• Adler, R. F., and Coauthors, 2003: The version 2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979–present). J. Hydrometeor., 4, 11471167, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, J. L., and Coauthors, 2004: The new GFDL global atmosphere and land model AM2–LM2: Evaluation with prescribed SST simulations. J. Climate, 17, 46414673, doi:10.1175/JCLI-3223.1.

    • Search Google Scholar
    • Export Citation
  • Delworth, T. L., and Coauthors, 2006: GFDL’s CM2 global coupled climate models. Part I: Formulation and simulation characteristics. J. Climate, 19, 643674, doi:10.1175/JCLI3629.1.

    • Search Google Scholar
    • Export Citation
  • He, Z., , and R. Wu, 2014: Indo-Pacific remote forcing in summer rainfall variability over the South China Sea. Climate Dyn.,42, 2323–2337, doi:10.1007/s00382-014-2123-7.

  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S. K. Yang, , J. J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 16311643, doi:10.1175/BAMS-83-11-1631.

    • Search Google Scholar
    • Export Citation
  • 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, 917932, doi:10.1175/1520-0442(1999)012<0917:RSSTVD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., , H. T. Wu, , and S. Yang, 1998: Hydrologic processes associated with the first transition of the Asian summer monsoon: A pilot satellite study. Bull. Amer. Meteor. Soc., 79, 18711882, doi:10.1175/1520-0477(1998)079<1871:HPAWTF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., , X. Li, , and M. T. Wu, 2002: Evolution of the large scale circulation, cloud structure and regional water cycle associated with the South China Sea monsoon during May–June, 1998. J. Meteor. Soc. Japan, 80, 11291147, doi:10.2151/jmsj.80.1129.

    • Search Google Scholar
    • Export Citation
  • Lin, S., 2004: A “vertically Lagrangian” finite-volume dynamical core for global models. Mon. Wea. Rev., 132, 22932307, doi:10.1175/1520-0493(2004)132<2293:AVLFDC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Mao, J. Y., , J. Chan, , and G. X. Wu, 2004: Relationship between the onset of the South China Sea summer monsoon and the structure of the Asian subtropical anticyclone. J. Meteor. Soc. Japan, 82, 845849, doi:10.2151/jmsj.2004.845.

    • Search Google Scholar
    • Export Citation
  • 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, 699706, doi:10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., , R. W. Reynolds, , T. C. Peterson, , and J. Lawrimore, 2008: Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006). J. Climate, 21, 22832296, doi:10.1175/2007JCLI2100.1.

    • Search Google Scholar
    • Export Citation
  • Tao, S. Y., , and L. X. Chen, 1987: A review of recent research on the East Asian summer monsoon in China. Monsoon Meteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 60–92.

  • Tomita, T., , and T. Yasunari, 1996: Role of the northeast winter monsoon on the biennial oscillation of the ENSO/monsoon system. J. Meteor. Soc. Japan, 74, 399413.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , R. Wu, , and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13, 15171536, doi:10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , R. Wu, , and T. Li, 2003: Atmosphere–warm ocean interaction and its impacts on Asian–Australian monsoon variation. J. Climate, 16, 11951211, doi:10.1175/1520-0442(2003)16<1195:AOIAII>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Wang, D., , Q. Xie, , Y. Du, , W. Q. Wang, , and J. Chen, 2002: The 1997–1998 warm event in the South China Sea. Chin. Sci. Bull., 47, 12211227, doi:10.1007/BF02907614.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Zhou, , and T. Li, 2009: Seasonally evolving dominant interannual variability mode over the East Asia. J. Climate, 22, 29923005, doi:10.1175/2008JCLI2710.1.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Li, , and T. Zhou, 2010a: Asymmetry of atmospheric circulation anomalies over the western North Pacific between El Niño and La Niña. J. Climate, 23, 48074822, doi:10.1175/2010JCLI3222.1.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Li, , and T. Zhou, 2010b: Relative contributions of the Indian Ocean and local SST anomalies to the maintenance of the western North Pacific anomalous anticyclone during El Niño decaying summer. J. Climate, 23, 29742986, doi:10.1175/2010JCLI3300.1.

    • Search Google Scholar
    • Export Citation
  • Wu, G. X., , and Y. S. Zhang, 1998: Tibetan Plateau forcing and the monsoon onset over South Asia and the South China Sea. Mon. Wea. Rev., 126, 913927, doi:10.1175/1520-0493(1998)126<0913:TPFATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., 2002: Processes for the northeastward advance of the summer monsoon over the western North Pacific. J. Meteor. Soc. Japan, 80, 6783, doi:10.2151/jmsj.80.67.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. Wang, 2000: Interannual variability of summer monsoon onset over the western North Pacific and the underlying processes. J. Climate, 13, 24832501, doi:10.1175/1520-0442(2000)013<2483:IVOSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. Wang, 2001: Multi-stage onset of the summer monsoon over the western North Pacific. Climate Dyn., 17, 277289, doi:10.1007/s003820000118.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. P. Kirtman, 2007: Regimes of local air-sea interactions and implications for performance of forced simulations. Climate Dyn., 29, 393410, doi:10.1007/s00382-007-0246-9.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , Z.-Z. Hu, , and B. P. Kirtman, 2003: Evolution of ENSO-related rainfall anomalies in East Asia. J. Climate, 16, 37423758, doi:10.1175/1520-0442(2003)016<3742:EOERAI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , B. P. Kirtman, , and K. Pegion, 2006: Local air–sea relationship in observations and model simulations. J. Climate, 19, 49144932, doi:10.1175/JCLI3904.1.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , S. Yang, , Z. P. Wen, , G. Huang, , and K. Hu, 2012: Interdecadal change in the relationship of southern China summer rainfall with tropical Indo-Pacific SST. Theor. Appl. Climatol., 108, 119133, doi:10.1007/s00704-011-0519-4.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , G. Huang, , Z. Du, , and K. Hu, 2013: Cross-season relation of the South China Sea precipitation variability between winter and summer. Climate Dyn., doi:10.1007/s00382-013-1820-y, in press.

    • Search Google Scholar
    • Export Citation
  • Xie, S. P., , Q. Xie, , D. 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.

    • Search Google Scholar
    • Export Citation
  • Yoo, S. H., , S. Yang, , and C. H. Ho, 2006: Variability of the Indian Ocean sea surface temperature and its impacts on Asian-Australian monsoon climate. J. Geophys. Res., 111, D03108, doi:10.1029/2005JD006001.

    • Search Google Scholar
    • Export Citation
  • Zhou, L. T., , C. Y. Tam, , W. Zhou, , and J. C. L. Chan, 2010: Influence of South China Sea SST and ENSO on winter rainfall over south China. Adv. Atmos. Sci., 27, 832844, doi:10.1007/s00376-009-9102-7.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Spatial distributions of the (a) first and (b) second EOF modes of precipitation anomalies (contours, mm day−1) over the South China Sea in AMJ and (c) their principal components PC1 and PC2. The box in (a) denotes the region for calculation of area-mean rainfall in Fig. 2.

  • View in gallery

    Time series of the leading mode as in Fig. 1b (solid line) and AMJ area-mean rainfall anomalies (dashed line, mm day−1) within the box in Fig. 1a. Gray lines represent ±0.5 standard deviation of normalized anomalies.

