• Chu, P-S., 1988: Extratropical forcing and the burst of equatorial westerlies in the western Pacific: A synoptic study. J. Meteor. Soc. Japan, 66 , 549564.

    • Search Google Scholar
    • Export Citation
  • Dickinson, M., , and J. Molinari, 2002: Mixed Rossby–gravity waves and western Pacific tropical cyclogenesis. Part I: Synoptic evolution. J. Atmos. Sci., 59 , 21832196.

    • Search Google Scholar
    • Export Citation
  • Harrison, D. E., , and P. S. Schopf, 1984: Kelvin wave induced anomalous advection and the onset of surface warming in El Niño events. Mon. Wea. Rev., 112 , 923933.

    • Search Google Scholar
    • Export Citation
  • Holland, G. J., 1995: Scale interaction in the western Pacific monsoon. Meteor. Atmos. Phys., 56 , 5779.

  • Kiladis, G. N., , G. A. Meehl, , and K. M. Weickmann, 1994: Large-scale circulation associated with westerly wind bursts and deep convection over the western equatorial Pacific. J. Geophys. Res., 99 , 1852718544.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., , and K. M. Weickmann, 1987: 30–60 day atmospheric oscillations: Composite life cycles of convection and circulation anomalies. Mon. Wea. Rev., 115 , 14071436.

    • Search Google Scholar
    • Export Citation
  • Lander, M. A., 1994: An exploratory analysis of the relationship between tropical storm formation in the western North Pacific and ENSO. Mon. Wea. Rev., 122 , 636651.

    • Search Google Scholar
    • Export Citation
  • Lau, K-H., , and N-C. Lau, 1990: Observed structure and propagation characteristics of tropical summertime synoptic scale disturbances. Mon. Wea. Rev., 118 , 18881913.

    • Search Google Scholar
    • Export Citation
  • Lau, K-H., , and N-C. Lau, 1992: The energetics and propagation dynamics of tropical summertime synoptic-scale disturbances. Mon. Wea. Rev., 120 , 25232539.

    • Search Google Scholar
    • Export Citation
  • Lengaigne, M., , J-P. Boulanger, , C. Menkes, , S. Masson, , G. Madec, , and P. Delecluse, 2002: Ocean response to the March 1997 westerly wind event. J. Geophys. Res., 107 , 80158031.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , H. H. Hendon, , and J. D. Glick, 1994: The relationship between tropical cyclones of the western Pacific and Indian Oceans and the Madden–Julian oscillation. J. Meteor. Soc. Japan, 72 , 401412.

    • Search Google Scholar
    • Export Citation
  • Lin, X., , and R. H. Johnson, 1996: Kinematic and thermodynamic characteristics of the flow over the western Pacific warm pool during TOGA COARE. J. Atmos. Sci., 53 , 695715.

    • Search Google Scholar
    • Export Citation
  • Love, G., 1985: Cross-equatorial influence of winter hemisphere subtropical cold surges. Mon. Wea. Rev., 113 , 14871498.

  • Madden, R. A., , and P. R. Julian, 1971: Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28 , 702708.

    • Search Google Scholar
    • Export Citation
  • Madden, R. A., , and P. R. Julian, 1972: Description of global-scale circulation cells in the tropics with a 40–50 day period. J. Atmos. Sci., 29 , 11091123.

    • Search Google Scholar
    • Export Citation
  • Madden, R. A., , and P. R. Julian, 1994: Observations of the 40–50-day tropical oscillation. A review. Mon. Wea. Rev., 122 , 814837.

  • Maloney, E. D., , and D. L. Hartmann, 2000: Modulation of eastern North Pacific hurricanes by the Madden–Julian oscillation. J. Climate, 13 , 14511460.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and D. L. Hartmann, 2001: The Madden–Julian oscillation, barotropic dynamics, and North Pacific tropical cyclone formation. Part I: Observation. J. Atmos. Sci., 58 , 25452558.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and M. J. Dickinson, 2003: The intraseasonal oscillation and the energetics of summertime tropical western North Pacific synoptic-scale disturbances. J. Atmos. Sci., 60 , 21532168.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and S. K. Esbensen, 2003: The amplification of east Pacific Madden–Julian oscillation convection and wind anomalies during June–November. J. Climate, 16 , 34823497.

    • Search Google Scholar
    • Export Citation
  • McPhaden, M. J., 1999: Genesis and evolution of the 1997–98 El Niño. Science, 283 , 950954.

  • McPhaden, M. J., 2004: Evolution of the 2002/03 El Niño. Bull. Amer. Meteor. Soc., 85 , 677694.

  • McPhaden, M. J., , H. P. Freitag, , S. P. Hayes, , B. A. Taft, , Z. Chen, , and K. Wyrtki, 1988: The response of the equatorial Pacific Ocean to a westerly wind burst in May 1986. J. Geophys. Res., 93 , 1058910603.

    • Search Google Scholar
    • Export Citation
  • McPhaden, M. J., , F. Bahr, , Y. Du Penhoat, , E. Firing, , S. P. Hayes, , P. P. Niiler, , P. L. Richardson, , and J. M. Toole, 1992: The response of the western equatorial Pacific Ocean to westerly wind bursts during November 1989 to January 1990. J. Geophys. Res., 97 , 1428914303.

    • Search Google Scholar
    • Export Citation
  • Nakazawa, T., 1988: Tropical super clusters within intraseasonal variations over the western Pacific. J. Meteor. Soc. Japan, 66 , 823839.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., 1972: Energy budget of wave disturbances over the Marshall Islands during the years of 1956 and 1958. J. Meteor. Soc. Japan, 50 , 7184.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., , Y. Nakagomi, , Y. Suzuki, , N. Hasegawa, , and A. Kadokura, 1985: Global analysis of the lower tropospheric disturbances in the tropics during the northern summer of the FGGE year. Part I: Global features of the disturbances. J. Meteor. Soc. Japan, 63 , 119.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., , N. A. Rayner, , T. M. Smith, , D. C. Stokes, , and W. Wang, 2002: An improved in situ and satellite sst analysis for climate. J. Climate, 15 , 16091625.

    • Search Google Scholar
    • Export Citation
  • Seiki, A., , and Y. N. Takayabu, 2007: Westerly wind bursts and their relationship with intraseasonal variations and ENSO. Part I: Statistics. Mon. Wea. Rev., 135 , 33253345.

    • Search Google Scholar
    • Export Citation
  • Slingo, J. M., , D. P. Rowell, , K. R. Sperber, , and F. Nortley, 1999: On the predictability of the interannual behavior of the Madden–Julian oscillation and its relationship with El Niño. Quart. J. Roy. Meteor. Soc., 125 , 583609.

    • Search Google Scholar
    • Export Citation
  • Sobel, A. H., , and C. S. Bretherton, 1999: Development of synoptic-scale disturbances over the summertime tropical northwest Pacific. J. Atmos. Sci., 56 , 31063127.

    • Search Google Scholar
    • Export Citation
  • Sobel, A. H., , and E. D. Maloney, 2000: Effect of ENSO and the MJO on western North Pacific tropical cyclones. Geophys. Res. Lett., 27 , 17391742.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., 1994: Large-scale cloud disturbances associated with equatorial waves. Part II: Westward-propagating inertio–gravity waves. J. Meteor. Soc. Japan, 72 , 451465.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , and M. Murakami, 1991: The structure of super cloud clusters observed in 1–20 June 1986 and their relationship to easterly waves. J. Meteor. Soc. Japan, 69 , 105125.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , and J. C. L. Chan, 2002: How strong ENSO events affect tropical storm activity over the western North Pacific. J. Climate, 15 , 16431658.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., , and H. R. Chang, 1988: Equatorial energy accumulation and emanation regions: Impacts of a zonally varying basic state. J. Atmos. Sci., 45 , 803829.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , and M. M. Rienecker, 1998: Evidence of an extratropical atmospheric influence during the onset of the 1997–98 El Niño. Geophys. Res. Lett., 25 , 35373540.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , R. A. Weller, , and W. T. Liu, 2003: Case analysis of a role of ENSO in regulating the generation of westerly wind bursts in the western equatorial Pacific. J. Geophys. Res., 108 , 31283147.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Lag correlation between 5-month running-mean WWB frequencies and 5-month running-mean SST anomalies in the Niño-3 region. WWB frequencies are counted over the western–central Pacific. Negative lag represents WWB occurrences preceding SST anomalies. Significance levels are indicated by solid lines (99%), dotted–dashed lines (95%), and dotted–dotted–dashed lines (90%). This figure is the same as Fig. 6 in Part I of this study, except that the correlation for the western and central Pacific is calculated together. The shaded region indicates the “pre–El Niño” periods.

