• 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, doi:10.1175/2009JCLI2392.1.

    • Search Google Scholar
    • Export Citation
  • Berg, W., , C. Kummerow, , and C. A. Morales, 2002: Differences between east and west Pacific rainfall systems. J. Climate, 15, 36593672, doi:10.1175/1520-0442(2002)015<3659:DBEAWP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , S. Fueglistaler, , G. Lesins, , and T. Mitovski, 2008: A low-level circulation in the tropics. J. Atmos. Sci., 65, 10191034, doi:10.1175/2007JAS2463.1.

    • Search Google Scholar
    • Export Citation
  • Grabowski, W., , J.-I. Yano, , and M. W. Moncrieff, 2000: Cloud resolving modeling of tropical circulations driven by large-scale SST gradients. J. Atmos. Sci., 57, 20222039, doi:10.1175/1520-0469(2000)057<2022:CRMOTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hagos, S., , and C. Zhang, 2010: Diabatic heating, divergent circulation and moisture transport in the African monsoon system. Quart. J. Roy. Meteor. Soc., 136 (S1), 411425, doi:10.1002/qj.538.

    • Search Google Scholar
    • Export Citation
  • Hagos, S., and Coauthors, 2010: Estimates of tropical diabatic heating profiles: Commonalities and uncertainties. J. Climate, 23, 542558, doi:10.1175/2009JCLI3025.1.

    • Search Google Scholar
    • Export Citation
  • Hartmann, D. L., , H. H. Hendon, , and R. A. Houze Jr., 1984: Some implications of the mesoscale circulations in tropical cloud clusters for large-scale dynamics and climate. J. Atmos. Sci., 41, 113121, doi:10.1175/1520-0469(1984)041<0113:SIOTMC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Iguchi, T., , T. Kozu, , R. Meneghini, , J. Awaka, , and K. Okamoto, 2000: Rain-profiling algorithm for the TRMM precipitation radar. J. Appl. Meteor., 39, 20382052, doi:10.1175/1520-0450(2001)040<2038:RPAFTT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., and Coauthors, 2011: Vertical diabatic heating structure of the MJO: Intercomparison between recent reanalyses and TRMM estimates. Mon. Wea. Rev., 139, 32083223, doi:10.1175/2011MWR3636.1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kubar, T. L., , D. L. Hartmann, , and R. Wood, 2007: Radiative and convective driving of tropical high clouds. J. Climate, 20, 55105526, doi:10.1175/2007JCLI1628.1.

    • Search Google Scholar
    • Export Citation
  • Larson, K., , and D. L. Hartmann, 2003: Interactions among cloud, water vapor, radiation, and large-scale circulation in the tropical climate. Part II: Sensitivity to spatial gradients of sea surface temperature. J. Climate, 16, 14411455, doi:10.1175/1520-0442-16.10.1441.

    • Search Google Scholar
    • Export Citation
  • Ling, J., , and C. Zhang, 2011: Structural evolution in heating profiles of the MJO in global reanalyses and TRMM retrievals. J. Climate, 24, 825842, doi:10.1175/2010JCLI3826.1.

    • Search Google Scholar
    • Export Citation
  • Ling, J., , and C. Zhang, 2013: Diabatic heating profiles in recent global reanalyses. J. Climate, 26, 33073325, doi:10.1175/JCLI-D-12-00384.1.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. J. Zipser, , D. J. Cecil, , S. W. Nesbitt, , and S. Sherwood, 2008: A cloud and precipitation feature database from nine years of TRMM observations. J. Appl. Meteor. Climatol., 47, 27122728, doi:10.1175/2008JAMC1890.1.

    • 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, doi:10.1175/1520-0469(1972)029<1109:DOGSCC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., , R. Cifelli, , and S. A. Rutledge, 2006: Storm morphology and rainfall characteristics of TRMM precipitation features. Mon. Wea. Rev., 134, 27022721, doi:10.1175/MWR3200.1.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., , C. Zhang, , and S.-H. Chen, 2007: Dynamics of the shallow meridional circulation around intertropical convergence zones. J. Atmos. Sci., 64, 22622285, doi:10.1175/JAS3964.1.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., , S. W. Powell, , C. Zhang, , and B. E. Mapes, 2010: Idealized simulations of the intertropical convergence zone and its multilevel flows. J. Atmos. Sci., 67, 40284053, doi:10.1175/2010JAS3417.1.

    • Search Google Scholar
    • Export Citation
  • Onogi, K., and Coauthors, 2007: The JRA-25 Reanalysis. J. Meteor. Soc. Japan, 85, 369432, doi:10.2151/jmsj.85.369.

  • Raymond, D. J., and Coauthors, 2004: EPIC2001 and the coupled ocean–atmosphere system of the tropical east Pacific. Bull. Amer. Meteor. Soc., 85, 13411354, doi:10.1175/BAMS-85-9-1341.

    • Search Google Scholar
    • Export Citation
  • Rienecker, M. M., and Coauthors, 2011: MERRA: NASA’s Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, 36243648, doi:10.1175/JCLI-D-11-00015.1.

    • Search Google Scholar
    • Export Citation
  • Schneider, E. K., 1977: Axially symmetric steady-state models of the basic state for instability and climate studies. Part II: Nonlinear calculations. J. Atmos. Sci., 34, 280296, doi:10.1175/1520-0469(1977)034<0280:ASSSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schneider, E. K., , and R. S. Lindzen, 1977: Axially symmetric steady-state models of the basic state for instability and climate studies. Part I: Linearized calculations. J. Atmos. Sci., 34, 263279, doi:10.1175/1520-0469(1977)034<0263:ASSSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , and R. A. Houze Jr., 2003: Stratiform rain in the tropics as seen by the TRMM Precipitation Radar. J. Climate, 16, 17391756, doi:10.1175/1520-0442(2003)016<1739:SRITTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , R. A. Houze Jr., , and I. Kraucunas, 2004: The tropical dynamical response to latent heating estimates derived from the TRMM precipitation radar. J. Atmos. Sci., 61, 13411358, doi:10.1175/1520-0469(2004)061<1341:TTDRTL>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , P. E. Ciesielski, , and M. H. Zhang, 2008: Tropical cloud heating profiles: Analysis from KWAJEX. Mon. Wea. Rev., 136, 42894300, doi:10.1175/2008MWR2275.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , W.-K. Tao, , and D. E. Johnson, 2004: Spectral retrieval of latent heating profiles from TRMM PR data. Part I: Development of a model-based algorithm. J. Appl. Meteor., 43, 10951113, doi:10.1175/1520-0450(2004)043<1095:SROLHP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , W.-K. Tao, , and C.-L. Shie, 2007: Spectral retrieval of latent heating profiles from TRMM PR data. Part II: Algorithm improvement and heating estimates over tropical ocean regions. J. Appl. Meteor. Climatol., 46, 10981124, doi:10.1175/JAM2510.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , and W.-K. Tao, 2008: Spectral retrieval of latent heating profiles from TRMM PR data. Part III: Estimating apparent moisture sink profiles over tropical oceans. J. Appl. Meteor. Climatol., 47, 620640, doi:10.1175/2007JAMC1738.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , S. Kida, , W.-K. Tao, , X. Zeng, , C. Yokoyama, , and T. L’Ecuyer, 2009: Spectral retrieval of latent heating profiles from TRMM PR data. Part IV: Comparisons of lookup tables from two- and three-dimensional cloud-resolving model simulations. J. Climate, 22, 55775594, doi:10.1175/2009JCLI2919.1.

    • Search Google Scholar
    • Export Citation
  • Stachnik, J. P., , and C. Schumacher, 2011: A comparison of the Hadley circulation in modern reanalyses. J. Geophys. Res., 116, D22102, doi:10.1029/2011JD016677.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , J. Yokomori, , and K. Yoneyama, 2006: A diagnostic study on interactions between atmospheric thermodynamic structure and cumulus convection over the tropical western Pacific Ocean and over the Indochina peninsula. J. Meteor. Soc. Japan, 84A, 151169, doi:10.2151/jmsj.84A.151.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , S. Shige, , W.-K. Tao, , and N. Hirota, 2010: Shallow and deep latent heating modes over tropical oceans observed with TRMM PR spectral latent heating data. J. Climate, 23, 20302046, doi:10.1175/2009JCLI3110.1.

    • Search Google Scholar
    • Export Citation
  • Tomas, R. A., , and P. J. Webster, 1997: The role of inertial instability in determining the location and strength of near-equatorial convection. Quart. J. Roy. Meteor. Soc., 123, 14451482, doi:10.1002/qj.49712354202.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , D. P. Stepaniak, , and J. M. Caron, 2000: The global monsoon as seen through the divergent atmospheric circulation. J. Climate, 13, 39693993, doi:10.1175/1520-0442(2000)013<3969:TGMAST>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, Y., , S.-P. Xie, , B. Wang, , and H. Xu, 2005: Large-scale atmospheric forcing by southeast Pacific boundary-layer clouds: A regional model study. J. Climate, 18, 934951, doi:10.1175/JCLI3302.1.

