• Andersson, A., S. Bakan, K. Fennig, H. Grassl, C-P. Klepp, and J. Schulz, 2007: Hamburg ocean atmosphere parameters and fluxes from satellite data—HOAPS-3—Twice daily composite. World Data Center for Climate. [Available online at http://cera-www.dkrz.de/WDCC/ui/Compact.jsp?acronym=HOAPS3_DAILY].

    • Search Google Scholar
    • Export Citation
  • Annamalai, H., and J. M. Slingo, 2001: Active/break cycles: Diagnosis of the intraseasonal variability of the Asian summer monsoon. Climate Dyn., 18 , 85102.

    • Search Google Scholar
    • Export Citation
  • Back, L. E., and C. S. Bretherton, 2009: On the relationship between SST gradients, boundary layer winds, and convergence over the tropical oceans. J. Climate, 22 , 41824196.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and J. Srinivasan, 2006: Comment on “Structures and mechanisms of the northward-propagating boreal summer intraseasonal oscillation”. J. Climate, 19 , 47384743.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and A. H. Sobel, 2008a: Instability of the axisymmetric monsoon flow and intraseasonal oscillation. J. Geophys. Res., 113 , D07108. doi:10.1029/2007JD009291.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and A. H. Sobel, 2008b: Poleward-propagating intraseasonal monsoon disturbances in an intermediate-complexity axisymmetric model. J. Atmos. Sci., 65 , 470489.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., A. H. Sobel, and J. Vialard, 2008: Ocean–atmosphere coupling in the monsoon intraseasonal oscillation: A simple model study. J. Climate, 21 , 52545270.

    • Search Google Scholar
    • Export Citation
  • Chao, W. C., 2000: Multiple quasi equilibria of the ITCZ and the origin of monsoon onset. J. Atmos. Sci., 57 , 641651.

  • Chou, C., L-F. Huang, L. Tseng, J-Y. Tu, and P-H. Tan, 2009: Annual cycle of rainfall in the western North Pacific and East Asian sector. J. Climate, 22 , 20732094.

    • Search Google Scholar
    • Export Citation
  • Drbohlav, H-H., and B. Wang, 2005: Mechanism of the northward-propagating intraseasonal oscillation: Insights from a zonally symmetric model. J. Climate, 18 , 952972.

    • Search Google Scholar
    • Export Citation
  • Duvel, J-P., and J. Vialard, 2007: Indo-Pacific sea surface temperature perturbations associated with intraseasonal oscillations of tropical convection. J. Climate, 20 , 30563082.

    • Search Google Scholar
    • Export Citation
  • Fu, X., B. Wang, and T. Li, 2002: Impacts of air–sea coupling on the simulation of mean Asian summer monsoon in the ECHAM4 model. Mon. Wea. Rev., 130 , 28892904.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., 2005: South Asian monsoon. Intraseasonal Variability in the Atmosphere–Ocean Climate System, W. K. M. Lau and D. E. Waliser, Eds., Springer, 19–61.

    • Search Google Scholar
    • Export Citation
  • Hoffman, R. N., and S. M. Leidner, 2005: An introduction to the near-real-time QuikSCAT data. Wea. Forecasting, 20 , 476493.

  • Hsu, H-H., 2005: East Asian monsoon. Intraseasonal Variability in the Atmosphere–Ocean Climate System, W. K. M. Lau and D. E. Waliser, Eds., Springer, 63–94.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., and C-H. Weng, 2001: Northwestward propagation of the intraseasonal oscillation in the western North Pacific during the boreal summer: Structure and mechanism. J. Climate, 14 , 38343850.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., C-H. Weng, and C-H. Wu, 2004: Contrasting characteristics between the northward and eastward propagation of the intraseasonal oscillation during the boreal summer. J. Climate, 17 , 727743.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., T. Li, and B. Wang, 2004: Structures and mechanisms of the northward-propagating boreal summer intraseasonal oscillation. J. Climate, 17 , 10221039.

    • Search Google Scholar
    • Export Citation
  • Kanamitsu, M., W. Ebisuzaki, J. Woollen, S-K. Yang, J. J. Hnilo, M. Fiorino, and G. L. Potter, 2002: NCEP-DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Kemball-Cook, S., and B. Wang, 2001: Equatorial waves and air–sea interaction in the boreal summer intraseasonal oscillation. J. Climate, 14 , 29232942.

    • Search Google Scholar
    • Export Citation
  • Klingaman, N. P., H. Weller, J. N. Slingo, and P. M. Inness, 2008: The intraseasonal variability of the Indian summer monsoon using TMI sea surface temperature and ECMWF reanalysis. J. Climate, 21 , 25192539.

    • Search Google Scholar
    • Export Citation
  • Krishnamurti, T. N., and D. Subrahmanyam, 1982: The 30–50-day mode at 850 mb during MONEX. J. Atmos. Sci., 39 , 20882095.

  • Krishnamurti, T. N., P. K. Jayakumar, J. Sheng, N. Surgi, and A. Kumar, 1985: Divergent circulations on the 30- to 50-day time scale. J. Atmos. Sci., 42 , 364375.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., and Coauthors, 2000: The status of the Tropical Rainfall Measuring Mission (TRMM) after two years in orbit. J. Appl. Meteor., 39 , 19651982.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., and P. H. Chan, 1986: Aspects of the 40–50-day oscillation during the northern summer as inferred from outgoing longwave radiation. Mon. Wea. Rev., 114 , 13541367.

    • Search Google Scholar
    • Export Citation
  • Lawrence, D. M., and P. J. Webster, 2002: The boreal summer intraseasonal oscillation: Relationship between northward and eastward movement of convection. J. Atmos. Sci., 59 , 15931606.

    • Search Google Scholar
    • Export Citation
  • Lindzen, R. S., and S. Nigam, 1987: On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics. J. Atmos. Sci., 44 , 24182436.

    • Search Google Scholar
    • Export Citation
  • LinHo, and B. Wang, 2002: The time-space structure of the Asian-Pacific summer monsoon: A fast annual cycle view. J. Climate, 15 , 20012019.

    • 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., 112 , 814837.

  • Maloney, E. D., and D. L. Hartmann, 1998: Frictional moisture convergence in a composite life cycle of the Madden–Julian oscillation. J. Climate, 11 , 23872403.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., and J. T. Kiehl, 2002: MJO-related SST variations over the tropical eastern Pacific during Northern Hemisphere summer. J. Climate, 15 , 675689.

    • Search Google Scholar
    • Export Citation
  • Rajendran, K., and A. Kitoh, 2006: Modulation of tropical intraseasonal oscillations by ocean–atmosphere coupling. J. Climate, 19 , 366391.

    • Search Google Scholar
    • Export Citation
  • Roxy, M., and Y. Tanimoto, 2007: Role of SST over the Indian Ocean in influencing the intraseasonal variability of the Indian summer monsoon. J. Meteor. Soc. Japan, 85 , 349358.

    • Search Google Scholar
    • Export Citation
  • Sengupta, D., B. N. Goswami, and R. Senan, 2001: Coherent intraseasonal oscillations of ocean and atmosphere during the Asian summer monsoon. Geophys. Res. Lett., 28 , 41274130.

    • Search Google Scholar
    • Export Citation
  • Tsou, C-H., P-C. Hsu, W-S. Kau, and H-H. Hsu, 2005: Northward and northwestward propagation of 30–60-day oscillation in the tropical and extratropical western North Pacific. J. Meteor. Soc. Japan, 83 , 711726.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and D. E. Harrison, 2002: Monsoon break and subseasonal sea surface temperature variability in the Bay of Bengal. J. Climate, 15 , 14851493.

    • Search Google Scholar
    • Export Citation
  • Waliser, D. E., R. Murtugudde, and L. E. Lucas, 2004: Indo-Pacific ocean response to atmospheric intraseasonal variability: 2. Boreal summer and the intraseasonal oscillation. J. Geophys. Res., 109 , C03030. doi:10.1029/2003JC002002.

    • Search Google Scholar
    • Export Citation
  • Wang, B., and X. Xie, 1997: A model for the boreal summer intraseasonal oscillation. J. Atmos. Sci., 54 , 7286.