  • View in gallery

    Vertical structures along the equator of (a) the composite total zonal winds with intervals of 2.0 m s−1 and (b) the composite eddy zonal winds relative to the 11-day running-mean daily zonal winds with intervals of 1.0 m s−1 for WWB events over the western–central Pacific. Solid (dashed) contours indicate westerly (easterly) and dark gray contours are 0 m s−1. The abscissa is relative longitude, as mentioned in section 2c; i.e., 0° RLO indicates the longitude where each WWB takes its maximum wind anomalies. An average longitude for 0° RLO is 161.4°E. Shaded regions indicate more than the 95% significance level.

  • View in gallery

    Horizontal structures of composite eddy wind fields at 850, 500, and 200 hPa for WWB events over the western–central Pacific. Composites of the eddy zonal winds are shown in contours with intervals of 1.0 m s−1. Solid (dashed) lines represent positive (negative) values on and over (under) 1.0 (−1.0) m s−1. Shaded regions indicate more than the 95% significance level. Vectors represent composite eddy wind fields where either the zonal or meridional component is significant at the 95% level. The abscissa is relative longitude, as in Fig. 2.

  • View in gallery

    Composite EKE at 850 hPa for WWB events over the western–central Pacific in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values on and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level. The abscissa is relative longitude, as in Fig. 2.

  • View in gallery

    Composites of (a) the barotropic energy conversion at 850 hPa, (c) the conversion from the eddy available potential energy at 300 hPa, and (e) energy redistribution by convergence of the eddy geopotential flux at 850 hPa for WWB events over the western-central Pacific with intervals of 2.0, 6.0, and 2.0 × 10−5 m2 s−3, respectively. Right panels show the vertical cross sections. Positive values are represented by solid lines. Shaded regions indicate more than the 95% significance level. The abscissa is the relative longitude, as in Fig. 2.

  • View in gallery

    The composite vertical structures of each term (10−5 m2 s−3) for WWB events averaged between 10°N and 10°S, −10° and +10° RLO. Solid black and gray lines represent KmKe and PeKe, and dashed black and gray lines show GKe and KR, respectively. Error bars for each term shown on the right and left sides indicate 95% confidence limits.

  • View in gallery

    Composite EKE at 850 hPa for the top-50 events during the MJOs’ (a) westerly and (b) easterly phases in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values at and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level.

  • View in gallery

    Composite environmental wind fields (m s−1) at 850 hPa defined as the 11-day running mean, SST (°C), and environmental OLR (W m−2) for the top-50 MJO events in the westerly phases. Composite results during the (right) pre–El Niño and (left) others periods. The contours indicate the 95% significance level. Vectors represent composite environmental winds where either the zonal or the meridional component is significant at the 95% level.

  • View in gallery

    Composite EKE at 850 hPa for the top-50 MJO events during the westerly phases on day 0 in the (a) pre–El Niño and (b) others periods in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values on and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level. (c) The difference between the two periods in contours with intervals of 1.0 m2 s−2. Positive values are represented by solid lines, which indicates larger EKE in the correlated periods. Dark (light) shades indicate positive (negative) values larger (smaller) than 3.0 (−3.0) m2 s−2.

  • View in gallery

    The composite vertical structures of each term (10−5 m2 s−3) for the top-50 MJO events during the westerly phases on day −4 averaged between 10°N–10°S and 160°E–180° in the (a) pre–El Niño and (b) others periods. Solid black and gray lines represent KmKe and PeKe, and dashed black and gray lines show GKe and R, respectively. Error bars for each term shown on the right and left sides indicate 95% confidence limits.

  • View in gallery

    Time evolution of each term (10−5 m2 s−3) averaged for 10°N–10°S, 160°E–180° in the pre–El Niño periods. The abscissa represents the time lag from the day of the MJO westerly maximum. Gray line represents PeKe at 300 hPa with a left ordinate. Solid black and dashed lines show KmKe and GKe at 850 hPa, respectively, with a right ordinate. Bars on the bottom indicate the number of the WWB occurrences in each day. The smallest bars indicate one occurrence.

  • View in gallery

    Composites of the difference in (a) KmKe at 850 hPa, (b) PeKe at 300 hPa, and (c) GKe at 850 hPa between the pre–El Niño and others periods of the top-50 MJO events during the westerly phases on day −4 with intervals of 2.0, 6.0, and 3.0 × 10−5 m2 s−3, respectively. Positive values are represented by solid lines, which indicate larger values in the pre–El Niño periods. Dark (light) shades indicate positive (negative) values larger (smaller) than 2.0, 6.0, and 3.0 (−2.0, −6.0, and −3.0) × 10−5 m2 s−2. Boldface, dashed counters indicate composite environmental OLR of 210 W m−2 in the correlated periods on day −4.

  • View in gallery

    Typical cases of eddy vorticity at 850 hPa with intervals of 1.0 × 10−5 s−1 in the (a) pre–El Niño and (b) others periods. Dark (light) shading indicates positive (negative) values larger (smaller) than 1.0 (−1.0) × 10−5 s−1. Boldface, dashed counters indicate composite environmental OLR of 210 W m−2 on day −4.

  • View in gallery

    Schematics showing spatial distributions of environmental convection and of synoptic-scale disturbances overlaid on composite zonal winds and wind vectors (m s−1) during the (a) pre–El Niño and (b) others periods. Red (blue) circles indicate synoptic-scale cyclonic (anticyclonic) disturbances. White clouds show the large-scale MJO convection. Gray clouds represent synoptic-scale convection.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 154 154 39
PDF Downloads 150 150 44

Westerly Wind Bursts and Their Relationship with Intraseasonal Variations and ENSO. Part II: Energetics over the Western and Central Pacific

View More View Less
  • 1 Institute of Observational Research for Global Change, Japan Agency for Marine–Earth Science and Technology, Yokosuka, Kanagawa, Japan
  • | 2 Center for Climate System Research, University of Tokyo, Kashiwa, Chiba, and Institute of Observational Research for Global Change, Japan Agency for Marine–Earth Science and Technology, Yokosuka, Kanagawa, Japan
© Get Permissions
Full access

Abstract

The mechanism of synoptic-scale eddy development in the generation of westerly wind bursts (WWBs) over the western–central Pacific, and their relationship with the El Niño–Southern Oscillation (ENSO) and the Madden–Julian oscillation (MJO), were examined. In the WWB occurrences, barotropic structures of equatorial eddy westerlies with cyclonic disturbances were found from the surface to the upper troposphere. The dominant contributions to substantial eddy kinetic energy (EKE) were the barotropic energy conversion (KmKe) in the lower and middle tropospheres and the conversion from eddy available potential energy (PeKe) in the upper troposphere. Low-frequency environmental westerlies centered near the equator preceded strong zonal convergence and meridional shear, resulting in the substantial KmKe. The activation of synoptic convection also contributed to an increase in EKE through PeKe. These energies were redistributed to the lower-equatorial troposphere through energy flux convergence (GKe). These results showed that environmental fields contribute to the EKE increase near the equator and are important factors in WWB occurrences. Next, eddy growth was compared under different phases of MJO and ENSO. The MJO westerly phases of strong MJO events were classified into two groups, in terms of ENSO phases. Higher EKE values were found over the equatorial central Pacific in the WWB–ENSO correlated (pre–El Niño) periods. The energetics during these periods comported with those of the WWB generations. In the uncorrelated periods, the enhancement of eddy disturbances occurred far from the equator near the Philippines, where the activities of the easterly wave disturbances are well known. It is noteworthy that the enhanced region of the disturbances in the pre–El Niño periods coincided with the vicinity of large-scale MJO convection. It is suggested that coincidence corresponds with an enhancement of the internal disturbances embedded in the MJO, which is found only when the environmental conditions are favorable in association with ENSO.