    • Search Google Scholar
    • Export Citation
  • Wu, Z., 2003: A shallow CISK, deep equilibrium mechanism for the interaction between large-scale convection and large-scale circulations in the tropics. J. Atmos. Sci., 60, 377392, doi:10.1175/1520-0469(2003)060<0377:ASCDEM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Yokoyama, C., , and Y. N. Takayabu, 2012: Relationships between rain characteristics and environment. Part I: TRMM precipitation features and the large-scale environment over the tropical Pacific. Mon. Wea. Rev., 140, 28312840, doi:10.1175/MWR-D-11-00252.1.

    • Search Google Scholar
    • Export Citation
  • Yuter, S. E., , and R. A. Houze Jr., 2000: The 1997 Pan American Climate Studies Tropical Eastern Pacific Process Study. Part I: ITCZ region. Bull. Amer. Meteor. Soc., 81, 451481, doi:10.1175/1520-0477(2000)081<0451:TPACST>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Yuter, S. E., , Y. L. Serra, , and R. A. Houze Jr., 2000: The 1997 Pan American Climate Studies Tropical Eastern Pacific Process Study. Part II: Stratocumulus region. Bull. Amer. Meteor. Soc., 81, 483490, doi:10.1175/1520-0477(2000)081<0483:TPACST>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , and S. M. Hagos, 2009: Bi-modal structure and variability of large-scale diabatic heating in the tropics. J. Atmos. Sci., 66, 36213640, doi:10.1175/2009JAS3089.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , M. McGauley, , and N. A. Bond, 2004: Shallow meridional circulation in the tropical eastern Pacific. J. Climate, 17, 133139, doi:10.1175/1520-0442(2004)017<0133:SMCITT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , P. Woodworth, , and G. Gu, 2006: The seasonal cycle in the lower troposphere over West Africa from sounding observations. Quart. J. Roy. Meteor. Soc., 132, 25592582, doi:10.1256/qj.06.23.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , D. S. Nolan, , C. D. Thorncroft, , and H. Nguyen, 2008: Shallow meridional circulation in the tropical atmosphere. J. Climate, 21, 34533470, doi:10.1175/2007JCLI1870.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., and Coauthors, 2010: MJO signals in latent heating: Results from TRMM retrievals. J. Atmos. Sci., 67, 34883508, doi:10.1175/2010JAS3398.1.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Examples of TRMM PF observation for the (a),(b) mostly deep, (c),(d) intermediate, and (e),(f) mostly shallow periods. (left) Plan views of TRMM PR surface radar echoes (dBZ; color shades) and TRMM Visible and Infrared Scanner (VIRS) brightness temperature (°C; black-and-white shades). (right) Vertical distributions of ratio of the number of 20-dBZ area at each height to the near-surface raining area over each 2° × 2° reference grid (shown with broken rectangular in the plan view). In (a),(c), and (e), black contours indicate the areas with greater than or equal to 20 dBZ at 6 km, and slant broken lines indicate the PR overpass. The orbit number and observation date are shown at the top left corner of each plan view.

  • View in gallery

    Rain contribution (%) binned into rain volume for the mostly shallow (solid), the intermediate (dashed), and the mostly deep (dash–dotted) periods. Abscissa indicates rain volume (mm h−1 km2) in logarithmic scale.

  • View in gallery

    Distribution of rain volume (mm h−1 km2) from four PF types, which are defined in section 4, as a function of the ratios of 20-dBZ area at 6 km to the total rain area for each TRMM overpass over each grid. Solid, dotted, dashed, and dash–dotted lines indicate types 1, 2, 3, and 4, respectively. Starting from the left, shaded regions indicate the mostly shallow (ratios of 0–0.05), intermediate (ratios of 0.25–0.3), and mostly deep (ratios of 0.5–1) periods.

  • View in gallery

    Frequency (%) of 20-dBZ echo-top heights (km) over the reference grids associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods. The figure is made based on all TRMM PR2A25 nadir data in each reference grid.

  • View in gallery

    Mean profiles of (a) the TRMM SLH, (b) reanalysis latent heating (K day−1), and (c) reanalysis vertical pressure velocity (Pa s−1). Profiles are averaged over 8°–12°N, 130°–100°W during boreal autumn (September–November) for 1998–2010. In (b) and (c), profiles for JRA25-JCDAS, MERRA, and ERA-I are denoted with solid, dashed, and dash–dotted lines, respectively.

  • View in gallery

    Composite occurrence frequencies of precipitation rates (mm h−1) associated with the mostly shallow (black bars), intermediate (blue bars), and mostly deep (red bars) periods for (a) JRA25-JCDAS, (b) MERRA, and (c) ERA-I.

  • View in gallery

    Composite profiles of (left) latent heating (K day−1) and (right) vertical pressure velocity (Pa s−1) associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, (e) TRMM SLH, and (f) ERA-I. Gray lines indicate 90% confidence intervals. The composite is performed for profiles averaged over the 5° × 5° region around the center of each reference grid.

  • View in gallery

    Composite meridional–pressure cross sections of latent heat (K day−1; colors), vertical pressure velocity (Pa s−1; contours), and meridional circulation (arrows) for the (top) mostly deep, (middle) intermediate, and (bottom) mostly shallow periods. Cross sections for (left) JRA25-JCDAS, (middle) MERRA, and (right) ERA-I are shown. Contours are plotted at intervals of 0.02 Pa s−1. Solid contours are for negative values, and thick solid lines indicate 0 Pa s−1. The composite is performed for meridional cross sections averaged over 5° in longitudes at the center of each reference grid. Abscissa and ordinate axes indicate latitudes relative to the composite center and pressure levels, respectively.

  • View in gallery

    Composite profiles of meridional winds (m s−1), which are averaged over 5° in longitude at the center of each reference grid, and from the latitudinal center of each reference grid to (left) 10° south of convection or (right) 10° north of convection, for the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, and (e),(f) ERA-I. Gray lines indicate 90% confidence intervals.

  • View in gallery

    Composite profiles of divergence (×10−5 s−1), which is averaged over the 5° × 5° region around the center of each reference grid, for the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a) JRA25-JCDAS, (b) MERRA, and (c) ERA-I. Gray lines indicate 90% confidence intervals.

  • View in gallery

    Composite profiles of (left) latent heating anomaly (K day−1) and (right) divergence anomaly (×10−5 s−1) associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, (e) TRMM SLH, and (f) ERA-I. Gray lines indicate 90% confidence intervals. The composite is performed for profiles averaged over the 5° × 5° region around the center of each reference grid.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 27 27 7
PDF Downloads 23 23 6

TRMM-Observed Shallow versus Deep Convection in the Eastern Pacific Related to Large-Scale Circulations in Reanalysis Datasets

View More View Less
  • 1 Department of Atmospheric Sciences, University of Utah, Salt Lake City, Utah
© Get Permissions
Full access

Abstract

Over the eastern Pacific, recent studies have shown that a shallow large-scale meridional circulation with its return flow just above the boundary layer coexists with a deep Hadley circulation. This study examines how the vertical structure of large-scale circulations is related to satellite-observed individual precipitation properties over the eastern Pacific in boreal autumn. Three reanalysis datasets are used to describe differences in their behavior. The results are compared among reanalyses and three distinctly different convection periods, which are defined according to their radar echo depths. Shallow and deep circulations are shown to often coexist for each of the three periods, resulting in the multicell circulation structure. Deep (shallow) circulations preferentially appear in the mostly deep (shallow) convection period of radar echo depths. Thus, depth of convection basically corresponds to which circulation branch is dominant. This anticipated relationship between the circulation structure and depths of convection is common in all three reanalyses. Notable differences among reanalyses are found in the mid- to upper troposphere in either the time-mean state or the composite analysis based on the convection periods. Reanalyses have large variations in characteristics associated with deep circulations such as the upper-tropospheric divergence and outflows and the midlevel inflows, which are consistent with their different profiles of latent heating in the mid- to upper troposphere. On the other hand, discrepancies in shallow circulations and shallow convection are also found, but they are not as large as those in deep ones.

Current affiliation: Atmosphere and Ocean Research Institute, University of Tokyo, Chiba, Japan.

Corresponding author address: Chie Yokoyama, Atmosphere and Ocean Research Institute, University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8568, Japan. E-mail: chie@aori.u-tokyo.ac.jp

Abstract

Over the eastern Pacific, recent studies have shown that a shallow large-scale meridional circulation with its return flow just above the boundary layer coexists with a deep Hadley circulation. This study examines how the vertical structure of large-scale circulations is related to satellite-observed individual precipitation properties over the eastern Pacific in boreal autumn. Three reanalysis datasets are used to describe differences in their behavior. The results are compared among reanalyses and three distinctly different convection periods, which are defined according to their radar echo depths. Shallow and deep circulations are shown to often coexist for each of the three periods, resulting in the multicell circulation structure. Deep (shallow) circulations preferentially appear in the mostly deep (shallow) convection period of radar echo depths. Thus, depth of convection basically corresponds to which circulation branch is dominant. This anticipated relationship between the circulation structure and depths of convection is common in all three reanalyses. Notable differences among reanalyses are found in the mid- to upper troposphere in either the time-mean state or the composite analysis based on the convection periods. Reanalyses have large variations in characteristics associated with deep circulations such as the upper-tropospheric divergence and outflows and the midlevel inflows, which are consistent with their different profiles of latent heating in the mid- to upper troposphere. On the other hand, discrepancies in shallow circulations and shallow convection are also found, but they are not as large as those in deep ones.

