1. Introduction
During boreal winter, the South China Sea (SCS) displays climatologically warm sea surface temperatures (SSTs) strongly regulated by surface heat fluxes involving air–sea interaction (Liu et al. 2004; Koseki et al. 2013; Wu and Chen 2015; Thompson et al. 2016; Seow and Tozuka 2019; Xiao et al. 2018). Located in the center of the Asian–Australian monsoon system (Chang et al. 2005b), SSTs over the SCS are strongly forced by the East Asian winter monsoon, but the SSTs in turn influence the regional monsoon (Koseki et al. 2013; Wu and Chen 2015; Wu 2016).
A key feature of the SCS winter SST pattern is an elongated cool SST pool extending from southern China to the Vietnam coast and to peninsular Malaysia. This is known as the “cold tongue” (CT) (Liu et al. 2004; Varikoden et al. 2010; Koseki et al. 2013; Thompson et al. 2016). Climatologically, from November to February, the SCS winter monsoon results in latent heat loss and southward advection of cold water over the western SCS, leading to the formation of CT (Seow and Tozuka 2019; Fig. 1a). On the interannual time scale, the northern CT (i.e., 10°–20°N) variability is mainly controlled by latent heat, while the southern CT (i.e., 2°–10°N) variability is controlled by horizontal advection. These are contrasted with the basinwide SST variability, which is mainly controlled by latent heat flux anomalies under El Niño–Southern Oscillation (ENSO) events, as noted by Wang et al. (2006) and Liu et al. (2014).
(a) NDJ mean 850-hPa wind and DJF mean SST over the SCS region averaged in the reanalysis/observation data. The (b) first and (c) second modes of MEOF structures of NDJ mean 850-hPa wind and DJF mean SST anomalies over the 1982–2015 period expressed as covariances between PC with SST and wind. Percentages of explained total variance for each mode of MEOF are indicated in parentheses. SST and wind data within the green box covering the SCS (0°–20°N, 100°–120°E) in (b) and (c) are the input data for the MEOF analysis. Covariances between unscaled first PC time series with NDJ (d) net shortwave radiation, (e) latent heat flux, and (f) horizontal advection and surface oceanic current. (g) First mode PC time series, with dashed lines indicating the one standard deviation thresholds and blue (red) solid circles indicating strong (weak) CT years.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
At the same time, the winter monsoon is strongly modulated by various tropical remote forcings, such as ENSO (Wang et al. 2000; Wang and Zhang 2002) in the tropical Pacific and the Indian Ocean basin mode (IOBM; Watanabe and Jin 2002; Annamalai et al. 2005) and Indian Ocean dipole (IOD; Yang et al. 2010; Yuan et al. 2012; M. Chen et al. 2016) in the tropical Indian Ocean. Those adjacent oceans affect the SCS atmospheric conditions through the Matsuno–Gill response (Matsuno 1966; Gill 1980). For example, different ENSO types influence or are influenced by the Indian Ocean, which in turn control the strength of anticyclonic wind anomaly over the SCS and western Pacific (e.g., M. Chen et al. 2019; Xiao et al. 2020). However, it is still difficult to quantify the relative contributions of these interannual remote forcings on atmospheric circulation anomalies over the SCS and the CT variability. This is partly because the tropical Pacific and Indian Oceans are known to exert influences on each other. For instance, the ENSO can force both positive IOBM and IOD via atmospheric equatorial Rossby waves (Ashok et al. 2003; Annamalai et al. 2005; M. Chen et al. 2019) and the IOD can influence the western tropical Pacific via an atmospheric Kelvin wave that in turn affects the ENSO evolution (Luo et al. 2010; Wu et al. 2012).
Another factor is the atmospheric internal variability. For example, many studies have characterized convectively coupled equatorial waves occurring at the intraseasonal time scale, like atmospheric Kelvin waves (Baranowski et al. 2016; W. T. Chen et al. 2019), tropical depression-type waves (Fukutomi 2019), and the Madden–Julian oscillation (MJO; Madden and Julian 1971; Chang et al. 2005a; Lim et al. 2017). However, no study to date has investigated the impact of atmospheric internal variability on the SCS atmospheric circulation and CT at the interannual time scale.
Motivated by the above, we aim to understand how each tropical remote forcing affects the SCS atmospheric circulation that controls the CT variability on the interannual time scale using observational data, reanalysis products, and coupled model experiments. We first provide a description of the data and model experiment designs in sections 2 and 3, respectively. Section 4 outlines the multivariate empirical orthogonal function (MEOF) method used to capture the coupled air–sea modes of variability. Section 5 discusses the winter atmospheric patterns that control the CT variability via reanalysis and observational data analyses. Section 6 analyzes the relative contributions of different tropical remote forcings through coupled model experiments. Section 7 summarizes the key results and suggests potential future research directions.
2. Data
We focus on the study period when satellite observations are available for the SST, which spans from 1982 to 2015. For all the datasets described next, monthly mean data are used. The SST data with a 1° × 1° horizontal resolution are from the National Oceanic and Atmospheric Administration (NOAA) Optimum Interpolation Sea Surface Temperature (OISST), version 2, dataset. It is based on the Pathfinder Advanced Very High Resolution Radiometer (AVHRR) infrared satellite data from 1982 to 2005 and operational AVHRR data from 2006 onward (Reynolds et al. 2002). The atmospheric data of 850-hPa zonal and meridional winds and surface heat fluxes are from the ERA-Interim reanalysis dataset of the European Centre for Medium-Range Weather Forecasts (ECMWF). They have a T255 spectral resolution and 60 vertical levels from the surface to 0.1 hPa (Dee et al. 2011). The zonal and meridional surface ocean current data with a 1° × 1° horizontal resolution were obtained from the Ocean Reanalysis System 4 (ORAS4), as implemented by the ECMWF based on the Nucleus for European Modeling of the Ocean (NEMO) model (Balmaseda et al. 2013). The rainfall data of a 2.5° × 2.5° horizontal resolution were obtained from the Global Precipitation Climatology Project (GPCP), version 2.3, dataset, which combines observations by satellite over land and oceans, and those of land rain gauges and soundings (Adler et al. 2003). Anomalies are calculated by removing the monthly climatologies, and the long-term linear trend is removed via a least squares fit in all observational and reanalysis data and model results.
3. Model and experiment design
We use the version 2 of the Scale Interaction Experiment Frontier (SINTEX-F2) coupled model (Masson et al. 2012). The atmospheric component is ECHAM 5.3, which has a horizontal resolution of T106 with 31 vertical levels (Roeckner et al. 2003). A mass flux scheme (Tiedtke 1989) is applied for cumulus convection with modifications for penetrative convection (Nordeng 1994). The oceanic component is NEMO, which has a horizontal resolution of ORCA05 with 31 vertical levels without any further refinement over the tropics (Madec 2008). The model is also embedded with the Louvain-la-Neuve Sea Ice Model, version 2 (LIM2; Timmermann et al. 2005). The atmospheric and oceanic fluxes are exchanged every 2 h with no flux correction using the Ocean Atmosphere Sea Ice Soil, version 3 (OASIS3), coupler (Valcke 2013). Previous studies (e.g., Terray et al. 2012; Doi et al. 2016) have shown that the model can realistically simulate tropical and midlatitude interannual variability over the Indian and Pacific Oceans.