Corresponding author address: Ayako Seiki, Institute of Observational Research for Global Change, JAMSTEC, 2-15 Natsushimachou, Yokosuka, Kanagawa 237-0061, Japan. Email: aseiki@jamstec.go.jp

Abstract

The mechanism of synoptic-scale eddy development in the generation of westerly wind bursts (WWBs) over the western–central Pacific, and their relationship with the El Niño–Southern Oscillation (ENSO) and the Madden–Julian oscillation (MJO), were examined. In the WWB occurrences, barotropic structures of equatorial eddy westerlies with cyclonic disturbances were found from the surface to the upper troposphere. The dominant contributions to substantial eddy kinetic energy (EKE) were the barotropic energy conversion (KmKe) in the lower and middle tropospheres and the conversion from eddy available potential energy (PeKe) in the upper troposphere. Low-frequency environmental westerlies centered near the equator preceded strong zonal convergence and meridional shear, resulting in the substantial KmKe. The activation of synoptic convection also contributed to an increase in EKE through PeKe. These energies were redistributed to the lower-equatorial troposphere through energy flux convergence (GKe). These results showed that environmental fields contribute to the EKE increase near the equator and are important factors in WWB occurrences. Next, eddy growth was compared under different phases of MJO and ENSO. The MJO westerly phases of strong MJO events were classified into two groups, in terms of ENSO phases. Higher EKE values were found over the equatorial central Pacific in the WWB–ENSO correlated (pre–El Niño) periods. The energetics during these periods comported with those of the WWB generations. In the uncorrelated periods, the enhancement of eddy disturbances occurred far from the equator near the Philippines, where the activities of the easterly wave disturbances are well known. It is noteworthy that the enhanced region of the disturbances in the pre–El Niño periods coincided with the vicinity of large-scale MJO convection. It is suggested that coincidence corresponds with an enhancement of the internal disturbances embedded in the MJO, which is found only when the environmental conditions are favorable in association with ENSO.

Corresponding author address: Ayako Seiki, Institute of Observational Research for Global Change, JAMSTEC, 2-15 Natsushimachou, Yokosuka, Kanagawa 237-0061, Japan. Email: aseiki@jamstec.go.jp

1. Introduction

Westerly wind bursts (WWBs) are synoptic-scale disturbances represented by strong westerly winds near the equator and have been known to trigger or enhance El Niño development (Harrison and Schopf 1984; McPhaden et al. 1988; McPhaden et al. 1992; McPhaden 1999; Lengaigne et al. 2002; McPhaden 2004).

In Seiki and Takayabu (2007, hereafter Part I) of this study, significant lag correlations were found between Niño-3 region sea surface temperature (SST) anomalies and WWB frequencies over the western, central, and eastern Pacific and the Indian Ocean. A common environment for WWB occurrences among all regions included the existence of active convection of the Madden–Julian oscillation (MJO; Madden and Julian 1971, 1972, 1994) with the intensified Rossby wave response and equatorial background westerly winds in association with the El Niño–Southern Oscillation (ENSO) phases. In addition, it was suggested that large MJO amplitude is a favorable, but not a sufficient condition for WWB generations. As the second part of this study, we focus on how the background wind patterns, associated with ENSO, bring the WWB occurrences during MJO active convection.

Whereas the MJO has been recognized as a planetary-scale disturbance, an observational study using geostationary satellite data found that there was a hierarchy in MJO convection (Nakazawa 1988). In his study, the eastward-moving MJO with zonal wavenumbers 1–3 was composed of several eastward-moving super–cloud clusters, which further consisted of westward-moving mesoscale clusters. Takayabu and Murakami (1991) suggested that the westward-moving clusters corresponded to the vortex-type easterly waves over the western Pacific. Takayabu (1994), on the other hand, suggested that an MJO observed during the Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA COARE) consisted of westward-propagating inertio–gravity waves with a period of ∼2 days. Namely, MJOs are aggregations of smaller-scale disturbances or clusters, though the mean MJO scale has a zonal scale of wavenumbers 1–3. Therefore, it is important to focus on the development of synoptic-scale disturbances to examine the intensification of the MJO convection. In this study, we focus on synoptic-scale phenomena to examine the fluctuation of internal disturbances the compose the MJO in the generation of WWBs.

For tropical cyclones, one of synoptic-scale disturbances, a genesis region of tropical cyclones was shown to shift eastward over the western North Pacific during El Niño (e.g., Lander 1994). Wang and Chan (2002) indicated the northwest–southeast dipole pattern in the peak season of tropical cyclone formations associated with the intensity of ENSO episodes.

As for the relationship between MJO and tropical cyclones, the number of tropical cyclones significantly increases during the MJO westerly phases compared to the easterly phases over the western North Pacific (Liebmann et al. 1994) and over the eastern North Pacific (Maloney and Hartmann 2000). These studies suggest that variations in large-scale circulation and convection, modulated by the MJO, could contribute to the development of synoptic-scale disturbances such as the easterly waves over the western North Pacific, and tropical cyclones over the eastern North Pacific. This has been attributed to a barotropic energy conversion from mean kinetic energy to eddy kinetic energy (EKE) in the lower troposphere and an energy conversion from eddy available potential energy to EKE in the upper troposphere (Lau and Lau 1992; Maloney and Dickinson 2003; Maloney and Hartmann 2001; Maloney and Esbensen 2003). In addition, mixed Rossby gravity waves have been shown to effect the development of tropical cyclones (Dickinson and Molinari 2002).

Other studies indicated that the barotropic wave accumulation through convergence of the low-level zonal winds contributes to the development of synoptic disturbances (Webster and Chang 1988; Holland 1995; Sobel and Bretherton 1999). Sobel and Maloney (2000) showed that the tongue of wave activity flux convergence extended farther eastward during El Niño years than it did in La Niña years over the western North Pacific.

The purpose of this study is to examine the mechanism of the synoptic-scale eddy developments, which are associated with the generation of WWBs, and its relationship with the MJO and ENSO phases. The domain of focus is the western and central Pacific from 120°E to the date line.

First, the vertical structures of WWBs and the energetics of synoptic-scale disturbances in the generation of WWBs are examined in a composite analysis in section 3. Section 4 is devoted to the analysis on the relationship among synoptic eddy developments, the MJO, and ENSO in the WWB generation process. A summary and discussion are given in section 5.

2. Data and analysis methods

a. Data

Primary data used in this study were from the European Centre for Medium-Range Weather Forecasts (ECMWF) 40-yr reanalysis dataset (ERA-40) for the period January 1979–August 2002. Zonal and meridional winds (u and υ), pressure vertical velocity (ω), pressure (p), temperature (T), and geopotential (Φ) at pressure levels from 1000 to 100 hPa and 10-m wind data were recorded four times a day. The daily means of these variables were used in this study. Daily outgoing longwave radiation (OLR) data derived by National Oceanic and Atmospheric Administration (NOAA) on 2.5° × 2.5° grids for the period of 1979–2002 were also used as an index for tropical deep convection. Weekly 1° × 1° gridded optimal interpolation SST (OISST) data (Reynolds et al. 2002) from 1981 to 2002 were also used.

To examine the effect of the MJO on synoptic-scale disturbances under different MJO phases, we used bandpass-filtered 200-hPa velocity potential (bpf-χ200) with half-power frequency cutoffs at 20 and 100 days, calculated from the ERA-40 winds, for MJO indices. It is known that bpf-χ200 represents a large-scale divergence in the upper troposphere and is a good indicator of MJO signals (Knutson and Weickmann 1987; Slingo et al. 1999).

The study region was the western and central Pacific (120°E–180°). The western and central Pacific, which were discussed separately in Part I, were combined here because of their similar characteristics.

b. Definition of WWBs

A WWB was identified when following three conditions were satisfied.

  1. The surface zonal wind anomalies averaged between 2.5°N and 2.5°S met or exceeded 5 m s−1. This meridional average width was chosen with reference to the oceanic radius of deformation.
  2. The area that satisfied the above zonal wind criteria extended zonally over at least 10° longitude.
  3. The above two conditions lasted for at least 2 days, and the center longitudes of two continuous days were 7.5° or closer.

Wind anomalies were the deviations from the seasonal climatology defined as 91-day running means using 10-m zonal wind data. See section 3 of Part I for further details.

c. Composite procedure based on the WWBs

In section 3, the vertical structure of the wind fields and the energetics of the WWB occurrences were examined. Composites were based on the point at which each WWB event attained its maximum amplitude in zonal wind anomalies on the longitude–time section. Thus, 0° relative longitude on day 0 corresponds to the longitude when the equatorial WWB attained its maximum zonal wind anomaly. Real longitude, corresponding to 0° RLO, varies from 135° to 177.5°E, with an average of 161.4°E over the western–central Pacific. The composite number was 116. Statistical significances were examined with a Student’s t test, using the 95% significance level.

d. Extracting the top-50 MJO amplitude events and the composite procedure

Another composite analysis was performed referring to large-scale variations: the westerly and easterly phases of MJO (section 4), and the westerly phase of the MJO classified into two groups regarding the ENSO phases (section 5).

