CHEN Xiong, LI Chongyin, 2), *, and TAN Yanke

?

The Influence of El Ni?o on MJO over the Equatorial Pacific

CHEN Xiong1), LI Chongyin1), 2), *, and TAN Yanke1)

1),,211101,2),,,100029,

In this paper, the influence of El Ni?o event on the Madden-Julian Oscillation (MJO) over the equatorial Pacific is studied by using reanalysis data and relevant numerical simulation results. It is clearly shown that El Ni?o can reduce the intensity of MJO. The kinetic energy of MJO over the equatorial Pacific is stronger before the occurrence of the El Ni?o event, but it is reduced rapidly after El Ni?o event outbreak, and the weakened MJO even can continue to the next summer. The convection over the central-western Pacific is weakened in El Ni?o winter. The positive anomalous OLR over the central-western Pacific has opposite variation in El Ni?o winter comparing to the non-ENSO cases. The vertical structure of MJO also affected by El Ni?o event, so the opposite direction features of the geopotential height and the zonal wind in upper and lower level troposphere for the MJO are not remarkable in the El Ni?o winter and tend to be barotropic features. El Ni?o event also has an influence on the eastward propa- gation of the MJO too. During El Ni?o winter, the eastward propagation of the MJO is not so regular and unanimous and there exists some eastward propagation, which is faster than that in non-ENSO case. Dynamic analyses suggest that positive SSTA (El Ni?o case) affects the atmospheric thickness over the equatorial Pacific and then the excited atmospheric wave-CISK mode is weakened, so that the intensity of MJO is reduced; the combining of the barotropic unstable mode in the atmosphere excited by external forcing (SSTA) and the original MJO may be an important reason for the MJO vertical structure tending to be barotropic during the El Ni?o.

El Ni?o; MJO; intensity; vertical structure; eastward propagation

1 Introduction

El Ni?o is an anomalous rising event of the sea surface temperature (SST) in the equatorial central-eastern Pacific caused by large-scale atmosphere-ocean interaction (Bjerknes, 1966; Rasmussen and Carpenter, 1982), which would lead to evident and wide anomalies of the atmospheric circulation and climate around the world (Charles, 1997; Serrano, 2011). It has become an important scientific problem since 1980s (Rasmussen and Wallace, 1983; Jin, 1997; Li and Mu, 2000; Chen, 2004). In general, the averaged period of El Ni?o cycle is about 3–4 years, and the El Ni?o has been regarded as an important factor for the interannual variation of climate and its prediction. (Bhalme and Jadhav, 1984; Gadgil, 1988; Zhang, 1996; Tao and Zhang, 1998).

Another important scientific problem in the climate variability is the intraseasonal oscillation (ISO) in the atmosphere, especially the Madden-Julian Oscillation (MJO), which is first found by Madden and Julian (1971, 1972). A lot of studies have shown that the MJO is very obvious in the equatorial atmosphere and relates closely to monthly/seasonal variations of climate (Murakami, 1984; Lau and Chan, 1985; Knutson and Weickmann, 1987; Li, 1991; Madden and Julian, 1994; Matthews, 2008; Lin and Brunet, 2011; Zhang, 2013). The activity of the MJO over the tropical, especially over the eastern India Ocean and western Pacific, is one of the most influential factors for the monsoon rainfall in summer and snow in winter over South Asia and East Asia (Pai, 2011; Jia and Liang, 2011; Jia, 2011). It can also modify the weather in North America (Riddle, 2013; Moon, 2012).