Current affiliation: Atmosphere and Ocean Research Institute, University of Tokyo, Chiba, Japan.

Corresponding author address: Chie Yokoyama, Atmosphere and Ocean Research Institute, University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8568, Japan. E-mail: chie@aori.u-tokyo.ac.jp

1. Introduction

Diabatic heating associated with tropical convection plays an important role in redistributing the heat near the surface throughout the troposphere, and therefore affects large-scale circulation. Especially, latent heating associated with precipitation processes plays the principal role in determining profiles of diabatic heating. Many studies have investigated how vertical profiles of latent heating influence the circulations (e.g., Hartmann et al. 1984; Schumacher et al. 2004). Because different profiles of heating result from different characteristics of precipitation systems (Schumacher et al. 2008), quantitative investigation of relationships between various characteristics of precipitation systems and the environment is essential to understand the effects of tropical rain on large-scale circulations. In the tropics, Takayabu et al. (2010) showed that deep and shallow modes of diabatic heating, which correspond to deep convection and congestus, respectively, are dominant. Zhang and Hagos (2009) also applied a rotated empirical orthogonal function analysis to tropical sounding data from field campaigns to show that the two leading modes (one deep and one shallow) account for a large fraction of the total variance in diabatic heating data.

Over the tropical Pacific, satellite observations have revealed that there are significant differences in precipitation characteristics between the eastern and western Pacific. More shallow rain as well as more stratiform rain is found over the eastern Pacific than the western Pacific (Berg et al. 2002; Schumacher and Houze 2003; Nesbitt et al. 2006; Kubar et al. 2007; Yokoyama and Takayabu 2012). Over the eastern Pacific intertropical convergence zone (ITCZ), both congestus with echo-top heights (ETHs) below 8 km and moderately deep (8–14 km) organized precipitation systems contribute more to the total rainfall during the boreal autumn than over the western Pacific warm pool, where deeper precipitation systems are dominant (Yokoyama and Takayabu 2012).

Consistent with the large amount of shallow rain, it has been shown that a shallow meridional circulation (SMC) exists in addition to a deep Hadley circulation over the eastern Pacific. Zhang et al. (2004) used in situ observations to show the coexistence of the SMC with the northerly shallow return flow, located at 2–4 km in height, just above the boundary layer. The SMCs are found in global reanalyses (Tomas and Webster 1997; Trenberth et al. 2000; Zhang et al. 2008) and numerical simulations (Nolan et al. 2007, 2010). Nolan et al. (2007) show that shallow return flows have importance in terms of the moisture transport in and out of the ITCZ, so it is meaningful to understand the behavior of the large-scale circulation over the eastern Pacific in detail. Note that the SMCs appear not only over the eastern Pacific but also over the central and western Pacific, Atlantic, Africa, and the Indian subcontinent (e.g., Takayabu et al. 2006; Zhang et al. 2006, 2008). In addition, SMCs are observed not only over ITCZ regions, which we will examine in this study, but also in monsoon heat low regions such as the West Africa monsoon region (Hagos and Zhang 2010).

As one of the mechanisms that cause the SMCs, some studies emphasize the importance of significant surface temperature gradients. The SMCs appear together with deep circulations in simulations with a significant surface temperature gradient (Schneider and Lindzen 1977; Schneider 1977; Grabowski et al. 2000; Larson and Hartmann 2003). Nolan et al. (2007) considered the sea-breeze model to show that the observed surface pressure and temperature gradients around the eastern Pacific ITCZ can lead to a reversal of the pressure gradient above the boundary layer, and thus the shallow return flow is driven. They also performed the idealized full-physics simulation driven by the SST gradient to reproduce an SMC. Their simulations also show that the shallow return flow is strongest when deep convection is inactive, and is suppressed when strong deep convection develops around the ITCZ. In addition, as another mechanism, Wang et al. (2005) used a regional model to show that in response to the cloud-induced cooling in the inversion layer, an in situ anomalous high pressure system develops in the boundary layer and an anomalous shallow meridional circulation develops over the equatorial eastern Pacific.

On the other hand, SMCs are also potentially driven by shallow convection. Takayabu et al. (2006) showed the existence of a significant meridional divergence around 500–600 hPa, which almost coincides with the diagnosed level of detrainment, in addition to the deep meridional cell of the local Hadley circulation over the western Pacific. In addition, Wu (2003) showed that shallow heating profiles effectively drive shallow circulations, which range from the surface to ~500 hPa. Zhang and Hagos (2009) also simulated the circulation with multiple overturning cells, using a linear balance model forced by three composited heating profiles (shallow bottom heavy, deep middle heavy, and stratiform-like top heavy heating profiles) and showed that shallow bottom heavy heating profiles drive significant inflows in the boundary layer and outflows just above it. Therefore, it is important to show relationships between meridional circulations and actual individual observed precipitation properties.

However, it is not well known how meridional circulation structures are associated with individual periods of observed precipitation characteristics. Folkins et al. (2008) examine divergence profiles of the equatorial region (~95°E–170°W), depending upon the instantaneous rainfall rate within the region. They showed that the upper-tropospheric divergence increases as the average rain rate increases, while the magnitude of the divergence between 3 and 8 km tends to decrease. However, it has not been shown how instantaneous circulation patterns vary with observed instantaneous vertical extent of precipitation. In this study, we will investigate the relationships between such precipitation properties and large-scale circulation properties over the eastern Pacific, where much shallow rain is observed.

In this study, we utilize reanalysis data to examine the environment. One should carefully analyze variables such as precipitation and moisture, because they may be strongly influenced by model characteristics such as cumulus parameterization (Kalnay et al. 1996). Especially in the tropics, Rienecker et al. (2011) showed relatively large discrepancies between two reanalyses not only for thermodynamical variables but for dynamical variables such as tropospheric winds.

In recent years, some comparisons of reanalysis datasets have been published (e.g., Zhang et al. 2010; Ling and Zhang 2011, 2013; Jiang et al. 2011). Hagos et al. (2010) compared diabatic heating profiles among reanalyses, satellite-based products, and sounding data to find that the primary uncertainty is in the amount of shallow heating over the tropical oceans. Zhang et al. (2008) compared SMCs over some tropical regions among three reanalyses. While their comparison provided some confidence that SMCs in the reanalyses are not artifacts, discrepancies among reanalyses can be substantial over open oceans. Stachnik and Schumacher (2011) compared the mean tropical Hadley circulation intensity among reanalyses to show that the reanalyses with stronger (weaker) circulation intensities tend to have more (less) tropical precipitation than the remaining datasets. In addition, they showed that some reanalyses produce a higher circulation center compared to the remaining datasets, which may be explained by different profiles of latent heat release among the reanalyses. In this study, we are also motivated to describe commonalities and differences among the behavior of reanalysis datasets over the eastern Pacific.

The objectives of this study are as follows: 1) to determine relationships between meridional circulations and actual observed properties of precipitation systems, 2) to specifically examine the relationship between the existence and strength of the shallow meridional circulation and the depth of precipitation systems, and 3) to describe commonalities and differences between reanalyses of the shallow and deep meridional circulations over the eastern Pacific.

2. Data

Characteristics of precipitation systems are examined by using the Tropical Rainfall Measuring Mission (TRMM) radar precipitation feature (PF) level-2 data of the University of Utah database (Liu et al. 2008) based on the TRMM version 6 data. The radar PFs are defined by grouping contiguous pixels with the PR 2A25 near-surface rain greater than 0 mm h−1, where PR 2A25 is the product containing the orbital data of radar reflectivity and precipitation rate profiles observed by the TRMM Precipitation Radar (PR; Iguchi et al. 2000). The TRMM PR has horizontal resolution of 4.3 km (before the orbit boost in August 2001) or 5 km (after the boost). Only PFs with at least four rain pixels are analyzed in this study. Each PF includes many variables. Simple average latitude/longitude, number of rain pixels, rain volume, number of pixels with reflectivity greater than or equal to 20 dBZ at 1-km intervals in height, and maximum height of 20-dBZ radar reflectivity are used. Rain volume of each PF is defined as the product of the average rain rate (mm h−1) and area (km2).