We have performed five experiments to understand the relative contributions of different remote forcings to the SCS atmospheric circulation governing the CT variability. Each experiment is integrated for 100 years, but the first 30 years are treated as the spinup run, and only the last 70 years are analyzed. In the control run, the ocean and atmosphere are freely and globally coupled. In the tropical Pacific and Maritime Continent (TPMC) [tropical Indian Ocean and Maritime Continent (TIOMC)] run, the ocean and atmosphere are freely coupled only over the tropical Pacific (Indian Ocean) and Maritime Continent, whereas the SST in other areas is strongly nudged to the monthly SST climatology of the control run using the surface heat flux. The ocean and atmosphere are freely coupled only over the Maritime Continent, and the SST in other areas is strongly nudged to the control run climatology in the Maritime Continent (MC) run. Last, in the no tropical Indian Ocean–tropical Pacific (noTIOTP) run, the ocean and atmosphere are freely coupled everywhere except over the tropical Pacific and Indian Oceans. The air–sea coupling and SST nudging areas are illustrated in Fig. 2, and the remote forcings that influence the SCS atmospheric circulation and CT variabilities for each experiment are summarized in Table 1. By comparing results of all five experiments, we can quantitatively assess the individual impacts of each tropical remote forcing within this model. Such decoupled experiments using SINTEX-F2 have been successfully used to understand tropical remote forcing impacts in other regions (e.g., Morioka et al. 2014; Prodhomme et al. 2015; Crétat et al. 2018; Kataoka et al. 2018), but this is the first time that the method is applied to the SCS.
Air–sea coupling (shading) and SST nudging (white) areas for (a) TPMC, (b) TIOMC, (c) MC, and (d) noTIOTP runs.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
Types of remote forcing for the control, TPMC, TIOMC, MC, and noTIOTP runs.
4. MEOF analysis
5. Local and remote atmospheric impacts on CT variability
a. Local atmospheric impact on CT
Since the SST lags the winter mixed layer temperature tendency by one month in the SCS (Seow and Tozuka 2019), we perform an MEOF analysis on the November–January (NDJ) mean wind and December–February (DJF) mean SST data. The leading mode of MEOF, expressed as covariances between the leading mode PC time series with SST and 850-hPa wind, captures strong CT associated with cyclonic wind anomalies at the center of SCS (Fig. 1b). The spatial pattern shows a zonal SST contrast that defines the CT feature, consistent with the CT definition adopted by Varikoden et al. (2010) and Seow and Tozuka (2019). We also calculate the CT SST index defined based on the DJF SST mean contrast between the CT (2°–20°N, 103°–113°W) and non-CT (2°–10°N, 113°–120°E) regions, which is the area-averaged SST difference between the CT and non-CT regions, with the long-term linear trend removed (Seow and Tozuka 2019). The CT SST index is strongly correlated with the first PC time series at −0.82, which is statistically significant at the 99% confidence level based on the Student’s t test. The first MEOF mode explains 50% of the total variance, while the second mode (Fig. 1c), capturing the basinwide SST variation linked to the prevailing northeast monsoon variability, explains one-fifth of the total variance. Since the eigenvalues between the first two modes are well separated according to the standard error criterion by North et al. (1982) and the first PC time series is strongly correlated with the CT SST index, we focus on the first mode as a measure of CT and its associated atmospheric forcing variabilities.
b. Remote atmospheric impacts—Tropical Pacific and Indian Oceans
To identify combined impacts of air–sea coupling over the tropical Pacific and Indian Oceans on the SCS cyclonic wind anomaly and CT variations, we analyze the NDJ mean Indo-Pacific 850 hPa wind, rainfall, and SST anomaly composite maps of strong CT years (Figs. 3b,d). We refer to the first PC time series and define a strong (weak) CT year in which the PC value lies above (below) the one (negative one) standard deviation threshold (Fig. 1g). We also calculate how significantly different the anomalies are between the strong and nonstrong (nonweak and weak) CT years composites based on the nonparametric two-tailed Mann–Whitney U test at the 90% confidence level. Since composites of strong and weak CT years are nearly symmetric to each other, we focus only on the strong CT years throughout the rest of this paper.
Composites of SST and 850-hPa wind anomalies for strong CT years in (a) SON and (b) NDJ. (c),(d) As in (a) and (b), but for rainfall anomalies for SON [in (c)] and NDJ [in (d)]. A total of 5 years (1995/96, 1996/97, 1998/99, 2005/06, and 2007/08 September–January) out of 33 years of the study period are included in the composites. Only SST and rainfall anomalies significantly different from nonstrong CT years based on the two-tailed Mann–Whitney U nonparametric test at the 90% confidence level are shown.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
In Fig. 3b, we observe negative (positive) SST anomalies over the central-eastern (western) equatorial Pacific, implying that strong CT years are accompanied by canonical La Niña episodes. Over the Indian Ocean, a negative IOD pattern is seen during the September–November (SON) period (Fig. 3a), whereas basinwide SST cooling over the equatorial Indian Ocean forced by La Niña, representing the negative IOBM, can also be seen (Fig. 3b; Chowdary and Gnanaseelan 2007). A pair of off-equatorial cyclonic wind anomalies straddles over the SCS and northwest Australia, around 120°E longitudinal line. The cyclonic wind anomalies are accompanied by positive (negative) rainfall anomalies that represent enhanced (suppressed) deep convection and diabatic heating (cooling) anomalies occurring over the Maritime Continent and equatorial western Pacific (western equatorial Indian Ocean) as seen in Fig. 3d. Westerly and easterly wind anomalies lie to the west and east of the cyclonic wind anomaly pair, respectively. These anomalies represent the classical Matsuno–Gill pattern (Matsuno 1966; Gill 1980), where diabatic heating anomalies over the western equatorial Pacific force easterly anomalies over the central equatorial Pacific as a Kelvin wave response and a pair of cyclonic wind anomalies over the SCS and northwest Australia as a Rossby wave response. On the other hand, diabatic cooling anomalies in the western and central equatorial Indian Ocean induce an atmospheric Kelvin wave and equatorial ridge anomalies in the form of westerly wind anomalies over the central and eastern equatorial Indian Ocean. As explained by Xie et al. (2009), the equatorial ridge anomalies associated with the Kelvin wave enhances off-equatorial tropospheric positive (negative) wind curl anomalies, convergence, and convection over the western Pacific (northwest Australia). The diabatic heating anomalies arising from the enhanced off-equatorial convergence excite a Rossby wave response that amplifies the cyclonic wind anomaly pair under the circulation–convection feedback (Xie et al. 2009). The atmospheric Kelvin wave mechanism is validated by the model experiments in which air–sea coupling over the western Pacific and northwest Australian basins are suppressed, as discussed later in section 6c. Thus, it seems that both the equatorial Pacific and Indian Oceans contribute to the cyclonic wind anomaly pair during La Niña winters, as corroborated by Z. Chen et al. (2016).
We further confirm the relationship between the first PC time series with ENSO, IOBM, and IOD via a correlation analysis. We define the ENSO index as the area-averaged NDJ mean of SST anomalies in the Niño-3.4 region (5°S–5°N, 170°–120°W). The IOBM index is the area-averaged NDJ mean of SST anomalies in the tropical Indian Ocean basin (10°S–10°N, 40°–100°E). The IOD index is defined as the difference of area-averaged SON SST anomalies over the western (10°S–10°N, 50°–70°E) and eastern (10°S–0°, 90°–110°E) Indian Ocean. The correlations of the first PC time series with the ENSO, IOBM, and IOD indices are −0.74, −0.55, and −0.68, respectively. These correlations are statistically significant at the 99% confidence level based on the Student’s t test.