Both El Ni?o and MJO are important factors to cause interannual variation and monthly/seasonal climate variations. What is the relationship between these two kinds of the climate mode with different time-scale? Lau and Peng (1987a) proved an inference based on energy transfer from MJO scale to interannual scale: the MJO in the trop- ical atmosphere could excite the El Ni?o event. Some investigations on the excited mechanism of the El Ni?o event have indicated that frequent activities of stronger East Asian trough in winter (stronger anomalies of the East Asian winter monsoon) will enhance the cumulus convections and reduce the trade winds in the equatorial central-western Pacific area (Li, 1989; Yu and Rienecker, 1998; Li and Mu, 2000); The joint effects of stronger MJO caused by the strong cumulus convection and anom- alous oceanic Kelvin waves caused by the weakened trade winds over the equatorial central-western Pacific will lead to the occurrence of El Ni?o event (Li and Liao, 1998; Seo and Yan, 2005; Mcphaden, 2006). In other words, the important effect of MJO in the equatorial atmosphere on El Ni?o event has been well studied. In this paper the influence of El Ni?o on MJO will be investigated by using diagnose analysis and numerical simulation results. Many studies have illustrated that the extratropical influence is important for the initiation of MJO (Ray, 2010; Ray and Zhang, 2010; Wang, 2012), and the average state of ocean is not critical for its initiation; but the displacement of SST also has an effect on the MJO’s intensity, propagation and predictability (Pegion and Kirtman, 2008a, 2008b; Kim, 2010). Seiki(2010) indicated that El Ni?o could lead to an anomalous extratropical high over the north Pacific during its developing period, and the northeasterly trade surges originating from the high would moisturize the atmosphere over the western north Pacific by convergence, which will lead to a strong convection in these areas, and the intensity of MJO to the north of equator is enhanced consequently. But whether and how El Ni?o influences MJO during its mature phase still needs to be further investigated.

The present analysis documents the influence of El Ni?o on MJO over the equatorial Pacific, including the influence on the intensity of MJO (Section 3), MJO’s vertical structure (Section 4) and its eastward propagation (Section 5). A simple dynamic analysis and the conclusion are given in Section 6 and Section 7 respectively.

2 The Data Used

The atmospheric data used in this paper are the daily mean reanalysis data of National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) (Kalany, 1996), including daily wind and geopotential height covering the period of 1 January 1948 to 31 December 2011. This dataset has a 2.5?×2.5? horizontal resolution and with 17 vertical pressure levels from 1000hPa to 10hPa. The daily outgoing longwave radiation (OLR) data with a 2.5?×2.5? horizontal resolution are from the National Oceanic and Atmospheric Administration (NOAA) (Liebmann and Smith, 1996) during 1 January 1979 to 31 December 2011. The monthly mean SST data are from Hadley Center Sea Ice and Sea Surface Temperature datasets (HadISST1) of the UK Meteorological Office (Rayner, 2003) with a horizontal resolution of 1?×1?. Although the SST datasets are available from 1870 to the present, we choose the period of 1950–2011 for our analyses.

To isolate the MJO, the atmospheric data have been 30–60 day band-pass filtered with a 200 point Lanczos bandpass filter (Duchon, 1979).

3 Influence of El Ni?o on MJO Intensity over the Equatorial Pacific

Using ECMWF grid data (1981–1988) Li and Li (1998) not only found the obvious enhancement of MJO prior to El Ni?o event, but also revealed the reduction of MJO during the mature phase of El Ni?o, which means the exciting effect of MJO in the tropical atmosphere (especially in the equatorial central-western Pacific) on the El Ni?o event and the reducing effect of El Ni?o on the MJO. Based on the longer NCEP dataset (1950–2011), the temporal variation of the normalized SST in Ni?o 3.4 area (5?N–5?S, 170?W–120?W) and the normalized kinetic energy of MJO at 200hPa over the equatorial Pacific region (10?N–10?S, 130?E–120?W) in winter (December–February) from 1950 to 2010 are shown in Fig.1. It is very clear that the MJO kinetic energy over the equatorial Pacific is weaker during El Ni?o winter, such as 1972/73, 1982/83, 1997/98 winters. The correlation coefficient between them is ?0.46 (above 0.001 confidence level), indicating that the higher SST in Ni?o 3.4 region corresponds to the weaker MJO over equatorial Pacific. Similar results are also found over the equatorial Pacific region at 500hPa and 850hPa levels (figures are not given). Therefore, it is suggest that El Ni?o has a reducing effect on the MJO intensity over the equatorial Pacific.