To diagnose the properties of the large-scale environment, we utilize three reanalysis datasets: the Japanese 25-year Reanalysis and Japan Meteorological Agency (JMA) Climate Data Assimilation System (jointly JRA25-JCDAS; Onogi et al. 2007), the National Aeronautics and Space Administration (NASA) Modern-Era Retrospective Analysis for Research and Applications (MERRA; Rienecker et al. 2011), and the Interim European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-Interim, hereafter ERA-I; Dee et al. 2011). We utilize horizontal wind, vertical pressure velocity (ω), precipitation rate, and latent heating in the reanalyses. The sum of convective heating and large-scale condensation rates is used as latent heating for JRA25-JCDAS, while temperature tendency from moist physics is utilized for MERRA. Because no direct output of heating rate is provided for ERA-I, we do not include its heating rates in this study.

The variables except precipitation rate and latent heating are available on a 1.25° horizontal grid for JRA25-JCDAS and MERRA, and 1.5° for ERA-I. These data are obtained 6 hourly for JRA25-JCDAS and ERA-I. Although these variables for the MERRA data are provided 3 hourly, we utilize the data 6 hourly in the same manner as the other datasets. Latent heating data are on a 2.5° grid for JRA25-JCDAS and a 1.25° grid for MERRA. The latent heating data are provided as 6-h mean data for JRA25-JCDAS. The MERRA latent heating data, which are originally 3-h mean, are averaged over 6 h. For JRA25-JCDAS, 6-h mean precipitation rate data are provided on the model grid using 320 longitudinal and 160 latitudinal Gaussian grid points. Precipitation data, which are on ⅔° in longitude and ½° in latitude for MERRA, and on a 1.5° grid for ERA-I, are converted to 6-h mean data. JRA25-JCDAS, MERRA, and ERA-I have 12 levels, 25 levels, and 27 levels from 1000 to 100 hPa, respectively.

Additionally, we use the TRMM latent heat research product, which is based on the spectral latent heating (SLH) algorithm (Shige et al. 2004, 2007, 2008, 2009). The algorithm uses the TRMM PR2A25 product to retrieve heating profiles utilizing the look-up tables, which are derived from tropical precipitation and latent heating data simulated by a cloud-resolving model. The look-up tables are based on rain top heights for convective rain and shallow stratiform rain, while they are based on rain rates at the melting level for deep stratiform rain. Note that the look-up tables are derived from model runs using Tropical Ocean and Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA COARE) data. Thus, one should keep it in mind that the lookup table may be biased toward the western Pacific, where convection is deeper than that over the eastern Pacific. The horizontal resolution is 0.5°. All data used in this study are from boreal autumn (September–November) for 13 years from 1998 to 2010.

3. Methodology

The composite analysis is performed to capture the environment associated with deep and shallow precipitation observed with the TRMM satellite in boreal autumn over the eastern Pacific ITCZ region (2°–14°N, 100°–130°W), according to the following procedure. First, we divide the region into 2° × 2° grids, and calculate ratios of the total area with reflectivity greater than or equal to 20 dBZ at 1-km intervals in height to the total rain area for each TRMM overpass over each grid. Ratios are obtained by integrating PFs, which have the average longitude and latitude of PFs in each grid. Only grids with at least 100 rain pixels are used for analysis, because those with <100 pixels have a very small contribution to total rainfall. Second, based on the ratios at 6 km, the above grids are divided into “mostly shallow” (0% ≤ ratios < 5%), “intermediate” (25% ≤ ratios < 30%), and “mostly deep” (50% ≤ ratios < 100%) periods. We define the intermediate period as well as the other two periods, because organized precipitation systems over the eastern Pacific ITCZ tend to include lower echo tops than those over the western Pacific (Yokoyama and Takayabu 2012). It is interesting to investigate what patterns of large-scale circulations are found for the intermediate convection period, and how they are different from those for the mostly deep convection period.

As examples of TRMM PR snapshots, plan views and vertical distributions of percentage of the number of 20-dBZ areas at each height to the near-surface raining area over each 2° × 2° reference grid for three convection periods are shown in Fig. 1. Note that the TRMM PR often crosses only part of the reference grid. If a PF has the average longitude and latitude in a given reference grid, even though part of it resides outside of the grid, we consider that the PF belongs to the reference grid and use the whole of the PF to calculate the PF properties. Finally, we make composites of the environmental variables. The mostly deep, intermediate, and mostly shallow periods consist of 192, 580, and 3095 cases, respectively.

Fig. 1.
Fig. 1.

Examples of TRMM PF observation for the (a),(b) mostly deep, (c),(d) intermediate, and (e),(f) mostly shallow periods. (left) Plan views of TRMM PR surface radar echoes (dBZ; color shades) and TRMM Visible and Infrared Scanner (VIRS) brightness temperature (°C; black-and-white shades). (right) Vertical distributions of ratio of the number of 20-dBZ area at each height to the near-surface raining area over each 2° × 2° reference grid (shown with broken rectangular in the plan view). In (a),(c), and (e), black contours indicate the areas with greater than or equal to 20 dBZ at 6 km, and slant broken lines indicate the PR overpass. The orbit number and observation date are shown at the top left corner of each plan view.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

4. Characteristics of the three convection periods

The methodology mentioned in section 3 enables us to examine relationships between convection and the environment such as the meridional circulation over the eastern Pacific, where there have been few field campaigns and observation of cumulus convection is mostly by satellites. To know what kind of PFs are characteristic of each period, mean rain volume and mean ratios of 20-dBZ area at 4, 6, 8, 10, and 12 km height to the total rain area for the three convection periods are shown in Table 1. Obviously, the mostly deep and shallow periods have PFs with substantially different characteristics. On average, ~21% of the total rain area for the mostly deep period has heights greater than or equal to 8 km, and ~72% of the total rain area over each grid for the mostly shallow period have 20-dBZ echo-top heights less than 4 km. Figure 2 also shows rain contribution from reference grids with each rain volume for the three convection periods. There is more rain from grids with smaller (larger) rain volume for the mostly shallow period (the mostly deep and intermediate periods). Note that it is not always true that reference grids for the mostly shallow period have more shallow rain amount than those for the mostly deep and intermediate periods. This is because organized systems, which are dominant for the mostly deep and intermediate periods (as will be shown in Fig. 3), tend to include significant shallow rain as well as deep rain.

Table 1.

Mean rain volume and mean ratios at 4, 6, 8, 10, and 12 km for the mostly deep, intermediate, and mostly shallow periods. Ratio at each height indicates the ratio of 20-dBZ area at each height to the total rain area for each TRMM overpass over each grid. The three periods are defined depending on the ratios at 6 km (see text).

Table 1.
Fig. 2.
Fig. 2.

Rain contribution (%) binned into rain volume for the mostly shallow (solid), the intermediate (dashed), and the mostly deep (dash–dotted) periods. Abscissa indicates rain volume (mm h−1 km2) in logarithmic scale.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

Fig. 3.
Fig. 3.

Distribution of rain volume (mm h−1 km2) from four PF types, which are defined in section 4, as a function of the ratios of 20-dBZ area at 6 km to the total rain area for each TRMM overpass over each grid. Solid, dotted, dashed, and dash–dotted lines indicate types 1, 2, 3, and 4, respectively. Starting from the left, shaded regions indicate the mostly shallow (ratios of 0–0.05), intermediate (ratios of 0.25–0.3), and mostly deep (ratios of 0.5–1) periods.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

More closely, Fig. 3 shows rain volume from the four PF types, which are classified according to areas and maximum heights of 20 dBZ (MAXHT20), as a function of the ratio of 20-dBZ area at 6 km to the total rain area of each reference grid. The four PF types are defined in a similar manner to Yokoyama and Takayabu (2012) except that they used maximum storm heights from the TRMM 2A23 algorithm instead of MAXHT20 here: type 1 (area < 103.5 km2, MAXHT20 < 8 km), type 2 (area < 103.5 km2, MAXHT20 ≥ 8 km), type 3 (area ≥ 103.5 km2, 8 km ≤ MAXHT20 < 14 km), and type 4 (area ≥ 103.5 km2, 14 km ≤ MAXHT20 ≤ 20 km). Types 1 and 2 represent congestus and small tall systems, respectively. Both types 3 and 4 characterize organized systems, but they are different in their MAXHT20s. In Fig. 3, rainfall from type-3 and type-4 PFs is dominant in the mostly deep period (ratios of 0.5–1), while type-1 rain is dominant in the mostly shallow period (ratios of 0–0.05). Rainfall in the intermediate period (ratios of 0.25–0.3) is largely contributed by type-3 PFs, which are moderately deep organized systems. In addition, Fig. 4 shows frequency of 20-dBZ echo-top heights, which is calculated based on all TRMM PR2A25 nadir pixels in each reference grid. It is shown that the three convection periods indicate striking differences in frequency of 20-dBZ echo-top heights.

Fig. 4.
Fig. 4.

