Recent Changes of Northern Indian Ocean Summer Rainfall Based on CMIP5 Multi-Model

YANG Yali1), 2), DU Yan1), *, ZHANG Yuhong1), 2), and CHENG Xuhua1)



Recent Changes of Northern Indian Ocean Summer Rainfall Based on CMIP5 Multi-Model

YANG Yali, DU Yan, ZHANG Yuhong, and CHENG Xuhua

1),,,510301,.2),,100049,..

This study evaluates the simulation of summer rainfall changes in the Northern Indian Ocean (NIO) based on the fifth phase of Coupled Model Intercomparison Project (CMIP5). The historical runs of 20 CMIP5 coupled General Circulation Models (GCMs) are analyzed. The Multi-Model ensemble (MME) of the CMIP5 models well reproduces the general feature of NIO summer rainfall. For a short period 1979-2005, 14 out of 20 models show an increased trend in the mean rainfall and a similar spatial distribution to the Global Precipitation Climatology Project (GPCP) observations in MME. The increasing of the convergence in the equatorial IO results in the increase of rainfall significantly. The equatorial rainfall trend patterns seem modulated by the SST warming in the tropical Indian Ocean, which confirm the mechanism of ‘warmer-get-wetter’ theory. For a long period 1950-2005, the trend of monsoon rainfall over India shows a decrease over the most parts of the India except an increase over the south corn er of the Indian Peninsula, due to a weakened summer monsoon circulation. The pattern is well simulated in half of the CMIP5 models. The rainfall over the north India is different for a short period, in which rainfall increases in 1979-2005, implying possible decadal variation in the NIO summer climate.

NIO summer rainfall; Indian summer monsoon; inter-decadal changes

1 Introduction

The Indian Ocean (IO) is a typical monsoon region, and the monsoon rainfall is crucial to the social and economic activities of local residents. It is important to know the rainfall change in the past and whether the precipitation will increase or decrease in the future under global warming scenario.

The IO has been experienced a basin-wide warming trend since the 1950s from observational and modeling evidence (Alory., 2007; Du and Xie, 2008; Luffman., 2010). Xie. (2010a) found that the SST trend patterns play a key role in determining the precipitation changes. The tropical mean sea surface temperature (SST) warming leads to an upward trend in the convective threshold over the past 30 years, triggering a global precipitation distribution adjustment (Johnson and Xie, 2010). Levermann. (2009) and Zickfeld. (2005) showed that the Indian summer monsoon can operate in two stable regimes: enhanced summer monsoon or a low rainfall over India. The increase of tropical IO SST may enhance the convection through the troposphere and con-vective outflow at the upper levels, leading to a trend in subsidence over the Indian continent monsoon region (Luffman., 2010). Some studies supposed that the warming is caused by emissions of greenhouse gases from human activities, and Indian summer monsoon rainfall is likely to increase with increased carbon dioxide (Meehl and Washington, 1993; Kitoh., 1997; Hu., 2000). The increased rainfall is related to an enhanced land-ocean thermal gradient driven by increased surface air temperatures over Eurasia in winter and spring (Kumar., 1999; Hu., 2000). Moreover, a larger moisture flux convergence resulting from a warming Indian Ocean can lead to the intensification of the mean rainfall (Meehl., 2003; Ueda., 2006).

In addition, Hoerling. (2010) found that the major features of regional trend in annual precipitation during 1977-2006 are consistent with an atmospheric response to observed SST variability. They also proposed that the relationships between SST and rainfall changes are not generally symptomatic of human-induced emissions of greenhouse gases. In other researches, the India rainfall showed an obvious decreased trend from the 1950s (Ramesh and Goswami, 2007; Guhathakurta and Rajeevan, 2008; Hoerling., 2010). Several studies indicated that the decadal variations of Indian summer monsoon is strongly correlated with the relationships between El Niño-Southern Oscillation (ENSO) and IO climate (Xie., 2010b; Chowdary., 2012; Krishnamurthy and Goswami, 2000; Torrence and Webster, 1999; Kumar and L’Heureux, 2010). Cai. (2010) showed that the asymmetry in ENSO connection is an important factor, which can affect the regional rainfall at a multi-decadal time scale. Moreover, the higher frequency of El Niño Modoki (Ashok., 2007) in recent years is considered as a factor, which can not be ignored in investigating the climate changes (Weng., 2007). Recently, there are studies showing a slowdown of the Walker circulation in the past few decades (Deser., 2010; Vecchi and Soden, 2007; Tokinaga., 2012). However, an opposite conclusion indicated an enhanced east-west Walker circulation (Wang., 2012; Luo., 2012) in recent 2–3 decades. How the rainfall patterns changes due to the convective modulation is still an open question.