Fig.1 The temporal variation of the normalized SST in Ni?o 3.4 (red line) and kinetic energy of MJO (black line) at 200hPa over the equatorial Pacific region in winters from 1950 to 2010 (the time 1950 presenting the winter of 1950/51).

The outgoing longwave radiation (OLR) is an efficient physical quantity to describe atmospheric motion, particularly the cumulus convection in the tropics; it is also used to investigate the activity of MJO in the equatorial atmosphere (Lau and Chan, 1985; Murakami, 1986). In order to demonstrate the variability of OLR under different conditions, temporal variations of the composite monthly OLR anomalies over the equatorial central-western Pacific region (10?S–10?N，130?E–180?) for the El Ni?o cases (1982/83, 1986/87, 1991/92, 1994/95, 1997/98, 2002/03 and 2009/10) and the non-ENSO cases (1980/81, 1989/90, 1990/91, 1992/93, 1993/94, 2001/02 and 2003/04) are shown in Fig.2. It is clear that the OLR anomalies are all positive in El Ni?o winter and are negative in non-ENSO winter. Furthermore, the temporal variation of OLR anomalies during El Ni?o is converse to the analogue in non-ENSO. These results mean that the convections over the equatorial central-western Pacific is weaker than normal in El Ni?o winter. Since the convection over the equatorial central-western Pacific is closely associated with the MJO, and the convective heating feedback is an important mechanism to excite the MJO (Li, 1985; Lau and Peng, 1987b). The results also show that the El Ni?o event can affect the intensity of MJO activity over the equatorial Pacific, especially over the central-western Pacific.

Fig.2 Temporal variation of monthly OLR anomalies over the equatorial central-western Pacific region (10?S–10?N, 130?E–180?) for El Ni?o cases (red line) and non-ENSO cases (black line), the number ‘0’ and ‘1’ represent the year and the next year of El Ni?o occurrences respectively (units: Wm?2).

In order to further investigate the influence of El Ni?o on MJO, Fig.3 gives the monthly evolution of the composite MJO kinetic energy at 200hPa associated with the El Ni?o years (1975/58, 1965/66, 1972/73, 1982/83, 1986/87, 1991/92, 1997/98, 2002/03 and 2009/10) and non-ENSO years (1952/53, 1959/60, 1960/61, 1978/79, 1980/81, 1992/93, 1993/94, 2001/02 and 2003/04). The El Ni?o event has always occurred in June-July and matured in winter in general. It is very clear in Fig.3 that the MJO is stronger in the winter and spring (especially in the spring) before the occurrence of El Ni?o, but it reduces rapidly associated with the El Ni?o outbreak and the weakened MJO can continue to the next summer. Although the decrease of MJO can be regarded as the energy transform to and maintain the El Ni?o event, it also shows the influence of El Ni?o on MJO. The results showed here are consistent with that indicated in previous studies (Li and Li, 1998).

Fig.3 Evolutions of the MJO kinetic energy at 200ha over the equatorial Pacific (10?S–10?N, 130?E–120?W) associated with El Ni?o cases (red line) and non-ENSO cases (black line), the number ‘?1’, ‘0’ and ‘1’ representing the previous year, the year and the next year of El Ni?o occurrences respectively (units: m2s?2).

In the 1990s El Ni?o events occurred several times, how about the variation of MJO kinetic energy during this period? Fig.4 shows the temporal variation of the monthly MJO kinetic energy at 200hPa in the equatorial Pacific (10?S–10?N, 130?E–120?W) during January 1991–December 1997, the El Ni?o events being represented by black lines in the figure. It is clear that the kinetic energy of MJO over the equatorial Pacific is reduced after the occurrence of El Ni?o. The reduction of the MJO kinetic energy in July 1991, August 1993 and June 1997 corresponds to the El Ni?o events in 1991/92, 1993/94, 1997/98 winter respectively. The decrease of the MJO kinetic energy in March 1993 is associated with the ascending of the SST in Ni?o3.4 region in the spring of 1993. Though it did not develop to a real El Ni?o event, the reduction of the MJO kinetic energy is also very obvious in accompanying the ascending of the SST. And the higher kinetic energy of MJO in the summer of 1996 might be related with the La Ni?a event in 1995/96 winter, which needs to be further studies. These results suggest that El Ni?o is one of the major causes for the MJO weakening although there are stronger MJO in 1991/92 winter and in 1995 spring, which may be caused by another unknown factor.