  • View in gallery

    Distributions of correlation coefficient of (a) precipitation (shading) and 850-hPa wind (vectors) and (b) SST (shading) in AMJ with respect to the time series of the leading mode of AMJ precipitation anomalies. Thick contours denote regions where the precipitation and SST correlations are significant at the 95% confidence level according to the Student’s t test. Only wind vectors that are significant at the 95% confidence level are plotted. The three rectangular boxes in (b) denote the key SST regions that are used to construct area-mean SST anomalies in Fig. 4.

  • View in gallery

    Normalized time series of the PC1 of AMJ SCS rainfall (gray bars), and AMJ SST averaged over the tropical Indian Ocean (TIO, solid line), the western North Pacific (WNP, long-dash line) and the equatorial Pacific (EP, long–short dash line). Gray lines denote ±0.5 standard deviation of normalized anomalies. The numbers at the top right are the simultaneous correlation coefficients with the PC1.

  • View in gallery

    Composite anomalies of (a) precipitation (shading, mm day−1) and 850-hPa wind (vectors, m s−1) and (b) SST (shading, °C) during AMJ for type-1 cases. Black contours denote regions where the composite precipitation and SST anomalies are significant at the 95% confidence level according to one sample t test. Only wind anomalies that are significant at the 95% confidence level are plotted.

  • View in gallery

    Composite anomalies of (a) 850- and (b) 200-hPa velocity potential (shading, 10−6 m2 s−1) and divergent winds (vectors, m s−1) and (c) 500-hPa vertical pressure velocity (shading, Pa s−1) during AMJ for type-1 cases. Black contours and green vectors denote regions where the composite anomalies are significant at the 95% confidence level according to one sample t test.

  • View in gallery

    Composite anomalies of (a) precipitation (shading, mm day−1) and 850-hPa wind (vectors, m s−1), (b) SST (shading, °C), and (c) instability index (shading, K hPa−1) during AMJ for type-2 cases. Black contours denote regions where the composite anomalies are significant at the 95% confidence level according to one sample t test. Only wind anomalies that are significant at the 95% confidence level are plotted.

  • View in gallery

    As in Fig. 6, but type-2 cases.

  • View in gallery

    As in Fig. 7, but for type-3 cases and without panel (c).

  • View in gallery

    As in Fig. 6, but type-3 cases.

  • View in gallery

    The SST anomalies specified in the AGCM experiments. The SST anomalies are the same as those in Fig. 9b in the tropical Indo-Pacific oceans but only the negative anomalies are plotted. Red, blue, and green boxes denote the WSIO, the ESIO, and the ETP, respectively.

  • View in gallery

    The responses of AMJ rainfall (shaded, mm day−1) and 850-hPa winds (vectors, m s−1) to anomalous SST forcing in (a) SIO, (b) SIO plus ETP, (c) ESIO, and (d) WSIO as differences of responses to negative minus positive SST anomalies divided by 2. The vectors depict winds with speed over 0.2 m s−1.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 12 12 4
PDF Downloads 4 4 1

Relation of the South China Sea Precipitation Variability to Tropical Indo-Pacific SST Anomalies during Spring-to-Summer Transition

View More View Less
  • 1 State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
  • 2 Institute of Space and Earth Information Science, and Department of Geography and Resource Management, and Shenzhen Research Institute, Chinese University of Hong Kong, Hong Kong, China
  • 3 Center for Monsoon System Research, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
© Get Permissions
Full access

Abstract

The present study investigates the relationship of South China Sea (SCS) precipitation to tropical Indo-Pacific sea surface temperature (SST) during April–June (AMJ), which is the transition season from spring to summer. It is revealed that SCS rainfall anomalies in AMJ are influenced by SST anomalies in the equatorial Pacific (EP), tropical Indian Ocean (TIO), and western North Pacific (WNP). Three types of SST-influenced cases are obtained based on different combinations of SST anomalies in the above three regions. When same-sign EP and TIO SST anomalies are accompanied by opposite WNP SST anomalies, both anomalous cross-equatorial flows from the southwestern TIO induced by negative SST anomalies there and an anomalous Walker circulation forced by negative EP SST anomalies contribute to enhanced convection over the SCS and the surrounding regions with additional contribution from positive WNP SST anomalies via a Rossby wave–type response. In the cases of combined effects of EP and WNP SST anomalies, above-normal SST in the WNP is a direct cause of above-normal SCS rainfall though the WNP SST anomalies are induced by EP SST forcing. In the cases of combined impacts of TIO and EP SST anomalies, the accompanying coastal Sumatra SST anomalies contribute to the SCS rainfall variability via an anomalous cross-equatorial vertical circulation. The negative SST anomalies near the Sumatra coast induce descent over the southeastern TIO and ascent over the SCS and WNP. Model experiments with an atmospheric model confirm the impacts of southern TIO and EP SST anomalies on AMJ rainfall variation over the SCS.

Corresponding author address: Dr. Renguang Wu, Fok Ying Tung Remote Sensing Science Building, Chinese University of Hong Kong, Shatin, NT, Hong Kong, China. E-mail: renguang@cuhk.edu.hk

Abstract

The present study investigates the relationship of South China Sea (SCS) precipitation to tropical Indo-Pacific sea surface temperature (SST) during April–June (AMJ), which is the transition season from spring to summer. It is revealed that SCS rainfall anomalies in AMJ are influenced by SST anomalies in the equatorial Pacific (EP), tropical Indian Ocean (TIO), and western North Pacific (WNP). Three types of SST-influenced cases are obtained based on different combinations of SST anomalies in the above three regions. When same-sign EP and TIO SST anomalies are accompanied by opposite WNP SST anomalies, both anomalous cross-equatorial flows from the southwestern TIO induced by negative SST anomalies there and an anomalous Walker circulation forced by negative EP SST anomalies contribute to enhanced convection over the SCS and the surrounding regions with additional contribution from positive WNP SST anomalies via a Rossby wave–type response. In the cases of combined effects of EP and WNP SST anomalies, above-normal SST in the WNP is a direct cause of above-normal SCS rainfall though the WNP SST anomalies are induced by EP SST forcing. In the cases of combined impacts of TIO and EP SST anomalies, the accompanying coastal Sumatra SST anomalies contribute to the SCS rainfall variability via an anomalous cross-equatorial vertical circulation. The negative SST anomalies near the Sumatra coast induce descent over the southeastern TIO and ascent over the SCS and WNP. Model experiments with an atmospheric model confirm the impacts of southern TIO and EP SST anomalies on AMJ rainfall variation over the SCS.

Corresponding author address: Dr. Renguang Wu, Fok Ying Tung Remote Sensing Science Building, Chinese University of Hong Kong, Shatin, NT, Hong Kong, China. E-mail: renguang@cuhk.edu.hk

1. Introduction

The South China Sea (SCS) is a semienclosed marginal ocean basin, and water exchanges between the SCS and the Pacific and Indian Oceans. Its unique location determines that the climate variability in the SCS region is influenced by anomalous states of tropical Indian and Pacific Oceans (Wang et al. 2000, 2006; Xie et al. 2003; Wu et al. 2012, 2013). In addition, anomalous state of the SCS can affect East and Southeast Asian climate (Tomita and Yasunari 1996; Wang et al. 2002; Zhou et al. 2010). Wang et al. (2002) pointed out that a strong warm event in the SCS is accompanied by severe flood in the valley of the Yangtze River. Zhou et al. (2010) indicated that anomalous SCS sea surface temperature (SST) and the El Niño–Southern Oscillation (ENSO) have an impact on winter rainfall over south China. Therefore, revealing the climate variability in SCS and its factors can improve our understanding of the influences of tropical Indo-Pacific oceans on East and Southeast Asian climate.