  • Wang, B., and LinHo, 2002: Rainy season of the Asian-Pacific summer monsoon. J. Climate, 15 , 386398.

  • Wang, B., P. Webster, K. Kikuchi, T. Yasunari, and Y. Qi, 2006: Boreal summer quasi-monthly oscillation in the global tropics. Climate Dyn., 27 , 661675.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 1997: A well-calibrated ocean algorithm for special sensor microwave/imager. J. Geophys. Res., 102 , 87038718.

  • Wentz, F. J., C. Gentemann, D. Smith, and D. Chelton, 2000: Satellite measurements of sea-surface temperature through clouds. Science, 288 , 847850.

    • Search Google Scholar
    • Export Citation
  • Wheeler, M., and G. N. Kiladis, 1999: Convectively coupled equatorial waves: Analysis of clouds and temperature in the wavenumber-frequency domain. J. Atmos. Sci., 56 , 374399.

    • Search Google Scholar
    • Export Citation
  • Yasunari, T., 1981: Structure of an Indian summer monsoon system with around 40-day period. J. Meteor. Soc. Japan, 59 , 336354.

  • Yu, L., and R. A. Weller, 2007: Objectively analyzed air–sea fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., 2005: Madden–Julian oscillation. Rev. Geophys., 43 , RG2003. doi:10.1029/2004RG000158.

  • Zhang, Y., W. B. Rossow, A. A. Lacis, V. Oinas, and M. I. Mishchenko, 2004: Calculation of radiative fluxes from the surface to top of atmosphere based on ISCCP and other global datasets: Refinement of the radiative transfer model and the input data. J. Geophys. Res., 109 , D19105. doi:10.1029/2003JD004457.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    The 30–90-day-filtered daily precipitation (mm day−1) averaged over 5°S–5°N for the summer period (May–September) of 2000–07. The IO and the WNP are marked by black dashed lines.

  • View in gallery

    The 30–90-day-filtered daily precipitation (mm day−1) averaged over the IO (75°–90°E) for the summer period (May–September) of 2000–07. Major northward-propagating ISOs are marked by black solid lines.

  • View in gallery

    As in Fig. 2, but averaged over the WNP (130°–170°E).

  • View in gallery

    A composite of eight phases for one cycle of the northward-propagating ISO over the IO. Shaded is precipitation (mm day−1), and the vectors are surface wind (m s−1). The blue dot denotes cold SST anomalies, and the red cross indicates warm SST anomalies. All variables are after 30–90-day filtering.

  • View in gallery

    Composites of the northward-propagating ISOs over the IO based on the ISO center marked in Fig. 2 for (a) precipitation (mm day−1), (b) the longitudinal component of surface winds (m s−1), (c) surface vorticity (10−6 s−1), and (d) surface divergence (10−6 s−1). All variables are after 30–90-day filtering.

  • View in gallery

    As in Fig. 5, but for (a) SST (°C), (b) SST tendency (°C day−1), (c) surface latent heat flux (W m−2), (d) surface wind speed (m s−1), (e) surface solar radiation (W m−2), and (f) pseudohorizontal moisture advection (mm day−1) (see text).

  • View in gallery

    As in Fig. 4, but for surface latent heat flux from the OAFlux dataset (W m−2; shaded), SST tendency (°C day−1; blue dot and red cross), and net surface solar radiation (W m−2; contours).

  • View in gallery

    Mean state of the summer circulation over the IO: (a) vertical wind shear (200 − 850-hPa zonal winds; m s−1), and (b) differences of meridional winds between 200 and 850 hPa (200 − 850 hPa; m s−1).

  • View in gallery

    As in Fig. 4, but for the WNP.

  • View in gallery

    As in Fig. 5, but for the WNP.

  • View in gallery

    As in Fig. 6, but for the WNP.

  • View in gallery

    As in Fig. 7, but for the WNP.

  • View in gallery

    As in Fig. 8, but for the WNP.

  • View in gallery

    Seasonal variation of mean SST (2000–07; °C) over the WNP.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 172 105 5
PDF Downloads 76 42 4

Mechanisms of Northward-Propagating Intraseasonal Oscillation—A Comparison between the Indian Ocean and the Western North Pacific

View More View Less
  • 1 Research Center for Environmental Changes, Academia Sinica, and Department of Atmospheric Sciences, National Taiwan University, Taipei, Taiwan
  • | 2 Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan
Full access

Abstract

Mechanisms of northward-propagating intraseasonal oscillations (ISOs) over the Indian Ocean (IO) and the western North Pacific (WNP) are examined for the possibility of their existence in observations. They include the following: 1) the vorticity advection effect, which is associated with the advection of anomalous baroclinic vorticity by mean baroclinic meridional winds; 2) the vertical wind shear effect, which is the vertical advection associated with the meridional gradient of baroclinic divergence and mean easterly vertical wind shear; 3) the moisture advection effect induced by mean flow; and 4) the air–sea interaction via surface latent heat flux. Because of differences in mean state, the influence of each mechanism on the northward-propagating ISOs is different between the IO and the WNP. The vorticity advection effect is consistently found over both the IO and the WNP, while the air–sea interaction has different impacts on the northward-propagating ISOs over the IO and the WNP. The vertical wind shear effect and the moisture advection effect are relatively important over the IO but not over the WNP. Processes to determine changes in SST are also different between the IO and the WNP. Over the IO, SST is mainly associated with surface solar radiation. Wind-stirring effects, surface latent heat flux, and subsurface water entrainment are secondary. Over the WNP, wind-stirring effects become important, but surface solar radiation is secondary.

Corresponding author address: Chia Chou, Research Center for Environmental Changes, Academia Sinica, P.O. Box 1-48, Taipei 11529, Taiwan. Email: chiachou@rcec.sinica.edu.tw

Abstract

Mechanisms of northward-propagating intraseasonal oscillations (ISOs) over the Indian Ocean (IO) and the western North Pacific (WNP) are examined for the possibility of their existence in observations. They include the following: 1) the vorticity advection effect, which is associated with the advection of anomalous baroclinic vorticity by mean baroclinic meridional winds; 2) the vertical wind shear effect, which is the vertical advection associated with the meridional gradient of baroclinic divergence and mean easterly vertical wind shear; 3) the moisture advection effect induced by mean flow; and 4) the air–sea interaction via surface latent heat flux. Because of differences in mean state, the influence of each mechanism on the northward-propagating ISOs is different between the IO and the WNP. The vorticity advection effect is consistently found over both the IO and the WNP, while the air–sea interaction has different impacts on the northward-propagating ISOs over the IO and the WNP. The vertical wind shear effect and the moisture advection effect are relatively important over the IO but not over the WNP. Processes to determine changes in SST are also different between the IO and the WNP. Over the IO, SST is mainly associated with surface solar radiation. Wind-stirring effects, surface latent heat flux, and subsurface water entrainment are secondary. Over the WNP, wind-stirring effects become important, but surface solar radiation is secondary.

Corresponding author address: Chia Chou, Research Center for Environmental Changes, Academia Sinica, P.O. Box 1-48, Taipei 11529, Taiwan. Email: chiachou@rcec.sinica.edu.tw

1. Introduction

The northward propagation of intraseasonal oscillations (ISOs) is commonly found over the Indian Ocean (IO) and the western North Pacific (WNP) during the boreal summer (e.g., Yasunari 1981; Krishnamurti and Subrahmanyam 1982; Lau and Chan 1986; Annamalai and Slingo 2001; Goswami 2005; Hsu 2005). To explain the northward-propagating ISOs, several mechanisms have been proposed (e.g., Wang and Xie 1997; Hsu et al. 2004; Jiang et al. 2004; Goswami 2005; Wang et al. 2006; Bellon and Sobel 2008a). In general, the origin for ISOs propagating northward is associated with low-level convergence a few degrees north of the convection maximum. This low-level convergence could be caused by barotropic vorticity in the free atmosphere above (Jiang et al. 2004; Goswami 2005; Bellon and Sobel 2008a). However, what causes this barotropic vorticity in the free atmosphere is still an unsettled issue. Jiang et al. (2004) proposed two mechanisms. First, off the equator, the vertical shear mechanism is responsible for such barotropic vorticity occurring north of the convection maximum, particularly over the IO. This mechanism has been simulated in a zonally symmetric model (Drbohlav and Wang 2005), but the scale selection of the unstable mode is unclear (Bellon and Srinivasan 2006). Second, near the equator, the moisture–convection feedback mechanism, which is associated with low-level moisture convergence, is the main mechanism. In this mechanism, anomalous moisture convergence induced by mean and anomalous meridional winds is generated north of the convection maximum, which could induce the northward-propagating ISOs. Bellon and Sobel (2008a) proposed a different mechanism that is associated with the vorticity advection mechanism. In their idealized aquaplanet model, the meridional advection of anomalous baroclinic vorticity by the mean meridional baroclinic flow can induce barotropic vorticity north of the convection maximum.