Two numerical experiments in the CSIRO-GCM to investigate the influence of El Ni?o on MJO had been completed (Li and Smith, 1995). One experiment is the control experiment (CE) in which the SST used climatological data with temporal variation; the other is the anomalous experiment (AE) in which the SST used the observational data in 1983 (representing El Ni?o event). The longitudinal distribution of kinetic energy of MJO in the upper troposphere in winter obtained from CE and AE is shown in Fig.5. It is very clear that the kinetic energy of MJO in AE (dashed line) is smaller than that in CE (solid line) for either the zonal mean or the maximum. The numerical simulation in GCM shows that the MJO in the tropical atmosphere is reduced under the El Ni?o condition. Therefore, the data analyses and numerical simulation all show that the occurrence of El Ni?o will reduce the intensity of MJO and reveal the influence of El Ni?o on MJO.

Fig.4 Temporal variation of MJO kinetic energy at 200hPa in the equatorial Pacific (10?S–10?N, 130?E–120?W) region during 1991–1997 (units: m2s?2).

Fig.5 Longitudinal distribution of kinetic energy of the simulated MJO in GCM averaged between 11.1?S–11.1?N at 350hPa. Solid and dashed lines represent the results in the CE and the AE respectively (Li and Smith, 1995).

4 Influence of El Ni?o on the Vertical Structure of MJO

Studies (Murakami and Nakazawa, 1985; Li, 1991) have proved that the vertical structure of MJO in the tropical atmosphere shows a baroclinic feature and its wind field and pressure field change their signs from the lower to the upper troposphere. In Fig.6, the longitudinal distribution of the geopotential height for MJO at 850hPa and 200hPa averaged between 10?S–10?N in the El Ni?o cases (1982/83 winter, 1986/87 winter and 1997/98 winter) is shown respectively. We can see that the troughs (ridges) at 850hPa and 200hPa do not obviously change their signs from the lower to upper level. This means that the baroclinic feature of vertical structure is not clear for the MJO in the El Ni?o cases. Moreover, the zonal wind and pressure for the MJO in the El Ni?o case also show a vertical structure feature tending to be barotropic. For example, the longitudinal distribution of zonal wind for the MJO at 850hPa and 200hPa averaged between 10?S–10?N in 1982/83 winter, 1986/87 winter and 1997/98 winter is shown in Fig.7, respectively. Obviously, the opposite char- acteristics of zonal wind in upper and lower troposphere for the MJO are not remarkable in El Ni?o case. Therefore, it can be suggested that the El Ni?o has a remarkable influence on the vertical structure of MJO, which leads to the vertical structure of MJO tend to be barotropic.

Fig.6 Longitudinal distributions of geopotential height for the MJO at 850hPa and 200hPa averaged between 10?S–10?N in 1982/83 winter (a), 1986/87 winter (b) and 1997/98 winter (c) respectively. Sold and dashed lines represent the cases of 200hPa and 850hPa respectively (the winter is from December to next year’s February).

Fig.7 The same as in Fig.6, but for zonal wind.

The numerical simulated results in CSIRO-GCM also show the influence of El Ni?o on the vertical structure of MJO (figure is omitted). The simulated results show that the longitudinal distributions of pressure and temperature (that of geopotential height is similar) for the MJO have general baroclinic features in the control run (CE), which is similar to general MJO; but in the anomalous experiment (AE), the simulated MJO tends to be barotropic features, since the opposite direction feature is not clear. In other words, the El Ni?o condition (positive SSTA in the equatorial Easter Pacific) in the GCM will make the MJO tend to be barotropic structures.