Previous studies have shown that amplitudes of synoptic disturbances were enhanced in the MJO westerly phases over the western and eastern North Pacific (Lau and Lau 1992; Maloney and Hartmann 2001). In our study, first, the 50 largest-amplitude MJO events were extracted from bpf-χ200, averaged over the western–central Pacific (10°N–10°S, 120°E–180°), to examine the amplitude effect of the MJOs in the equatorial Tropics. These events are hereafter called the top-50 MJO events. The weakest MJO among these top-50 events had an amplitude of 1.7 times the standard deviation. Next, the westerly and easterly phases in each MJO were determined with respect to the maximum amplitude of the bpf zonal winds at 850 hPa (bpf-U850). Considering the eastward-moving MJO structure, in which convection preceding the low-level westerlies, the nearest maximum (minimum) bpf-U850 following (preceding) the minima bpf-χ200 was determined to be the westerly (easterly) phase.

In Part I of this study, it was shown that WWBs occur in association with MJO-like eastward propagation of convection, which nearly coincided with low-level westerlies, and were correlated with ENSO. However, variations in MJO amplitudes were not correlated with WWB frequency. To elucidate the effects of ENSO on WWB generations, we classified the top-50 MJO events of the westerly phases into two groups with respect to ENSO phases, and composites based on these classes were compared.

Figure 1 shows the lag correlations between WWB frequency and Niño-3 sea surface temperature anomalies (SSTAs). Significant lag correlations above the 99% significance level were found from lag −10 to +1 months, which correspond to the pre–El Niño periods. MJO westerly events in this correlated period were classified as MJOs in the “pre–El Niño periods” represented by shading in Fig. 1. The other MJO events during uncorrelated periods were classified as MJOs in “the others periods,” which are shown without shading.

3. Energetics in WWB occurrences

a. The vertical structure of WWBs

First, we examined the vertical structure of the atmospheric disturbances associated with WWBs. Composite analyses based on WWBs were performed (see section 2c for details).

The vertical structures of the composite total zonal winds and eddy zonal winds on day 0 of the WWB maxima are shown in Fig. 2. The total westerlies extended from the surface to above 400 hPa (Fig. 2a). The maxima in westerlies tilt slightly westward with height, but extend almost vertically. Deep westerlies reaching the upper troposphere are quite exceptional in the equatorial region. In addition, baroclinic structures were found in the total zonal winds such that easterlies were located in the upper troposphere above the westerlies, and the easterlies were especially strong above 200 hPa. These baroclinic structures illustrate well the mean structure of the MJO (Knutson and Weickmann 1987). This composite vertical structure of WWBs in total zonal winds using ERA-40 data resembles a WWB event observed during TOGA COARE on 30 December 1992, which was discussed in Lin and Johnson (1996). A good correspondence of the composite structure with this TOGA COARE case assures the further analysis of WWB structure using ERA-40 data.

Figure 2b depicts a composite structure of “eddy zonal winds,” which are defined as the deviation from the 11-day running-mean daily zonal winds, to describe the components with periods shorter than the MJO. Hereafter, “eddy” and “synoptic” refer to this deviation in this study. It is notable that the depths of the eddy westerlies are deeper than those of the total westerlies. A barotropic structure of eddy westerlies was found throughout the entire depth of the troposphere, in contrast to the results for the total zonal winds (Fig. 2a), which had a baroclinic structure. The horizontal scale of the eddy disturbances was about 30° in longitude, corresponding to ∼3300 km along the equator.

Next, the horizontal structures of the disturbances at each level were examined. In Fig. 3, eddy wind fields at 850, 500, and 200 hPa on day 0 are shown. Twin-cyclonic flows straddling the equator in the lower and middle troposphere, and divergent flow in the upper troposphere, were present around the composite center. Thus, deep eddy westerlies in the WWB generation (Fig. 2b) were associated with deep cyclonic disturbances from the surface to the middle troposphere.

b. The eddy kinetic energy budget

As shown in the previous subsection, deep synoptic disturbances at a spatial scale of about 3000 km and a time scale of several days were present in association with the WWBs. In this section, we examine the energetics of these disturbances.

The energy budget equations used in this study were the same as those in Lau and Lau (1992). A variable with an overbar indicates its low-frequency component defined as an 11-day running mean, that is, periods longer than the synoptic-scale disturbances, and are referred to as low frequency or environmental. The eddy component was calculated as a residual, and is referred to as eddy or synoptic.

The eddy kinetic energy equation is
i1520-0493-135-10-3346-e1
i1520-0493-135-10-3346-e2
where K′ is the eddy kinetic energy (EKE), V is the three-dimensional velocity vector, Vh is the horizontal velocity vector, and R is the gas constant for dry air. The first term on the right-hand side in Eq. (1) denotes the barotropic conversion to EKE from low-frequency kinetic energy (KmKe). The second and third terms represent the advection of EKE by the environmental flow (AmKe) and by the eddy flow (AeKe), respectively. The fourth term represents the conversion from eddy available potential energy (EAPE) to EKE through a rising or sinking motion of warm or cold air parcels (PeKe). The fifth term corresponds to the divergence of the eddy geopotential flux (GKe). Dissipation or subgrid-scale effects are represented by the last term (KR). The terms AmKe, AeKe, and GKe mainly represent the spatial redistribution of EKE, so these terms are not considered to be real sources or sinks.

Figure 4 shows the composite EKE at 850 hPa on day 0 of the WWB maxima. A substantial EKE value in the Northern Hemisphere was located along the northwest–southeast band from 20°N, −20° RLO to the composite center near the equator. Higher EKE was also found in the Southern Hemisphere around the composite center. These substantial EKE amounts around the composite center in both hemispheres reflect the twin-cyclone structure straddling the equator (Fig. 3).

Next, we performed an energy budget analysis to examine the energy source of the EKE, that is, the source of the synoptic disturbances bearing WWBs. The dominant contributions were attributed to KmKe, PeKe, GKe, and KR.

Composites of KmKe at 850 hPa are shown in Fig. 5a. The distribution of KmKe at tropical latitudes agrees with that of the high-EKE region, with maxima of KmKe located around 5°N. Strong energy conversions were found along the northwest–southeast band in the Northern Hemisphere from 15°N, −30° RLO to 5°N, +10° RLO. Large values of KmKe, exceeding 2.0 × 10−5 m2 s−3, extended from the surface to above 400 hPa over the central Pacific (Fig. 5b).

The contributions to KmKe were dominated generally by the −u/∂x term and the −u/∂y term (not shown), which are related to the zonal convergence of the environmental zonal wind, and the meridional shear of the environmental zonal wind, respectively. The dominance of EKE generation by the −u/∂x term was consistent with barotropic wave accumulation by the mean flow, as has been noted in previous studies (Webster and Chang 1988; Holland 1995; Sobel and Bretherton 1999). Generation by the −u/∂y term indicated that eddy activity was enhanced by the conversion from the meridional shear of the environmental zonal winds.

Horizontal structures of environmental wind fields defined as the 11-day running mean (not shown) were similar to those of the 91-day running means shown in Part I: equatorial westerlies penetrated into the easterly trade winds near the equator around the composite center. The barotropic wave accumulation was the dominant contributor to the generation of EKE over the central Pacific. On the other hand, over the western Pacific, substantial EKE was primarily attributed to the shear conversion, and secondarily to the wave accumulation. It was shown in Part I that the location of the WWB center was found to be at the eastern edge of the environmental equatorial westerly region over the central Pacific, whereas it was located around the center of the westerlies over the western Pacific. The WWB occurrences at the eastern edge of the westerlies over the central Pacific were consistent with EKE generation by the barotropic wave accumulation.

Next, we examined the contribution of PeKe (Fig. 5), which is related to synoptic convection. In previous studies of synoptic-scale disturbances over the central and western North Pacific (Nitta 1972; Lau and Lau 1992), PeKe peaks have been observed in the upper troposphere, corresponding to the peak level of upward vertical flow and the large diabatic heating within the tropical convection. In agreement with these studies, large values of PeKe extended from the middle to the upper troposphere from 500 to 200 hPa, and peaked at 300 hPa (Fig. 5d). Horizontal distributions of PeKe at 300 hPa (Fig. 5c) resembled those of KmKe at 850 hPa. This suggests that the coexistence of the amplification of disturbances through dynamical effects (KmKe) and convective effects (PeKe) are essential. Relatively large differences between KmKe and PeKe were found over the central Pacific in the Southern Hemisphere, where large PeKe values were found from 5° to 20°S, from −5° to +15° RLO, but the KmKe was small. Higher SSTs around the WWB center found in Part I can be associated with this active convection.