Besides the internal atmospheric dynamics discussed above, air–sea interaction also plays a role in northward-propagating ISOs. Several observations show signs of intraseasonal oscillations in SST (e.g., Sengupta et al. 2001; Vecchi and Harrison 2002; Goswami 2005; Rajendran and Kitoh 2006; Wang et al. 2006; Duvel and Vialard 2007; Roxy and Tanimoto 2007), which is related to the atmospheric ISOs. Warm (cold) SST anomalies lead positive (negative) precipitation anomalies, whereas cold (warm) SST anomalies lag positive (negative) precipitation anomalies. These SST anomalies are determined by a surface heat flux exchange and changes in the mixed-layer depth via wind-stirring processes (Waliser et al. 2004; Bellon et al. 2008). Studies show that air–sea interaction has influences on the ISO characteristics (e.g., Kemball-Cook and Wang 2001; Fu et al. 2002; Waliser et al. 2004), but it is not essential to induce the northward-propagating ISOs. Some studies, however, do show that air–sea interaction can enhance the instability of the monsoon flow (Jiang et al. 2004; Bellon et al. 2008).

In this study, we aim to examine those existing mechanisms in observations to find which mechanism is the most essential or most common for the northward-propagating ISOs by comparing ISOs between the IO and the WNP. Section 2 describes data that we used here, most of which are satellite observations with high temporal and spatial resolutions. Most northward-propagating ISOs over the IO and the WNP in the boreal summer are related to the eastward-propagating ISOs near the equator (e.g., Lawrence and Webster 2002), so the relationship between these eastward- and northward-propagating ISOs is briefly discussed in section 3. The northward-propagating ISOs over the IO and the WNP are examined in sections 4 and 5, respectively. A comparison of the results between the IO and the WNP is in section 6, followed by the conclusions.

2. Data and analysis method

a. Data

In this study, most of the data, including precipitation, SST, surface winds, and water vapor, are derived from satellite observations. Daily precipitation (mm day−1) is derived from the Tropical Rainfall Measuring Mission (TRMM) satellite 3B42 3-hourly rain-rate (mm h−1) data (Kummerow et al. 2000); SST is from the TRMM Microwave Imager (TMI) version 4 daily data (Wentz et al. 2000); surface winds come from the Quick Scatterometer (QuikScat or QSCAT) daily data (Hoffman and Leidner 2005); and column-integrated water vapor is from the Special Sensor Microwave Imager (SSM/I) version 6 daily data (Wentz 1997). The satellite data all have 0.25° × 0.25° resolution, and the period of 2000–07 is chosen as this period is common to all of the datasets. Some other data with relatively coarser spatial resolutions are also used here. Two daily latent heat fluxes are used. The first is derived from the Hamburg Ocean Atmosphere Parameters and Fluxes from Satellite Data version 3 (HOAPS3) twice daily data (Andersson et al. 2007) with 1° × 1° resolution. Because of data availability, only the period of 2000–05 is used for latent heat flux. Compared to the same latent heat flux but averaged over a longer period (1998–2005), the result is similar. The other latent heat flux is from Objectively Analyzed Air–Sea Fluxes (OAFlux) daily data (Yu and Weller 2007), which is 1° × 1° in spatial resolution. The period of 2000–07 is used. Daily surface shortwave radiation is averaged from the International Satellite Cloud Climatology Project (ISCCP) flux data-surface (FD-SRF) 3-hourly data (Zhang et al. 2004) on a 2.5° × 2.5° grid. Besides surface winds, winds and specific humidity at 1000, 925, 850, 700, and 200 hPa are also used for calculating vertical wind shear and low-level moisture advection. Here, we have chosen the National Centers for Environmental Prediction (NCEP)–Department of Energy (DOE) Global Reanalysis 2 dataset (NCEP R2; Kanamitsu et al. 2002). The spatial resolutions of the NCEP R2 data are 2.5° × 2.5° horizontally and 17 standard pressure layers vertically. The NCEP R2 6-hourly data in 1979–2008 are used to calculate climatology with a pentad resolution.

b. Analysis method

All data except for those used for climatology (i.e., Figs. 8, 13, and 14) were first filtered by a 30–90-day bandpass filter, based on a simple Fourier expansion. ISOs are identified by maximum precipitation anomalies averaged over the region of interest, such as the equator (5°S–5°N), the IO (75°–90°E), and the WNP (130°–170°E). The 14 strongest northward-propagating ISOs in the IO and the 15 strongest in the WNP were chosen for the composites discussed below. Using the precipitation anomalies around 15°N in the IO and 7.5°N in the WNP, the cycle of each northward-propagating ISO can be identified between two minimum precipitation anomalies. The averaged ISO cycle is around 42 days in IO and 43 days in the WNP. Each ISO cycle was then equally divided into eight phases for a two-dimensional composite, with the interval around 5 days. At each given day of these northward-propagating ISO cycles, the latitude with maximum zonally averaged precipitation anomalies was chosen as a reference latitude (zero latitude) for a one-dimensional composite.

3. The boreal summer intraseasonal oscillation

Figure 1 shows the 30–90-day-filtered daily precipitation averaged over 5°S–5°N for the period of 2000–07. Eastward propagations of ISOs, starting in the IO, are clearly seen in the boreal summer, which has been discussed in many studies (e.g., Madden and Julian 1972, 1994; Wheeler and Kiladis 1999; Maloney and Kiehl 2002; Zhang 2005). A few possible westward propagations can also be found near the date line, such as in 2001. The eastward-propagating ISOs show interannual variability—stronger and more evident in 2001, 2004, and 2007 and weaker and less evident in 2005 and 2006. In some years, such as in 2002 and 2003, ISOs are stronger in early summer, whereas in some other years, such as in 2000, ISOs are stronger in late summer.

In the IO (75°–90°E), besides an eastward-propagating component, a northward-propagating component of ISOs is also found (Fig. 2). These northward-propagating ISOs can reach as far north as around 25°N, the foothills of the Himalayas. Some slightly southward propagations are found near the equator, such as in June and July 2005, but northward-propagating ISOs still dominate the IO. The northward-propagating ISOs over the IO also exhibit a weak interannual variability, similar to those in Fig. 1. Northward-propagating ISOs are also found over the WNP (130°–170°E), which is shown in Fig. 3. In this region, northward-propagating ISOs can reach as far north as around 20°N. Some southward-propagating convection systems are also found. As in the IO, the northward-propagating ISOs dominate the WNP, but the occurrence of the southward-propagating ISOs is much more frequent in the WNP than in the IO. These southward-propagating systems usually merge with the northward-propagating ISOs, such as in 2005. The northward-propagating ISOs over the WNP show a relatively clear interannual variability, with stronger amplitudes in 2002 and 2004 and weaker amplitudes in 2006 and 2007.

Most northward-propagating ISOs over the IO and the WNP (Figs. 2 and 3) are related to the eastward-propagating ISOs near the equator (Fig. 1), which has been discussed in previous studies (e.g., Krishnamurti et al. 1985; Lau and Chan 1986; Wang and Xie 1997; Maloney and Hartmann 1998; Lawrence and Webster 2002; Wang et al. 2006). For instance, when an eastward-propagating ISO reached the IO domain in early July of 2004 (Fig. 1), a northward-propagating component of ISO appeared. It continued to propagate to 25°N around mid-August (Fig. 2). When this eastward-propagating ISO moved eastward to the WNP in late July and early August (Fig. 1), a component of ISO started to propagate northward to around 20°N in late August (Fig. 3). In other words, the eastward-propagating ISOs can have northward-propagating components when they pass through the IO and the WNP during the boreal summer. Thus, the northward-propagating ISOs show an approximate 20-day difference between the IO and the WNP. Figures 1 –3 also hint at a possible correlation in magnitude between the eastward-propagating ISOs at the equator and the northward-propagating ISOs over the IO and the WNP, which will be examined in the future.