Previous studies showed that ENSO contributes to the interannual variability of rainfall over the SCS and the surrounding regions in boreal summer and winter (Wang et al. 2000, 2003; Wu et al. 2003; Wu et al. 2009, 2010a). Wu et al. (2003) demonstrated that ENSO-induced positive precipitation anomalies are located in southern China during winter and move to central China in the following summer, while negative precipitation anomalies occur over northern China during the summer and fall of an ENSO developing year. Wu et al. (2009) indicated that an anomalous anticyclone over the western North Pacific (WNP) in El Niño decaying summers is associated with positive precipitation anomalies from central China to southern Japan. Other factors, such as regional SST anomalies, also affect the climate variability in the SCS. For example, Wu et al. (2010b) showed that the Indian Ocean basin mode plays a crucial role in late summer by the Kelvin wave–induced anticyclonic shear and boundary layer divergence. Wu et al. (2012) indicated the influence of the SST anomalies in the tropical southeast Indian Ocean on the vertical motion over the SCS in boreal summer. Wu et al. (2013) examined the cross-season connections of rainfall variability in the SCS between winter and summer. The results showed that the in-phase relation from winter to summer occurs more often in El Niño and La Niña decaying years, and out-of-phase relation from summer to winter appears more frequently in El Niño and La Niña developing years.

Most of the above studies focused on boreal summer or winter climate and little attention has been paid to the transition period from spring to summer. In the present study, we focus on the SCS rainfall variability during April–June (AMJ) that is used to represent the transition season. This period is chosen because the onset of SCS summer monsoon occurs mostly in May (Wu and Wang 2000, 2001) and it signifies the start of the East Asian summer monsoon (Tao and Chen 1987). In this study, there are two issues to be addressed: 1) Which regional SST anomalies contribute to the SCS AMJ rainfall variability? 2) What are the processes for the influences of tropical Indian and Pacific SST anomalies? We will address these two issues using both observational analysis and numerical experiments.

The rest of the paper is organized as follows. The datasets and method are described in section 2. The leading mode of AMJ rainfall over SCS and its relationship with SST in the Indo-Pacific region are presented in section 3. Section 4 investigates three types of influences of tropical Indo-Pacific SST anomalies on SCS AMJ rainfall variability. The model results are shown in section 5. At last, summary and discussion are given in section 6.

2. Data and methods

This study utilizes monthly mean rainfall from version 2 of the Global Precipitation Climatology Project (GPCP; Adler et al. 2003). This dataset is on a 2.5° × 2.5° grid and is available from 1979 to 2012. (The GPCP rainfall is obtained through anonymous ftp at ftp://precip.gsfc.nasa.gov/pub/gpcp-v2/.)

The SST dataset used in the present study is the National Oceanic and Atmospheric Administration (NOAA) extended reconstructed SST, version 3 (ERSST3; Smith et al. 2008). ERSST3 is provided by NOAA’s Office of Oceanic and Atmospheric Research (OAR) Earth System Research Laboratory (ESRL) Physical Science Division (PSD), Boulder, Colorado (available online at http://www.esrl.noaa.gov/psd/data/gridded/data.noaa.ersst.html). The ERSST3 is on a 2° × 2° grid from 1979 to 2012.

We use monthly mean horizontal winds and vertical pressure velocity from the National Centers for Environmental Prediction–U.S. Department of Energy (NCEP–DOE) Reanalysis 2 (Kanamitsu et al. 2002) on a regular 2.5° × 2.5° grid for the period 1979–2012. The reanalysis product is provided by the NOAA/Cooperative Institute for Research in Environmental Sciences (CIRES) Climate Diagnostics Center, Boulder, Colorado (available via anonymous ftp at ftp://ftp.cdc.noaa.gov/).

To better capture the rainfall variability in the SCS and the influences of SST anomalies, the least squares linear trends have been removed from all of the above variables before the analysis conducted in the following. An empirical orthogonal function (EOF) and composite analyses are applied in this study. The significance level of correlation in the present study is determined based on the Student’s t test.

3. Relationship between AMJ SCS rainfall and Indo-Pacific SST

An EOF analysis is performed for AMJ precipitation anomaly in the SCS region. Figure 1 shows the spatial patterns and principal components (PCs) of the first and second EOF modes. The first and second modes account for about 34.5% and 14.2% of the total variance, respectively. According to the rule of North et al. (1982), the leading mode is separated from the other modes. The first EOF mode shows a large positive loading over the central SCS (Fig. 1a). A negative loading is confined to the southern coast of China (Fig. 1b). The correlation coefficient between the PC1 and area-mean AMJ rainfall in the region of 7°–17°N, 110°–120°E (box in Fig. 1a) is as high as 0.935. The time series of the leading mode and area-mean rainfall are very close to each other (Fig. 2). Thus, the leading mode (PC1) well represents the rainfall variability over the central SCS in AMJ. We use the 0.5 standard deviation as a criterion to determine an anomalous year of rainfall or SST. According to this criterion, there are 13 positive and 10 negative anomalous rainfall years during 1979–2012 (Fig. 2), which is the basis for the analysis of same-sign and opposite-sign relationships of SCS rainfall with SST in the Indo-Pacific region.

Fig. 1.
Fig. 1.

Spatial distributions of the (a) first and (b) second EOF modes of precipitation anomalies (contours, mm day−1) over the South China Sea in AMJ and (c) their principal components PC1 and PC2. The box in (a) denotes the region for calculation of area-mean rainfall in Fig. 2.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

Fig. 2.
Fig. 2.

Time series of the leading mode as in Fig. 1b (solid line) and AMJ area-mean rainfall anomalies (dashed line, mm day−1) within the box in Fig. 1a. Gray lines represent ±0.5 standard deviation of normalized anomalies.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The distributions of correlation coefficient of precipitation, 850-hPa wind, and SST anomalies in AMJ with PC1 are shown in Fig. 3. Significant positive correlation of precipitation covers the Indochinese Peninsula, most parts of the SCS, and the western North Pacific. Three negative correlation regions of precipitation are located over western tropical Indian Ocean, northwest of Australia, and the equatorial central Pacific. A cyclonic circulation is located over the SCS, which consists of northeasterly winds from south China and westerly winds from the Indian Ocean. Strong easterly winds prevail over the equatorial central Pacific (Fig. 3a). The SST correlation field features two negative regions and one positive region over the tropical Indian Ocean (TIO), the equatorial Pacific (EP), and the WNP, respectively (Fig. 3b).

Fig. 3.
Fig. 3.