Kripalani. (2007) evaluated the coupled climate model in simulating the variability of south Asian summer monsoon rainfall under the Fourth Assessment Report (AR4) of the Intergovernmental Panel on Climate Change(IPCC). Their results revealed a significant increase in mean monsoon rainfall and a possible extension of the monsoon period, although the projected summer monsoon circulation appeared to weaken. Nevertheless, there are still discrepancies in numerical models to simulate the observed rainfall on a regional scale (Gadgil and Sajani, 1998; Kang., 2002; Waliser., 2003; Wang., 2004). Using the high-resolution atmospheric general circulation models (AGCMs) and the CMIP5 dataset, Hsu. (2012, 2013) concluded that the global monsoon precipitation (GMP) has an increasing trend over the past three decads and in the future warmer climate. The intensity of GMP is mainly attributed to the increases in moisture convergence and surface evaporation under global warming. They also found that although the water vapor plays a positive role in thermodynamics, it is offset to a certain extent by a weakened monsoon circulation.

The present study aims to evaluate the simulation of the NIO rainfall changes in the past decades based on the CMIP5 Multi-Model ensemble (MME). We first test whether CMIP5 models can reproduce the rainfall changes similar to the observations, and then verify the ability of CMIP5 models to simulate the long-term variability of the tropical atmospheric circulation.

The rest of the paper is organized as follows. Section 2 describes CMIP5 datasets. Section 3 assesses the recent 30 years changes of the NIO rainfall based on CMIP5 models. Section 4 explores the land monsoon rainfall trend in the near past 60 years. Section 5 is a summary and discussion.

2 Data

The CMIP5 multi-model products of World Climate Research Programme (WCRP), collected and archived by the Working Group on Coupled Models (WGCM) Climate simulation panel and the Program for Climate Model Diagnosis and Inter-comparison (PCMDI), are used in this study. To compare with observations, the analysis focuses on the climate of the 20th century historical runs, which are forced with the observed history of greenhouse gas (GHG) concentrations, solar radiation, and other climate forcing. At the present stage, we have obtained the necessary data from 20 models of them. Table 1 lists the names, institutions, and resolution of the models. Monthly mean outputs, including SST, 10m wind, and rainfall, are used. The scenario experiments span from 1870 to 2005, but this study focuses on the period from 1950 to 2005.

Table 1 The WCRP CMIP5 models used in this study. In the following figures, the ensemble mean of all models is marked MME

The rainfall data is the GPCP version 2.2 monthlycombined precipitation, with a resolution of 2.5˚×2.5˚ for the period 1979-2010 (Adler., 2003). For the land rainfall, we use the University of Delaware precipitation for the period from 1900 to 2008 (http: //www.esrl.noaa.gov/psd/data/gridded/data.UD-el_AirT_Precip.html). We also use ERA-40 rainfall from 1958 to 2001 for supplemental reference (Uppala., 2005). The extended reconstructed SST (ERSST) by the National Oceanic and Atmospheric Administration (NOAA) is available on a 2˚ grid from 1860 to 2011 (Smith., 2008). The Wave- and Anemometer-based Sea-surface Wind (WASWind) data sets are used (Tokinaga and Xie, 2011) with a resolution of 4˚×4˚ for the period 1950-2009. For sea level pressure (SLP), we use the Hadley Centre’s mean SLP data sets version 2 (HadSLP2) with a resolution of 2.5˚×2.5˚ from 1850 to 2004 (Allan and Ansell, 2006). We calculate the linear trend of the summer (June-September) rainfall and surface wind in the NIO for the period of 1979-2005, and obtain the summer rainfall trend of Indian region for the period of 1950-2005. The NIO is defined as the region of 50˚E-100˚E, 0˚-20˚N.