4. The Indian Ocean sector

a. Observations

A two-dimensional composite of the 14 strongest northward-propagating ISOs over the IO (marked by black lines in Fig. 2) is constructed for each of the eight phases (Fig. 4). Similar to Fig. 3 in Jiang et al. (2004) and Fig. 2 in Wang et al. (2006), positive precipitation anomalies (green shading) propagate northward, particularly in phases 3 and 4. They reach the mature stage in phase 5 and gradually disappear in phase 8. As the positive precipitation anomalies reach their maximum strength, new negative precipitation anomalies also start to appear near the equator. A low-level cyclonic circulation closely coincides with these positive precipitation anomalies and moves northward. Besides precipitation and low-level winds, SSTs also show a northward movement, which is slightly lagging (leading) the precipitation anomalies. In Figs. 4d–4g, cold (warm) SST anomalies move northward from the equator to 20°N in the Bay of Bengal, lagging (leading) positive precipitation anomalies. This implies a possible ocean–atmosphere coupling for northward-propagating ISOs over the IO. More details will be discussed later.

Furthermore, we composed one-dimensional composites for various variables. Figure 5a shows precipitation anomalies that are roughly symmetric in the meridional direction, with the maximum at zero latitude, that is, the ISO center. Surface zonal winds and near-surface vorticity (Figs. 5b and 5c) imply that the center of the associated cyclonic circulation (see Fig. 4) is located around 2° north of the convection center, which is consistent with other observation analyses (Jiang et al. 2004; Goswami 2005; Wang et al. 2006) and is believed to be the origin of the northward-propagating ISOs over the IO (Jiang et al. 2004; Bellon and Sobel 2008a). The divergence center shown in Fig. 5d is not clear, because of the distortion of some small-scale variations, but it should also be north of the precipitation maximum, according to the sign change of the zonal winds shown in Fig. 5b.

We next examine the possible ocean–atmosphere interaction found in Fig. 4. Some studies (e.g., Vecchi and Harrison 2002; Goswami 2005; Wang et al. 2006; Bellon et al. 2008) show a SST signal that is associated with ISOs over the Bay of Bengal, with about a 10-day lead to the precipitation maximum. Figure 6a shows that warm SST anomalies lead the precipitation maximum by around 7°, which is consistent with previous studies. On the other hand, Fig. 6a also implies that cold SST anomalies are around 7° south of the maximum precipitation anomalies, similar to those found in Klingaman et al. (2008). We note that the sign change of the SST anomalies coincides well with the precipitation maximum. The SST tendency (Fig. 6b) is very similar to the distribution of the precipitation anomalies shown in Fig. 5a but with an opposite sign. It implies that the surface flux associated with convection might be a main cause in determining the SST changes. First, by examining the surface latent heat flux in Fig. 6c, it can be seen that the pattern is different from that of the SST tendency. Positive latent heat flux means that the energy transfers from the ocean to the atmosphere and vice versa for negative values. Main positive latent heat flux occurs south of the maximum precipitation, whereas negative latent heat flux occurs north of the maximum precipitation. This implies that the SST changes are not controlled by surface latent heat flux. Obviously, this latent heat flux is also not controlled by the SST anomalies because positive (negative) latent heat flux coincides with cold (warm) SST anomalies. In fact, changes in the latent heat flux are mainly determined by changes in surface wind speed (Fig. 6d), which is enhanced south of the precipitation maximum and reduced north of the precipitation maximum, under a background of westerlies. We note that the patterns of the latent heat flux between two datasets are slightly different. The maximum anomalies of the OAFlux latent heat flux are closer to the ISO center than the HOAPS3 latent heat flux. We next examine net surface solar radiation, which is shown in Fig. 6e. Positive (negative) net surface solar radiation is downward (upward), that is, from the atmosphere to the ocean. The pattern of net surface solar radiation is similar to those of the precipitation anomalies and the SST tendency. Larger precipitation anomalies are associated with more cloudiness, which reflects more solar radiation and reduces net surface solar radiation. Moreover, the change of net surface solar radiation, with a maximum around 35 W m−2, is much larger than the change in surface latent heat flux, with a maximum around 10 W m−2. Based on spatial distribution and amplitude, the changes in net surface solar radiation should be the main factor in determining the SST tendency over the IO. This also implies that ocean heat transport via subsurface entrainment (e.g., Duvel and Vialard 2007; Bellon and Sobel 2008a) might not be as important as solar radiation. However, a further study to directly calculate the subsurface entrainment is needed to confirm this hypothesis. We further examine the evolution of the surface heat fluxes and the SST tendency in one complete ISO cycle (Fig. 7). Because of data length, here we only present the latent heat flux from the OAFlux dataset. When ISOs propagate northward in phases 3–6, positive latent heat flux is slightly lagging the SST cooling tendency, while negative net surface solar radiation roughly coincides with the SST tendency. The lag of latent heat flux to the SST tendency shown in Fig. 7 is more evident in the HOAPS3 dataset (not shown).

b. Mechanisms

Considering mechanisms that drive ISOs northward, ocean–atmosphere interaction could play a role. In Fig. 6c, main positive and negative latent heat flux anomalies are located south and north of the maximum precipitation anomalies, respectively. This implies that the changes in surface latent heat flux might slow down the northward propagation of ISOs over the IO, which would be consistent with some previous studies (e.g., Hsu et al. 2004; Bellon et al. 2008). In Fig. 7, the positive latent heat flux anomalies lag the ISO convection mainly in the beginning of the northward propagation, that is, phases 3 and 4, so the slow down could occur in this period. Another possible mechanism for the northward-propagating ISOs is associated with surface sensible heat flux (Hsu et al. 2004; Wang et al. 2006; Klingaman et al. 2008). Warm SST anomalies usually induce positive sensible heat flux anomalies, which can destabilize the atmosphere and induce low-level convergence and favor the ISOs moving northward. Figure 6a indeed shows that warm SST anomalies are a few degrees to the north of the maximum precipitation anomalies, thus implying that they could drive ISOs northward. However, the sensible heat flux anomalies are relatively small (not shown) because of a negative contribution from surface wind speed anomalies, that is, reduced surface wind speed north of the maximum precipitation anomalies; thus, this mechanism involving changes in sensible heat flux should not be the main mechanism driving ISOs northward. Besides via latent heat and sensible heat fluxes, SST could also induce low-level convergence via its gradient (Back and Bretherton 2009; Lindzen and Nigam 1987), which then affects the northward-propagating ISOs. The distribution of SST anomalies shown in Fig. 6a, however, implies that the possible convergence should occur much farther north of the ISO center, which is not consistent what Fig. 5c shows. Thus, the SST gradient should not be a dominant factor here, but it does create a favorable condition for ISOs propagating northward.