5 Influence of El Ni?o on the Eastward Propagation of MJO

The eastward propagation along the equator is a major feature of MJO. But the propagation of MJO in El Ni?o case shows some differences with that in non-ENSO case. This suggests that the El Ni?o event has a certain influence on the eastward propagation of MJO. The time-longitude sections of averaged low frequency (30–60day) zonal wind (10?S–10?N) at 850hPa in non-ENSO winters (1952/53, 1960/61, 1989/90) are shown in Fig.8. They exhibit eastward propagation features in general: obvious unanimous eastward propagating along the equator and faster after crossing the date line, which is the same as the result of Madden and Julian (1994). But in El Ni?o winters there are some differences (Fig.9) with those in Fig.8. At first, eastward propagation is not so regular and unanimous, some eastward propagation in El Ni?o case is faster than that in non-ENSO cases, and even a few west- ward propagation of MJO in El Ni?o cases are found. Comparing Fig.8 and Fig.9, it is also shown that the MJO is weaker in El Ni?o cases than the one in non-ENSO cases. This result is the same as the preceding one, and also means that the results obtained earlier are authentic. The differences of propagation and intensity are more remarkable from December to March in El Ni?o winter than the analogue in non-ENSO winter. Although the analyses in this study just show the results in some cases, it can be suggested in a certain sense that the El Ni?o event has influence on eastward propagation of the MJO.

Fig.8 The time-longitude sections of averaged low frequency (30–60 days) zonal wind between 10?S–10?N for the MJO at 850hPa in non-ENSO winters (1952/53, 1960/61, 1989/90) (units: ms?1).

Fig.9 The same as in Fig.8, but for El Ni?o winters (1982/83, 1986/87, 1997/98).

6 A Simple Dynamic Analysis of El Ni?o Impacts on MJO

An exciting mechanism of tropical intraseasonal oscillation to El Ni?o event has been studied (Li and Liao, 1998); how about the influence of El Nino on the MJO? A simple dynamic analysis will be presented as follows.

Comparison analyses show that the distribution of tropical convection is changed prominently and the convection is weaker only in eastern India Ocean and central-western Pacific in El Ni?o winter. This means that the influence of El Ni?o on the MJO is not prominent through the feedback of cumulus convection over the whole equatorial Pacific.

The dynamical equations related to the MJO (or the ISO in the tropical atmosphere) can be written as follows (Li and Li, 1996):

, (2)

, (3)

where,andare the components of wind in,and-directions respectively;geopotential disturbance;the Brunt-Vaisala frequency.

The first term on the right hand side of Eq. (4) represents the cumulus convection heating (is heating profile) and the second term () is the effect of anomalous SST on the atmospheric motion.

Data analyses (Li and Li, 1996) show that the thickness between the 200hPa level and the 850hPa level () in El Ni?o event is greater than that in non-ENSO case. We do not simply think that the SSTA is only a heating factor in the atmosphere, but there just show the relationship between the atmospheric thickness and the SSTA. In the case without the heating, the greater (smaller) atmospheric thickness corresponds to lower (higher) tem- perature. Therefore, the temperature variation is opposite to the thickness representing the SSTA. In a 2-level model, we can write Eqs. (1)–(4) as follows:

, (6)

where ? is the height interval in the 2-level model.

Finally, the wave-CISK mode, which can drive the MJO (or the ISO in the tropical atmosphere), with the influence of SSTA can be obtained and its growth rate and frequency written as follows:

, (8)

whereSis a parameter related to anomalous SST.

From Eq. (7), we can find that the positive SSTA (+S) will reduce the wave-CISK mode; the Eq. (8) shows that the SSTA, whether positive or negative, will slow down the propagation of the wave-CISK mode. We suggest that the positive SSTA (El Ni?o case) will affect atmospheric thickness over the equatorial Pacific and then the excited atmospheric wave-CISK mode is weaker, so that the MJO is reduced. Obviously, dynamical analysis is consistent with the results in the data diagnose and numerical simulation. This further shows that El Ni?o event can reduce the intensity of MJO.