Distributions of correlation coefficient of (a) precipitation (shading) and 850-hPa wind (vectors) and (b) SST (shading) in AMJ with respect to the time series of the leading mode of AMJ precipitation anomalies. Thick contours denote regions where the precipitation and SST correlations are significant at the 95% confidence level according to the Student’s t test. Only wind vectors that are significant at the 95% confidence level are plotted. The three rectangular boxes in (b) denote the key SST regions that are used to construct area-mean SST anomalies in Fig. 4.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

Figure 3a suggests that the rainfall variability in the SCS region is a part of the large-scale rainfall pattern of the northwestern Pacific. The correlation coefficient of grid precipitation with the precipitation in the central SCS region (7°–17°N, 110°–120°E) reaches 0.6 in the northwestern Pacific over 0°–20°N, 125°–150°E. However, there is still a large part of interannual variations that are not the same in SCS as in WNP. In addition, the regional air–sea relationship displays difference between the SCS and northwestern Pacific. As seen from Fig. 3, the positive rainfall anomaly corresponds to the negative SST anomaly in the SCS region, signifying atmospheric forcing of the SST in the SCS (Wu et al. 2006; Wu and Kirtman 2007), whereas the positive rainfall anomaly over the northwestern Pacific features a Rossby wave–type response to the positive SST anomaly to the east. Thus, it is necessary to distinguish the rainfall variability in the SCS and WNP regions.

We have also analyzed the distribution of precipitation, 850-hPa wind, and SST anomalies corresponding to PC2 (figures not shown). An anomalous anticyclone with a negative precipitation anomaly is located over the northern SCS and south China. An anomalous cyclone is seen over the low latitudes of the WNP. The wind anomalies over TIO and the central and eastern tropical Pacific appear weak. The SST anomalies feature a tripole pattern in the SCS and tropical Pacific: negative in southern SCS, positive in the equatorial western Pacific with northeast and southeast extension, and negative in the equatorial eastern Pacific. The SST anomalies in the TIO are small. The cyclonic and anticyclonic winds over the western Pacific, SCS, and south China appear as Rossby wave–type responses to positive SST anomalies in the equatorial western Pacific and negative SST anomalies in the southern SCS.

Based on the distribution of the SST correlation, we define three key SST regions, as shown by the rectangular boxes in Fig. 3b, to conduct the following comparisons. Figure 4 shows the time series of AMJ SST anomalies averaged over the three boxes shown in Fig. 3b. According to the criterion of anomalous year, there are 10 (10), 9 (11), and 11 (11) positive (negative) years of TIO, WNP, and EP SST anomalies, respectively. The correlation coefficients of these area-mean SST anomalies with AMJ SCS PC1 are −0.607, 0.67, and −0.682, respectively, all of which are significant at the 95% confidence level according to the Student’s t test. From the correlation, it appears that AMJ SCS rainfall variability is related to the TIO, WNP, and EP SST anomalies. From Fig. 4, the SST anomalies in the above regions vary coherently in some years, but not in the other years. In the next section, we perform a conditional composite analysis to address the respective influences of TIO, WNP, and EP SST anomalies on the SCS rainfall variability.

Fig. 4.
Fig. 4.

Normalized time series of the PC1 of AMJ SCS rainfall (gray bars), and AMJ SST averaged over the tropical Indian Ocean (TIO, solid line), the western North Pacific (WNP, long-dash line) and the equatorial Pacific (EP, long–short dash line). Gray lines denote ±0.5 standard deviation of normalized anomalies. The numbers at the top right are the simultaneous correlation coefficients with the PC1.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

4. Composite analysis of the influence of Indo-Pacific SST anomalies

The previous section suggests that the TIO, WNP, and EP SST anomalies may contribute to AMJ SCS rainfall variability either coherently or individually. It also indicates that the TIO and EP SST anomalies tend to have the opposite sign and WNP SST anomalies tend to have the same sign as SCS AMJ rainfall anomalies. Based on the sign relationship of these SST anomalies and SCS rainfall anomalies, we determine in which year TIO, WNP, and EP SST anomalies may have influences on AMJ SCS rainfall based on the criterion of 0.5 standard deviation (see Table 1). From Table 1, three specific types of SST influence cases can be determined. In the first type (called type 1), a positive/negative SCS rainfall anomaly corresponds to a negative/positive TIO and EP SST anomaly and a positive/negative WNP SST anomaly. In these cases, there appears to be combined influences of TIO, WNP, and EP SST anomalies. Type-1 cases contain three positive (1984, 1999, and 2000) and four negative SCS rainfall years (1983, 1987, 1991, and 2005). In the second type (type 2), TIO SST is normal and a positive/negative SCS rainfall anomaly corresponds to a negative/positive EP SST anomaly and/or a positive/negative WNP SST anomaly. In these cases, SCS rainfall is influenced mainly by the Pacific SST anomalies. Type-2 cases include three positive (1981, 1988 and 2001) and four negative (1992, 1993, 1995, and 1997) SCS rainfall years. In the third type (type 3), the WNP SST anomaly is small and a positive/negative SCS rainfall anomaly corresponds to a negative/positive TIO and EP SST anomaly. In these cases, TIO and EP SST anomalies have combined influences on the SCS rainfall. Type-3 cases consist of three positive (1985, 2008, and 2011) and one negative (1998) SCS rainfall years.

Table 1.

Anomalous years when AMJ TIO and EP SST anomalies have the opposite sign to SCS AMJ rainfall and AMJ WNP SST anomalies have the same sign as SCS AMJ rainfall based on the criterion of 0.5 standard deviation. Entries in boldface denote years of combined influences of TIO, WNP, and EP SST anomalies (type 1). Entries in italic denote years of influence of the Pacific SST anomalies (type 2). Entries in bold and italic denote years of combined influence of TIO and EP SST anomalies (type 3).

Table 1.

In the following, we construct composite anomalies for the above three types of cases separately to understand the influences of SST anomalies in different regions on the SCS rainfall variability. As the number of either positive or negative rainfall anomaly cases in each type is small, we combine the positive and negative rainfall anomaly cases in the composite analysis to increase the robustness of the results. In constructing the composite, we multiply the anomalies by −1 when AMJ SCS rainfall anomalies are negative. One sample t test is used to determine whether the composite anomalies are significant. A parallel analysis has been conducted using one standard deviation as the criterion. The obtained composite anomalies display spatial distributions similar to those based on the 0.5 standard deviation.

a. Combined influences of TIO, WNP, and EP SST anomalies

Figure 5 shows composite anomalies of precipitation, 850-hPa winds, and SST in AMJ for type-1 cases when TIO, WNP, and EP SST anomalies may act in concert. There are two negative rainfall anomaly regions: one over the southwestern tropical Indian Ocean and the other over the equatorial central Pacific (Fig. 5a). Positive rainfall anomalies are observed over the Indochinese Peninsula, most parts of the SCS, the western North Pacific, and the Maritime Continent. These are the regions of lower-level wind convergence between anomalous westerlies from the Indian Ocean and anomalous easterlies from the Pacific or East Asia. Anomalous easterlies over the equatorial Pacific appear as response to anomalous cooling associated with negative SST anomalies in the equatorial central and eastern Pacific (Fig. 5b). Anomalous westerlies over the north Indian Ocean are connected with anomalous cross-equatorial flows from the south Indian Ocean. These cross-equatorial winds appear to be caused by negative SST anomalies in the southwestern tropical Indian Ocean where negative rainfall anomalies are observed (Fig. 5a). The anomalous winds over the tropical Indian Ocean closely resemble those corresponding to the asymmetric mode during boreal spring (Wu et al. 2008). Negative SST anomalies in the north Indian Ocean are likely a response of the ocean to the atmosphere as anomalous westerlies increase surface wind speed, enhancing surface evaporation and oceanic mixing. Our results agree with Yoo et al. (2006) who showed that the southern Indian Ocean SST anomalies influence the Asian summer monsoon more strongly than their northern Indian Ocean counterpart. Positive SST anomalies in the western North Pacific appear to contribute complementarily to the above-normal rainfall and the anomalous cyclone around the SCS via a Rossby wave–type response.