Frequency (%) of 20-dBZ echo-top heights (km) over the reference grids associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods. The figure is made based on all TRMM PR2A25 nadir data in each reference grid.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

5. Results

a. Overall properties of reanalyses

We first refer to the time-mean profiles of latent heating and ω, which are averaged around the center of the eastern Pacific ITCZ (8°–12°N, 130°–100°W) during boreal autumn from 1998 to 2010 (Fig. 5). The TRMM SLH has a peak in the upper troposphere and a secondary peak in the lower troposphere, whereas reanalyses have larger latent heating in the lower troposphere than in the upper troposphere. The lower peaks of latent heating for JRA25-JCDAS and MERRA are found at 850 and 925 hPa (not far from cloud base), respectively. We are uncertain how to interpret the peaks of reanalysis latent heating below 800 hPa, because more rain is likely to fall from clouds reaching 4–5 km than 2 km. In addition, very strong cooling occurs near the surface in the reanalyses. The cooling is possibly caused by evaporation of rainfall near the surface, but it is much stronger than the cooling shown in the TRMM estimates.

Fig. 5.
Fig. 5.

Mean profiles of (a) the TRMM SLH, (b) reanalysis latent heating (K day−1), and (c) reanalysis vertical pressure velocity (Pa s−1). Profiles are averaged over 8°–12°N, 130°–100°W during boreal autumn (September–November) for 1998–2010. In (b) and (c), profiles for JRA25-JCDAS, MERRA, and ERA-I are denoted with solid, dashed, and dash–dotted lines, respectively.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

In Fig. 5b, it is shown that levels of latent heating peaks are different between JRA25-JCDAS and MERRA. Ling and Zhang (2013) pointed out that there are significant differences in the number and level of peaks of tropical diabatic heating among reanalyses. According to their result, over the eastern Pacific ITCZ in July, MERRA has multiple peaks of diabatic heating: one below 850 hPa, one between 850 and 700 hPa, one between 700 and 500 hPa, one between 500 and 400 hPa, and one between 400 and 300 hPa. In Fig. 5b, on the other hand, MERRA produces peaks of latent heating below 900 hPa, and around 600 hPa, as well as a faint peak around 400 hPa. One of the reasons for differences in the level of heating peak between this study and Ling and Zhang (2013) may be because we investigate latent heating instead of total diabatic heating, which is examined in their study. Radiative heating and eddy heat flux convergence may cause some differences between latent heating and total diabatic heating.

It is also worth noting that more significant discrepancies in the profiles of latent heating among reanalyses are found in the upper troposphere than in the lower troposphere. Upper-tropospheric latent heating is larger in JRA25-JCDAS than in MERRA. The similar discrepancies among reanalyses are also found in ω profiles (Fig. 5c): upper-tropospheric ω is larger (smaller) in JRA25-JCDAS (MERRA). ERA-I has upper-tropospheric ω with magnitudes between those of JRA25-JCDAS and MERRA. These discrepancies among reanalyses will repeatedly appear in the subsequent composite analyses based on the three convection periods.

Looking at composite frequencies of precipitation rates, which are averaged over the 5° × 5° region around the center of each reference grid for the mostly deep, intermediate, and mostly shallow convection periods (Fig. 6), the reanalyses have large discrepancies, which are consistent with those shown in the time-mean state. Specifically, JRA25-JCDAS tends to produce heavier precipitation more frequently compared to the other reanalyses, even for the mostly shallow convection period. On the other hand, MERRA often produces lighter precipitation for the mostly deep and intermediate convection periods. ERA-I tends to produce heavier precipitation than MERRA, and lighter precipitation than the JRA25-JCDAS. Note that reanalyses do not always reproduce heavy rainfall when TRMM observes heavy rainfall, and the reproducibility of reanalyses is different from one case to another (not shown). But our main focus is on the statistics of numerous cases, and comparing the statistics of shallow, intermediate, and deep convection. We believe that it is reasonable to demonstrate that each of the reanalyses tends to produce heavier precipitation for the mostly deep and intermediate convection periods and lighter precipitation for the mostly shallow convection periods. Thus, the reanalyses basically produce precipitation in a manner consistent with the TRMM observations in a relative sense. Thus, we next examine properties of large-scale circulations based on the three periods with three different depths of convection, although we should keep in mind that the reanalyses have significant uncertainties, especially in the upper troposphere.

Fig. 6.
Fig. 6.

Composite occurrence frequencies of precipitation rates (mm h−1) associated with the mostly shallow (black bars), intermediate (blue bars), and mostly deep (red bars) periods for (a) JRA25-JCDAS, (b) MERRA, and (c) ERA-I.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

b. Circulation properties associated with three convection periods

In this subsection, we examine how reanalyses represent convection associated with three convection periods. Figure 7 (left) shows composites of profiles of latent heating for JRA25-JCDAS, MERRA, and the TRMM estimates, which are averaged over the 5° × 5° region around the center of each reference grid, associated with the mostly deep, intermediate, and mostly shallow periods of radar echo depths. Note that we cannot expect that TRMM estimates are quantitatively comparable to reanalysis latent heating. While the TRMM estimates are instantaneous data, reanalysis latent heating is averaged for 6 h. However, the TRMM SLH can inform us of what kind of convection occurs at the reference time of the composites, and we can expect that the 6-h mean reanalysis latent heating is affected by that convection.

Fig. 7.
Fig. 7.

Composite profiles of (left) latent heating (K day−1) and (right) vertical pressure velocity (Pa s−1) associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, (e) TRMM SLH, and (f) ERA-I. Gray lines indicate 90% confidence intervals. The composite is performed for profiles averaged over the 5° × 5° region around the center of each reference grid.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

For both reanalyses, the profiles for the mostly deep and intermediate periods are strikingly different from those for the mostly shallow period; latent heating in the upper troposphere is much stronger in the mostly deep and intermediate periods than the mostly shallow period. That indicates that our categorizations successfully capture the relative characteristics of cumulus convection even in reanalyses. On the other hand, the differences in the reanalysis profiles between the mostly deep and intermediate periods are subtle, and much smaller compared to those in the TRMM profiles. One possible reason may be that latent heating in reanalysis datasets is averaged for several hours, while the convection periods are determined based on the TRMM snapshots. Convection for the moderately deep and intermediate periods may often be linked to each other as part of the life cycle of convection, but data averaged for several hours may not have enough time resolution to capture such shifts in the convection cycle.

It is notable that the reanalysis latent heating profiles are very different from the TRMM latent heating profiles. Overall, magnitudes of upper-tropospheric latent heating are much larger in the TRMM estimates than in reanalyses. This may come from the difference between reanalysis latent heating averaged for several hours and instantaneous TRMM SLH, as already mentioned. In addition, the disagreement in magnitudes of latent heating may be caused by the fact that the TRMM SLH is averaged over only part of the 5° × 5° region around each reference grid (see Fig. 1). Consistent with the time-mean state, on the other hand, ratios of shallow to deep latent heating are larger in reanalyses than in the TRMM estimates for the three convection periods. This discrepancy is consistent with the findings of Jiang et al. (2011), which showed shallow latent heating is smaller in TRMM estimates than in reanalyses during the eastward propagation of the Madden–Julian oscillation (MJO; Madden and Julian 1972). Low sensitivity of the TRMM PR to light rain may partly cause the discrepancy shown in our study, and from the related issue that small showers may fill only part of the PR footprint. There is also a possibility that reanalyses underestimate stratiform heating, which has the maximum in the upper troposphere.

The lower peaks of latent heating are found at 850 hPa (925 hPa) for the JRA25-JCDAS (MERRA) for all three convection periods. They are very robust and similar in magnitudes between the two reanalyses. The TRMM estimates show that shallower peaks for the mostly deep and intermediate periods appear around 3 km, and are 2 km higher than that for the mostly shallow period. Such a difference in shallower peak heights among three periods is not shown in the reanalyses. The very strong cooling near the surface is always shown in the reanalyses regardless of the convection periods.

The most notable differences between the two reanalysis datasets are found in the mid- to upper troposphere for the mostly deep and intermediate periods. The upper peaks for JRA25-JCDAS are found around 400 hPa. MERRA has a significant peak around 600 hPa, where JRA25-JCDAS indicates the minima, and a faint peak around 400 hPa. The TRMM estimates indicate no corresponding peak near 600 hPa. This midtropospheric (600 hPa) peak of latent heating for MERRA is also shown in the results of some previous studies (Hagos et al. 2010; Jiang et al. 2011). For the mostly shallow period, on the other hand, JRA25-JCDAS and MERRA have a secondary maximum at 300 and 600 hPa, respectively. Overall, it seems that JRA25-JCDAS tends to represent stronger latent heating in the upper troposphere than the MERRA.

In Fig. 7 (right), lower peaks of ω around 850 hPa (850–800 hPa) and higher peaks around 300 hPa (350–300 hPa) are found for JRA25-JCDAS (MERRA and ERA-I). Every reanalysis can successfully capture different characteristics of large-scale circulations associated with different convection periods. Again, there are small differences in the lower-tropospheric ω among reanalyses but significant differences in the upper troposphere. JRA25-JCDAS produces stronger upward motions in the upper troposphere even for the mostly shallow period, compared to the other reanalyses. MERRA produces downward motions in the upper troposphere for the mostly shallow period, and it has bottom-heavy profiles of ω in the deeper convection periods. Note that these discrepancies in latent heating and ω among reanalyses appear regardless of the size of areas being averaged.