3 Variability of the NIO Rainfall During 1979-2005

To evaluate the recent changes of rainfall in the NIO based on the 20 CMIP5 models, the time series of sum- mer (June-September) rainfall anomalies from 1950 to 2005 are calculated and shown in Fig.1. Meanwhile, the Models’ results are compared to the GPCP and ERA-40 rainfall data. Note that the GPCP data from 1979 to 2005, and the ERA-40 data spans from 1958 to 2001. In the observations, the summer rainfall in the NIO shows an increased trend from 1979 to 2005. The ERA-40 rainfall, although different in interannual variability, indicates a significant upward trend in recent 22 years. We focus on the long term changes in this work. 14 out of 20 CMIP5 models reproduce a similar rainfall trend to what in observations, and only 2 models display decrease trend from 1979 to 2005, which is opposite to the GPCP rainfall trend. In addition, the rainfall trend in 4 models (IPSL-CM5A-LR; IPSL-CM5A-MR; MIROC5; Nor-ESM1-M) are not significant. In Fig.1, the average of all models is marked as MME (Multi-Model ensemble) composite. Although the magnitude is small, the MME indicates that most models simulate the prominent feature of rainfall trend in the NIO successfully.

Fig.1 Interannual variability and linear trend of the summer (June-September) precipitation (unit: mmd-1) in theNorthern Indian Ocean.

Fig.2 shows the spatial distribution of linear trend in regional rainfall and surface wind. Owing to limited satel- lite observations, the rainfall trend is calculated for 1979-2005. The MME trend is calculated for the last 27 year period over both land and oceans. In the observations (Fig.2a), the enhanced rainfall occurs in most parts of the oceanic monsoon regions: the equatorial IO, the Arabian Sea, and the northern Bay of Bangle (BOB). It can explain the upward trend of mean summer rainfall in the NIO (Fig.1). Besides, an increased rainfall occurs in the southeast of the Asian continent and part of the Pacific, mainly in the northwestern, northeastern and southern tropical Pacific. However, a decrease rainfall trend shows at the equatorial zonal regions of Pacific. The drying condition occurs over the southern Indian Peninsula, BOB, and central and eastern equatorial Pacific. Note that the increased rainfall trend is more remarkable in western Pacific than in eastern Pacific. The simulated rainfall trends are consistent with previous studies. (Wang., 2012; Luo., 2012; Luffman., 2010; Hoerling., 2010).

Fig.2 Spatial distributions of the linear trends in summer (June-September) precipitation (unit: mmd-1 per 27year) and surface wind (unit: ms-1 per 27year) over 1979-2005 for (a) observations and (b) Multi-Model ensemble.

The MME reproducesan upward rainfall trend in the NIO, similar to observations. However, the increased rain- fallcovers almost all land of the South Asian region and the central and eastern equatorial Pacific, different from observations. Likewise, an increased rainfall in Somali coast is opposite to the observed drought. Specifically half of the 20 models (ACCESS1.0; BCC-CSM1.1; Can- ESM2; FGOALS-S2; GISS-E2-H; HadGEM2-ES; IPSL-CM5A-LR; MIROC-ESM; MPI-ESM-LR; MRI-CGCM3) can reproduce resemble rainfall patterns in the NIO and tropical Pacific, and 7 models (CanESM2; CSIRO-MK3.6; GFDL-ESM2G; GFDL-ESM2M; MIROC-ESM; MPI-ESM-LR; NorESM1-M) reconstruct similar drying condition in the Southern Indian peninsula (Figure not shown).

Although the patterns of rainfall trend in the MME are similar to the observations, the background circulation is different (Fig.2). Observations indicate a weakening of the summer monsoon circulation in the NIO, with strong easterly in the BOB and northeasterly wind in the Arabian Sea. On the contrary, there is strengthened southeasterly in the southeastern IO and the airflow cross the equator. In the NIO, southwesterly or southerly wind blows over the Arabian Sea and the BOB. The surface wind trend in MME and observations suggest that the similar rainfall trend patterns may be due to different mechanisms associated with the IO. In observations and the MME, the easterlies over equatorial IO are against the prevailing monsoonal wind, and can help to strengthen the airflow convergence in the equatorial IO, resulting in the increase of rainfall (Luffman., 2010). A southwesterly or southerly wind may also be able to bring more water vapor to NIO to intensify rainfall over the Arabian Sea and the BOB in the MME. However, the observed winds weak in strength and without obvious airflow convergence over the Arabian Sea, the BOB and western North Pacific indicate that the the water vapor transport or airflow convergence is not a solid factor contributing to the increased rainfall. Recent researches (Hsu., 2012, 2013) using diagnosis of a column-integrated moisture budget reveal that the increase of surface evaporation plays a positive role in global monsoon rainfall under global warming, which may supply a possible interpretation for the rainfall anomaly.