Over the western equatorial Pacific, we can interpret the winter diabatic heating anomalies as a response to the local positive autumn and winter SST anomalies underneath them (Figs. 3a,b). However, over the equatorial Indian Ocean, the winter atmosphere is sensitive to the winter SST only over the western basin instead of the whole basin. This is seen from the diabatic heating anomaly dipole pattern accompanying the underlying basinwide SST cooling over the Indian Ocean during winter (Fig. 3b). In agreement with Kohyama and Tozuka (2016), observational analyses show that during winter the local SST controls the convection strength only over the western equatorial Indian Ocean, whereas it is the opposite over the eastern equatorial Indian Ocean. While the westerly wind anomalies prevailing over the equatorial Indian Ocean and Maritime Continent are forced by the diabatic heating anomaly dipole pattern, the negative IOBM is not behind such diabatic heating anomaly dipole pattern. This is confirmed by M. Chen et al. (2016) in their SST sensitivity experiments; the westerly wind anomalies cannot be forced under the tropical Indian Ocean basinwide SST cooling. The dipole pattern of diabatic heating anomalies seem to be related to the preceding autumn negative IOD, when SST tends to force convection over the whole tropical Indian Ocean in autumn that persists into winter (Fig. 3a; M. Chen et al. 2016).
c. Remote atmospheric impact—Atmospheric internal variability
Apart from the contributions from both equatorial Pacific and Indian Oceans, atmospheric internal variability may also play a role in the SCS cyclonic wind anomalies during the strong CT years. We investigate this by first removing ENSO and IOBM years and repeating the MEOF analysis. Given that the heating anomaly dipole pattern is related to the IOD, we also remove ENSO and IOD years and repeat the MEOF analysis. Years in which the ENSO index is below the −0.5 (above the 0.5) standard deviation threshold are defined as La Niña (El Niño); years in which the IOBM index is below (above) the −1 (1) standard deviation threshold are defined as negative (positive) IOBM events. Years with the IOD index below (above) the −1 (1) standard deviation threshold are defined as negative (positive) IOD years. We obtain 6 (10) years out of 33 years of the study period that are non-ENSO and non-IOBM (non-IOD) years. As the sample size after removing ENSO, IOBM and IOD years is small, we only show the MEOF patterns but not composite analyses of strong events without ENSO, IOBM, and IOD.
As shown in Fig. 4, results for the removal of ENSO and IOBM, and ENSO and IOD years are consistent with each other. The first MEOF patterns in Fig. 4 show a CT feature with aloft cyclonic wind anomalies similar to Fig. 1b. This suggests that atmospheric internal variability may also be responsible for the cyclonic wind anomalies.
As in Fig. 1b, but after removing (a) ENSO and IOBM years and (b) ENSO and IOD years prior to MEOF analysis. A total of 6 (10) years out of 33 years of the study period are non-ENSO and non-IOBM (non-IOD) years.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
6. Relative contributions of different tropical remote forcings
Since observational data analysis cannot totally distinguish contributions from both the tropical forcings and atmospheric internal variability, we further investigate their relative contributions using model experiments. Comparing the SCS SST and 850-hPa wind climatology in the control run (Fig. 5a) with those in the OISST and ERA-interim (Fig. 1a), the model can well simulate the wind climatology. Although the simulated SST climatology is warmer than the OISST data, the model well captures the winter CT structure. By carrying out the same MEOF analysis outlined in section 4a on the control run results, the first mode shows a stronger CT accompanied by stronger cyclonic wind anomalies over the SCS (Fig. 5b), despite the fact that the center of cyclonic wind anomalies is located over the east of Philippines, not over the SCS as in the observation (Fig. 1b). The first mode also explains nearly half of the total variance, which is consistent with the observation. It is also found that NDJ mean 850-hPa wind, rainfall, and SST anomaly composites in the control run (Fig. S1 in the online supplemental material) are largely consistent with those in the observation (Fig. 3).
(a) As in Fig. 1a, but for the last 70 years of the control run. (b)–(f) As in Fig. 1b, but for the control [in (b)], TPMC [in (c)], TIOMC [in (d)], MC [in (e)], and noTIOTP [in (f)] runs.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
a. Interexperimental differences in SCS cyclonic wind anomaly and CT variabilities
The NDJ means of SCS SST, rainfall, and 850-hPa wind anomalies in the TPMC, TIOMC, MC, and noTIOTP runs differ only slightly from those of the control run (Fig. S2). Also, the leading MEOF mode of the four sensitivity experiments shows similar spatial structures, where the CT feature is accompanied by aloft cyclonic wind anomalies (Figs. 5c–f). For all five experiments, including the control run, the eigenvalues between the first two modes are well separated according to the standard error criterion by North et al. (1982). However, the center of the cyclonic wind anomalies is different; it is located over the SCS in the TIOMC, MC, and noTIOTP runs, while it is located over the Philippines in the TPMC run. We note that the percentage of explained variance is only useful for knowing how dominant the first PC is compared to the rest of PCs within each experiment, and it cannot be used to directly compare between different experiments.
b. Qualitative analysis of tropical remote forcings
Next, we identify the influences of various tropical remote forcings in the four sensitivity experiments onto the SCS cyclonic wind anomaly and CT variabilities. For this purpose, composites of strong CT years (defined when the first PC time series lie above the one standard deviation threshold) are prepared for SON and NDJ mean 850-hPa wind, rainfall, and SST anomalies over the wider tropical Indo-Pacific region (Figs. 6–9). In the TPMC run, a pair of cyclonic wind anomalies with significantly positive rainfall anomalies over the Philippines and northwest Australia and strong westerly (easterly) anomalies over the equatorial Indian Ocean (western Pacific) are seen in Figs. 6c and 6d. Also, the Pacific SST anomaly pattern features a canonical La Niña pattern (Figs. 6a,b). All of these are consistent with the observation. Moreover, the SCS cyclonic wind anomalies may be generated as a Rossby wave response to diabatic heating anomalies over the warmer-than-normal western equatorial Pacific and reinforced by an atmospheric Kelvin wave response to diabatic cooling anomalies over the air–sea decoupled tropical Indian Ocean (Figs. 6c,d). Such diabatic cooling anomalies could be related to both the Walker circulation’s subsiding anomalies caused by La Niña and atmospheric internal variability.