Besides air–sea interaction, other mechanisms could also induce the northward propagation of ISOs. The first is associated with low-level moisture convergence induced by horizontal winds (Jiang et al. 2004), which usually occurs near the equator. Since most atmospheric water vapor concentrates at the lower troposphere, SSM/I column-integrated water vapor can be used to roughly estimate low-level horizontal moisture advection (Fig. 6f), combined with the QSCAT surface winds. We refer to this quantity as pseudohorizontal moisture advection. Figure 6f shows positive pseudohorizontal moisture advection (−v · q > 0) north of the maximum precipitation anomalies. We also directly calculated low-level horizontal moisture advection in the NCEP R2 reanalysis and similar changes were found (Fig. 6f). These anomalies are dominated by mean flow and anomalous water vapor, that is, −v · q′, not anomalous flow and mean water vapor, that is, −v′ · q (not shown). Thus, only one of the two mechanisms of low-level moisture convergence proposed by Jiang et al. (2004) is found here. This positive moisture advection can be easily understood. Maximum moisture anomalies induced by an ISO usually coincide well with the precipitation maximum (not shown). With the background of southerly winds over the IO, a convergence of moisture flux tends to occur north of the ISO precipitation maximum, while a divergence of moisture flux should occur south of the ISO precipitation maximum. Precipitation then tends to increase north of the precipitation maximum and ISOs could move northward, so this mechanism could favor northward-propagating ISOs, as discussed in Jiang et al. (2004). Away from the equator, a mechanism associated with vertical wind shear could be responsible for northward-propagating ISOs (e.g., Jiang et al. 2004; Wang et al. 2006). Under a background of mean easterly vertical wind shear, positive barotropic vorticity is generated north of the ISO precipitation maximum via the vertical advection that is associated with meridional gradient of baroclinic divergence and this background flow. The free atmosphere barotropic vorticity then induces moisture convergence in the atmospheric boundary layer (ABL). Figure 8a shows a strong easterly vertical wind shear, which extends the farthest north around August; so mean easterly vertical wind shear could indeed be a cause for the northward propagation of ISOs over the IO.

Bellon and Sobel (2008a) proposed another mechanism associated with meridional advection of baroclinic vorticity in the free atmosphere. In this mechanism, positive barotropic vorticity north of the ISO precipitation maximum is induced by the advection of anomalous baroclinic vorticity via mean baroclinic meridional winds. This barotropic vorticity then creates convergence in the ABL via a Coriolis effect. Figure 8b shows differences of mean meridional winds averaged over 75°–90°E between 200 and 850 hPa. The southerly and northerly winds in the lower and upper troposphere are consistent with the mechanism of the meridional advection of vorticity in Bellon and Sobel (2008a). We note that small baroclinic meridional winds occurring near 20°N might weaken the contribution of this mechanism to the northward-propagating ISOs, particularly between 15° and 25°N.

5. The western North Pacific sector

a. Observations

Besides the IO, ISOs can also propagate northward over the WNP. The 15 strongest northward-propagating ISOs marked by black lines in Fig. 3 were composed here. Figure 9 shows eight phases of one complete northward-propagating ISO cycle. Positive precipitation anomalies propagate eastward from the IO into the western Pacific (phases 1–3), which is consistent with the Madden–Julian oscillation propagation shown in Fig. 1. The positive precipitation anomalies start to propagate northward in phases 4–5. The northward propagation of the positive precipitation anomalies is not only found over the WNP, but they also extend northwestward to the South China Sea (SCS), the Bay of Bengal, and the northwestern coast of the Indian Peninsula, similar to Fig. 2 in Wang et al. (2006). This implies a possible association of the northward-propagating ISOs between the WNP and the IO. Comparing Fig. 3 with Fig. 2, 11 of 15 ISOs in the WNP, which is around 73%, are connected to the ISOs in the IO. Why do the northward-propagating ISOs in the IO and the WNP connect to each other so closely? This is an interesting question, which will be examined in the future. An anomalous cyclone is found slightly north of the maximum precipitation anomalies. Warm SST anomalies occur north of the ISO precipitation maximum and extend much farther north, more than 20° away from the ISO precipitation maximum. Cold SST anomalies are also found south of the positive precipitation anomalies.

From phase 6 to 7, the positive precipitation anomalies and the corresponding cyclone continue to propagate northward, but the associated area shrinks westward slightly, similar to Fig. 2 in Hsu and Weng (2001). Warm SST anomalies can still be found north of the positive precipitation anomalies (north of 20°N), while cold SST anomalies coincide relatively well with the positive precipitation anomalies, which is slightly different from the previous phase. In other words, the cold SST anomalies start to catch up with the positive precipitation anomalies. The area with the cold SST anomalies is broader than the area of the positive precipitation anomalies.

To further investigate the northward-propagating ISOs over the WNP, we used one-dimensional composites for different variables. Figure 10a shows the precipitation composite. As expected, the maximum precipitation anomalies are located at zero latitude (a reference latitude), that is, the ISO convection center, and are roughly symmetric between the southern and northern sides. Zonal winds shown in Fig. 10b imply that a cyclonic circulation coincides with ISO, with its center around 3° north of the precipitation maximum, which is consistent with the results of Fig. 9. Maximum vorticity anomalies and minimum divergence anomalies shown in Figs. 10c and 10d are also around a few degrees north of the precipitation maximum, which supports the main cause for northward-propagating ISOs over the WNP (e.g., Hsu et al. 2004; Tsou et al. 2005).

Figure 11a shows SST anomalies relative to the position of the precipitation maximum. Weak warm SST anomalies are found more than 5° north of the precipitation maximum—much farther away from the precipitation maximum, compared to those in the IO. Most areas with positive precipitation anomalies coincide with cold SST anomalies, but the minimum SST anomalies are located around 5° south of the precipitation maximum. This result is consistent with the finding in Fig. 9. Warm SST anomalies are located relatively farther away from the ISO center, whereas cold SST anomalies are much closer to the ISO center. The SST tendency shown in Fig. 11b implies a cooling over almost the entire domain that we are interested in, with stronger cooling more north than south of the precipitation maximum. Latent heat flux also shows similar changes with a reversed sign, that is, positive anomalies over the entire domain with stronger magnitudes north of the precipitation maximum (Fig. 11c), which is similar to the result of Hsu and Weng (2001). We note that both latent heat fluxes, that is, the HOAPS3 and the OAFlux, are very similar. These latent heat flux changes have a similar pattern to those changes in surface wind speed shown in Fig. 11d. In other words, the latent heat flux over this region is mainly controlled by surface winds, not SST anomalies. Unlike surface wind speed in the IO, enhanced surface wind speed occurs almost over the entire WNP. This implies that the mean monsoon trough over the WNP strongly links to the intraseasonal oscillation of surface winds (not shown). Figure 11e shows net solar radiation at the surface, which is clearly consistent with the precipitation anomalies. Both latent heat flux and surface solar radiation indicate a cooling for oceans. Examining the distributions of both surface fluxes, the latent heat flux is clearly more similar to the SST tendency than the surface solar radiation. This might imply that latent heat flux is more dominant in affecting SST than solar radiation over this region. We further examine the evolution of surface heat fluxes and the SST tendency shown in Fig. 12. It is evident that the SST tendency coincides much better with the latent heat flux anomalies than the net surface solar radiation anomalies when ISOs propagate northward from phase 4 to 7. When examining their amplitudes, however, the changes in surface solar radiation are slightly stronger than the changes in latent heat flux. This result implies that net surface solar radiation is at least as important as latent heat flux even though its spatial distribution (Fig. 11e) is inconsistent with the pattern of the SST tendency (Fig. 11b). Another term that could affect SST is the entrainment of subsurface water. Several studies (e.g., Duvel and Vialard 2007; Bellon et al. 2008) show that the entrainment is associated with surface winds. Thus, the changes of the subsurface water entrainment should be similar to the surface wind speed changes shown in Fig. 11d. In other words, the combination of the changes in surface latent heat flux and subsurface water entrainment could be larger than the changes in surface solar radiation, so the SST tendency is similar to the latent heat flux anomalies, not the net surface solar radiation anomalies. We note that a direct calculation of ocean heat transport is still needed to verify the hypothesis of the subsurface water entrainment effect on SST.

b. Mechanisms

To understand mechanisms for northward-propagating ISOs over the WNP, we first examine processes associated with air–sea interaction, including latent heat flux and sensible heat flux. In latent heat flux, positive anomalies are found over almost the entire domain (Figs. 11c and 12), so the convection associated with ISOs should be enhanced. Comparing the latent heat flux anomalies located south and north of the ISO convection center, stronger positive anomalies are found north of the precipitation maximum (Figs. 11c and 12), which could favor the ISOs propagating northward. This is more evident from phase 5 to 7 when ISOs start to propagate northward (Fig. 12). In sensible heat flux, warm SST anomalies north of the ISO precipitation maximum are too weak to induce significant sensible heat flux anomalies even though surface wind speed is increased over this region. Thus, it should not be the main mechanism to make ISOs propagate northward. The SST gradient derived from Fig. 11c implies that low-level convergence is much farther north, similar to that in the IO, excluding it from being a dominant factor for ISOs propagating northward. Overall, air–sea interaction could speed up the northward propagation of ISOs over the WNP, via surface latent heat flux in particular. Another mechanism is associated with horizontal moisture advection, which has been proposed to explain the northward-propagating ISOs in the IO (Jiang et al. 2004) and in the WNP (Hsu and Weng 2001). In this study, however, the anomalies of both pseudo and low-level horizontal moisture advection are almost all negative over the entire domain (Fig. 11f). Over the WNP, mean flow is associated with the Asian monsoon trough, so the horizontal moisture advection is not guaranteed to be positive north of the ISO precipitation maximum, in contrast to the IO. This dry advection indicates that horizontal moisture advection should not induce the northward-propagating ISOs over the WNP. Instead, it would suppress the ISO.

