 2011: Wave- and Anemometer-Based Sea Surface Wind (WASWind) for climate change analysis. J. Climate, 24, 267–285, doi:10.1175/2010JCLI3789.1.
Tokinaga, H., S.-P. Xie, C. Deser, Y. Kosaka, and Y. M. Okumura, 2012a: Slowdown of the Walker circulation driven by tropical Indo-Pacific warming. Nature, 491, 439–443, doi:10.1038/nature11576.
Tokinaga, H., S.-P. Xie, A. Timmermann, S. McGregor, T. Ogata, H. Kubota, and Y. M. Okumura, 2012b: Regional patterns of tropical Indo-Pacific climate change: Evidence of the Walker circulation weakening. J. Climate, 25, 1689–1710, doi:10.1175/JCLI-D-11-00263.1.
Trenberth, K. E., J. M. Caron, D. P. Stepaniak, and S. Worley, 2002: Evolution of El Niño–Southern Oscillation and global atmospheric surface temperatures. J. Geophys. Res., 107 (D8), doi:10.1029/2000JD000298.
Wang, B., and S.-I. An, 2005: A method for detecting season-dependent modes of climate variability: S-EOF analysis. Geophys. Res. Lett., 32, L15710, doi:10.1029/2005GL022709.
Wang, B., R. Wu, and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13, 1517–1536, doi:10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.
Wang, B., R. Wu, and T. Li, 2003: Atmosphere–warm ocean interaction and its impacts on Asian–Australian monsoon variation. J. Climate, 16, 1195–1211, doi:10.1175/1520-0442(2003)16<1195:AOIAII>2.0.CO;2.
Wang, C., R. H. Weisberg, and J. I. Virmani, 1999: Western Pacific interannual variability associated with the El Niño–Southern Oscillation. J. Geophys. Res., 104 (C3), 5131–5149, doi:10.1029/1998JC900090.
Wang, C., W. Wang, D. Wang, and Q. Wang, 2006: Interannual variability of the South China Sea associated with El Niño. J. Geophys. Res., 111, C03023, doi:10.1029/2005JC003333.
Weisberg, R. H., and C. Wang, 1997: A western Pacific oscillator paradigm for the El Niño–Southern Oscillation. Geophys. Res. Lett., 24, 779–782, doi:10.1029/97GL00689.
Willmott, C. J., and K. Matsuura, 2001: Terrestrial air temperature and precipitation: Monthly and annual time series (1950–1999). [Available online at http://climate.geog.udel.edu/~climate/html_pages/README.ghcn_ts2.html.]
Woodruff, S. D., and Coauthors, 2011: ICOADS release 2.5: Extensions and enhancements to the surface marine meteorological archive. Int. J. Climatol., 31, 951–967, doi:10.1002/joc.2103.
Worley, S. J., S. D. Woodruff, R. W. Reynolds, S. J. Lubker, and N. Lott, 2005: ICOADS release 2.1 data and products. Int. J. Climatol., 25, 823–842, doi:10.1002/joc.1166.
Wu, R., B. P. Kirtman, and V. Krishnamurthy, 2008: An asymmetric mode of tropical Indian Ocean rainfall variability in boreal spring. J. Geophys. Res., 113, D05104, doi:10.1029/2007JD009316.
Wu, R., W. Chen, G. Wang, and K. Hu, 2014: Relative contribution of ENSO and East Asian winter monsoon to the South China Sea SST anomalies during ENSO decaying years. J. Geophys. Res. Atmos., 119, 5046–5064, doi:10.1002/2013JD021095.
Wu, Y.-L., Y. Du, Y.-H. Zhang, and X.-T. Zheng, 2012: Interannual variability of sea surface temperature in the northern Indian Ocean associated with ENSO and IOD. Atmos. Oceanic Sci. Lett., 5, 295–300.
Xie, S.-P., Q. Xie, D. X. Wang, and W. T. Liu, 2003: Summer upwelling in the South China Sea and its role in regional climate variations. J. Geophys. Res., 108, 3261, doi:10.1029/2003JC001867.
Xie, S.-P., K. 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, doi:10.1175/2008JCLI2544.1.
Xie, S.-P., Y. Du, G. Huang, X.-T. Zheng, H. Tokinaga, K. Hu, and Q. Liu, 2010: Decadal shift in El Niño influences on Indo–western Pacific and East Asian climate in the 1970s. J. Climate, 23, 3352–3368, doi:10.1175/2010JCLI3429.1.
Yang, J., Q. Liu, S.-P. Xie, Z. Liu, and L. Wu, 2007: Impact of the Indian Ocean SST basin mode on the Asian summer monsoon. Geophys. Res. Lett., 34, L02708, doi:10.1029/2006GL028571.