In the central and eastern Pacific, wind appears to be an important mechanism for rainfall variations (Xie., 2010a). As in the southeastern IO, the strong south-easterlies in the southeastern Pacific cross the equator and converge in the central and eastern equatorial Pacific, leading to an increased rainfall over the region in the MME. The absence of observed anomalous southeasterly and divergence results in rainfall decrease over the central and eastern equatorial Pacific. In addition, the consistency of increasing rainfall with surface convergence exists in the eastern North Pacific.

Previous studies indicated that the spatial patterns of SST warming plays a key role in determining rainfall changes, based on a ‘warmer-get-wetter’ pattern (Xie., 2010; Luffman., 2010). As illustrated in Fig.3, 15 models and the observations confirm the rainfall changes following the above mechanism. 12 out of the 15 models indicate that the rainfall increase and the local SST warms above the tropical average. The rest 3 models show a reduced rainfall because the local warming falls below the tropical average.

4 Variability of the Land Monsoon Rainfall from 1950-2005

Many studies (.., Gadgil and Sajani, 1998; Kang., 2002; Waliser., 2003; Wang., 2004) suggested that there are significant shortages in reproducing the mean monsoon climate and the difficulty to capture major features of the Asian summer monsoon in the present GCMs. Nevertheless, some work indicated that using coupled ocean-atmosphere models, instead of atmospheric models only, can improve the simulation and prediction of the Indian monsoon significantly (Kumar., 2005). In most of the Indian region, the summer monsoon rainfall accounts for nearly 80% of the annual rainfall (Ramesh and Goswami, 2007). The University of Delaware rainfall data is used here to evaluate the long-term changes of simulated land rainfall of India in CMIP5 models.

The climatology is prepared for the monsoon season (June-September) in the last 56 model years (correspon- ding to the period 1950-2005) and compared with the observed climatology. The spatial distribution and ampli- tude of monsoon climatological seasonal rainfall in Indian monsoon region is shown in Fig.4a. From the MME we can find that the models capture well the major features of the observed climatology rainfall over India and its neighboring region. Especially in the west coast of the Indian peninsula and along the northeast coast of the BOB, the observed rainfall-rich region caused by to- pographic lifting (Xie., 2006) is reproduced precisely by the MME despite some discrepancy in amplitude. The proportion of summer rainfall in total annual rainfall ranges from 55% to 75% (Fig.4b), all being smaller than observations. The simulated monthly rainfall by the 8 models (BCC-CSM1.1; CanESM2; CNRM-CM5; FGOALS-S2; GFDL-ESM2G; GFDL-ESM2M; INMC M4; MPI-ESM-LR; NorESM1-M) is close to observations,.., nearly 30mm per day. Although the rainfall by some models is small in amplitude, the simulated results of mean monsoon climate by the 20 CMIP5 models are encouraging.

Fig.5 Spatial distribution of the linear trends in land precipitation (unit: mmd-1 per 56year) and surface wind (unit: ms-1 per 56year) in summer (June-September).

A few recent studies suggested that the Indian monsoon undergoes abrupt shifts and weakens in the past years (.., Ramesh and Goswami, 2007; Guhathakurta and Rajeevan, 2008; Hoerling., 2010; Cook and Vizy, 2010). For the last 5-6 decades, the spatial and temporal extent of continental monsoon rainfall is reducing, even though at larger scale the total mean rainfall may even increase (Ramesh and Goswami, 2007). Fig.5 is a spatial distribution of linear trend in summer (June-September) rainfall and surface wind. In observations, the rainfall trend in India shows an opposite polarity, with rainfall decreased over most part of central and Northern India and increased over the south corner of Indian Peninsula. The Indian summer monsoon is weakening because of lacking obvious cross-equatorial flow and northerly in the Arabian Sea and easterly in the BOB respectively, which blow against the prevailing southwesterly. The MME reproduces the rainfall trend patterns in the southern Indian Peninsula, but an increased trend contrary to observations in the central and northern India. The surface wind in the MME also indicates a reduced Indian summer monsoon, similar to observations and consistent with the weakening rainfall patterns. However, some models (.. INMCM4) indicate a strengthened monsoon. We note that overestimated southeasterlies which prevail over the southeastern IO in 5 models (ACCESSM1.0; CanESM2; HadGEM2-ES; MPI-ESM-LR; MRI-CGCM3); extend southerly wind to the South Asian continent and make the land wet. In some studies the observed rainfall pattern is attributed to the resemblance to the lagged response to ENSO. It is associated with the SST pattern that characterizes the positive anomalies over the NIO (Mishra., 2012). The rainfall pattern may also relate to the land-sea temperature gradient, through enhancing the trades from ocean to land to increase the rainfall. Similar results had been reproduced in previous studies (Meehl and Washington, 1993; Hu., 2000; Kumar., 1999; Meehl., 2003; Ueda., 2006), which explored the potential impact of anthropogenic forcing on the Indian monsoon. The anomalous flows in the MME of Fig.2 suggest discrepancy in simulating the Indian monsoon trend with particular models.