The distributions of GKe are shown in Figs. 5e and 5f. Since GKe included negative PeKe in its components mathematically, the vertical structures of GKe around 300 hPa were similar to those of PeKe with a reversed sign (Figs. 5d and 5f). The vertical structure of GKe indicated that the EKE generated by PeKe was redistributed into the lower troposphere below 400 hPa and the upper troposphere above 200 hPa by divergent flows. It is also interesting to note that horizontal distributions of GKe at 850 hPa (Fig. 5e) show large GKe values near the equator around the composite center, whereas the two conversion terms were weak on the equator but strong away from the equator. These results indicate that the EKE away from the equator was redistributed by GKe into the low-level equatorial region.

The area-averaged vertical structures of the dominant four terms are compared in Fig. 6. As mentioned above, KmKe (PeKe) generated EKE in the lower (upper) troposphere, and GKe redistributed EKE into the upper troposphere and the boundary layer. In addition to this, the residual term (KR) shows strong dissipation, especially in the boundary layer. This suggests that EKE was transferred from the atmosphere to the ocean in the WWB generation, resulting in inducing the oceanic Kelvin waves.

To summarize, the special distribution of the environmental wind field (equatorial westerlies in the easterly trade winds) and convective field influences the generation/amplification of lower cyclonic eddy disturbances, which gather eddy energy toward the equator, resulting in WWBs.

4. Comparisons of energetics during the MJO westerly phase between the different ENSO phases

So far in this study, we have performed a composite analysis based on WWBs. However, the manner in which the environment affects the enhancement and suppression of WWBs is still unclear. Moreover, from Part I, we have seen that WWBs were associated with MJO active convection but large MJO amplitude was not a sufficient condition for WWBs. In this section, therefore, we focus on the relationship between the phases of strong MJO events and WWB occurrences, and on their relationship to ENSO.

a. MJO westerly and easterly phases

First, we confirm the results of the previous studies: the influence of MJO westerly–easterly phases on the development of synoptic-scale disturbances.

A composite analysis was performed for the westerly and easterly phases of the top-50 MJO events, which were detected with bpf-χ200, as explained precisely in section 2d.

The composite EKE at 850 hPa in the MJOs’ westerly and easterly phases is shown in Fig. 7. In the westerly phase (Fig. 7a), a higher EKE band was located across the broad area from the northwest to the southeast, around or to the east of the Philippines. This band extended to the east of New Guinea and Australia. Higher EKE values around the equator were found from 140°E to 180°, though these values were smaller than they were at higher latitudes. Lau and Lau (1992) and Maloney and Dickinson (2003) have previously demonstrated the presence of higher EKE in the western North Pacific around the Philippines in the MJO westerly phase during the boreal summer. In contrast, in the easterly phases, EKE was not concentrated at low latitudes (Fig. 7b).

These composite results indicate that higher EKE in the Tropics can be found during the MJO westerly phase compared to the easterly phase in all regions. Nitta et al. (1985) and Lau and Lau (1990) indicated the existence of strong synoptic-scale disturbances in the Tropics during the northern summer over the western North Pacific, the Bay of Bengal, the eastern North Pacific, and the Atlantic–Caribbean region. In addition to the above higher-EKE regions, higher EKE values were also found over the equatorial central Pacific during the westerly phase of the MJO in Fig. 7a.

b. Environmental fields

Next, we focused on the impact of ENSO on the developmental process of synoptic disturbances associated with the MJO westerly phases, by comparing MJOs in the WWB–ENSO-correlated (pre–El Niño) periods with those in the uncorrelated (others) periods. See section 2d for the precise description of the analysis method.

The composites of environmental wind fields defined as the 11-day running mean at 850 hPa and SST show clear differences in the basic fields between the two periods. Environmental westerlies extended to the date line from 120°E near the equator during the pre–El Niño periods (Fig. 8a). During the others periods, equatorial westerlies did not extend as far eastward. Instead, strong westerlies were located over the region from the eastern part of the Bay of Bengal to the South China Sea (Fig. 8b). In pre–El Niño periods, the warm pool above 29°C extended beyond the date line even at the equator and reached 150°W at 10°S because of pre–El Niño conditions (Fig. 8c). In the others periods (Fig. 8d), on the other hand, the warm SST was confined to the west of the date line in the equatorial region, but extended northward from 120° to 150°E to the east of the Philippines. Although there were regional differences in the wind field, environmental convection associated with the MJO was centered on 150°E, east of Papua New Guinea in both cases (Figs. 8e and 8f). The amplitude was somewhat larger by about 10 W m−2 in the pre–El Niño periods than it was in the others periods, and more active convection was found west of the Philippines in the others periods.

A large fraction of top-50 MJO westerly events was detected from May to June, and from November to December. However, significant differences in seasonality were not observed between the numbers of MJOs in the pre–El Niño and others periods, suggesting that interannual variations have a greater influence than do seasonal variations of the environmental fields in this region.

c. The eddy kinetic energy budget

Figure 9 depicts composites of EKE at 850 hPa for the westerly phase of the top-50 MJO events, comparing between the two periods. High EKE peaks in the pre–El Niño periods (Fig. 9a) were observed around 10°N, 145°E in the Northern Hemisphere and around 15°S, 170°E in the Southern Hemisphere. In contrast, a substantial amount of EKE was located around 20°N, 130°E in the others periods (Fig. 9b). The differences in EKE between the two periods (Fig. 9c) shows that an area of higher EKE was located over the equatorial central Pacific (15°N–20°S, 150°E–180°) in the pre–El Niño periods, whereas it was found over the western North Pacific (10°–25°N, 120°–150°E) in the others periods. Note that higher EKE regions around the equator during the pre–El Niño periods corresponded to the area of frequent WWB occurrences (160°E–180°). This suggests that the EKE increases around the equator result in WWB occurrences.

To examine the energy source of EKE under different large-scale conditions, an analysis of the eddy kinetic energy budget was performed. The eddy kinetic energy budget equation is the same as Eq. (1) in section 3b. Here, we focus on the four dominant terms: KmKe, PeKe, GKe, and KR. Figure 10 shows the vertical distributions of each composite term averaged between 10°N–10°S and 160°E–180° during the pre–El Niño and others periods. Notable is that vertical structures during pre–El Niño periods (Fig. 10a) closely resemble those of the WWB occurrences as shown in Fig. 6. That is, EKE generated by KmKe in the lower to middle troposphere and by PeKe in the upper troposphere was redistributed into the boundary layer (GKe) during the pre–El Niño periods. In addition, strong dissipation in the boundary layer (KR) can indicate the large momentum transfer from the atmosphere to the ocean as mentioned in the WWB composite. Horizontal distributions (not shown) were similar to those of EKE (Fig. 9) between 20°N and 20°S; large KmKe and PeKe values were found off the equator in the lower and upper troposphere, respectively; and GKe had the largest values near the equator. In the others periods (Fig. 10b), on the other hand, the amplitude of each term was much smaller than that in the pre–El Niño periods. This is due to fact that the eddy-enhanced region was located away from the equator around the western North Pacific as shown in Fig. 9b.

The contributions in KmKe were dominated by barotropic wave accumulation and by the shear conversion. In the pre–El Niño periods, EKE generation over the central Pacific was largely attributed to the wave accumulation, whereas EKE north of Papua New Guinea was generated by shear conversion (not shown). The equatorial environmental westerlies shown in Fig. 8a during the pre–El Niño periods were related to the intensification of a zonal flow gradient over the central Pacific and a meridional gradient of zonal flow north of Papua New Guinea. This large-scale wind pattern produced a strong conversion near the equator. In contrast in the others periods, EKE generations by both the wave accumulation and shear conversion were comparable around the Philippines, which is consistent with the results of Maloney and Hartmann (2001), though they did not consider interannual variations. The equatorial westerlies shown in Fig. 8b do not extend as far eastward as they do in the pre–El Niño periods, and the zonal gradient was weaker. Instead, strong westerlies, located over the South China Sea, were associated with a large horizontal gradient and a meridional shear of zonal flow around the Philippines. Thus, variations in the large-scale circulation associated with the ENSO phases can control where the zonal convergence and meridional shear are intensified and exerts an influence on the distribution of EKE generation by KmKe in the MJO westerly phases.

As for PeKe, the warm pool above 29°C extended eastward beyond the date line during the pre–El Niño periods, whereas in the others periods the warm equatorial SST was confined to the west of the date line, but extended northward into the western North Pacific (Fig. 8). These differences in SST distribution may also affect the location of eddy convection by providing sufficient moisture, although this does not completely explain the regional differences in the PeKe distributions.