Besides the air–sea interaction and the thermodynamic effect discussed above, some dynamic mechanisms that were found for northward-propagating ISOs in the IO might also be dominant over the WNP. The first possible mechanism is associated with vertical wind shear (Jiang et al. 2004). Figure 13a shows the vertical wind shear between 200 and 850 hPa. A weak easterly wind shear is found south of 10°N after mid-June. For the rest of WNP, the vertical wind shear is either near zero or westerly, which could not induce barotropic vorticity anomalies north of the ISO precipitation maximum via the vertical advection induced by the easterly wind shear (Jiang et al. 2004). In other words, the vertical wind shear effect over the WNP cannot be as dominant as in the IO for inducing northward-propagating ISOs. The other dynamic mechanism is associated with horizontal advection of vorticity (Bellon and Sobel 2008a), in which anomalous baroclinic vorticity is advected by mean baroclinic meridional winds. Figure 13b shows differences of mean meridional winds between 200 and 850 hPa. Most areas are dominated by negative values, which indicate southerly winds at 850 hPa and northerly winds at 200 hPa. This baroclinic structure of meridional winds can induce horizontal advection of baroclinic vorticity, especially after June. The horizontal vorticity advection then further induces positive barotropic vorticity anomalies north of the ISO precipitation maximum, so ISOs can propagate northward. A very weak baroclinic structure of meridional winds along 10°N during early summer, which is associated with the southerly winds at 200 hPa and weak southerly winds at 850 hPa, could limit the northward propagation of ISOs. The seasonal variation of the northward extent of ISOs will be examined in the future.

6. Discussion

In previous studies, several mechanisms for the northward-propagating ISOs over the IO and WNP have been proposed. Most of them have also been proved by simplified model studies (e.g., Drbohlav and Wang 2005; Bellon and Sobel 2008a). However, is there a dominant mechanism? Or are they all equally important? Here, we examined a large-scale environment and discussed the possibility of each mechanism for inducing northward-propagating ISOs. According to the previous two sections, most mechanisms differ in the strength of their influence between the IO and the WNP. For instance, the moisture advection effect is found in the IO, but it is much weaker in the WNP. The vertical wind shear effect dominates in the IO but not quite as well in the WNP. The air–sea interaction via surface latent heat flux tends to slow down the northward propagation of ISOs over the IO, but it accelerates the northward propagation of the ISOs over the WNP. However, there is one exception: the baroclinic advection of baroclinic vorticity (Bellon and Sobel 2008a). The baroclinic meridional winds are mainly associated with the local Hadley circulation. Thus, to have the advection of baroclinic vorticity, the establishment of the boreal summer Hadley circulation is crucial; in other words, the main ascending branch must be located in the Northern Hemisphere, and the main descending branch must be south of the ascending branch.

Over the IO, the local Hadley circulation is linked to the South Asian summer monsoon. The South Asian summer monsoon system not only results in the baroclinic structure of mean meridional winds, but it also creates the easterly vertical wind shear. Both have been proved to be favorable conditions for inducing the northward propagation of ISOs. Over the WNP, on the other hand, the origin of the local Hadley circulation is related to an oceanic summer monsoon, the WNP summer monsoon, which occurs during the second-half of the boreal summer (LinHo and Wang 2002; Wang and LinHo 2002). During this period, the Asian summer monsoon trough extends much farther eastward (e.g., Chou et al. 2009), but the associated easterly vertical wind shear is much weaker and does not extend as far northward as over the IO. This oceanic summer monsoon could be associated with the seasonal migration of warm SST (Chao 2000). The location of the major monsoon rainband is determined by the earth’s rotation and the location of the SST peak. Figure 14 shows the seasonal variation of mean SST averaged over the WNP. Warm SST (higher than 29°C) jumps northward to almost 20°N in late May and reaches the farthest north and its maximum strength in July, which is just before the grand onset of the WNP summer monsoon (LinHo and Wang 2002). As the warm SST extends northward with the progress of the season, the monsoon flow eventually becomes unstable and ISOs start to propagate northward from the equatorial rainband to the monsoon rainband (Bellon and Sobel 2008b). This implies that the northward propagation of ISOs might be associated with the seasonal variation of the mean state, particularly the WNP summer monsoon onset.

Another interesting finding from the comparison of the northward-propagating ISOs between the IO and the WNP is related to air–sea interaction. In general, subsurface water entrainment and surface heat fluxes, such as latent heat, sensible heat, and longwave and shortwave radiation, could affect SST. The partition of these contributions, however, may differ from region to region. Here, we have found that the main causes for SST changes are different between the IO and the WNP. Over the IO, in the region of 75°–90°E that includes the eastern Indian Ocean and the Bay of Bengal, SST changes are dominated by surface heat fluxes—surface solar radiation in particular. Over the WNP, on the other hand, SST changes are dominated by surface wind–induced forcings, latent heat flux, and subsurface water entrainment via changes of the mixed-layer depth. A further study is needed since no direct evidence of subsurface water entrainment changes is provided here.

The differences in air–sea interaction between the IO and the WNP have different impacts on the northward-propagating ISOs. Over the IO, positive (negative) latent heat flux anomalies lag (lead) the associated ISOs, so the process associated with surface latent heat flux tends to slow down ISOs propagating northward, which is consistent with the finding of Bellon et al. (2008). Over the WNP, on the other hand, relatively strong positive latent heat flux anomalies lead the ISOs and relatively weak positive latent heat flux anomalies lag the ISOs, so the effect of surface latent heat flux tends to speed up the northward propagation of the ISOs. Since the surface latent heat flux anomalies are associated with surface wind speed anomalies over both the IO and the WNP, mean low-level flow plays a very important role here. The southwesterly winds associated with the South Asian summer monsoon circulation dominate almost the entire IO, while the southwesterly and southeasterly winds associated with the WNP summer monsoon (e.g., Chou et al. 2009; Wang and LinHo 2002) dominate the southern and northern parts of the WNP, respectively. These different mean flows over the IO and the WNP create different latent heat flux anomalies, which have different impacts on northward-propagating ISOs.

In this study, we chose the region of 130°–170°E to represent the WNP because the positive precipitation anomalies extend eastward to as far as 170°E during the peak phase (Fig. 9). This region is wider than what most changes of other variables indicate. We did another set of analysis for a narrower region of 120°–150°E and only two more ISOs will be chosen for composites. The associated composites, such as shown in Figs. 9 –12, are almost unchanged. For the region of 120°–150°E, the easterly vertical wind shear does become stronger and extends a little more northward, but it is still much weaker than that in the IO. The baroclinic meridional winds, on the other hand, become much stronger, more than double of that in the IO. Overall, it is clear that the vorticity advection effect is more dominant in the WNP than in the IO, while the vertical wind shear effect becomes less important.