5 Summary

This study evaluates the simulation of the NIO summer rainfall changes with the CMIP5. The IO shows an upward summer rainfall trend for the period 1979–2005, especially a remarkable increased rainfall in the equatorial IO (Figs.1-2). Most of the models show an upward trend as do the observations. The MME of the CMIP5 models has a similar spatial distribution of rainfall trend in the NIO and an easterly surface wind trend over the equatorial IO. Both in observations and the MME, the increased rainfall trend patterns in the equatorial IO are controlled by the SST warming, which confirms the mechanism of ‘warmer-get-wetter’ theory (Xie., 2010a). The warming leads to a strengthened convergence over the equatorial IO. In observations, however, neither the horizontal advection nor the water convergence can explain the positive rainfall anomaly in the Arabian Sea, the BOB and the west Northern Pacific. For a longer period 1950-2005, the land monsoon rainfall trend patterns are compared. The rainfall shows a decreasing trend pattern over major part of India and a increasing one in the south conner of Indian Peninsula in observations. The MME simulates an opposite rainfall trend in the central and northern India. Half of the models have a similar pattern to that in observations. The surface wind trends in observations and the MME both indicate weakened monsoon circulation. The difference in north India rainfall, which increases in the 1979-2005 period and decreases in the 1950-2005 period, suggests a multi-decadal variation in rainfall variation in the NIO.

Acknowledgements

This work was supported by the National Basic Research Program of China (2012CB955603, 2010CB- 950302), and the Chinese Academy of Sciences (XDA 05090404, LTOZZ1202).

Adler, R. F., Huffman G. J., Chang, A., Ferraro, R., Xie, P. P., Janowiak, J., Rudolf, B., Schneider, U., Curtis, S., Bolvin, D., Gruber, A., Susskind, J., Arkin, P., and Nelkin, E., 2003. The version 2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present)., 4: 1147-1167.

Allan, R., and Ansell, T., 2006. A new globally complete monthly historical gridded mean sea level pressure dataset (HadSLP2): 1850–2004., 19: 5816-5842.

Alory, G., Wijffels, S., and Meyers, G., 2007. Observed temperature trends in the Indian Ocean over 1960–1999 and associated mechanisms., 34 (2), L02606, DOI: 10.1029/2006GL028044.

Ashok, K., Behera, S. K., Rao, S. A., Weng, H. Y., and Yamagata, T., 2007. El Niño Modoki and its possible teleconnection., 112: C11007, 27pp, DOI:10.1029/2006JC003798.

Cai, W. J., van Rensch, P., Cowan, T., and Sullivan, A., 2010. Asymmetry in ENSO teleconnection with regional rainfall, its multidecadal variability, and impact., 23 (18): 4944-4955.

Chowdary, J. S., Xie, S. P., Tokinaga, H., Okumura, Y. M., Kubota, H., Johnson, N. C., and Zheng, X. T., 2012. Inter-decadal variations in ENSO teleconnection to the Indo-western Pacific for 1870–2007., 25: 1722-744.

Cook, K. H., and Vizy, E. K., 2010. Hydrodynamics of the Caribbean low-level jet and its relationship to precipitation., 23 (6): 1477-1494.

Deser, C., Phillips, A. S., and Alexander, M. A., 2010. Twentieth century tropical sea surface temperature trends revisited., 37, L10701, DOI: 10.1029/2010GL043321.

Du, Y., and Xie, S. P., 2008. Role of atmospheric adjustments in the tropical Indian Ocean warming during the 20th century in climate models., 35, L08712.

Gadgil, S., and Sajani, S., 1998. Monsoon precipitation in the AMIP runs., 14: 659-689.