Next, composite meridional–pressure cross sections of meridional circulations, vertical pressure velocity (ω), and latent heating are shown associated with three convection periods (Fig. 8). These variables are averaged over 5° in a zonal direction around each reference grid, but not in a meridional direction. Figure 8 shows that significant differences in meridional circulation properties between the mostly deep and shallow periods exist, even though they are composited with respect to the TRMM-based precipitation properties. The cross sections associated with the intermediate period are relatively similar to those related to the mostly deep period, although there are some small coherent differences in the environment between these two periods.

Fig. 8.
Fig. 8.

Composite meridional–pressure cross sections of latent heat (K day−1; colors), vertical pressure velocity (Pa s−1; contours), and meridional circulation (arrows) for the (top) mostly deep, (middle) intermediate, and (bottom) mostly shallow periods. Cross sections for (left) JRA25-JCDAS, (middle) MERRA, and (right) ERA-I are shown. Contours are plotted at intervals of 0.02 Pa s−1. Solid contours are for negative values, and thick solid lines indicate 0 Pa s−1. The composite is performed for meridional cross sections averaged over 5° in longitudes at the center of each reference grid. Abscissa and ordinate axes indicate latitudes relative to the composite center and pressure levels, respectively.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

Figure 9 shows composite profiles of meridional wind averaged over 5° in longitude at the center of each reference grid, and from the latitudinal center of each grid to 10° south (left) or 10° north (right) of convection. Surprisingly, there is little difference in magnitudes of meridional winds among the reanalyses and the convection periods. It is also noticed that the profiles are not symmetric about the convection center. To the south of convection, both upper-tropospheric return flows around 250–150 hPa and shallow return flows around 700–500 hPa coexist for all periods. To the north of convection, weak southerlies are found throughout the layer above ~800 hPa with two peaks around 800–650 and 200 hPa for the mostly deep and intermediate periods, while weak northerlies are found in the layer of ~800–400 hPa for the mostly shallow period. That means that shallow return flows are observed on the both sides of convection for the mostly deep and intermediate periods, but only to the south of convection for the mostly shallow period. For JRA25-JCDAS and ERA-I, midlevel inflows around 400 hPa are also shown only to the south of convection. The north–south symmetry in the structure of meridional velocity may be related to the difference in inertial stability between the north and south of the ITCZ (Tomas and Webster 1997).

Fig. 9.
Fig. 9.

Composite profiles of meridional winds (m s−1), which are averaged over 5° in longitude at the center of each reference grid, and from the latitudinal center of each reference grid to (left) 10° south of convection or (right) 10° north of convection, for the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, and (e),(f) ERA-I. Gray lines indicate 90% confidence intervals.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

Importantly, the shallow return flows to the south of convection tend to be more preferentially shown for the mostly shallow period than for the mostly deep and intermediate periods. In contrast, the upper-tropospheric return flows of the deep circulations tend to be more dominant for the mostly deep and intermediate periods. Thus, the depths of convection may basically correspond to which branches of meridional circulations are dominant, although whether convection drives the circulations or the circulations drive convection is still an open question.

While the above features are relatively robust among reanalyses, there are also some significant differences in the meridional circulation structure among reanalyses, which alerts us to utilize reanalysis data carefully. Figure 8 shows that JRA25-JCDAS has two peaks of the upward motion in the lower and upper troposphere, while MERRA has a single peak in the lower troposphere. ERA-I shows obvious changes in the ω profiles depending on convective periods. In addition, JRA25-JCDAS has a stronger and deeper ascent core around the convection center and deeper downward motions to the north and south of convection than other datasets, which is reasonable considering stronger latent heating in the upper troposphere in JRA25-JCDAS.

Significant differences in meridional winds among reanalyses are also found especially to the south of convection (Figs. 8 and 9). MERRA and the ERA-I have more robust shallow return flows for all periods in terms of strength, thickness, and latitudinal extent, compared to the JRA25-JCDAS. Upper-tropospheric return flows are strongest (weakest) in JRA25-JCDAS (MERRA), and ERA-I is between these two datasets. Thus, the circulation properties such as dominance of upper-tropospheric (shallow) return flows, which are associated with the mostly deep (shallow) convection periods, tend to be most pronounced in JRA25-JCDAS (MERRA), where signals of deep convection tend to be strongest (weakest). Again, ERA-I is intermediate to the two extremes. It appears that there is at least consistency between discrepancies among reanalyses and differences among convection periods.

To evaluate the dominance of large-scale circulation patterns associated with convection periods, we show composite profiles of divergence (Fig. 10), which are averaged over the 5° × 5° region around the center of each reference grid. Upper-tropospheric divergence is stronger in the mostly deep and intermediate periods than in the mostly shallow period, while middle-lower tropospheric divergence is stronger in the mostly shallow period, which is consistent with the above relationships between the meridional circulations and convection periods. There is little difference in magnitudes of the peak of the shallow divergence among the reanalyses, but the vertical extent and structure of the shallow divergence are different among the reanalyses. In addition, all reanalysis datasets show that low-level convergence is stronger and deeper in the mostly deep and intermediate periods, compared to the mostly shallow period.

Fig. 10.
Fig. 10.

Composite profiles of divergence (×10−5 s−1), which is averaged over the 5° × 5° region around the center of each reference grid, for the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a) JRA25-JCDAS, (b) MERRA, and (c) ERA-I. Gray lines indicate 90% confidence intervals.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

Moreover, midlevel convergence appears around 400 hPa for all periods. It is shown that the midlevel convergence is less (more) significant in MERRA (JRA25-JCDAS), where weaker (stronger) deep convection appears, compared to JRA25-JCDAS (MERRA). Nolan et al. (2007) showed that midlevel inflows are stronger when the ITCZ convection is weaker, while Nolan et al. (2010) showed that they are stronger when the ITCZ convection is stronger. Thus, this study is consistent with the later results of Nolan et al. (2010). However, we should note that each reanalysis does not capture differences in midlevel convergence among the three convection periods.

At the end of this subsection, we quantify how often deep and shallow circulations occur in each convection period. We here utilize divergence at 200 (div200) and 700 hPa (div700) averaged over the 5° × 5° region as indices of the deep and shallow circulations, because Fig. 10 shows that divergence peaks around these levels in all reanalyses. Note that the low-level convergence is not considered in the indices of circulations, because there is little change in the low-level convergence among the reanalyses and the convection periods (Fig. 10). In addition, we here focus only on the mostly deep and shallow periods, because divergence profiles for the intermediate period are basically similar to those for the deep period.

Table 2 summarizes frequencies of the cases, which are categorized according to the sign of div200 and div700. The categories with +div700 and +div200, with −div700 and +div200, with −div700 and −div200, and with +div700 and −div200 represent the coexistence of shallow and deep circulations, the deep circulations only, no or very weak circulations, and the shallow circulations only, respectively. For all reanalyses, the category for the coexistence of shallow and deep circulations dominates in the mostly deep period. In comparison with JRA25-JCDAS and ERA-I, MERRA less frequently indicates the coexistence of shallow and deep circulations, and more frequently indicates the categories with negative div200. For the mostly shallow period, JRA25-JCDAS tends to most frequently indicate the coexistence of shallow and deep circulations, while MERRA tends to indicate shallow circulations only. ERA-I seems to represent the decrease in the frequency of the coexistence of shallow and deep circulations and the increase in the frequency of the shallow circulations only, associated with a shift from the mostly deep to shallow periods. The discrepancies among reanalyses, which are found throughout this study, are also confirmed in terms of the frequency of each circulation pattern.

Table 2.

The numbers of profiles, which are classified according to the sign of divergence at 700 and 200 hPa, for the mostly deep and shallow periods in JRA25-JCDAS, MERRA, and ERA-I. Numbers in parentheses indicate the frequencies (%). Divergence is averaged over the 5° × 5° region around the center of each reference grid.

Table 2.

c. Composites of anomalies from the time-mean state

While each reanalysis can represent anticipated differences in the depths of large-scale circulations for observed convection periods to some extent, there are still large discrepancies among reanalyses themselves during different types of convection periods (Figs. 5 and 710). In this subsection, in order to focus on differences among convection periods, we perform composite analysis of anomalies of latent heating and divergence from the time-mean state (Fig. 11).

Fig. 11.
Fig. 11.

Composite profiles of (left) latent heating anomaly (K day−1) and (right) divergence anomaly (×10−5 s−1) associated with the mostly shallow (solid), intermediate (dashed), and mostly deep (dash–dotted) periods: (a),(b) JRA25-JCDAS, (c),(d) MERRA, (e) TRMM SLH, and (f) ERA-I. Gray lines indicate 90% confidence intervals. The composite is performed for profiles averaged over the 5° × 5° region around the center of each reference grid.