As in Fig. 3, but for the TPMC run. A total of 13 years out of 70 years of model simulation are included in the composites.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
As in Fig. 3, but for the TIOMC run. A total of 8 years out of 70 years of model simulation are included in the composites.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
As in Fig. 3, but for the MC run over the last 70 years of model simulation. A total of 14 years out of 70 years of model simulation are included in the composites.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
As in Fig. 3, but for the noTIOTP run over the last 70 years of model simulation. A total of 14 years out of 70 years of model simulation are included in the composites.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
In the TIOMC run, we see a pair of cyclonic wind anomalies with significant positive rainfall anomalies over the SCS and northwest Australia with strong westerly (easterly) wind anomalies over the equatorial Indian Ocean (western equatorial Pacific) in Fig. 7d. However, the cyclonic wind anomalies appear to be a Rossby wave response to diabatic heating anomalies in the western Pacific, as well as reinforced by an atmospheric Kelvin wave response to diabatic cooling anomalies over the eastern equatorial Indian Ocean (Fig. 7d), which is inconsistent with the observation (Fig. 3d). The diabatic cooling anomalies are flanked by zonally diverging wind anomalies (Fig. 7d). No significant SST and rainfall signals associated with the IOD are seen in the SON mean composites (Figs. 7a,c). This suggests that the NDJ mean negative rainfall anomalies over the eastern equatorial Indian Ocean are likely to be related to atmospheric internal variability, as illustrated in the MC run later. Also, we can infer that the rainfall anomaly dipole and negative IOD over the equatorial Indian Ocean can contribute to the SCS cyclonic wind anomalies in the observation only under La Niña conditions. In other words, at least in this model, the tropical Indian Ocean air–sea coupling or ENSO-independent negative IOD alone is unlikely to contribute to the SCS cyclonic wind anomalies, as explained in the next section.
In the MC run, a pair of cyclonic wind anomalies with significantly positive rainfall anomalies during the NDJ period is seen (Fig. 8d). Given that the suppressed air–sea coupling over the western Pacific, as discussed in section 5b, the circulation–convection feedback explains the cyclonic wind anomalies (Xie et al. 2009). The cyclonic wind anomalies are an atmospheric Kelvin wave response to diabatic cooling anomalies, flanked by diverging zonal wind anomalies, and a surface equatorial ridge over the equatorial Indian Ocean and SCS, which enhances off-equatorial convergence and convection over the western Pacific. The enhanced convection produces diabatic heating anomalies over the western Pacific and SCS, which in turn excites a Rossby wave that strengthens the cyclonic wind anomalies. Such rainfall anomalies are relatively weak in SON when diabatic cooling anomalies over the equatorial Indian Ocean are weak (Fig. 8c). Therefore, in the absence of air–sea coupling outside the MC, we can confirm that the dry phase of atmospheric internal variability (i.e., the phase with negative rainfall anomalies) over the Indian Ocean can induce the SCS cyclonic wind anomalies that strengthen the CT.
The results for the noTIOTP run in Fig. 9 mirror those of the MC run. Similar diabatic cooling anomalies over the equatorial Indian Ocean are seen during NDJ (Fig. 9d). Given that air–sea coupling is suppressed over both tropical Pacific and Indian Oceans, even with tropical Atlantic and extratropical oceans air–sea coupling, the atmospheric internal variability still contributes to the pair of cyclonic wind anomalies. Air–sea coupling in the tropical Atlantic and extratropical oceans appears to regulate the contribution of atmospheric internal variability to the SCS cyclonic wind anomalies, as discussed in the next section.
c. Quantitative analysis of tropical remote forcings
Since the eigenvector of the first mode for each run has unit magnitude and its associated eigenvalue determines its magnitude, we can quantify the strength of SCS cyclonic wind anomaly and CT variabilities across all runs.
To overcome the abovementioned concerns, we adopt two computation methods. The first method uses the eigenvalues based on the MEOF analysis of the last 70 years of data in Fig. 10 and calculate the forcing contribution using Eq. (5). The second method uses the eigenvalues based on various subset data of shorter period in Fig. 10. We select a 30-yr period as an example. For each experiment, 41 different subsets of a 30-yr period within the model’s last 70 years of data are prepared by shifting the time series by one year forward each time (e.g., the first subset contains the data from the 31st to 60th years, the second subset contains data from the 32nd to 61st years, …, and the 41st subset contains the data from the 71st to 100th years). There are variations in the eigenvalues among the subsets (Fig. 10). We calculate the percentage contributions of a forcing based on eigenvalues of subsets from various experiments across all different possible periods obtained via permutation. For example, to calculate PTP using Eq. (5b), besides using the eigenvalues of first subset of control, TPMC, and MC runs, we also choose the first subset of control run, second subset of TPMC run, and second subset of MC run, and so on. As such, we obtain 412 possible values of PAIV+Local from Eq. (5a), and 413 possible values of PTP, PTIO, and PTAETO from Eqs. (5b)–(5d). For each forcing, we compute their mean of all percentage contributions to remove the random error arising from the model variability between subsets from different experiments and periods.
Eigenvalues of first MEOF mode based on SST, and 850-hPa zonal and meridional wind variables for the control, TPMC, TIOMC, MC, and noTIOTP runs from the last 70 years of model simulation. Eigenvalues for the last 70 years of model simulation are represented by shaded bars. Means and ranges of eigenvalues for 41 different subsets of 30-yr period are indicated by hatched bars and vertical error bars, respectively.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
The percentage contributions to the SCS cyclonic wind anomaly are shown in Fig. 11. The percentage contribution by nonlinear interaction is just the residual, which is the difference between 100% and sum of percentage contributions of all forcings. Both methods agree in that the sum of atmospheric internal variability and local air–sea interaction contributes the most to the SCS cyclonic wind anomaly and CT variabilities in the control run at 80%–100% (Fig. 11). This is followed by the tropical Pacific air–sea coupling at 40%–60% and the tropical Atlantic and extratropical oceans air–sea coupling at 20%–30% according to Eqs. (5b) and (5d). In contrast, the tropical Indian Ocean air–sea coupling reduces the variabilities by 20%–40% according to Eq. (5c). From Eqs. (5a)–(5d), the nonlinear interaction reduces the variabilities by 20%–50%. Hence, we infer that the nonlinear interactions between different forcings in the control run undermine the contributions of individual forcing. Our results appear to be consistent with Kajtar et al. (2017), who found that the Indian Ocean variability damps the Pacific and Atlantic Ocean variabilities via their partially coupled experiments with another coupled model.
Relative percentage contributions of air–sea coupling over the tropical Pacific (TP), tropical Indian Ocean (TIO), tropical Atlantic and extratropical oceans (TAETO), atmospheric internal variability and local air–sea interaction (AIV + Local), and nonlinear interaction to the SCS cyclonic wind anomaly and CT variabilities in the control run calculated from Eq. (5). Percentages for the last 70 years of model simulation are represented by shaded bars. Percentages calculated from mean eigenvalues as explained in the text are indicated by hatched bars.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
We now discuss the reason behind the reduced variabilities by the tropical Indian Ocean coupling. The earlier observational data analysis suggests that the diabatic heating anomaly dipole pattern over the tropical Indian Ocean linked to the negative IOD may enhance the SCS cyclonic wind anomalies (Fig. 3). However, composites of the strong CT years in the TIOMC run in Fig. 7 show the absence of diabatic heating anomaly dipole pattern and negative IOD, and the tropical Indian Ocean air–sea coupling reduces the SCS cyclonic wind anomaly and CT variabilities. The reason behind the absence of diabatic heating anomaly dipole pattern and reduced variabilities by the tropical Indian Ocean air–sea coupling may be the northwest Pacific monsoon (Wang et al. 2001; Kajikawa et al. 2003; Huang and Shukla 2007). Without air–sea coupling over the tropical Pacific in the TIOMC, MC, and noTIOTP runs, anticyclonic wind anomalies are visible during SON (Figs. 7a,c, 8a,c, and 9a,c) for the MC and noTIOTP runs and only during September for the TIOMC run (figure not shown) in the strong CT years. Since there is no Pacific air–sea coupling in the TIOMC and MC runs, the anticyclonic wind anomalies must be triggered by the atmospheric internal variability that controls the northwest Pacific monsoon (Wang et al. 2001).