Fig. 5.
Fig. 5.

Composite anomalies of (a) precipitation (shading, mm day−1) and 850-hPa wind (vectors, m s−1) and (b) SST (shading, °C) during AMJ for type-1 cases. Black contours denote regions where the composite precipitation and SST anomalies are significant at the 95% confidence level according to one sample t test. Only wind anomalies that are significant at the 95% confidence level are plotted.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The corresponding composite anomalies of 850- and 200-hPa velocity potential and divergent winds and 500-hPa vertical velocity in AMJ are shown in Fig. 6. Above-normal precipitation over the SCS, western North Pacific, and the Maritime Continent seen in Fig. 5a corresponds to lower-level convergence, upper-level divergence, and vertical ascent motion. In contrast, there are lower-level divergence, upper-level convergence, and descent over the southwestern tropical Indian Ocean and the equatorial central Pacific, corresponding to negative SST anomalies (Fig. 5b) and below-normal precipitation (Fig. 5a). The downward vertical motion is strong over the equatorial central Pacific (Fig. 6c). The above anomalies imply an anomalous Walker circulation between the equatorial central Pacific and the Maritime Continent–Philippines. This anomalous Walker circulation links the SCS climate to the equatorial central and eastern Pacific SST anomalies. While the ascending motion over the western North Pacific (Fig. 6c) may be partially attributed to positive SST anomalies there (Fig. 5b), lower-level southwesterly winds indicate reduced surface evaporation that may contribute to SST warming. Thus, positive SST anomalies in the WNP appear as a response to the atmospheric change induced by remote SST forcing in the equatorial central and eastern Pacific. This suggests air–sea interaction processes in the western North Pacific (Wang et al. 2000; Wu et al. 2013).

Fig. 6.
Fig. 6.

Composite anomalies of (a) 850- and (b) 200-hPa velocity potential (shading, 10−6 m2 s−1) and divergent winds (vectors, m s−1) and (c) 500-hPa vertical pressure velocity (shading, Pa s−1) during AMJ for type-1 cases. Black contours and green vectors denote regions where the composite anomalies are significant at the 95% confidence level according to one sample t test.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The above analysis suggests that the SCS rainfall variability is affected by both TIO and EP SST anomalies in type 1. Negative SST anomalies in the southwestern tropical Indian Ocean induce anomalous cross-equatorial flows that enhance monsoon westerlies over the north Indian Ocean through the SCS. Negative SST anomalies in the equatorial central Pacific induce an anomalous Walker circulation that enhances ascent over the Maritime Continent and the SCS. Together, they lead to above-normal precipitation over the SCS. When opposite SST anomalies occur in the above regions, below-normal precipitation is expected over the SCS. In addition, wind anomalies over the western North Pacific in response to equatorial central Pacific SST forcing induce regional SST anomalies, which, in turn, may have a complementary contribution to the SCS rainfall anomalies via a Rossby wave–type response as demonstrated by Wu et al. (2013) using the atmospheric model experiment with specified SST anomalies in the WNP.

b. Influences of the Pacific SST anomalies

Geographically, the effects of WNP or EP SST anomalies could be combined as the influence of the Pacific SST. Figure 7 shows composite anomalies of precipitation, 850-hPa wind, SST, and the instability index in AMJ for type-2 cases. Here, the instability index is defined as the difference of saturated equivalent potential temperature between 1000 and 700 hPa (divided by the pressure difference) to illustrate the convective instability region. The pattern of SST anomaly distribution indicates that there is hardly any large negative anomaly in the tropical Indian Ocean (Fig. 7b). This confirms the lack of the Indian Ocean SST impact. In the Pacific Ocean, there is a contrast of the SST anomaly between the equatorial central and eastern Pacific and the western North Pacific. The contrast of SST anomaly signifies an eastward extension of the western Pacific warm pool along 10°–20°N. The response of the atmosphere to negative SST anomalies in the equatorial Pacific features easterly winds over the equatorial region and a large anticyclone over the North Pacific (Fig. 7a). Consequently, negative and positive rainfall anomalies appear over the equatorial central and western Pacific, respectively (Fig. 7a). There are anomalous westerlies over the north Indian Ocean and cross-equatorial southerlies over the western Indian Ocean. These appear to be in response to the anomalous heating over the SCS and the surrounding regions as inferred from above-normal precipitation extending from the Philippine Sea to the Bay of Bengal.

Fig. 7.
Fig. 7.

Composite anomalies of (a) precipitation (shading, mm day−1) and 850-hPa wind (vectors, m s−1), (b) SST (shading, °C), and (c) instability index (shading, K hPa−1) during AMJ for type-2 cases. Black contours denote regions where the composite anomalies are significant at the 95% confidence level according to one sample t test. Only wind anomalies that are significant at the 95% confidence level are plotted.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The corresponding composite anomalies of lower-level and upper-level velocity potential and divergent winds as well as 500-hPa vertical velocity are shown in Fig. 8. Lower-level divergence, upper-level convergence, and descent are seen over the equatorial central Pacific; and lower-level convergence, upper-level divergence, and ascent are observed to extend from the Maritime Continent and the Philippine Sea through the SCS and the Indochinese Peninsula to the Bay of Bengal. Compared to type 1, both rainfall and wind anomalies over the SCS are weaker. This difference is partly due to the lack of the TIO SST effect, and partly due to the relatively weak EP SST anomalies.

Fig. 8.
Fig. 8.

As in Fig. 6, but type-2 cases.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The positive SST anomalies in the western North Pacific cover a much larger area compared to type 1 and extend to the SCS (Fig. 7b vs Fig. 5b). Higher SST means warmer surface condition, which may destabilize the lower troposphere (Wu and Wang 2000, 2001). This is confirmed by the change in the stability index shown in Fig. 7c. A large area of a positive instability index change is seen over the WNP and southwestern Pacific, which corresponds well to positive SST anomalies (Fig. 7b). It indicates that warmer SST in the WNP indeed increases the convective instability in the atmosphere. This would provide a favorable condition for more precipitation. Although anomalous lower-level winds in type 2 are not so obvious over the SCS (Fig. 7a), the stability change may induce above-normal rainfall even if there is no strong convergence. This suggests that the WNP SST anomalies may play an important role in the AMJ SCS rainfall variability when the EP and TIO SST influences are relatively weak.

c. Combined influences of TIO and EP SST anomalies

Now we examine the four cases with combined influences of TIO and EP SST anomalies (type 3). Figure 9 shows composite anomalies of rainfall, 850-hPa wind, and SST in AMJ for type 3. Negative rainfall anomalies are observed over the equatorial central Pacific and tropical south Indian Ocean and positive rainfall anomalies are seen over the equatorial western Pacific, the Philippine Sea, and the SCS (Fig. 9a). Consistently, there are anomalous easterlies over the equatorial central Pacific, anomalous westerlies over the north Indian Ocean through the Philippine Sea, and anomalous cross-equatorial southerlies over the tropical Indian Ocean. Over southern China and the subtropical western North Pacific, a band of negative rainfall anomalies can be seen, which is accompanied by anomalous northerly winds to the south side. The precipitation and wind anomalies in the tropics can be attributed to negative SST anomalies in the equatorial central and eastern Pacific and tropical Indian Ocean (Fig. 9b) via the processes as in type 1. In type 3, negative SST anomalies are also observed in the SCS to the subtropical western North Pacific (Fig. 9b). These negative SST anomalies in the SCS are responses to atmospheric change. Above-normal precipitation is expected to reduce the shortwave radiation reaching the ocean surface. Anomalous westerlies increase surface wind speed, which may enhance surface evaporation and oceanic mixing. These effects may contribute to SST cooling. In turn, these local SST anomalies may have a negative feedback on the atmosphere, leading to smaller precipitation anomalies in the SCS region compared to the other two types of cases (Figs. 5a and 7a).