7 Concluding Remarks and Discussion

The above mentioned data analyses and numerical simulation clearly show that the influence of El Ni?o event on the MJO is very obvious: El Ni?o will reduce the intensity of MJO so that the MJO is weaker after El Ni?o occurrence; the occurrence of El Ni?o event will lead the vertical structure of the MJO tend to be barotropic; and the El Ni?o event has an influence on the eastward propagation of MJO too.

In general, the kinetic energy of atmospheric motion can be divided into three parts in time scale: synoptic system (<10d), low-frequency oscillation (major is ISO) and quasi-stationary wave (>90d). The enhancement of MJO (especially over the equatorial central-western Pacific region) is just an energy transformation from synoptic scale system (convective cloud cluster) to MJO, which has been studied as a dynamical mechanism of MJO in the tropical atmosphere (Li, 1985; Lau and Peng, 1987b). After the occurrence of El Ni?o event, the energy of low-frequency oscillation will transfer to the quasi-stationary perturbation and the energy of MJO is reduced. Therefore, the MJO is weaker during El Ni?o event. The detailed discussion in relation to this kind of the energy transformation will be given in another special study.

The influence of El Ni?o on MJO is not prominent through the feedback of cumulus convection. Generally, the positive SSTA (El Ni?o case) will affect the atmospheric thickness over the equatorial Pacific and then the excited atmospheric wave-CISK mode is weaker, so that the MJO is reduced. Obviously, dynamical analysis is consistent with the results in the data diagnose and numerical simulation.

In relation to the influence of El Ni?o on the vertical structure of MJO, which tends to be barotropic features, a brief discussion can be as follows: the SSTA can be regarded as an external forcing for the atmospheric motion, while some studies have shown that the atmospheric response to external forcing tends to produce barotropic unstable modes in the atmosphere (Hoskins, 1983). Therefore, we suggest that the combining of the barotropic unstable mode in the atmosphere excited by external forcing and the original MJO may be an important reason for the MJO tending to be barotropic structure during El Ni?o.

Acknowledgements

The authors thank the anonymous reviewers who provided constructive comments on an early version of the manuscript, which were helpful for improving the overall quality of the paper. This study was supported by the National ‘973’ Programme (No. 2013CB956203) and the National Natural Science Foundation of China (No. 41275086).

Bhalme, H. N., and Jadhav, S. K., 1984. The southern oscillation and its relation to the monsoon rainfall., 4: 509-520.

Bjerknes, J., 1966. A possible response of the atmospheric Hadley circulation to equatorial anomalies of ocean temperature., 18: 820-829.

Charles, C. D., Hunter, D. E., and Fairbanks, R. G., 1997. Interaction between the ENSO and the Asian monsoon in a coral record of tropical climate., 277: 925-928, DOI: 10.1126/science.277.5328.925.

Chen, D., Cane, M. A., Kaplan, A., Zebiak, S. E., and Huang. D., 2004. Predictability of El Ni?o over the past 148 years., 428: 733-736.

Duchon, C. E., 1979. Lanczos filtering in one and two dimensions., 18: 1016-1022.

Gadgil, S., 1988. Recent advances in monsoon research with particular reference to the Indian monsoon., 36: 193-204.

Hoskins, B., James, I., and White, G., 1983. The shase propagation and mean-flow interaction of large-scale weather systems., 40: 1595-1612.

Jia, X. L., and Liang, X. Y., 2011. Possible impacts of the MJO on the severe ice-snow weather in November of 2009 in China., 27 (5): 639-648 (in Chinese).

Jia, X. L., Chen, L. J., Ren, F. M., and Li, C. Y., 2011. Impacts of the MJO on winter rainfall and circulation in China., 28 (3): 521-533.

Jin, F. F., 1997. An equatorial ocean recharge paradigm for ENSO. Part I: conceptual model., 54: 811-829.

Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu, Y., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K. C., Ropelewski, C., Wang, J., Leetma, A., Reynolds, R., Jenne, R., and Joseph, D., 1996. The NCEP/NCAR 40-year reanalysis project., 77: 437-471.