The tropical Indian Ocean air–sea coupling that generates the IOD as the main mode of climate variability in SON is followed by the strengthening anticyclonic wind anomalies over the northwest Pacific. We show this via an MEOF analysis of SON mean of the tropical Indian Ocean (10°S–10°N, 40°–100°E) SST and 850 hPa wind of the TIOMC run and constructing a monthly mean composite of years when their first PC time series values lie above one standard deviation (Fig. 12). The first MEOF mode, which explains 64% of the total variance, captures a negative IOD developing in July–September (JAS) and terminating by NDJ (Figs. 12c–e and 12h–j). Previous studies (e.g., Kajikawa et al. 2003; Huang and Shukla 2007) noted that under neutral ENSO conditions, the northwest Pacific anticyclonic (cyclonic) wind anomaly appearing from summer and lasting through autumn is accompanied by meridionally asymmetric cyclonic (anticyclonic) wind anomalies over the southern Indian Ocean, which triggers a negative (positive) IOD. This is consistent with Figs. 12c,d,h,i, showing that the negative IOD is accompanied by anticyclonic wind anomalies over northern SCS and northwest Pacific and negative wind curl anomalies over the southern tropical Indian Ocean. At the same time, positive rainfall and diabatic heating anomalies over the eastern equatorial Indian Ocean from JAS can strengthen the northwest Pacific anticyclonic wind anomalies by inducing an atmospheric Kelvin wave via easterly wind anomalies over the western tropical Pacific (Figs. 12h,i; Z. Chen et al. 2016). By NDJ, the northwest Pacific anticyclonic wind anomalies have dissipated and wind anomalies over the SCS are weak (Figs. 12e,j). While diabatic cooling anomalies over the western tropical Indian Ocean can excite cyclonic wind anomalies over the SCS via an atmospheric Kelvin wave response, they are weakened by Rossby waves from the diabatic cooling anomalies and anticyclonic wind anomalies over the northwest Pacific. Hence, when the northwest Pacific anticyclonic wind anomalies, negative IOD, and the dry phase of atmospheric internal variability over the equatorial Indian Ocean occur simultaneously, the SCS cyclonic wind anomaly and CT variabilities will be reduced under the tropical Indian Ocean air–sea coupling.
(a),(f) March–May (MAM), (b),(g) May–July (MJJ), (c),(h) JAS, (d),(i) SON, and (e),(j) NDJ 850-hPa wind anomalies and mean (left) SST and (right) rainfall composites of first PC time series values above the one standard deviation of MEOF analysis of SON 850-hPa wind and SST means of equatorial Indian Ocean over the last 70 years of model simulation in the TIOMC run. A total of 12 years out of 70 years of model simulation are included in the composites. Only SST and wind anomalies significantly different from nonnegative IOD years based on the two-tailed Mann–Whitney U nonparametric test at the 90% confidence level are shown.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
In contrast, under the tropical Pacific air–sea coupling and La Niña as in the observation (Fig. 3) and the control run (Fig. S1), the negative IOD can be forced by La Niña via westerly wind anomalies over the tropical Indian Ocean, which are a Rossby wave response to the diabatic heating anomalies over the warm equatorial western Pacific. Although an anomalous anticyclone is seen over the northwest Pacific in SON, it is weaker than that in Fig. 12 as it is subsequently weakened by the underlying positive SST anomalies during early winter onward. The positive rainfall and diabatic heating anomalies over the eastern pole of negative IOD can strengthen the anticyclonic wind anomalies, but the weakening impact of western tropical Pacific SSTs plays a more dominant role. Only without the strengthening of anticyclonic wind anomalies during La Niña, the impact of diabatic cooling anomalies over the cool western equatorial Indian Ocean during the decaying negative IOD in reinforcing the SCS cyclonic wind anomalies is evident as per observation.
7. Conclusions
In this paper, via the observational data analyses and model experiments, we identify that the SCS CT variability is controlled by the local SCS cyclonic wind anomaly variation at the interannual time scales. Such variabilities are captured by the first MEOF mode analysis of winter 850-hPa wind and SST anomalies over the SCS. The key qualitative results are summarized in Fig. 13. In the observation under the combined influences of air–sea coupling over all oceans and atmospheric internal variability, during strong CT years, the SCS cyclonic wind anomalies are generated by an equatorial Rossby wave response to diabatic heating and deep convection anomalies over the warmer-than-normal western equatorial Pacific during La Niña. Also, the diabatic cooling anomalies over cooler-than-normal western equatorial Indian Ocean during the decaying negative IOD induces an atmospheric Kelvin wave that generates eastward equatorial ridge anomalies, thereby enhancing off-equatorial convergence that strengthens the SCS cyclonic wind anomalies. Moreover, the cyclonic wind anomalies can be strengthened by an atmospheric Kelvin wave response to the dry phase of atmospheric internal variability over the equatorial Indian Ocean during winter. During neutral ENSO and IOD conditions, their dry phase is preceded by anticyclonic wind anomalies over the northwest Pacific.
(a) Strengthening impacts of both negative IOD and atmospheric internal variability–induced Kelvin wave and La Niña–induced Rossby wave on SCS cyclonic wind anomalies with NDJ SST and 850-hPa wind anomalies as in Fig. 3b. (b) Strengthening impact of negative IOD and atmospheric internal variability–induced Kelvin wave but weakening impact of northwest Pacific anticyclonic wind–induced Rossby wave on SCS cyclonic wind anomalies with SON SST and 850-hPa wind anomalies as in Fig. 12d.
Citation: Journal of Climate 34, 10; 10.1175/JCLI-D-20-0657.1
In the model, with only the tropical Pacific air–sea coupling in the TPMC run, the forcing of SCS cyclonic wind anomalies by La Niña is reproduced. With only the tropical Indian Ocean air–sea coupling in the TIOMC run and without both tropical Pacific and Indian Oceans air–sea coupling in both MC and noTIOTP runs, the SCS cyclonic wind anomalies are only forced by the dry phase of atmospheric internal variability over the equatorial Indian Ocean. The quantitative analysis and Fig. 12 suggest that the negative IOD that develops without ENSO influence weakens the positive contribution of dry phase of atmospheric internal variability over the Indian Ocean to the SCS cyclonic wind anomalies. This is due to the co-occurrence of the negative IOD and northwest Pacific anticyclonic wind anomalies that suppresses the SCS cyclonic wind anomalies triggered by dry phase of atmospheric internal variability when the tropical Pacific air–sea coupling is suppressed (Fig. 13b). In contrast, during La Niña, the northwest Pacific anticyclonic wind anomalies are suppressed by the underlying positive SST anomalies, which strengthen the impact of negative IOD on the SCS cyclonic wind anomalies (Fig. 13a). Therefore, both the tropical Pacific air–sea coupling and atmospheric internal variability positively contribute to the stronger SCS cyclonic wind anomaly and CT. In contrast, the tropical Indian Ocean air–sea coupling reduces those variabilities.