Fig. 9.
Fig. 9.

As in Fig. 7, but for type-3 cases and without panel (c).

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

While composite precipitation and wind anomalies in type 3 display similarity to those in type 1, there are several important differences to note. First, cross-equatorial southerlies over the Indian Ocean have a much broader eastward extension, covering the whole Indian Ocean and part of the Maritime Continent (Fig. 9a). This is consistent with the large eastward extension of negative rainfall anomalies over the tropical south Indian Ocean. The above differences are attributed to the eastward extension of negative SST anomalies to coastal Sumatra (Fig. 9b). The cross-equatorial southerlies converge with the northerlies from the subtropics, directly enhancing lower-level convergence and precipitation over the SCS. Second, obvious negative SST anomalies appear in the SCS as a response of the ocean to the atmosphere. Third, negative rainfall anomalies appear over southern China–subtropical western North Pacific, forming a north–south contrast pattern with positive precipitation anomalies over the SCS–Philippine Sea.

The corresponding composite anomalies of velocity potential, divergent winds, and vertical velocity are shown by Fig. 10. The most prominent feature is the large lower-level convergence and upper-level divergence over the western North Pacific (Figs. 10a,b). This is accompanied by upward motion (Fig. 10c). These contrast with opposite anomalies over the equatorial eastern Pacific and tropical south Indian Ocean. In addition, part of the upper-level divergent winds flow northward toward the subtropics where there is convergence and descent (Figs. 10b,c). At the lower level, there is divergence and anomalous southward flow (Fig. 10a). The magnitude of 500-hPa vertical velocity over the Maritime Continent is larger than that in type 1 (Fig. 10c vs Fig. 6c).

Fig. 10.
Fig. 10.

As in Fig. 6, but type-3 cases.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

The above analyses suggest an important role of coastal Sumatra SST anomalies. The strong downdraft induced by negative SST anomalies near coastal Sumatra causes the formation of an anomalous vertical circulation with upper-level cross-equatorial southwestward flow, lower-level cross-equatorial northeastward flow, and the upward branch over the eastern Maritime Continent and the western North Pacific. Such a vertical circulation links the SCS climate to the southeastern tropical Indian Ocean SST anomalies, consistent with Wu et al. (2012).

5. Results of model experiments

The above observational results indicate the essential roles of SST anomalies in the tropical Indo-Pacific oceans, especially the two key regions: the southern tropical Indian Ocean (SIO; 30°S–5°N, 30°–120°E) and the eastern tropical Pacific Ocean (ETP; 30°S–20°N, 155°E–90°W), in influencing the rainfall variability over the SCS during the transition season from spring to summer (AMJ). In this section, we conduct experiments with an atmospheric general circulation model (AGCM) to confirm the roles of the remote SST forcing in the above two regions in the SCS rainfall variations.

The AGCM used is the Geophysical Fluid Dynamics Laboratory (GFDL) Atmospheric Model, version 2.1 (AM2.1). The AM2.1 model is the atmospheric component of GFDL Coupled Model, version 2.1 (CM2.1), which employs the finite-volume dynamical core (Lin 2004) with a horizontal resolution of 2° latitude by 2.5° longitude and 24 levels with a hybrid coordinate in the vertical direction. A comprehensive description of CM2.1 can be found in Anderson et al. (2004) and Delworth et al. (2006).

At first, we conduct one 10-yr control integration that is forced with monthly climatological SST specified in the global oceans. Second, we carry out two sensitivity experiments with SST anomalies added in the two key ocean areas: the SIO and ETP (green box in Fig. 11). Two additional sensitivity experiments are performed with SST anomalies added in the southwestern and southeastern tropical Indian Ocean (red and blue boxes in Fig. 11), named as WSIO and ESIO runs, respectively. These additional experiments are used to understand the relative importance of SST anomalies in these two regions. These sensitivity experiments include four individual (SIO, ESIO, WSIO, and ETP) and one combined (SIO plus ETP) experiment. Each experiment is integrated for 10 years and forced with prescribed SST like in the control run except that monthly SST anomalies are added to monthly climatological SST during March–June in SIO, ESIO, WSIO, and ETP in individual experiments or in two regions in the combined SST experiment. A detailed description of numerical experiments is given in Table 2.

Fig. 11.
Fig. 11.

The SST anomalies specified in the AGCM experiments. The SST anomalies are the same as those in Fig. 9b in the tropical Indo-Pacific oceans but only the negative anomalies are plotted. Red, blue, and green boxes denote the WSIO, the ESIO, and the ETP, respectively.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

Table 2.

Description of numerical experiments with GFDL AM2.1. Refer to Fig. 11 for the domains of the eastern tropical Pacific Ocean (ETP), southern tropical Indian Ocean (SIO), western part of SIO (WSIO), and eastern part of SIO (ESIO).

Table 2.

The SST anomalies applied in these sensitivity experiments are shown in Fig. 11, which is the same as SST anomalies in Fig. 9b but only negative SST anomalies are adopted. In addition, simulations similar to the above sensitivity simulations but with opposite SST anomalies are performed to verify the sensitivity of the response to the sign of anomalous SST forcing. The difference between the sensitivity experiment and the control run is considered to be a response to the SST anomalies specified in the sensitivity experiment. The obtained responses to negative and positive SST anomalies display a similar feature (with opposite signs). In the following, we only discuss results of responses to negative minus positive SST anomalies (divided by 2) for some selected experiments.

In the AGCM, the SIO cooling (Fig. 12a) causes below-normal rainfall over most parts of SIO and above-normal rainfall over the northern Indian Ocean, SCS, and WNP. Correspondingly, low-level winds over the tropical Indian Ocean feature cross-equatorial flows and easterlies south of the equator and westerlies north of the equator. The westerlies extend eastward and form an anomalous cyclone over the northern SCS and the WNP. These features well match the composite anomalies (Fig. 9a) except for opposite rainfall over the western equatorial Indian Ocean and too eastward of an extension of the anomalous cyclone over the WNP. These differences may be due to the lack of remote forcing from the ETP.

Fig. 12.
Fig. 12.

The responses of AMJ rainfall (shaded, mm day−1) and 850-hPa winds (vectors, m s−1) to anomalous SST forcing in (a) SIO, (b) SIO plus ETP, (c) ESIO, and (d) WSIO as differences of responses to negative minus positive SST anomalies divided by 2. The vectors depict winds with speed over 0.2 m s−1.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-14-00089.1

In the SIO plus ETP run (Fig. 12b), remarkable easterly and negative rainfall anomalies are situated over the central and eastern tropical Pacific Ocean, suggesting the atmospheric response to the ETP cooling. The distributions of rainfall and low-level wind anomalies over the tropical Indian Ocean are similar to those in the SIO run except that the magnitude of these anomalies is weakened with the remote forcing of the ETP cooling. Meanwhile, the low-level anomalous cyclone and positive rainfall anomalies are enhanced and more concentrated over the SCS and WNP. These features indicate the contribution of the ETP cooling, which is obvious in the individual ETP run as well (figure not given).