Finally, we examined the composite time evolution of each term averaged for 10°N–10°S, 160°E–180° in the MJO westerly phase in the pre–El Niño periods (Fig. 11). The abscissa represents the time lag from the day of the MJO westerly maximum. Both KmKe and PeKe increased prior to the day of the MJO westerly maximum and peaked on day −4. GKe in the lower troposphere also increased before the MJO westerly reached its maximum, but it peaked on day 0 following the two conversion terms by a few days. The number of WWBs, shown by gray bars, increased after the two conversion terms attained their peaks on day −4. It is suggested that EKE starts to increase a couple of weeks before the MJO westerly maximum, through KmKe at lower levels and through PeKe at upper levels off of the equator. Then, the EKE is redistributed toward the equator at lower levels by GKe following the two conversion terms.

In summary, higher EKE associated with active energy conversions was found at equatorial latitudes in the area of frequent WWB occurrences during the pre–El Niño periods. EKE increases near the equator can result in WWB occurrences. Contrastingly in the others periods, enhanced eddy disturbances were found far from the equator around the western North Pacific. These differences are attributed to different distributions in environmental wind fields and synoptic convection between the two periods. Moreover, characteristics of these active eddy disturbances found far from the equator in the others periods coincided with the easterly wave disturbances that have been indicated in previous studies (e.g., Lau and Lau 1992).

d. Eddy cyclonic disturbances and their relationship with large-scale convective activities

Finally, in this section we examine the relationship between the large-scale convection, such as MJO, and the synoptic disturbances that account for substantial EKE generation.

The relationship between the large-scale (MJO) convection and the regional shift of the EKE evolutions in different ENSO phases is shown in Fig. 12. The differences of each term in the eddy kinetic energy budget between the pre–El Niño and others periods are indicated by thin contours and the boldface contour represents the region of MJO convection (composite environmental OLR). In the pre–El Niño periods, all three terms had larger values over the central Pacific at low latitudes (10°N–10°S, 160°E–180°), which are shown in dark shades. On the other hand, substantial energy conversions by KmKe and PeKe were found around the Philippines in the others periods, which are shown in light shades. Notable in this figure is that the region of EKE increase during the pre–El Niño periods coincides with that of MJO active convection. It has been indicated that the hierarchy in the eastward-propagating MJO convection consists of westward-moving mesoscale clusters or synoptic disturbances (Nakazawa 1988; Takayabu and Murakami 1991; Takayabu 1994). The above results indicate that enhanced synoptic disturbances during the pre–El Niño periods correspond to the disturbances embedded in MJO convections.

Figure 13 illustrates cases typical of the eddy vorticity for both periods. During the pre–El Niño periods (Fig. 13a), there were three cyclonic disturbances, positive in the Northern Hemisphere, with a synoptic spatial scale of 1000 km north of the equator from 135°E to 180° on 16 December 1994. At that time, composite MJO convection was located around the equator from 140°E to 180° near the three cyclonic disturbances.

On 25 August 2000, during the others periods (Fig. 13b), anticyclonic and cyclonic disturbances appeared alternately along the northwest–southeast axis north of Papua New Guinea to the southeastern coast of China. This region is famous for westward-propagating disturbances associated with the easterly waves. It is noteworthy that MJO convection in this case was located at almost the same position as it was during the pre–El Niño periods. In other words, the region, where synoptic and cyclonic disturbances are enhanced, changes in association with the ENSO phases, even though MJO convection is located at similar equatorial longitudes. This result strongly suggests that variations in large-scale circulation associated with the ENSO phases dynamically determine the regions in which eddy disturbances are enhanced.

5. Summary and discussion

In this study, we examined the mechanisms of synoptic-scale eddy developments during the course of WWB generation. We aimed to clarify how the MJO and ENSO phases affect the frequency of WWB generation.

First, we examined the WWB structure and its energy budgets. A composite of the total wind field revealed that equatorial westerly winds extended from the surface to the middle troposphere almost vertically, and strong easterlies were found above the westerlies in the western-central Pacific. This baroclinic structure agreed well with a typical MJO structure (Knutson and Weickmann 1987) as well as with a WWB event observed during TOGA COARE (Lin and Johnson 1996). When we focused on the eddy fields, barotropic structures, reaching the upper troposphere, emerged for the eddy westerlies. Twin-cyclonic disturbances straddling the equator were observed from the surface to the middle troposphere associated with these eddy westerlies. Thus, the characteristic structure of the WWB consisted of deep twin-cyclonic disturbances and equatorial westerlies in eddy components. Large values of the eddy kinetic energy (EKE) occurred around the WWB center at equatorial latitudes. The dominant contributions were the barotropic energy conversion (KmKe) in the lower and middle troposphere, the conversion from EAPE to EKE (PeKe) in the upper troposphere, and the redistribution term from the energy flux convergence (GKe) in the lower and upper troposphere. Strong dissipation (KR) was also observed in the boundary layer.

The energy conversions from the environmental wind fields and from synoptic convection were the main sources of WWB generation, and this energy was redistributed to the upper and lower troposphere by eddy divergence flow, and to the equator by eddy cyclonic flow, resulting in WWBs. Strong dissipation in the boundary layer suggests large amounts of momentum transfer from the atmosphere to the ocean.

In Part I, the favorable periods for WWB occurrences were found to be associated with ENSO. Moreover, the larger amplitudes of the MJO were favorable but not sufficient for WWB generation. It was not yet clear what roles the ENSO and MJO play in the generation processes of WWBs. Therefore, in this study, we focused on the MJO and ENSO phases to reveal the enhancement and suppression mechanisms of WWBs. First, a composite analysis was performed for the westerly–easterly phase of the top-50 MJO events with strongest amplitudes. It was confirmed that higher EKE can be found in the westerly phase of strong MJOs in the lower troposphere in the Tropics, which is consistent with previous studies (Lau and Lau 1992; Maloney and Hartmann 2001; Maloney and Dickinson 2003).

Next, to examine the influences of ENSO on the eddy energetics, the MJO westerly phases of the top-50 events were classified into two groups for comparison: MJO in the WWB–ENSO-correlated (pre–El Niño) periods and those in the uncorrelated (others) periods. Higher EKE was found over the equatorial central Pacific in the pre–El Niño periods. Distributions of substantial EKE values near the equator coincided with those of the WWB occurrences. In contrast in the others periods, large eddy disturbances were found distant from the equator in the western North Pacific around the Philippines, where previous studies have revealed the activity of easterly wave disturbances (e.g., Lau and Lau 1992). These results are similar to previous results (Lander 1994; Sobel and Maloney 2000), which showed that the zonal displacement of regions for strong wave accumulation and frequent tropical cyclones is found between the El Niño and La Niña years. We have shown that northwest–southeast, not just zonal, displacement of EKE occurred primarily between the pre–El Niño and others periods in association with WWB occurrences.

Results of the energy budget during pre–El Niño periods were similar to those of the WWB composites: large values of generation by KmKe in the lower and middle troposphere, PeKe in the upper troposphere, and GKe in the lower and upper troposphere, as well as strong dissipation in the boundary layer. Time evolutions showed that EKE started to increase a couple of weeks before the MJO westerly maximum by KmKe and by PeKe off the equator. Following the increase of these two conversion terms, substantial EKE values were redistributed vertically into the lower and upper troposphere and horizontally into the equator in the lower troposphere by GKe. These results indicate that both strong MJO westerlies and horizontal structures of environmental winds and SST associated with the pre–El Niño condition contribute to the EKE increase near the equator and become important factors for WWB occurrences.

A notable result was that the eddy disturbances were amplified around active large-scale MJO convection near the equator only in the pre–El Niño periods, while synoptic disturbances developed far from MJO convection in the others periods. Figure 14 compares schematic diagrams of eddy developments under these two conditions. The MJO convection (white clouds) occurred around 150°E in the MJO westerly phases during both periods. However, synoptic disturbances (red and blue circles) within the MJO were enhanced only in the pre–El Niño periods (Fig. 14a), resulting in the amplification of the “internal eddies” of the MJO. Such enhancement of the westward-propagating vortical disturbances may reduce the eastward-propagating phase speed of the MJO associated with the WWB occurrences as shown in Part I. In contrast, in the others periods (Fig. 14b), synoptic disturbances associated with the westerly phase of equatorial MJOs developed over the western North Pacific around the Philippines along the monsoon trough, as represented by the convergence zone, resulting in the enhancement of easterly waves outside the MJO.