Kim, H. M., Hoyos, D., Webster, J., and Kang, I. S., 2010. Ocean-atmosphere coupling and boreal winter MJO., 35: 771-784, DOI: 10.1007/s00382-009-0612-x.

Knutson, T. R., and Weickmann, K. M., 1987. 30-60 day atmospheric oscillation: Composite life cyclones of convection and circulation anomalies., 115: 1407-1436.

Lau, K. M., and Chan, P. H., 1985. Aspects of the 40-50 day oscillation during the northern winter as inferred from outgoing longwave radiation., 113: 1889-1909.

Lau, K. M., and Peng, L., 1987a. The 40-50 day oscillation and the El Ni?o/Southern Oscillation: A new perspective., 67: 533-534.

Lau, K. M., and Peng, L., 1987b. Origin of low frequency (intraseasonal) oscillation in the tropical atmosphere, Part I: The basic theory., 44: 950-972.

Li, C. Y., 1985. Actions of summer monsoon trough (ridges) and tropical cyclones over South Asia and the moving CISK mode., 28: 1197-1207.

Li, C. Y., 1989. Frequent activities of stronger aerotroughs in East Asia in wintertime and the occurrence of the El Ni?o event., 32: 976-985.

Li, C. Y., 1991.. China Meteorological Press, Beijing, 207pp.

Li, C. Y., and Li, G. L., 1996. A dynamical study of El Ni?o on intraseasonal oscillation in tropical atmosphere., 20: 159-168.

Li, C. Y., and Li, G. L., 1998. The activities of low-frequency waves in the tropical atmosphere and ENSO., 15: 193-203.

Li, C. Y., and Liao, Q. H., 1998. The exciting mechanism of tropical intraseasonal oscillation to El Ni?o event., 4: 113-121.

Li, C. Y., and Mu, M. Q., 2000. Relationship between East-Asian winter monsoon, warm pool situation and EVSO cycle., 45: 1448-1455.

Li, C. Y., and Smith, I., 1995. Numerical simulation of the tropical intraseasonal oscillation and the effect of warm SSTS., 9: 1-12.

Liebmann, B., and Smith, C. A., 1996. Description of a complete (interpolated) outgoing longwave radiation dataset., 77: 1275-1277.

Lin, H., and Brunet, G., 2011. Impact of the North Atlantic Oscillation on the forecast skill of the Madden-Julian Oscillation., 38, L02802, DOI: 10.1029/2010GL046131.

Madden, R. A., and Julian, P. R., 1971. Detection of a 40-50 day oscillation in the zonal wind in the tropical Pacific., 28: 702-708.

Madden, R. A., and Julian, P. R., 1972. Description of global scale circulation cells in the tropics with 40-50 day period., 29: 1109-1123.

Madden, R. A., and Julian, P. R., 1994. Observations of the 40–50-day tropical oscillation–A review., 122: 814-837.

Matthews, A. J., 2008. Primary and successive events in the Madden-Julian Oscillation., 134: 439-453, DOI: 10.1002/qj.224.

Mcphaden, M. J., Zhang, X. B., Hendon, H. H., and Wheeler, M. C., 2006. Larg scale dynamics and MJO forcing ENSO variability., 33, L16702, DOI: 10.1029/2006GL026786.

Moon, J. Y., Wang, B., and Ha, K. J., 2012. MJO modulation on 2009/10 winter snowstorms in the United States., 25: 978-991, DOI: 10.1175/JCLI-D-11-00033.1.

Murakami, T., and Nakazawa, M. T., 1985. Tropical 45 day oscil- lation during the 1979 northern hemisphere summer., 42: 1107-1122.

Murakami, T., Chen, L., Xie, A., and Shrestha, M., 1986. Eastward propagation of 30–60 day perturbations as revealed from outgoing longwave radiation data., 43: 961-971.

Murakami, T., Nakawa, M. T., and He, J. H., 1984. On the 40–50 day oscillation during the 1979 northern hemisphere summer., 62: 440-468.