There are some issues that need to be addressed in the future. First, the possible role of intraseasonal variations remains unclear. The atmospheric internal variability appears to resemble the MJO, although the spatial and temporal characteristics need to be verified using daily observational data, analysis product, and model output in a separate study. The atmospheric internal variability in our model can also be influenced by air–sea interaction over both Indian and Pacific Oceans, as is the case for the observed MJO (Shinoda and Han 2005; Marshall et al. 2008; Izumo et al. 2010; Suematsu and Miura 2018); such effects are regarded as part of the nonlinear interaction term. Also, previous studies (e.g., Lim and Chang 1981; Wu and Qian 2003; Zhao et al. 2013) demonstrated that the extratropical atmosphere and ocean influence the SCS summer and winter monsoons, as well as the MJO, at the intraseasonal time scale.
Despite the fact that the focus of this paper is on tropical Indo-Pacific forcings, the contribution of tropical Atlantic and extratropical oceans is briefly discussed here. The extratropical Pacific may influence the tropical Pacific through the seasonal footprinting mechanisms (Vimont et al. 2003), while the tropical Atlantic is known to influence the Walker circulation over the central equatorial Pacific (Ding et al. 2012). These, in turn, may influence the SCS via the Matsuno–Gill response. Also, Zhang et al. (2020) showed that atmospheric pressure variability in the southern Indian Ocean related to the southern annular mode may affect the tropics. Moreover, the East Asian winter monsoon, which affects the SCS, is influenced by an upstream atmospheric wave train forced by the North Atlantic SST (Qiao and Feng 2016; Jian et al. 2020). However, further investigations on the physical mechanisms behind the tropical Atlantic and extratropical oceans require a separate in-depth study.
Modeling over the Maritime Continent is challenging due to inherent model configuration issues (Xue et al. 2020). Hence, this study can be repeated using different coupled models to verify the findings. Apart from atmospheric influences, variations in the South China Sea Throughflow (SCSTF) also influence the SCS oceanic current anomalies (Qu et al. 2004, 2009; Wei et al. 2016), which in turn control the CT variability. Tozuka et al. (2009) compared OGCM experiments with and without the SCSTF and showed that it has strong impacts on SST and currents in the CT region. On the other hand, a recent study by Wang et al. (2020) suggested that the atmospheric influence is much more important than the Luzon Strait intrusion in controlling the SCS cyclonic oceanic currents. It will be illuminating to quantify both atmospheric and oceanic contributions using a coupled model with a higher horizontal resolution that can better resolve the SCSTF in a separate future study.
Acknowledgments
We thank three anonymous reviewers for their constructive comments, which helped us greatly to improve our manuscript. The SINTEX-F2 model was run on the Data Analyzer (DA) system at the Japan Agency for Marine-Earth Science and Technology (JAMSTEC). The NOAA OISST, version 2, dataset was downloaded from https://www.esrl.noaa.gov/psd/data/gridded/data.noaa.oisst.v2.highres.html. The ERA-Interim data were downloaded from the ECMWF public datasets via Python according to the instructions in https://confluence.ecmwf.int//display/WEBAPI/Access+ECMWF+Public+Datasets. The GPCP, version 2.3, dataset was downloaded from https://www.esrl.noaa.gov/psd/data/gridded/data.gpcp.html. The first author is financially supported by the Research Fellowship of Japan Society for the Promotion of Science (JSPS) under Grant-in-Aid for JSPS Fellows, Grant 19J20585.
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.
Annamalai, H., S. P. Xie, J. P. McCreary, and R. Murtugudde, 2005: Impact of Indian Ocean sea surface temperature on developing El Niño. J. Climate, 18, 302–319, https://doi.org/10.1175/JCLI-3268.1.
Ashok, K., Z. Guan, and T. Yamagata, 2003: A look at the relationship between the ENSO and the Indian Ocean dipole. J. Meteor. Soc. Japan, 81, 41–56, https://doi.org/10.2151/jmsj.81.41.
Balmaseda, M. A., K. Mogensen, and A. T. Weaver, 2013: Evaluation of the ECMWF ocean reanalysis system ORAS4. Quart. J. Roy. Meteor. Soc., 139, 1132–1161, https://doi.org/10.1002/qj.2063.
Baranowski, D. B., M. K. Flatau, P. J. Flatau, and A. J. Matthews, 2016: Phase locking between atmospheric convectively coupled equatorial Kelvin waves and the diurnal cycle of precipitation over the Maritime Continent. Geophys. Res. Lett., 43, 8269–8276, https://doi.org/10.1002/2016GL069602.
Chang, C. P., P. A. Harr, and H. J. Chen, 2005a: Synoptic disturbances over the equatorial South China Sea and western Maritime Continent during boreal winter. Mon. Wea. Rev., 133, 489–503, https://doi.org/10.1175/MWR-2868.1.
Chang, C. P., Z. Wang, J. McBride, and C. H. Liu, 2005b: Annual cycle of Southeast Asia—Maritime Continent rainfall and the asymmetric monsoon transition. J. Climate, 18, 287–301, https://doi.org/10.1175/JCLI-3257.1.
Chen, M., T. Li, X. Shen, and B. Wu, 2016: Relative roles of dynamic and thermodynamic processes in causing evolution asymmetry between El Niño and La Niña. J. Climate, 29, 2201–2220, https://doi.org/10.1175/JCLI-D-15-0547.1.
Chen, M., J. Y. Yu, X. Wang, and W. P. Jiang, 2019: The changing impact mechanisms of a diverse El Niño on the western Pacific subtropical high. Geophys. Res. Lett., 46, 953–962, https://doi.org/10.1029/2018GL081131.
Chen, W. T., S. P. Hsu, Y. H. Tsai, and C. H. Sui, 2019: The influences of convectively coupled Kelvin waves on multiscale rainfall variability over the South China Sea and Maritime Continent in December 2016. J. Climate, 32, 6977–6993, https://doi.org/10.1175/JCLI-D-18-0471.1.
Chen, Z., Z. Wen, R. Wu, X. Lin, and J. Wang, 2016: Relative importance of tropical SST anomalies in maintaining the western North Pacific anomalous anticyclone during El Niño to La Niña transition years. Climate Dyn., 46, 1027–1041, https://doi.org/10.1007/s00382-015-2630-1.
Chowdary, J. S., and C. Gnanaseelan, 2007: Basin-wide warming of the Indian Ocean during El Niño and Indian Ocean dipole years. Int. J. Climatol., 27, 1421–1438, https://doi.org/10.1002/joc.1482.
Crétat, J., P. Terray, S. Masson, and K. P. Sooraj, 2018: Intrinsic precursors and timescale of the tropical Indian Ocean dipole: Insights from partially decoupled numerical experiment. Climate Dyn., 51, 1311–1332, https://doi.org/10.1007/s00382-017-3956-7.
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.
Ding, H., N. S. Keenlyside, and M. Latif, 2012: Impact of the equatorial Atlantic on the El Niño Southern Oscillation. Climate Dyn., 38, 1965–1972, https://doi.org/10.1007/s00382-011-1097-y.
Doi, T., S. K. Behera, and T. Yamagata, 2016: Improved seasonal prediction using the SINTEX-F2 coupled model. J. Adv. Model. Earth Syst., 8, 1847–1867, https://doi.org/10.1002/2016MS000744.
Fukutomi, Y., 2019: Tropical synoptic-scale waves propagating across the Maritime Continent and northern Australia. J. Geophys. Res. Atmos., 124, 7665–7682, https://doi.org/10.1029/2018JD029795.
Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106, 447–462, https://doi.org/10.1002/qj.49710644905.
Huang, B., and J. Shukla, 2007: Mechanisms for the interannual variability in the tropical Indian Ocean. Part II: Regional processes. J. Climate, 20, 2937–2960, https://doi.org/10.1175/JCLI4169.1.
Izumo, T., S. Masson, J. Vialard, C. de Boyer Montegut, S. K. Behera, G. Madec, K. Takahashi, and T. Yamagata, 2010: Low and high frequency Madden–Julian oscillations in austral summer: Interannual variations. Climate Dyn., 35, 669–683, https://doi.org/10.1007/s00382-009-0655-z.
Jian, Y., X. Lin, W. Zhou, M. Jian, M. Y. Leung, and P. K. Cheung, 2020: Analysis of record-high temperature over southeast coastal China in winter 2018/19: The combined effect of mid- to high-latitude circulation systems and SST Forcing over the North Atlantic and tropical western Pacific. J. Climate, 33, 8813–8831, https://doi.org/10.1175/JCLI-D-19-0732.1.
Kajikawa, Y., T. Yasunari, and R. Kawamura, 2003: The role of the local Hadley circulation over the western Pacific on the zonally asymmetric anomalies over the Indian Ocean. J. Meteor. Soc. Japan, 81, 259–276, https://doi.org/10.2151/jmsj.81.259.
Kajtar, J. B., A. Santoso, M. H. England, and W. Cai, 2017: Tropical climate variability: Interactions across the Pacific, Indian, and Atlantic Oceans. Climate Dyn., 48, 2173–2190, https://doi.org/10.1007/s00382-016-3199-z.
Kataoka, T., S. Masson, T. Izumo, T. Tozuka, and T. Yamagata, 2018: Can Ningaloo Niño/Niña develop without El Niño–Southern Oscillation? Geophys. Res. Lett., 45, 7040–7048, https://doi.org/10.1029/2018GL078188.
Kohyama, T., and T. Tozuka, 2016: Seasonal variability of the relationship between SST and OLR in the Indian Ocean and its implications for initialization in a CGCM with SST nudging. J. Oceanogr., 72, 327–337, https://doi.org/10.1007/s10872-015-0329-x.
Koseki, S., T. Y. Koh, and C. K. Teo, 2013: Effects of the cold tongue in the South China Sea on the monsoon, diurnal cycle and rainfall in the Maritime Continent. Quart. J. Roy. Meteor. Soc., 139, 1566–1582, https://doi.org/10.1002/qj.2052.
Li, C. H., A. L. Lin, and T. Li, 2017: Effects of air–sea coupling on the eastward propagating boreal winter intraseasonal oscillation over the tropical Indian Ocean. Atmos. Oceanic Sci. Lett., 10, 51–57, https://doi.org/10.1080/16742834.2017.1239187.
Li, T., B. Wang, C. P. Chang, and Y. Zhang, 2003: A theory for the Indian Ocean dipole–zonal mode. J. Atmos. Sci., 60, 2119–2135, https://doi.org/10.1175/1520-0469(2003)060<2119:ATFTIO>2.0.CO;2.
Lim, H., and C. P. Chang, 1981: A theory for midlatitude forcing of tropical motions during winter monsoons. J. Atmos. Sci., 38, 2377–2392, https://doi.org/10.1175/1520-0469(1981)038<2377:ATFMFO>2.0.CO;2.
Lim, S. Y., C. Marzin, P. Xavier, C.-P. Chang, and B. Timbal, 2017: Impacts of boreal winter monsoon cold surges and the interaction with MJO on Southeast Asia rainfall. J. Climate, 30, 4267–4281, https://doi.org/10.1175/JCLI-D-16-0546.1.
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, https://doi.org/10.1029/2003JC002179.
Liu, Q. Y., D. Wang, X. Wang, Y. Shu, Q. Xie, and J. Chen, 2014: Thermal variations in the South China Sea associated with the eastern and central Pacific El Niño events and their mechanisms. J. Geophys. Res. Oceans, 119, 8955–8972, https://doi.org/10.1002/2014JC010429.
Luo, J. J., R. Zhang, S. K. Behera, Y. Masumoto, F. F. Jin, R. Lukas, and T. Yamagata, 2010: Interaction between El Niño and extreme Indian Ocean dipole. J. Climate, 23, 726–742, https://doi.org/10.1175/2009JCLI3104.1.
Madden, R. A., and P. R. Julian, 1971: Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28, 702–708, https://doi.org/10.1175/1520-0469(1971)028<0702:DOADOI>2.0.CO;2.
Madec, G., 2008: NEMO ocean engine. Note du Pôle de Modélisation de l’Institut Pierre-Simon Laplace 27, 209 pp.
Marshall, A. G., O. Alves, and H. H. Hendon, 2008: An enhanced moisture convergence–evaporation feedback mechanism for MJO air–sea interaction. J. Atmos. Sci., 65, 970–986, https://doi.org/10.1175/2007JAS2313.1.
Masson, S., P. Terray, G. Madec, J.-J. Luo, T. Yamagata, and K. Takahashi, 2012: Impact of intra-daily SST variability on ENSO characteristics in a coupled model. Climate Dyn ., 39, 681–707, https://doi.org/10.1007/s00382-011-1247-2.
Matsuno, T., 1966: Quasi-geostrophic motions in the equatorial area. J. Meteor. Soc. Japan, 44, 25–43, https://doi.org/10.2151/jmsj1965.44.1_25.
Morioka, Y., S. Masson, P. Terray, C. Prodhomme, S. K. Behera, and Y. Masumoto, 2014: Role of tropical SST variability on the formation of subtropical dipoles. J. Climate, 27, 4486–4507, https://doi.org/10.1175/JCLI-D-13-00506.1.
Nordeng, T. E., 1994: Extended versions of the convective parameterization scheme at ECMWF and their impact on the mean and transient activity of the model in the tropics. ECMWF Tech. Memo. 206, 41 pp.
North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699–706, https://doi.org/10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.
Prodhomme, C., P. Terray, S. Masson, G. Boschat, and T. Izumo, 2015: Oceanic factors controlling the Indian summer monsoon onset in a coupled model. Climate Dyn., 44, 977–1002, https://doi.org/10.1007/s00382-014-2200-y.
Qiao, S., and G. Feng, 2016: Impact of the December North Atlantic Oscillation on the following February East Asian trough. J. Geophys. Res. Atmos., 121, 10 074–10 088, https://doi.org/10.1002/2016JD025007.
Qu, T., Y. Y. Kim, M. Yaremchuk, T. Tozuka, A. Ishida, and T. Yamagata, 2004: Can Luzon Strait transport play a role in conveying the impact of ENSO to the South China Sea? J. Climate, 17, 3644–3657, https://doi.org/10.1175/1520-0442(2004)017<3644:CLSTPA>2.0.CO;2.
Qu, T., Y. T. Song, and T. Yamagata, 2009: An introduction to the South China Sea Throughflow: Its dynamics, variability, and application for climate. Dyn. Atmos. Oceans, 47, 3–14, https://doi.org/10.1016/j.dynatmoce.2008.05.001.
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, https://doi.org/10.1175/1520-0442(2002)015<1609:AIISAS>2.0.CO;2.
Roeckner, E., and Coauthors, 2003: The atmospheric general circulation model ECHAM 5. Part I: Model description. MPI Rep. 349, 127 pp.