Guhathakurta, P., and Rajeevan, M., 2008. Trends in the rainfall pattern over India., 28: 1453-1469.

Hoerling, M., Eischeid, J., and Perlwitz, J., 2010. Regional precipitation trends: distinguishing natural variability from anthropogenic forcing., 23 (8): 2131- 2145.

Hsu, P., Li, T., Luo, J. J., Murakami, H., Kitoh, A., and Zhao, M., 2012. Increase of global monsoon area and precipitation under global warming: A robust signal?, 39, L06701, DOI:10.1029/2012GL051037.

Hsu, P., Li, T., Murakami, H., and Kitoh, A., 2013. Future change of the global monsoon revealed from 19 CMIP5 models., 118: 1-14.

Hu, Z. Z., Latif, M., Roeckner, E., and Bengtsson, L., 2000. Intensified Asian summer monsoon and its variability in a coupled model forced by increasing greenhouse gas concentrations., 27: 2681-2684.

Johnson, N. C., and Xie, S.-P., 2010. Changes in the sea surface temperature threshold for tropical convection., 3 (12): 842-845.

Kang, I. S., Jin, K., Wang, B., Lau, K. M., Shukla, J., Krishnamurthy, V., Schubert, S. D., Wailser, D. E., Stern, W. F., Kitoh, A., Meehl, G. A., Kanamitsu, M., Galin, V. Y., Satyan, V., Park, C. K., and Liu, Y., 2002. Intercomparison of the climatological variations of Asian summer monsoon precipitation simulated by 10 GCMs., 19: 383-395.

Kitoh, A., Yukimoto, S., Noda, A., and Motoi, T., 1997. Simulated changes in the Asian summer monsoon at times of increased atmospheric CO., 75: 1019-1031.

Kripalani, R. H., Oh, J. H., Kulkarni, A., Sabade, S. S., and Chaudhari, H. S., 2007. South Asian summer monsoon precipitation variability: Coupled climate model simulations and projections under IPCC AR4., 90: 133-159.

Krishnamurthy, V., and Goswami, B. N., 2000. Indian monsoon-ENSO relationship on interdecadal timescale., 13 (3): 579-595.

Kumar, A., Jha, B., and L’Heureux, M., 2010. Are tropical SST trends changing the global teleconnection during La Niña?, 37, L12702, DOI: 10.1029/ 2010GL043394.

Kumar, K. K., Hoerling, M., and Rajagopalan, B., 2005. Advancing dynamical prediction of Indian monsoon rainfall., 32, L08704, DOI: 10.1029/ 2004GL021979.

Kumar, K. K., Rajagopalan, B., and Cane, M. A., 1999. On the weakening relationship between Indian monsoon and ENSO., 284: 2156-2159.

Levermann, A., Schewe, J., Petoukhov, V., and Held, H., 2009. Basic mechanism for abrupt monsoon transitions., 106: 20572-20577.

Luffman, J. J., Taschetto, A. S., and England, M. H., 2010. Global and regional climate response to late twentieth-century warming over the Indian Ocean., 23: 1660-1674.

Luo, J. J., Sasakia, W., and Masumotoa, Y., 2012. Indian Ocean warming modulates Pacific climate change.,109 (46): 18701-18706.

Meehl, G. A., and Washington, W. M., 1993. South Asian summer monsoon variability in a model with doubled atmospheric carbon dioxide concentration., 260: 1101- 1104.

Meehl, G. A., Arblaster, J., and Loschnigg, J., 2003. Coupled ocean-atmosphere dynamical processes in the tropical Indian and Pacific Oceans and the TBO., 16 (13): 2138-2158.

Mishra, V., Smoliak, B. V., Lettenmaier, D. P., and Wallace, J. M., 2012. A prominent pattern of year-to-year variability in Indian Summer Monsoon Rainfall., 109 (19): 7213-7217.

Ramesh, K. V., and Goswami, P., 2007. The shrinking Indian summer monsoon. CSIR Centre for Mathematical Modelling and Computer Simulation.0709.

Smith, T. M., Reynolds, R. W., Peterson, T. C., and Lawrimore, J., 2008. Improvements to NOAA’s historical merged land-ocean surface temperature analysis (1880–2006)., 21: 2283-2296.

Tokinaga, H., and Xie, S. P., 2011. Wave and Anemometer-based Sea Surface Wind (WASWind) for climate change analysis., 24: 267-285.