Pai, D. S., Sreejith, O. P., and Hatwar, H. R., 2011. Impact of MJO on the intraseasonal variation of summer monsoon rainfall over India., 36: 41-55, DOI: 10.1007/s00382-009-0634-4.

Pegion, K., and Kirtman, B. P., 2008a. The impact of air–sea interactions on the simulation of tropical intraseasonal variability., 21: 6616-6635, DOI: 10.1175/2008JCLI2180.1.

Pegion, K., and Kirtman, B. P., 2008b. The impact of air–sea interactions on the predictability of the tropical intraseasonal oscillation., 21: 5870-5886, DOI: 10.1175/2008JCLI2209.1.

Rasmussen, E. M., and Carpenter, T. H., 1982. Variations in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation/El Ni?o., 110: 354-384.

Rasmussen, E. M., and Wallace, J. M., 1983. Meteorological aspects of El Ni?o/Southern Oscillation., 222: 1195-1202.

Ray, P., and Zhang, C. D., 2010. A case study of the mechanics of extratropical influence on the initiation of the Madden-Julian Oscillation., 67: 515-528, DOI: 10.1175/2009JAS3059.1.

Ray, P., Zhang, C. D., Dudhia, J., and Chen, S. S., 2009. A numerical case study on the initiation of the Madden-Julian Oscillation., 66: 310-331, DOI: 10.1175/2008JAS2701.1.

Rayner, N. A., Parker, D. E., Horton, E. B., Folland, C. K., Alexander, L. V., Rowell, D. P., Kent, E. C., and Kaplan, A., 2003. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century., 108 (D14): 4407, DOI: 10.1029/2002JD002670.

Riddle, E. E., Stoner, M. B., Johnson, N. C., L’Heureux, M. L., Collins, D. C., and Feldstein, S. B., 2013. The impact of the MJO on clusters of wintertime circulation anomalies over North American region., 40: 1749-1766, DOI: 10.1007/s00382-012-1493-y.

Seiki, A., Takayabu, Y. N., Yoneyama, K., and Shirooka, R., 2010. The impact of trade suges on the Madden-Julian Oscillation under different ENSO conditions., 6: 49-52, DOI: 10.2151/sola.2010-012.

Seo, K. H., and Yan, X., 2005. MJO-related oceanic Kelvin waves and the ENSO cycle: Astudy with NCEP Global Ocean data Assimilation System., 32, L07712, DOI: 10.1029/2005GL022511.

Serrano, J. G., Fonseca, B. R., Bladé, I., Gotor, P. Z., and Cámara, A., 2011. Rotational atmospheric circulation during North Atlantic-European winter: The influences of ENSO.. 37: 1727-1743, DOI: 10.1007/s00382-010-0968-y.

Tao, S. Y., and Zhang, Q. D., 1998. The responses of Asian winter/summer monsoon to the ENSO event., 22: 399-407 (in Chinese).

Wang, L., Kodera, K., and Chen, W., 2012. Observed triggering of tropical convection by a cold surge: Implications for MJO initiation., 138: 1740-1750, DOI: 10.1002/qj.1905.

Yu, L., and Rienecker, M. M., 1998. Evidence of an extratropical atmospheric influence during the onset of the 1997–98 El Ni?o., 25: 3537-3540.

Zhang, C. D., 2013. Madden-Julian Oscillation: Bridging weather and climate.,DOI: http://dx.doi.org/10.1175/BAMS-D-12-00026.1.2013.

Zhang, R. H., Sumi, A., and Kimoto, M., 1996. Impactof El Ni?o on the Asian monsoon: A diagnostic study of the ‘86/87’ and ‘91/92’ events., 74: 49-62.

(Edited by Xie Jun)

DOI 10.1007/s11802-015-2381-y

ISSN 1672-5182, 2015 14 (1): 1-8

? Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2015

(May 1, 2013; revised July 31, 2013; accepted July 11, 2014)

* Corresponding author. E-mail: lcy@lasg.iap.ac.cn