Figures 12c and 12d display the rainfall and 850-hPa wind response to the respective anomalous SST forcing in ESIO and WSIO. The ESIO cooling (Fig. 12c) triggers cross-equatorial flows and antisymmetric winds over the tropical Indian Ocean similar to those in Fig. 12a but with a weakened intensity. The anomalous westerlies north of the equator appear to be discontinued over the SCS though an anomalous cyclone is present over the WNP. As such, rainfall anomalies over the SCS are weak. In the WSIO run (Fig. 12d), cross-equatorial flows are limited to the western tropical Indian Ocean and an anomalous anticyclone and below-normal rainfall appear over the SCS. Comparing Figs. 12c,d with Fig. 12a, it suggests that the coexisting SST anomalies in the ESIO and WSIO are more important than SST anomalies only in part of the SIO in producing the observed response over the SCS.

It is worth noting that the responses in the WSIO and ESIO runs do not add up to the response in the SIO run. This nonlinearity appears to be related to the response in the equatorial Indian Ocean. In the WSIO run, positive rainfall anomalies are induced over the equatorial Indian Ocean (Fig. 12d). In turn, the associated anomalous heating induces easterlies to the east via a Kelvin wave–type response. Consequently, anticyclonic wind anomalies and below-normal precipitation develop over the SCS, which are opposite to the case when SST anomalies are specified in the whole SIO region (Fig. 12a). In the ESIO run, negative precipitation anomalies are induced over the equatorial central and eastern Indian Ocean (Fig. 12c). These appear to overcome the positive precipitation anomalies due to negative SST anomalies in the southwestern tropical Indian Ocean (Fig. 12a). As such, the Kevin wave–type response as stated above is not seen in the SIO case. Thus, the nonlinearity may be attributed to the cancellation of the atmospheric response to the eastern and western SIO SST anomalies.

6. Summary and discussion

The present study documented the relationship between SCS precipitation and tropical Indo-Pacific SST during AMJ. Analysis of observations showed that the SCS rainfall variability in AMJ is closely related to SST anomalies in the TIO, WNP, and EP. Based on the sign relationship of SCS rainfall anomalies and the SST anomalies in the above three regions, we classified three different types of Indo-Pacific SST influence cases on the SCS rainfall variability. We emphasize the roles of the SST anomaly pattern that is a combination of SST anomalies in different regions. A conditional composite analysis was then conducted to reveal the influences of different combinations of TIO, WNP, and EP SST anomalies on the SCS rainfall.

In type 1, when TIO, WNP, and EP SST anomalies may act in concert, anomalous cross-equatorial flows over western Indian Ocean are induced by TIO SST anomalies, which modulate monsoon westerlies over the north Indian Ocean through the SCS. At the same time, an anomalous Walker circulation is induced over the western and central tropical Pacific by EP SST anomalies, which modulates trade winds over the tropical western Pacific. Together, they alter lower-level convergence and consequently upward motion and precipitation over the SCS. In type 1, the WNP SST anomalies play a complementary role in the SCS rainfall variability via a Rossby wave–type response.

In type 2 (influence of EP and/or WNP SST anomalies), large-scale circulation anomalies appear weaker due to the lack of TIO influence and weakness of EP SST influence. The WNP SST anomalies affect the SCS precipitation via modulating the thermal condition and atmospheric stability. The results indicate that the WNP SST anomalies play a more important role in type 2 than in type 1.

In type 3 (combined influences of TIO and EP SST anomalies), the impact of the coastal Sumatra SST anomalies is prominent. An anomalous cross-equatorial vertical circulation plays a role in linking coastal Sumatra SST anomalies to the SCS rainfall variability. Different from the other two types, notable SST anomalies develop in the SCS as a response to cloud and shortwave radiation changes induced by remote SST forcing. Another feature unique in type 3 is a north–south contrast of precipitation anomalies between the SCS–Philippine Sea and southern China–subtropical western North Pacific.

The observational analysis provides a clue to the key regions of SST anomalies that may influence the SCS rainfall variability in the transition from spring to summer. The results, however, are subject to the limited number of cases in the analysis period for different types of SST influences. The results of numerical experiments with an AGCM with specified SST forcing confirm the influences of SST anomalies in tropical Indo-Pacific regions on the SCS rainfall variability. The sensitivity experiments suggest that a combined influence of SST anomalies in different regions is more robust than SST anomalies in individual regions.

Distinguishing itself from previous studies, the present study focuses on the precipitation variability during the spring to summer transition in the SCS region. The present study also emphasizes the importance of the SST anomaly pattern and the combined influences of SST anomalies in different regions. This differs from most previous studies that consider the influences of SST anomalies in individual regions.

In all the three types of cases, significant SST anomalies are seen in the EP. This suggests that the EP SST anomalies may play a dominant role in the SCS rainfall variability. From Table 1, there are two cases (2004 and 2010) of influence of TIO SST anomalies and three cases of influence of WNP SST anomalies (1995, 2001, and 2010) when the EP SST anomalies are small. This suggests that the roles of TIO and WNP SST anomalies are independent of ENSO. On the other hand, previous studies indicated that the SST anomalies in the tropical Indian Ocean may be induced by preceding SST anomalies through the atmospheric bridge (Klein et al. 1999). Thus, these SST anomalies may serve as a medium for ENSO’s indirect influence on SCS rainfall variability (He and Wu 2014). This is the case for year 2010 during which positive TIO SST anomalies in AMJ are preceded by positive EP SST anomalies in the previous winter, but not for the year 2004 when negative TIO SST anomalies occur after weak positive EP SST anomalies. Further studies are needed to separate the ENSO’s direct and indirect influences on the SCS rainfall variability in the transition season.

The onset of the SCS summer monsoon is an important stage in the transition from spring to summer. As an important component of the East Asian summer monsoon, the onset of the SCS summer monsoon is signified by a wind shift from easterly to westerly in the lower troposphere and a burst of deep convection over the SCS (Lau et al. 1998; Wu and Zhang 1998; Wu and Wang 2001; Wu 2002; Lau et al. 2002; Mao et al. 2004). The present analysis shows that the wind direction over the SCS is influenced by the Indo-Pacific SST anomalies to a large extent. A question that needs to be answered is what exactly causes the final shift of winds from easterly to westerly. The relative importance of the southern and, northern Indian Ocean and Pacific SST anomalies in the season transition is an important issue that needs to be addressed in the future. As the intraseasonal oscillation also contributes to the time of the summer monsoon onset over the SCS (Wu and Wang 2001), another issue to investigate is the role of the intraseasonal oscillation in the seasonal transition from spring to summer.

Acknowledgments

The present study is supported by National Key Basic Research Program of China Grant (2014CB953902). RW acknowledges the support of Hong Kong Research Grants Council Grant (CUHK403612) and National Natural Science Foundation of China Grant (41275081). WH acknowledges the support of the CAS Project (XDA11010402) and National Natural Science Foundation of China Grant (91337216). YL acknowledges the support of the National Key Basic Research Program of China Grant (2013CB430201) and National Natural Science Foundation of China Grant (41230527).