Seow, M. X. C., and T. Tozuka, 2019: Ocean thermodynamics behind the asymmetry of interannual variation of South China Sea winter cold tongue strength. Climate Dyn., 52, 3241–3253, https://doi.org/10.1007/s00382-018-4320-2.
Shinoda, T., and W. Han, 2005: Influence of the Indian Ocean dipole on atmospheric subseasonal variability. J. Climate, 18, 3891–3909, https://doi.org/10.1175/JCLI3510.1.
Suematsu, T., and H. Miura, 2018: Zonal SST difference as a potential environmental factor supporting the longevity of the Madden–Julian oscillation. J. Climate, 31, 7549–7564, https://doi.org/10.1175/JCLI-D-17-0822.1.
Terray, P., K. Kamala, S. Masson, G. Madec, A. K. Sahai, J. J. Luo, and T. Yamagata, 2012: The role of the intra-daily SST variability in the Indian monsoon variability and monsoon–ENSO–IOD relationships in a global coupled model. Climate Dyn., 39, 729–754, https://doi.org/10.1007/s00382-011-1240-9.
Thompson, B., P. Tkalich, P. Malanotte-Rizzoli, B. Fricot, and J. Mas, 2016: Dynamical and thermodynamical analysis of the South China Sea winter cold tongue. Climate Dyn., 47, 1629–1646, https://doi.org/10.1007/s00382-015-2924-3.
Tiedtke, M., 1989: A comprehensive mass flux scheme for cumulus parameterization in large-scale models. Mon. Wea. Rev., 117, 1779–1800, https://doi.org/10.1175/1520-0493(1989)117<1779:ACMFSF>2.0.CO;2.
Timmermann, R., H. Goosse, G. Madec, T. Fichefet, C. Ethe, and V. Duliere, 2005: On the representation of high latitude processes in the ORCA-LIM global coupled sea ice–ocean model. Ocean Modell., 8, 175–201, https://doi.org/10.1016/j.ocemod.2003.12.009.
Tong, H. W., J. C. Chan, and W. Zhou, 2009: The role of MJO and mid-latitude fronts in the South China Sea summer monsoon onset. Climate Dyn., 33, 827–841, https://doi.org/10.1007/s00382-008-0490-7.
Tozuka, T., T. Qu, Y. Masumoto, and T. Yamagata, 2009: Impacts of the South China Sea Throughflow on seasonal and interannual variations of the Indonesian Throughflow. Dyn. Atmos. Oceans, 47, 73–85, https://doi.org/10.1016/j.dynatmoce.2008.09.001.
Valcke, S., 2013: The OASIS3 coupler: A European climate modelling community software. Geosci. Model Dev., 6, 373–388, https://doi.org/10.5194/gmd-6-373-2013.
Varikoden, H., A. A. Samah, and C. A. Babu, 2010: The cold tongue in the South China Sea during boreal winter and its interaction with the atmosphere. Adv. Atmos. Sci., 27, 265–273, https://doi.org/10.1007/s00376-009-8141-4.
Vimont, D. J., J. M. Wallace, and D. S. Battisti, 2003: The seasonal footprinting mechanism in the Pacific: Implications for ENSO. J. Climate, 16, 2668–2675, https://doi.org/10.1175/1520-0442(2003)016<2668:TSFMIT>2.0.CO;2.
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. Wu, and X. 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, B., R. Wu, and K. M. Lau, 2001: Interannual variability of the Asian summer monsoon: Contrasts between the Indian and the western North Pacific–East Asian monsoons. J. Climate, 14, 4073–4090, https://doi.org/10.1175/1520-0442(2001)014<4073:IVOTAS>2.0.CO;2.
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, https://doi.org/10.1029/2005JC003333.
Wang, Q., and Coauthors, 2020: Interannual variability of South China Sea winter circulation: Response to Luzon Strait transport and El Niño wind. Climate Dyn., 54, 1145–1159, https://doi.org/10.1007/s00382-019-05050-2.
Watanabe, M., and F. F. Jin, 2002: Role of Indian Ocean warming in the development of Philippine Sea anticyclone during ENSO. Geophys. Res. Lett., 29, 1478, https://doi.org/10.1029/2001GL014318.
Wei, J., P. Malanotte-Rizzoli, M.-T. Li, and H. Wang, 2016: Decomposition of thermal and dynamic changes in the South China Sea induced by boundary forcing and surface fluxes during 1970-2000. J. Geophys. Res. Oceans, 121, 7953–7972, https://doi.org/10.1002/2016JC012078.
Wu, B., T. Zhou, and T. Li, 2012: Two distinct modes of tropical Indian Ocean precipitation in boreal winter and their impacts on equatorial western Pacific. J. Climate, 25, 921–938, https://doi.org/10.1175/JCLI-D-11-00065.1.
Wu, R., 2016: Coupled intraseasonal variations in the East Asian winter monsoon and the South China Sea–western North Pacific SST in boreal winter. Climate Dyn., 47, 2039–2057, https://doi.org/10.1007/s00382-015-2949-7.
Wu, R., and Z. Chen, 2015: Intraseasonal SST variations in the South China Sea during boreal winter and impacts of the East Asian winter monsoon. J. Geophys. Res. Atmos., 120, 5863–5878, https://doi.org/10.1002/2015JD023368.
Wu, T., and Z. Qian, 2003: The relation between the Tibetan winter snow and the Asian summer monsoon and rainfall: An observational investigation. J. Climate, 16, 2038–2051, https://doi.org/10.1175/1520-0442(2003)016<2038:TRBTTW>2.0.CO;2.
Xiao, F., L. Zeng, Q.-Y. Liu, W. Zhou, and D. Wang, 2018: Extreme subsurface warm events in the South China Sea during 1998/99 and 2006/07: Observations and mechanisms. Climate Dyn ., 50, 115–128, https://doi.org/10.1007/S00382-017-3588-y.
Xiao, F., D. Wang, and M. Y. Leung, 2020: Early and extreme warming in the South China Sea during 2015/2016: Role of an unusual Indian Ocean dipole event. Geophys. Res. Lett., 47, e2020GL089936, https://doi.org/10.1029/2020GL089936.
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, https://doi.org/10.1175/2008JCLI2544.1.
Xue, P., P. Malanotte-Rizzoli, J. Wei, and E. A. Eltahir, 2020: Coupled ocean–atmosphere modeling over the Maritime Continent: A review. J. Geophys. Res. Oceans, 125, e2019JC014978, https://doi.org/10.1029/2019JC014978.
Yang, J., Q. Liu, and Z. Liu, 2010: Linking observations of the Asian monsoon to the Indian Ocean SST: Possible roles of Indian Ocean basin mode and dipole mode. J. Climate, 23, 5889–5902, https://doi.org/10.1175/2010JCLI2962.1.
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.
Zhang, L. Y., Y. Du, W. Cai, Z. Chen, T. Tozuka, J.-Y. Yu, 2020: Triggering the Indian Ocean dipole from the Southern Hemisphere. Geophys. Res. Lett., 47, e2020GL088648, https://doi.org/10.1029/2020GL088648.
Zhao, C., T. Li, and T. Zhou, 2013: Precursor signals and processes associated with MJO initiation over the tropical Indian Ocean. J. Climate, 26, 291–307, https://doi.org/10.1175/JCLI-D-12-00113.1.