Thus, large-scale variations in the atmospheric circulation and SST associated with ENSO can enhance the equatorial synoptic disturbances embedded in the MJO convection only in the pre–El Niño periods. We suggest that WWBs are born from this nonlinearly interacted enhancement of synoptic disturbances and MJO convection. Then, WWB occurrences over the Pacific can accelerate and maintain El Niño development as indicated in previous studies (e.g., McPhaden et al. 1988). Oceanic responses to the WWBs and feedback such as the contribution of the eastward displacement warm pool to the WWB generation are left for future studies. The coincident enhancement of different scale disturbances, such as synoptic and intraseasonal ones, does not occur during the unfavorable phases of ENSO. These enhanced disturbances, outside the MJO, did not generate WWBs, nor did they excite the oceanic Kelvin wave that may enhance the El Niño development.

In this study, we have considered synoptic disturbances, the MJO, and ENSO separately for linear thinking. However, further studies are needed to clarify the nonlinear interactions between synoptic- and planetary-scale disturbances: for example, how the enhancement of synoptic disturbances changes the characteristics of the planetary-scale MJO, such as reducing its phase speed. We should probably further examine the role of mesoscale convective systems considering the phase speed variations. Second, what determines the MJO amplitude is still unclear. The possible effects of ENSO on modifying MJO amplitudes needs further investigation, since the influence of the MJO amplitudes on an environmental wind field play an important role in the development of synoptic disturbances. In addition, the contribution of the magnitudes of ENSO events is left for further work.

In this study, the favorable ENSO phase for WWB generation, the pre–El Niño periods, was made a primary focused in order to elucidate the mechanism underlying WWB generation. On the other hand, there was an interesting relationship to the MJO of western North Pacific easterly waves, which existed in the others periods. Previous studies indicated that the easterly wave disturbances developed during the boreal summer. However, dominant seasonality was not found in the MJO events during the others periods. Precise studies are needed to investigate whether interannual or seasonal variations are predominant in the formation of favorable basic fields because the events in the others periods were defined in a manner similar to the rest of the events.

Statistical examinations of the impact of midlatitude forcing, such as cold surges, on WWBs over the western Pacific (Love 1985; Chu 1988; Kiladis et al. 1994) and their relationship to the MJO (Yu and Rienecker 1998) and to ENSO (Yu et al. 2003), are also left for future studies.

Acknowledgments

This work was part of the first author’s doctoral dissertation. The first author would like to express thanks to both Prof. Masahide Kimoto and Prof. Akimasa Sumi, Center for Climate System Research, University of Tokyo, for their helpful comments and discussions. She is also thankful to Prof. Jun Matsumoto, University of Tokyo, and Prof. Hiroshi Niino, Ocean Research Institute, University of Tokyo, for their valuable comments that improved the manuscript. She is also thankful to Dr. Toru Nozawa, National Institute for Environmental Studies, and Dr. Masayoshi Ishii, Meteorological Research Institute, for their help with obtaining and processing the ERA-40 data.

REFERENCES

  • Chu, P-S., 1988: Extratropical forcing and the burst of equatorial westerlies in the western Pacific: A synoptic study. J. Meteor. Soc. Japan, 66 , 549564.

    • Search Google Scholar
    • Export Citation
  • Dickinson, M., , and J. Molinari, 2002: Mixed Rossby–gravity waves and western Pacific tropical cyclogenesis. Part I: Synoptic evolution. J. Atmos. Sci., 59 , 21832196.

    • Search Google Scholar
    • Export Citation
  • Harrison, D. E., , and P. S. Schopf, 1984: Kelvin wave induced anomalous advection and the onset of surface warming in El Niño events. Mon. Wea. Rev., 112 , 923933.

    • Search Google Scholar
    • Export Citation
  • Holland, G. J., 1995: Scale interaction in the western Pacific monsoon. Meteor. Atmos. Phys., 56 , 5779.

  • Kiladis, G. N., , G. A. Meehl, , and K. M. Weickmann, 1994: Large-scale circulation associated with westerly wind bursts and deep convection over the western equatorial Pacific. J. Geophys. Res., 99 , 1852718544.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., , and K. M. Weickmann, 1987: 30–60 day atmospheric oscillations: Composite life cycles of convection and circulation anomalies. Mon. Wea. Rev., 115 , 14071436.

    • Search Google Scholar
    • Export Citation
  • Lander, M. A., 1994: An exploratory analysis of the relationship between tropical storm formation in the western North Pacific and ENSO. Mon. Wea. Rev., 122 , 636651.

    • Search Google Scholar
    • Export Citation
  • Lau, K-H., , and N-C. Lau, 1990: Observed structure and propagation characteristics of tropical summertime synoptic scale disturbances. Mon. Wea. Rev., 118 , 18881913.

    • Search Google Scholar
    • Export Citation
  • Lau, K-H., , and N-C. Lau, 1992: The energetics and propagation dynamics of tropical summertime synoptic-scale disturbances. Mon. Wea. Rev., 120 , 25232539.

    • Search Google Scholar
    • Export Citation
  • Lengaigne, M., , J-P. Boulanger, , C. Menkes, , S. Masson, , G. Madec, , and P. Delecluse, 2002: Ocean response to the March 1997 westerly wind event. J. Geophys. Res., 107 , 80158031.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , H. H. Hendon, , and J. D. Glick, 1994: The relationship between tropical cyclones of the western Pacific and Indian Oceans and the Madden–Julian oscillation. J. Meteor. Soc. Japan, 72 , 401412.

    • Search Google Scholar
    • Export Citation
  • Lin, X., , and R. H. Johnson, 1996: Kinematic and thermodynamic characteristics of the flow over the western Pacific warm pool during TOGA COARE. J. Atmos. Sci., 53 , 695715.

    • Search Google Scholar
    • Export Citation
  • Love, G., 1985: Cross-equatorial influence of winter hemisphere subtropical cold surges. Mon. Wea. Rev., 113 , 14871498.

  • Madden, R. A., , and P. R. Julian, 1971: Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28 , 702708.

    • Search Google Scholar
    • Export Citation
  • Madden, R. A., , and P. R. Julian, 1972: Description of global-scale circulation cells in the tropics with a 40–50 day period. J. Atmos. Sci., 29 , 11091123.

    • Search Google Scholar
    • Export Citation
  • Madden, R. A., , and P. R. Julian, 1994: Observations of the 40–50-day tropical oscillation. A review. Mon. Wea. Rev., 122 , 814837.

  • Maloney, E. D., , and D. L. Hartmann, 2000: Modulation of eastern North Pacific hurricanes by the Madden–Julian oscillation. J. Climate, 13 , 14511460.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and D. L. Hartmann, 2001: The Madden–Julian oscillation, barotropic dynamics, and North Pacific tropical cyclone formation. Part I: Observation. J. Atmos. Sci., 58 , 25452558.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and M. J. Dickinson, 2003: The intraseasonal oscillation and the energetics of summertime tropical western North Pacific synoptic-scale disturbances. J. Atmos. Sci., 60 , 21532168.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., , and S. K. Esbensen, 2003: The amplification of east Pacific Madden–Julian oscillation convection and wind anomalies during June–November. J. Climate, 16 , 34823497.

    • Search Google Scholar
    • Export Citation
  • McPhaden, M. J., 1999: Genesis and evolution of the 1997–98 El Niño. Science, 283 , 950954.

  • McPhaden, M. J., 2004: Evolution of the 2002/03 El Niño. Bull. Amer. Meteor. Soc., 85 , 677694.

  • McPhaden, M. J., , H. P. Freitag, , S. P. Hayes, , B. A. Taft, , Z. Chen, , and K. Wyrtki, 1988: The response of the equatorial Pacific Ocean to a westerly wind burst in May 1986. J. Geophys. Res., 93 , 1058910603.

    • Search Google Scholar
    • Export Citation
  • McPhaden, M. J., , F. Bahr, , Y. Du Penhoat, , E. Firing, , S. P. Hayes, , P. P. Niiler, , P. L. Richardson, , and J. M. Toole, 1992: The response of the western equatorial Pacific Ocean to westerly wind bursts during November 1989 to January 1990. J. Geophys. Res., 97 , 1428914303.

    • Search Google Scholar
    • Export Citation
  • Nakazawa, T., 1988: Tropical super clusters within intraseasonal variations over the western Pacific. J. Meteor. Soc. Japan, 66 , 823839.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., 1972: Energy budget of wave disturbances over the Marshall Islands during the years of 1956 and 1958. J. Meteor. Soc. Japan, 50 , 7184.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., , Y. Nakagomi, , Y. Suzuki, , N. Hasegawa, , and A. Kadokura, 1985: Global analysis of the lower tropospheric disturbances in the tropics during the northern summer of the FGGE year. Part I: Global features of the disturbances. J. Meteor. Soc. Japan, 63 , 119.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., , N. A. Rayner, , T. M. Smith, , D. C. Stokes, , and W. Wang, 2002: An improved in situ and satellite sst analysis for climate. J. Climate, 15 , 16091625.