7. Conclusions

Intraseasonal oscillation (ISO) of the Indian Ocean (IO) and the western North Pacific (WNP) has not only eastward-propagating components but also northward-propagating components. Here, we have focused on those major existing mechanisms that could induce the northward propagation of ISOs over these two regions. These mechanisms include the following: 1) the vorticity advection effect, which is associated with the meridional advection of anomalous baroclinic vorticity by mean baroclinic meridional winds; 2) the vertical wind shear effect, which is the vertical advection associated with meridional gradient of baroclinic divergence and mean easterly vertical wind shear; 3) the moisture advection effect associated with moisture convergence by mean horizontal winds; and 4) the air–sea interaction via surface latent heat. By comparing large-scale conditions between the IO and the WNP, we can understand which mechanism is relatively more dominant in inducing northward-propagating ISOs. Over the IO, the dominant mechanisms are the vorticity advection effect, the vertical wind shear effect, and the moisture advection effect. The vorticity advection effect and the air–sea interaction, on the other hand, dominate the WNP. Among these mechanisms, only the vorticity advection effect is commonly found over both regions.

These mechanisms are all associated with the boreal summer basic state over these two regions. The South Asian summer monsoon dominates the IO, whereas the WNP summer monsoon dominates the WNP. The South Asian summer monsoon is associated with a cyclonic circulation at the lower troposphere and an anticyclonic circulation at the upper troposphere. This vertical structure of the summer monsoon circulation implies baroclinic meridional winds associated with the local Hadley circulation and strong easterly vertical wind shear. The local Hadley circulation tends to induce the vorticity advection effect, which generates anomalous barotropic vorticity north of the ISO center. This anomalous barotropic vorticity then induces low-level convergence, so the ISO propagates northward (Bellon and Sobel 2008a). The easterly vertical wind shear, which is associated with the vertical wind shear effect, also creates anomalous barotropic vorticity north of the ISO center, which further induces low-level convergence; so the ISO propagates northward, such as discussed in Jiang et al. (2004). The WNP summer monsoon, on the other hand, is an oceanic monsoon system, which is associated with the eastward extension of the Asian summer monsoon trough, a cyclonic circulation at lower troposphere. The baroclinic structure of meridional winds associated with the local Hadley circulation is clear over the WNP, but the vertical wind shear is weak and could be either easterly or westerly. Thus, the vorticity advection effect becomes more dominant, but the vertical wind shear effect becomes less dominant, compared to that over the IO.

The differences in low-level circulations between the IO and the WNP produce very different impacts on the northward-propagating ISOs over these two regions via moisture convergence and air–sea interaction. Over the IO, the dominant southwesterly winds tend to induce a positive horizontal moisture advection north of the ISO convection center, so the moisture advection effect favors northward-propagating ISOs. This mean low-level flow also creates an air–sea interaction with a positive latent heat flux south of the ISO convection center and a negative latent heat flux north of the ISO center, which tends to slow down the northward propagation of ISOs. We also find that the changes of SST over the IO are mainly associated with the cloud-radiative effect via solar radiation—surface latent heat flux is secondary. Over the WNP, the mean low-level cyclonic circulation induces negative horizontal moisture advection, so the moisture advection effect does not induce the northward-propagating ISOs. However, this low-level cyclonic circulation enhances surface wind speed to induce positive latent heat flux, which is stronger at the north than the south of the ISO center, so the air–sea interaction tends to speed up the northward propagation of ISOs. The processes that are responsible for the changes of SST over the WNP are also different from those in the IO. The changes of the WNP SST are mainly determined by effects related to wind stirring, which include surface latent heat flux and subsurface water entrainment.

Overall, we have two interesting findings in this study. First, the vorticity advection effect is the most common mechanism for the northward propagation of ISOs, based on the comparison of observations between the IO and the WNP. Other mechanisms should also be important, but they may differ between the IO and the WNP. The other interesting finding is the different air–sea interactions in the intraseasonal time scale between the IO and the WNP.

Acknowledgments

This work was supported under the National Science Council Grant 98-2628-M-001-001. The authors thank three anonymous reviewers for improving the quality of this paper.

REFERENCES

  • Andersson, A., S. Bakan, K. Fennig, H. Grassl, C-P. Klepp, and J. Schulz, 2007: Hamburg ocean atmosphere parameters and fluxes from satellite data—HOAPS-3—Twice daily composite. World Data Center for Climate. [Available online at http://cera-www.dkrz.de/WDCC/ui/Compact.jsp?acronym=HOAPS3_DAILY].

    • Search Google Scholar
    • Export Citation
  • Annamalai, H., and J. M. Slingo, 2001: Active/break cycles: Diagnosis of the intraseasonal variability of the Asian summer monsoon. Climate Dyn., 18 , 85102.

    • Search Google Scholar
    • Export Citation
  • Back, L. E., and C. S. Bretherton, 2009: On the relationship between SST gradients, boundary layer winds, and convergence over the tropical oceans. J. Climate, 22 , 41824196.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and J. Srinivasan, 2006: Comment on “Structures and mechanisms of the northward-propagating boreal summer intraseasonal oscillation”. J. Climate, 19 , 47384743.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and A. H. Sobel, 2008a: Instability of the axisymmetric monsoon flow and intraseasonal oscillation. J. Geophys. Res., 113 , D07108. doi:10.1029/2007JD009291.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., and A. H. Sobel, 2008b: Poleward-propagating intraseasonal monsoon disturbances in an intermediate-complexity axisymmetric model. J. Atmos. Sci., 65 , 470489.

    • Search Google Scholar
    • Export Citation
  • Bellon, G., A. H. Sobel, and J. Vialard, 2008: Ocean–atmosphere coupling in the monsoon intraseasonal oscillation: A simple model study. J. Climate, 21 , 52545270.

    • Search Google Scholar
    • Export Citation
  • Chao, W. C., 2000: Multiple quasi equilibria of the ITCZ and the origin of monsoon onset. J. Atmos. Sci., 57 , 641651.

  • Chou, C., L-F. Huang, L. Tseng, J-Y. Tu, and P-H. Tan, 2009: Annual cycle of rainfall in the western North Pacific and East Asian sector. J. Climate, 22 , 20732094.

    • Search Google Scholar
    • Export Citation
  • Drbohlav, H-H., and B. Wang, 2005: Mechanism of the northward-propagating intraseasonal oscillation: Insights from a zonally symmetric model. J. Climate, 18 , 952972.

    • Search Google Scholar
    • Export Citation
  • Duvel, J-P., and J. Vialard, 2007: Indo-Pacific sea surface temperature perturbations associated with intraseasonal oscillations of tropical convection. J. Climate, 20 , 30563082.

    • Search Google Scholar
    • Export Citation
  • Fu, X., B. Wang, and T. Li, 2002: Impacts of air–sea coupling on the simulation of mean Asian summer monsoon in the ECHAM4 model. Mon. Wea. Rev., 130 , 28892904.

    • Search Google Scholar
    • Export Citation
  • Goswami, B. N., 2005: South Asian monsoon. Intraseasonal Variability in the Atmosphere–Ocean Climate System, W. K. M. Lau and D. E. Waliser, Eds., Springer, 19–61.

    • Search Google Scholar
    • Export Citation
  • Hoffman, R. N., and S. M. Leidner, 2005: An introduction to the near-real-time QuikSCAT data. Wea. Forecasting, 20 , 476493.

  • Hsu, H-H., 2005: East Asian monsoon. Intraseasonal Variability in the Atmosphere–Ocean Climate System, W. K. M. Lau and D. E. Waliser, Eds., Springer, 63–94.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., and C-H. Weng, 2001: Northwestward propagation of the intraseasonal oscillation in the western North Pacific during the boreal summer: Structure and mechanism. J. Climate, 14 , 38343850.

    • Search Google Scholar
    • Export Citation
  • Hsu, H-H., C-H. Weng, and C-H. Wu, 2004: Contrasting characteristics between the northward and eastward propagation of the intraseasonal oscillation during the boreal summer. J. Climate, 17 , 727743.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., T. Li, and B. Wang, 2004: Structures and mechanisms of the northward-propagating boreal summer intraseasonal oscillation. J. Climate, 17 , 10221039.

    • Search Google Scholar
    • Export Citation
  • Kanamitsu, M., W. Ebisuzaki, J. Woollen, S-K. Yang, J. J. Hnilo, M. Fiorino, and G. L. Potter, 2002: NCEP-DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Kemball-Cook, S., and B. Wang, 2001: Equatorial waves and air–sea interaction in the boreal summer intraseasonal oscillation. J. Climate, 14 , 29232942.