REFERENCES

  • Adler, R. F., and Coauthors, 2003: The version 2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979–present). J. Hydrometeor., 4, 11471167, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, J. L., and Coauthors, 2004: The new GFDL global atmosphere and land model AM2–LM2: Evaluation with prescribed SST simulations. J. Climate, 17, 46414673, doi:10.1175/JCLI-3223.1.

    • Search Google Scholar
    • Export Citation
  • Delworth, T. L., and Coauthors, 2006: GFDL’s CM2 global coupled climate models. Part I: Formulation and simulation characteristics. J. Climate, 19, 643674, doi:10.1175/JCLI3629.1.

    • Search Google Scholar
    • Export Citation
  • He, Z., , and R. Wu, 2014: Indo-Pacific remote forcing in summer rainfall variability over the South China Sea. Climate Dyn.,42, 2323–2337, doi:10.1007/s00382-014-2123-7.

  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S. K. Yang, , J. J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 16311643, doi:10.1175/BAMS-83-11-1631.

    • Search Google Scholar
    • Export Citation
  • 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, 917932, doi:10.1175/1520-0442(1999)012<0917:RSSTVD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., , H. T. Wu, , and S. Yang, 1998: Hydrologic processes associated with the first transition of the Asian summer monsoon: A pilot satellite study. Bull. Amer. Meteor. Soc., 79, 18711882, doi:10.1175/1520-0477(1998)079<1871:HPAWTF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., , X. Li, , and M. T. Wu, 2002: Evolution of the large scale circulation, cloud structure and regional water cycle associated with the South China Sea monsoon during May–June, 1998. J. Meteor. Soc. Japan, 80, 11291147, doi:10.2151/jmsj.80.1129.

    • Search Google Scholar
    • Export Citation
  • Lin, S., 2004: A “vertically Lagrangian” finite-volume dynamical core for global models. Mon. Wea. Rev., 132, 22932307, doi:10.1175/1520-0493(2004)132<2293:AVLFDC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Mao, J. Y., , J. Chan, , and G. X. Wu, 2004: Relationship between the onset of the South China Sea summer monsoon and the structure of the Asian subtropical anticyclone. J. Meteor. Soc. Japan, 82, 845849, doi:10.2151/jmsj.2004.845.

    • Search Google Scholar
    • Export Citation
  • 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, 699706, doi:10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., , R. W. Reynolds, , T. C. Peterson, , and J. Lawrimore, 2008: Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006). J. Climate, 21, 22832296, doi:10.1175/2007JCLI2100.1.

    • Search Google Scholar
    • Export Citation
  • Tao, S. Y., , and L. X. Chen, 1987: A review of recent research on the East Asian summer monsoon in China. Monsoon Meteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 60–92.

  • Tomita, T., , and T. Yasunari, 1996: Role of the northeast winter monsoon on the biennial oscillation of the ENSO/monsoon system. J. Meteor. Soc. Japan, 74, 399413.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , R. Wu, , and X. Fu, 2000: Pacific–East Asian teleconnection: How does ENSO affect East Asian climate? J. Climate, 13, 15171536, doi:10.1175/1520-0442(2000)013<1517:PEATHD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , R. Wu, , and T. Li, 2003: Atmosphere–warm ocean interaction and its impacts on Asian–Australian monsoon variation. J. Climate, 16, 11951211, doi:10.1175/1520-0442(2003)16<1195:AOIAII>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Wang, D., , Q. Xie, , Y. Du, , W. Q. Wang, , and J. Chen, 2002: The 1997–1998 warm event in the South China Sea. Chin. Sci. Bull., 47, 12211227, doi:10.1007/BF02907614.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Zhou, , and T. Li, 2009: Seasonally evolving dominant interannual variability mode over the East Asia. J. Climate, 22, 29923005, doi:10.1175/2008JCLI2710.1.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Li, , and T. Zhou, 2010a: Asymmetry of atmospheric circulation anomalies over the western North Pacific between El Niño and La Niña. J. Climate, 23, 48074822, doi:10.1175/2010JCLI3222.1.

    • Search Google Scholar
    • Export Citation
  • Wu, B., , T. Li, , and T. Zhou, 2010b: Relative contributions of the Indian Ocean and local SST anomalies to the maintenance of the western North Pacific anomalous anticyclone during El Niño decaying summer. J. Climate, 23, 29742986, doi:10.1175/2010JCLI3300.1.

    • Search Google Scholar
    • Export Citation
  • Wu, G. X., , and Y. S. Zhang, 1998: Tibetan Plateau forcing and the monsoon onset over South Asia and the South China Sea. Mon. Wea. Rev., 126, 913927, doi:10.1175/1520-0493(1998)126<0913:TPFATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., 2002: Processes for the northeastward advance of the summer monsoon over the western North Pacific. J. Meteor. Soc. Japan, 80, 6783, doi:10.2151/jmsj.80.67.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. Wang, 2000: Interannual variability of summer monsoon onset over the western North Pacific and the underlying processes. J. Climate, 13, 24832501, doi:10.1175/1520-0442(2000)013<2483:IVOSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. Wang, 2001: Multi-stage onset of the summer monsoon over the western North Pacific. Climate Dyn., 17, 277289, doi:10.1007/s003820000118.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , and B. P. Kirtman, 2007: Regimes of local air-sea interactions and implications for performance of forced simulations. Climate Dyn., 29, 393410, doi:10.1007/s00382-007-0246-9.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , Z.-Z. Hu, , and B. P. Kirtman, 2003: Evolution of ENSO-related rainfall anomalies in East Asia. J. Climate, 16, 37423758, doi:10.1175/1520-0442(2003)016<3742:EOERAI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , B. P. Kirtman, , and K. Pegion, 2006: Local air–sea relationship in observations and model simulations. J. Climate, 19, 49144932, doi:10.1175/JCLI3904.1.

    • Search Google Scholar
    • Export Citation
  • 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.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , S. Yang, , Z. P. Wen, , G. Huang, , and K. Hu, 2012: Interdecadal change in the relationship of southern China summer rainfall with tropical Indo-Pacific SST. Theor. Appl. Climatol., 108, 119133, doi:10.1007/s00704-011-0519-4.

    • Search Google Scholar
    • Export Citation
  • Wu, R., , G. Huang, , Z. Du, , and K. Hu, 2013: Cross-season relation of the South China Sea precipitation variability between winter and summer. Climate Dyn., doi:10.1007/s00382-013-1820-y, in press.

    • Search Google Scholar
    • Export Citation
  • Xie, S. P., , Q. Xie, , D. 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.

    • Search Google Scholar
    • Export Citation
  • Yoo, S. H., , S. Yang, , and C. H. Ho, 2006: Variability of the Indian Ocean sea surface temperature and its impacts on Asian-Australian monsoon climate. J. Geophys. Res., 111, D03108, doi:10.1029/2005JD006001.

    • Search Google Scholar
    • Export Citation
  • Zhou, L. T., , C. Y. Tam, , W. Zhou, , and J. C. L. Chan, 2010: Influence of South China Sea SST and ENSO on winter rainfall over south China. Adv. Atmos. Sci., 27, 832844, doi:10.1007/s00376-009-9102-7.

    • Search Google Scholar
    • Export Citation
Save