    • Search Google Scholar
    • Export Citation
  • Seiki, A., , and Y. N. Takayabu, 2007: Westerly wind bursts and their relationship with intraseasonal variations and ENSO. Part I: Statistics. Mon. Wea. Rev., 135 , 33253345.

    • Search Google Scholar
    • Export Citation
  • Slingo, J. M., , D. P. Rowell, , K. R. Sperber, , and F. Nortley, 1999: On the predictability of the interannual behavior of the Madden–Julian oscillation and its relationship with El Niño. Quart. J. Roy. Meteor. Soc., 125 , 583609.

    • Search Google Scholar
    • Export Citation
  • Sobel, A. H., , and C. S. Bretherton, 1999: Development of synoptic-scale disturbances over the summertime tropical northwest Pacific. J. Atmos. Sci., 56 , 31063127.

    • Search Google Scholar
    • Export Citation
  • Sobel, A. H., , and E. D. Maloney, 2000: Effect of ENSO and the MJO on western North Pacific tropical cyclones. Geophys. Res. Lett., 27 , 17391742.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., 1994: Large-scale cloud disturbances associated with equatorial waves. Part II: Westward-propagating inertio–gravity waves. J. Meteor. Soc. Japan, 72 , 451465.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , and M. Murakami, 1991: The structure of super cloud clusters observed in 1–20 June 1986 and their relationship to easterly waves. J. Meteor. Soc. Japan, 69 , 105125.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , and J. C. L. Chan, 2002: How strong ENSO events affect tropical storm activity over the western North Pacific. J. Climate, 15 , 16431658.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., , and H. R. Chang, 1988: Equatorial energy accumulation and emanation regions: Impacts of a zonally varying basic state. J. Atmos. Sci., 45 , 803829.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , and M. M. Rienecker, 1998: Evidence of an extratropical atmospheric influence during the onset of the 1997–98 El Niño. Geophys. Res. Lett., 25 , 35373540.

    • Search Google Scholar
    • Export Citation
  • Yu, L., , R. A. Weller, , and W. T. Liu, 2003: Case analysis of a role of ENSO in regulating the generation of westerly wind bursts in the western equatorial Pacific. J. Geophys. Res., 108 , 31283147.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Lag correlation between 5-month running-mean WWB frequencies and 5-month running-mean SST anomalies in the Niño-3 region. WWB frequencies are counted over the western–central Pacific. Negative lag represents WWB occurrences preceding SST anomalies. Significance levels are indicated by solid lines (99%), dotted–dashed lines (95%), and dotted–dotted–dashed lines (90%). This figure is the same as Fig. 6 in Part I of this study, except that the correlation for the western and central Pacific is calculated together. The shaded region indicates the “pre–El Niño” periods.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 2.
Fig. 2.

Vertical structures along the equator of (a) the composite total zonal winds with intervals of 2.0 m s−1 and (b) the composite eddy zonal winds relative to the 11-day running-mean daily zonal winds with intervals of 1.0 m s−1 for WWB events over the western–central Pacific. Solid (dashed) contours indicate westerly (easterly) and dark gray contours are 0 m s−1. The abscissa is relative longitude, as mentioned in section 2c; i.e., 0° RLO indicates the longitude where each WWB takes its maximum wind anomalies. An average longitude for 0° RLO is 161.4°E. Shaded regions indicate more than the 95% significance level.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 3.
Fig. 3.

Horizontal structures of composite eddy wind fields at 850, 500, and 200 hPa for WWB events over the western–central Pacific. Composites of the eddy zonal winds are shown in contours with intervals of 1.0 m s−1. Solid (dashed) lines represent positive (negative) values on and over (under) 1.0 (−1.0) m s−1. Shaded regions indicate more than the 95% significance level. Vectors represent composite eddy wind fields where either the zonal or meridional component is significant at the 95% level. The abscissa is relative longitude, as in Fig. 2.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 4.
Fig. 4.

Composite EKE at 850 hPa for WWB events over the western–central Pacific in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values on and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level. The abscissa is relative longitude, as in Fig. 2.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 5.
Fig. 5.

Composites of (a) the barotropic energy conversion at 850 hPa, (c) the conversion from the eddy available potential energy at 300 hPa, and (e) energy redistribution by convergence of the eddy geopotential flux at 850 hPa for WWB events over the western-central Pacific with intervals of 2.0, 6.0, and 2.0 × 10−5 m2 s−3, respectively. Right panels show the vertical cross sections. Positive values are represented by solid lines. Shaded regions indicate more than the 95% significance level. The abscissa is the relative longitude, as in Fig. 2.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 6.
Fig. 6.

The composite vertical structures of each term (10−5 m2 s−3) for WWB events averaged between 10°N and 10°S, −10° and +10° RLO. Solid black and gray lines represent KmKe and PeKe, and dashed black and gray lines show GKe and KR, respectively. Error bars for each term shown on the right and left sides indicate 95% confidence limits.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 7.
Fig. 7.

Composite EKE at 850 hPa for the top-50 events during the MJOs’ (a) westerly and (b) easterly phases in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values at and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 8.
Fig. 8.

Composite environmental wind fields (m s−1) at 850 hPa defined as the 11-day running mean, SST (°C), and environmental OLR (W m−2) for the top-50 MJO events in the westerly phases. Composite results during the (right) pre–El Niño and (left) others periods. The contours indicate the 95% significance level. Vectors represent composite environmental winds where either the zonal or the meridional component is significant at the 95% level.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 9.
Fig. 9.

Composite EKE at 850 hPa for the top-50 MJO events during the westerly phases on day 0 in the (a) pre–El Niño and (b) others periods in contours with intervals of 2.0 m2 s−2. Solid lines represent EKE values on and above 8.0 m2 s−2. Shaded regions indicate more than the 95% significance level. (c) The difference between the two periods in contours with intervals of 1.0 m2 s−2. Positive values are represented by solid lines, which indicates larger EKE in the correlated periods. Dark (light) shades indicate positive (negative) values larger (smaller) than 3.0 (−3.0) m2 s−2.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 10.
Fig. 10.

The composite vertical structures of each term (10−5 m2 s−3) for the top-50 MJO events during the westerly phases on day −4 averaged between 10°N–10°S and 160°E–180° in the (a) pre–El Niño and (b) others periods. Solid black and gray lines represent KmKe and PeKe, and dashed black and gray lines show GKe and R, respectively. Error bars for each term shown on the right and left sides indicate 95% confidence limits.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 11.
Fig. 11.

Time evolution of each term (10−5 m2 s−3) averaged for 10°N–10°S, 160°E–180° in the pre–El Niño periods. The abscissa represents the time lag from the day of the MJO westerly maximum. Gray line represents PeKe at 300 hPa with a left ordinate. Solid black and dashed lines show KmKe and GKe at 850 hPa, respectively, with a right ordinate. Bars on the bottom indicate the number of the WWB occurrences in each day. The smallest bars indicate one occurrence.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 12.
Fig. 12.

Composites of the difference in (a) KmKe at 850 hPa, (b) PeKe at 300 hPa, and (c) GKe at 850 hPa between the pre–El Niño and others periods of the top-50 MJO events during the westerly phases on day −4 with intervals of 2.0, 6.0, and 3.0 × 10−5 m2 s−3, respectively. Positive values are represented by solid lines, which indicate larger values in the pre–El Niño periods. Dark (light) shades indicate positive (negative) values larger (smaller) than 2.0, 6.0, and 3.0 (−2.0, −6.0, and −3.0) × 10−5 m2 s−2. Boldface, dashed counters indicate composite environmental OLR of 210 W m−2 in the correlated periods on day −4.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 13.
Fig. 13.

Typical cases of eddy vorticity at 850 hPa with intervals of 1.0 × 10−5 s−1 in the (a) pre–El Niño and (b) others periods. Dark (light) shading indicates positive (negative) values larger (smaller) than 1.0 (−1.0) × 10−5 s−1. Boldface, dashed counters indicate composite environmental OLR of 210 W m−2 on day −4.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Fig. 14.
Fig. 14.

Schematics showing spatial distributions of environmental convection and of synoptic-scale disturbances overlaid on composite zonal winds and wind vectors (m s−1) during the (a) pre–El Niño and (b) others periods. Red (blue) circles indicate synoptic-scale cyclonic (anticyclonic) disturbances. White clouds show the large-scale MJO convection. Gray clouds represent synoptic-scale convection.

Citation: Monthly Weather Review 135, 10; 10.1175/MWR3503.1

Save