    • Search Google Scholar
    • Export Citation
  • Klingaman, N. P., H. Weller, J. N. Slingo, and P. M. Inness, 2008: The intraseasonal variability of the Indian summer monsoon using TMI sea surface temperature and ECMWF reanalysis. J. Climate, 21 , 25192539.

    • Search Google Scholar
    • Export Citation
  • Krishnamurti, T. N., and D. Subrahmanyam, 1982: The 30–50-day mode at 850 mb during MONEX. J. Atmos. Sci., 39 , 20882095.

  • Krishnamurti, T. N., P. K. Jayakumar, J. Sheng, N. Surgi, and A. Kumar, 1985: Divergent circulations on the 30- to 50-day time scale. J. Atmos. Sci., 42 , 364375.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., and Coauthors, 2000: The status of the Tropical Rainfall Measuring Mission (TRMM) after two years in orbit. J. Appl. Meteor., 39 , 19651982.

    • Search Google Scholar
    • Export Citation
  • Lau, K. M., and P. H. Chan, 1986: Aspects of the 40–50-day oscillation during the northern summer as inferred from outgoing longwave radiation. Mon. Wea. Rev., 114 , 13541367.

    • Search Google Scholar
    • Export Citation
  • Lawrence, D. M., and P. J. Webster, 2002: The boreal summer intraseasonal oscillation: Relationship between northward and eastward movement of convection. J. Atmos. Sci., 59 , 15931606.

    • Search Google Scholar
    • Export Citation
  • Lindzen, R. S., and S. Nigam, 1987: On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics. J. Atmos. Sci., 44 , 24182436.

    • Search Google Scholar
    • Export Citation
  • LinHo, and B. Wang, 2002: The time-space structure of the Asian-Pacific summer monsoon: A fast annual cycle view. J. Climate, 15 , 20012019.

    • 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., 112 , 814837.

  • Maloney, E. D., and D. L. Hartmann, 1998: Frictional moisture convergence in a composite life cycle of the Madden–Julian oscillation. J. Climate, 11 , 23872403.

    • Search Google Scholar
    • Export Citation
  • Maloney, E. D., and J. T. Kiehl, 2002: MJO-related SST variations over the tropical eastern Pacific during Northern Hemisphere summer. J. Climate, 15 , 675689.

    • Search Google Scholar
    • Export Citation
  • Rajendran, K., and A. Kitoh, 2006: Modulation of tropical intraseasonal oscillations by ocean–atmosphere coupling. J. Climate, 19 , 366391.

    • Search Google Scholar
    • Export Citation
  • Roxy, M., and Y. Tanimoto, 2007: Role of SST over the Indian Ocean in influencing the intraseasonal variability of the Indian summer monsoon. J. Meteor. Soc. Japan, 85 , 349358.

    • Search Google Scholar
    • Export Citation
  • Sengupta, D., B. N. Goswami, and R. Senan, 2001: Coherent intraseasonal oscillations of ocean and atmosphere during the Asian summer monsoon. Geophys. Res. Lett., 28 , 41274130.

    • Search Google Scholar
    • Export Citation
  • Tsou, C-H., P-C. Hsu, W-S. Kau, and H-H. Hsu, 2005: Northward and northwestward propagation of 30–60-day oscillation in the tropical and extratropical western North Pacific. J. Meteor. Soc. Japan, 83 , 711726.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and D. E. Harrison, 2002: Monsoon break and subseasonal sea surface temperature variability in the Bay of Bengal. J. Climate, 15 , 14851493.

    • Search Google Scholar
    • Export Citation
  • Waliser, D. E., R. Murtugudde, and L. E. Lucas, 2004: Indo-Pacific ocean response to atmospheric intraseasonal variability: 2. Boreal summer and the intraseasonal oscillation. J. Geophys. Res., 109 , C03030. doi:10.1029/2003JC002002.

    • Search Google Scholar
    • Export Citation
  • Wang, B., and X. Xie, 1997: A model for the boreal summer intraseasonal oscillation. J. Atmos. Sci., 54 , 7286.

  • Wang, B., and LinHo, 2002: Rainy season of the Asian-Pacific summer monsoon. J. Climate, 15 , 386398.

  • Wang, B., P. Webster, K. Kikuchi, T. Yasunari, and Y. Qi, 2006: Boreal summer quasi-monthly oscillation in the global tropics. Climate Dyn., 27 , 661675.

    • Search Google Scholar
    • Export Citation
  • Wentz, F. J., 1997: A well-calibrated ocean algorithm for special sensor microwave/imager. J. Geophys. Res., 102 , 87038718.

  • Wentz, F. J., C. Gentemann, D. Smith, and D. Chelton, 2000: Satellite measurements of sea-surface temperature through clouds. Science, 288 , 847850.

    • Search Google Scholar
    • Export Citation
  • Wheeler, M., and G. N. Kiladis, 1999: Convectively coupled equatorial waves: Analysis of clouds and temperature in the wavenumber-frequency domain. J. Atmos. Sci., 56 , 374399.

    • Search Google Scholar
    • Export Citation
  • Yasunari, T., 1981: Structure of an Indian summer monsoon system with around 40-day period. J. Meteor. Soc. Japan, 59 , 336354.

  • Yu, L., and R. A. Weller, 2007: Objectively analyzed air–sea fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., 2005: Madden–Julian oscillation. Rev. Geophys., 43 , RG2003. doi:10.1029/2004RG000158.

  • Zhang, Y., W. B. Rossow, A. A. Lacis, V. Oinas, and M. I. Mishchenko, 2004: Calculation of radiative fluxes from the surface to top of atmosphere based on ISCCP and other global datasets: Refinement of the radiative transfer model and the input data. J. Geophys. Res., 109 , D19105. doi:10.1029/2003JD004457.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

The 30–90-day-filtered daily precipitation (mm day−1) averaged over 5°S–5°N for the summer period (May–September) of 2000–07. The IO and the WNP are marked by black dashed lines.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 2.
Fig. 2.

The 30–90-day-filtered daily precipitation (mm day−1) averaged over the IO (75°–90°E) for the summer period (May–September) of 2000–07. Major northward-propagating ISOs are marked by black solid lines.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 3.
Fig. 3.

As in Fig. 2, but averaged over the WNP (130°–170°E).

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 4.
Fig. 4.

A composite of eight phases for one cycle of the northward-propagating ISO over the IO. Shaded is precipitation (mm day−1), and the vectors are surface wind (m s−1). The blue dot denotes cold SST anomalies, and the red cross indicates warm SST anomalies. All variables are after 30–90-day filtering.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 5.
Fig. 5.

Composites of the northward-propagating ISOs over the IO based on the ISO center marked in Fig. 2 for (a) precipitation (mm day−1), (b) the longitudinal component of surface winds (m s−1), (c) surface vorticity (10−6 s−1), and (d) surface divergence (10−6 s−1). All variables are after 30–90-day filtering.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 6.
Fig. 6.

As in Fig. 5, but for (a) SST (°C), (b) SST tendency (°C day−1), (c) surface latent heat flux (W m−2), (d) surface wind speed (m s−1), (e) surface solar radiation (W m−2), and (f) pseudohorizontal moisture advection (mm day−1) (see text).

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 7.
Fig. 7.

As in Fig. 4, but for surface latent heat flux from the OAFlux dataset (W m−2; shaded), SST tendency (°C day−1; blue dot and red cross), and net surface solar radiation (W m−2; contours).

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 8.
Fig. 8.

Mean state of the summer circulation over the IO: (a) vertical wind shear (200 − 850-hPa zonal winds; m s−1), and (b) differences of meridional winds between 200 and 850 hPa (200 − 850 hPa; m s−1).

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 9.
Fig. 9.

As in Fig. 4, but for the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 10.
Fig. 10.

As in Fig. 5, but for the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 11.
Fig. 11.

As in Fig. 6, but for the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 12.
Fig. 12.

As in Fig. 7, but for the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 13.
Fig. 13.

As in Fig. 8, but for the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Fig. 14.
Fig. 14.

Seasonal variation of mean SST (2000–07; °C) over the WNP.

Citation: Journal of Climate 23, 24; 10.1175/2010JCLI3596.1

Save