Citation: Journal of Climate 27, 14; 10.1175/JCLI-D-13-00315.1

As a result, all reanalyses more obviously distinguish the mostly shallow period from the mostly deep period for the composite profiles of anomalies (Fig. 11) than for those of the total fields (Figs. 7 and 10). All reanalyses show top-heavy profiles for the mostly deep and intermediate periods and bottom-heavy profiles for the mostly shallow period, which are basically consistent with TRMM estimates. On the other hand, the differences between the intermediate convection period and the mostly deep period remain unclear. In addition, reanalyses still have uncertainties in the magnitudes of upper-tropospheric divergence anomalies and in the number and heights of peaks. In short, reanalyses have the ability to represent large-scale circulations, whose depths are reasonably varied depending on the passing shallow or deep convection. However, reanalyses have their own preferred large-scale state, which tends to make the differences among convection periods less clear.

6. Discussion and summary

This study clarifies some of the uncertainty in the amount of shallow heating over tropical oceans shown by Hagos et al. (2010). We show that the shallow latent heating is relatively robust among the reanalyses over the eastern Pacific, compared to the deep heating. Over the eastern Pacific, shallow rain from congestus is well correlated with shallow (1000–925 hPa) convergence over the tropical Pacific (Yokoyama and Takayabu 2012). In JRA25-JCDAS and MERRA, we also confirm that latent heating at 850 hPa correlates with the low-level convergence at 925 hPa with correlations beyond 0.7 (not shown). Latent heating at 925 hPa where MERRA has the peak is also correlated with the low-level convergence with correlation coefficients of 0.43 and 0.40 for the mostly deep and shallow periods, respectively (not shown). Thus, these reanalyses successfully simulate the abovementioned relationships between observed precipitation and the environment. There are only minor differences in the low-level convergence among the reanalyses, which may result in relatively robust shallow convection and shallow circulations in the reanalyses.

Notable discrepancies among reanalyses are found in deep heating, which we suspect are closely related to the discrepancies in the circulation structure in the mid- to upper troposphere. These discrepancies among reanalyses, which are similar to those shown in the time-mean state, are consistently found throughout this study. It is widely believed that the latent heating profiles can be sensitive to the cumulus parameterization. Both JRA25-JCDAS and MERRA utilize cumulus parameterizations based on the Arakawa–Schubert scheme. However, a slight modification may change the results even in the same parameterization.

In addition, the differences in latent heating profiles among reanalyses can be caused by the differences in the representation of boundary layer processes, which may be related to the vertical resolution, thermodynamic structures, and tropospheric relative humidity. Three reanalyses also utilize different data assimilation systems, which can be one of the reasons for discrepancies in the behavior. Further investigation is needed to elucidate which is more realistic and why, by comparison with field campaign data over the eastern Pacific such as the Tropical Eastern Pacific Process Study (TEPPS; Yuter and Houze 2000; Yuter et al. 2000) and the Eastern Pacific Investigations of Climate (EPIC) 2001 experiment (Raymond et al. 2004).

The four primary findings of this study are summarized as follows. First, in terms of individual convective systems that are based on TRMM-observed snapshots, the depths of convection are basically related to which branches of large-scale circulations are dominant. The deep circulations preferentially appear to be associated with the mostly deep period of radar echo top, while the shallow circulations are dominant for the mostly shallow period. The shallow and deep circulations frequently coexist together for both deep and shallow periods, resulting in the multicell structure of the large-scale circulations. All reanalyses successfully represent these reasonable relationships, which is basically consistent with Nolan et al. (2007). However, the changes in circulations among convection periods are smaller than their results based on a regional model. Reanalyses have their own preferred large-scale state, which tends to make the changes among convection periods less clear. It is suggested that passing convection can perturb circulations from the preferred state, but may not change them as much as in reality. In addition, reanalyses can hardly distinguish the intermediate convection period from the mostly deep period.

Second, strong shallow convergence in the boundary layer, which is primarily driven by strong SST gradients (Back and Bretherton 2009), almost constantly appears over the eastern Pacific during the boreal autumn, regardless of the depth of the convection. That is confirmed by the fact that there are only small differences in the low-level convergence among convection periods. One reason is possibly that scatterometer ocean surface winds are assimilated into all three reanalyses. Because the SST gradients apparently control the low-level convergence rather than any effect of convection, it seems that low-level convergence is well represented in all three reanalyses.

Third, there are notable differences in the circulation structure in the mid- to upper troposphere, which are largely related to those in latent heating profiles. The JRA25-JCDAS tends to indicate the strongest characteristics associated with deep circulations such as the divergence and outflows in the upper troposphere and the midlevel inflows around 400 hPa. In contrast, MERRA tends to have the weakest deep circulations. ERA-I is intermediate between these two datasets. On the other hand, JRA25-JCDAS and MERRA have strikingly different profiles of latent heating in the mid- to upper troposphere, especially for the mostly deep period, which is consistent with the abovementioned relationships between the circulation structure and the depth of convection.

Fourth, discrepancies in the shallow divergence and the corresponding shallow return flows among reanalyses and convection periods are found but not as large as those in the deep ones. In MERRA and ERA-I, the shallow divergence and return flows are more significant in terms of vertical and latitudinal extent, compared to JRA25-JCDAS. The strong SST gradients over the eastern Pacific can cause much shallow rain through strong convergence in the boundary layer and thus shallow return flows, while they also can cause shallow return flows through the sea-breeze-like process. We need further investigation to elucidate whether the sea-breeze-like process or shallow convection exerts more control on shallow return flows.

Acknowledgments

This study is supported by the NASA Precipitation Measurement Mission program led by Dr. Ramesh Kakar and the late Dr. Arthur Hou, Grant NNX10AG90G. The authors would also like to express their gratitude to three anonymous reviewers for their very helpful comments. The JRA25-JCDAS data are provided by the cooperative research project of the JRA-25 long-term reanalysis by the JMA and the Central Research Institute of Electric Power Industry. The MERRA data are generated by the NASA Global Modeling and Assimilation Office (GMAO) and disseminated by the Goddard Earth Sciences (GES) Data and Information Services Center (DISC). The ERA-Interim data are obtained from the ECMWF data server.

REFERENCES

  • 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, doi:10.1175/2009JCLI2392.1.

    • Search Google Scholar
    • Export Citation
  • Berg, W., , C. Kummerow, , and C. A. Morales, 2002: Differences between east and west Pacific rainfall systems. J. Climate, 15, 36593672, doi:10.1175/1520-0442(2002)015<3659:DBEAWP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , S. Fueglistaler, , G. Lesins, , and T. Mitovski, 2008: A low-level circulation in the tropics. J. Atmos. Sci., 65, 10191034, doi:10.1175/2007JAS2463.1.

    • Search Google Scholar
    • Export Citation
  • Grabowski, W., , J.-I. Yano, , and M. W. Moncrieff, 2000: Cloud resolving modeling of tropical circulations driven by large-scale SST gradients. J. Atmos. Sci., 57, 20222039, doi:10.1175/1520-0469(2000)057<2022:CRMOTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hagos, S., , and C. Zhang, 2010: Diabatic heating, divergent circulation and moisture transport in the African monsoon system. Quart. J. Roy. Meteor. Soc., 136 (S1), 411425, doi:10.1002/qj.538.

    • Search Google Scholar
    • Export Citation
  • Hagos, S., and Coauthors, 2010: Estimates of tropical diabatic heating profiles: Commonalities and uncertainties. J. Climate, 23, 542558, doi:10.1175/2009JCLI3025.1.

    • Search Google Scholar
    • Export Citation
  • Hartmann, D. L., , H. H. Hendon, , and R. A. Houze Jr., 1984: Some implications of the mesoscale circulations in tropical cloud clusters for large-scale dynamics and climate. J. Atmos. Sci., 41, 113121, doi:10.1175/1520-0469(1984)041<0113:SIOTMC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Iguchi, T., , T. Kozu, , R. Meneghini, , J. Awaka, , and K. Okamoto, 2000: Rain-profiling algorithm for the TRMM precipitation radar. J. Appl. Meteor., 39, 20382052, doi:10.1175/1520-0450(2001)040<2038:RPAFTT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Jiang, X., and Coauthors, 2011: Vertical diabatic heating structure of the MJO: Intercomparison between recent reanalyses and TRMM estimates. Mon. Wea. Rev., 139, 32083223, doi:10.1175/2011MWR3636.1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kubar, T. L., , D. L. Hartmann, , and R. Wood, 2007: Radiative and convective driving of tropical high clouds. J. Climate, 20, 55105526, doi:10.1175/2007JCLI1628.1.

    • Search Google Scholar
    • Export Citation
  • Larson, K., , and D. L. Hartmann, 2003: Interactions among cloud, water vapor, radiation, and large-scale circulation in the tropical climate. Part II: Sensitivity to spatial gradients of sea surface temperature. J. Climate, 16, 14411455, doi:10.1175/1520-0442-16.10.1441.