Tokinaga, H., Xie, S. P., Deser, C., Kosaka, Y., and Okumura, Y. M., 2012. Slowdown of the Walker circulation driven by tropical Indo-Pacific warming., DOI: 10.1038/nature11576.

Tokinaga, H., Xie, S. P., Timmermann, A., McGregor, S., Ogata, T., Kubota, H., and Okumura, Y. M., 2012. Regional patterns of tropical Indo-Pacific climate change: evidence of the walker circulation weakening., 25 (5): 1689-1710, DOI: 10.1175/JCLI-D-11-00263.1

Torrence, C., and Webster, P. J., 1999. Interdecadal Changes in the ENSO-Monsoon System., 12: 2679- 2690.

Ueda, H., Iwai, A., Kuwako, K., and Hori, M. E., 2006. Impact of anthropogenic forcing on the Asian summer monsoon as simulated by eight GCMs., 33 (6), DOI: 10.1029/2005GL025336.

Uppala, S. M., Kallberg, P. W. , Simmons, A. J., Andrae, U., Bechtold, V. D., Fiorino, M., Gibson, J. K., Haseler, J., Hernandez, A., Kelly, G. A., Li, X., Onogi, K., Saarinen, S., Sokka, N., Allan, R. P., Andersson, E., Arpe, K., Balmaseda, M. A., Beljaars, A. C. M., Van De Berg, L., Bidlot, J., Bormann, N., Caires, S., Chevallier, F., Dethof, A., Dragosavac, M., Fisher, M., Fuentes, M., Hagemann, S., Holm, E., Hoskins, B. J., Isaksen, L., Janssen, P. A. E. M., Jenne, R., McNally, A. P., Mahfouf, J. F., Morcrette, J. J., Rayner, N. A., Saunders, R. W., Simon, P., Sterl, A., Trenberth, K. E., Untch, A., Vasiljevic, D., Viterbo, P., and Woollen, J., 2005. The ERA-40 re-analysis.. 131 (612): 2961-3012, DOI: 10.1256/qj. 04.176.

Vecchi, G. A., and Soden, B. J. 2007. Global warming and the weakening of the tropical circulation., 20: 4316-4340.

Waliser, D. E., Jin, K., Kang, I. S., Stern, W. F., Schubert, S. D., Wu, M. L. C., Lau, K. M., Lee, M. I., Krishnamurthy, V., Kitoh, A., Meehl, G. A., Galin, V. Y., Satyan, V., Mandke, S. K., Wu, G., Liu, Y., and Park, C. K., 2003. AGCM simulations of intra-seasonal variability associated with the Asian summer monsoon., 21: 423-446.

Wang, B., Kang, I. S., and Lee, J. Y., 2004. Ensemble simulation of Asian–Australian monsoon variability by 11 AGCMs.,17: 699-710.

Wang, B., Liu, J., Kim, H. J., Webster, P. J., and Yim, S. Y., 2012. Recent change of the global monsoon precipitation (1979–2008)., 39: 1123-1135.

Weng, H., Ashok, K., Behera, S. K., Rao, S. A., and Yamagata, T., 2007. Impacts of recent El Niño Modoki on dry/wet conditions in the Pacific rim during boreal summer., DOI: 10.1007/ s00382-007-0234-0.

Xie, S. P., Deser, C., Vecchi, G. A., Ma, J., Teng, H., and Wittenberg, A. T., 2010a. Global warming pattern formation: Sea surface temperature and rainfall., 23: 966-986.

Xie, S. P., Du, Y., Huang, G., Zheng, X. T., Tokinaga, H., Hu, K., and Liu, Q., 2010b. Decadal shift in El Niño influences on Indo-western Pacific and East Asian climate in the 1970s., 23: 3352-3368.

Xie, S. P., Xu, H., Saji, N. H., Wang, Y. Q., and Liu, W. T., 2006. Role of narrow mountains in large-scale organization of Asian monsoon convection., 19: 3420- 3429.

Zickfeld, K., Knopf, B., Petoukhov, V., and Schellnhuber, H., 2005. Is the Indian summer monsoon stable against global change?, 32, L15707, DOI: 10.1029/2005GL022771.

(Edited by Xie Jun)

10.1007/s11802-013-2269-7

ISSN 1672-5182, 2013 12 (2): 201-208

. Tel: 0086-20-89023180 E-mail:duyan@scsio.ac.cn

(January 5, 2013; revised February 20, 2013; accepted March 27, 2013)

© Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2013