    • Search Google Scholar
    • Export Citation
  • Ling, J., , and C. Zhang, 2011: Structural evolution in heating profiles of the MJO in global reanalyses and TRMM retrievals. J. Climate, 24, 825842, doi:10.1175/2010JCLI3826.1.

    • Search Google Scholar
    • Export Citation
  • Ling, J., , and C. Zhang, 2013: Diabatic heating profiles in recent global reanalyses. J. Climate, 26, 33073325, doi:10.1175/JCLI-D-12-00384.1.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. J. Zipser, , D. J. Cecil, , S. W. Nesbitt, , and S. Sherwood, 2008: A cloud and precipitation feature database from nine years of TRMM observations. J. Appl. Meteor. Climatol., 47, 27122728, doi:10.1175/2008JAMC1890.1.

    • 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, doi:10.1175/1520-0469(1972)029<1109:DOGSCC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., , R. Cifelli, , and S. A. Rutledge, 2006: Storm morphology and rainfall characteristics of TRMM precipitation features. Mon. Wea. Rev., 134, 27022721, doi:10.1175/MWR3200.1.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., , C. Zhang, , and S.-H. Chen, 2007: Dynamics of the shallow meridional circulation around intertropical convergence zones. J. Atmos. Sci., 64, 22622285, doi:10.1175/JAS3964.1.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., , S. W. Powell, , C. Zhang, , and B. E. Mapes, 2010: Idealized simulations of the intertropical convergence zone and its multilevel flows. J. Atmos. Sci., 67, 40284053, doi:10.1175/2010JAS3417.1.

    • Search Google Scholar
    • Export Citation
  • Onogi, K., and Coauthors, 2007: The JRA-25 Reanalysis. J. Meteor. Soc. Japan, 85, 369432, doi:10.2151/jmsj.85.369.

  • Raymond, D. J., and Coauthors, 2004: EPIC2001 and the coupled ocean–atmosphere system of the tropical east Pacific. Bull. Amer. Meteor. Soc., 85, 13411354, doi:10.1175/BAMS-85-9-1341.

    • Search Google Scholar
    • Export Citation
  • Rienecker, M. M., and Coauthors, 2011: MERRA: NASA’s Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, 36243648, doi:10.1175/JCLI-D-11-00015.1.

    • Search Google Scholar
    • Export Citation
  • Schneider, E. K., 1977: Axially symmetric steady-state models of the basic state for instability and climate studies. Part II: Nonlinear calculations. J. Atmos. Sci., 34, 280296, doi:10.1175/1520-0469(1977)034<0280:ASSSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schneider, E. K., , and R. S. Lindzen, 1977: Axially symmetric steady-state models of the basic state for instability and climate studies. Part I: Linearized calculations. J. Atmos. Sci., 34, 263279, doi:10.1175/1520-0469(1977)034<0263:ASSSMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , and R. A. Houze Jr., 2003: Stratiform rain in the tropics as seen by the TRMM Precipitation Radar. J. Climate, 16, 17391756, doi:10.1175/1520-0442(2003)016<1739:SRITTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , R. A. Houze Jr., , and I. Kraucunas, 2004: The tropical dynamical response to latent heating estimates derived from the TRMM precipitation radar. J. Atmos. Sci., 61, 13411358, doi:10.1175/1520-0469(2004)061<1341:TTDRTL>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schumacher, C., , P. E. Ciesielski, , and M. H. Zhang, 2008: Tropical cloud heating profiles: Analysis from KWAJEX. Mon. Wea. Rev., 136, 42894300, doi:10.1175/2008MWR2275.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , W.-K. Tao, , and D. E. Johnson, 2004: Spectral retrieval of latent heating profiles from TRMM PR data. Part I: Development of a model-based algorithm. J. Appl. Meteor., 43, 10951113, doi:10.1175/1520-0450(2004)043<1095:SROLHP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , W.-K. Tao, , and C.-L. Shie, 2007: Spectral retrieval of latent heating profiles from TRMM PR data. Part II: Algorithm improvement and heating estimates over tropical ocean regions. J. Appl. Meteor. Climatol., 46, 10981124, doi:10.1175/JAM2510.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , and W.-K. Tao, 2008: Spectral retrieval of latent heating profiles from TRMM PR data. Part III: Estimating apparent moisture sink profiles over tropical oceans. J. Appl. Meteor. Climatol., 47, 620640, doi:10.1175/2007JAMC1738.1.

    • Search Google Scholar
    • Export Citation
  • Shige, S., , Y. N. Takayabu, , S. Kida, , W.-K. Tao, , X. Zeng, , C. Yokoyama, , and T. L’Ecuyer, 2009: Spectral retrieval of latent heating profiles from TRMM PR data. Part IV: Comparisons of lookup tables from two- and three-dimensional cloud-resolving model simulations. J. Climate, 22, 55775594, doi:10.1175/2009JCLI2919.1.

    • Search Google Scholar
    • Export Citation
  • Stachnik, J. P., , and C. Schumacher, 2011: A comparison of the Hadley circulation in modern reanalyses. J. Geophys. Res., 116, D22102, doi:10.1029/2011JD016677.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , J. Yokomori, , and K. Yoneyama, 2006: A diagnostic study on interactions between atmospheric thermodynamic structure and cumulus convection over the tropical western Pacific Ocean and over the Indochina peninsula. J. Meteor. Soc. Japan, 84A, 151169, doi:10.2151/jmsj.84A.151.

    • Search Google Scholar
    • Export Citation
  • Takayabu, Y. N., , S. Shige, , W.-K. Tao, , and N. Hirota, 2010: Shallow and deep latent heating modes over tropical oceans observed with TRMM PR spectral latent heating data. J. Climate, 23, 20302046, doi:10.1175/2009JCLI3110.1.

    • Search Google Scholar
    • Export Citation
  • Tomas, R. A., , and P. J. Webster, 1997: The role of inertial instability in determining the location and strength of near-equatorial convection. Quart. J. Roy. Meteor. Soc., 123, 14451482, doi:10.1002/qj.49712354202.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , D. P. Stepaniak, , and J. M. Caron, 2000: The global monsoon as seen through the divergent atmospheric circulation. J. Climate, 13, 39693993, doi:10.1175/1520-0442(2000)013<3969:TGMAST>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wang, Y., , S.-P. Xie, , B. Wang, , and H. Xu, 2005: Large-scale atmospheric forcing by southeast Pacific boundary-layer clouds: A regional model study. J. Climate, 18, 934951, doi:10.1175/JCLI3302.1.

    • Search Google Scholar
    • Export Citation
  • Wu, Z., 2003: A shallow CISK, deep equilibrium mechanism for the interaction between large-scale convection and large-scale circulations in the tropics. J. Atmos. Sci., 60, 377392, doi:10.1175/1520-0469(2003)060<0377:ASCDEM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Yokoyama, C., , and Y. N. Takayabu, 2012: Relationships between rain characteristics and environment. Part I: TRMM precipitation features and the large-scale environment over the tropical Pacific. Mon. Wea. Rev., 140, 28312840, doi:10.1175/MWR-D-11-00252.1.

    • Search Google Scholar
    • Export Citation
  • Yuter, S. E., , and R. A. Houze Jr., 2000: The 1997 Pan American Climate Studies Tropical Eastern Pacific Process Study. Part I: ITCZ region. Bull. Amer. Meteor. Soc., 81, 451481, doi:10.1175/1520-0477(2000)081<0451:TPACST>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Yuter, S. E., , Y. L. Serra, , and R. A. Houze Jr., 2000: The 1997 Pan American Climate Studies Tropical Eastern Pacific Process Study. Part II: Stratocumulus region. Bull. Amer. Meteor. Soc., 81, 483490, doi:10.1175/1520-0477(2000)081<0483:TPACST>2.3.CO;2.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , and S. M. Hagos, 2009: Bi-modal structure and variability of large-scale diabatic heating in the tropics. J. Atmos. Sci., 66, 36213640, doi:10.1175/2009JAS3089.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , M. McGauley, , and N. A. Bond, 2004: Shallow meridional circulation in the tropical eastern Pacific. J. Climate, 17, 133139, doi:10.1175/1520-0442(2004)017<0133:SMCITT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , P. Woodworth, , and G. Gu, 2006: The seasonal cycle in the lower troposphere over West Africa from sounding observations. Quart. J. Roy. Meteor. Soc., 132, 25592582, doi:10.1256/qj.06.23.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., , D. S. Nolan, , C. D. Thorncroft, , and H. Nguyen, 2008: Shallow meridional circulation in the tropical atmosphere. J. Climate, 21, 34533470, doi:10.1175/2007JCLI1870.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, C., and Coauthors, 2010: MJO signals in latent heating: Results from TRMM retrievals. J. Atmos. Sci., 67, 34883508, doi:10.1175/2010JAS3398.1.

    • Search Google Scholar
    • Export Citation
Save