• Ackerman, T. P., , K-N. Liou, , F. P. J. Valero, , and L. Pfister, 1988: Heating rates in tropical anvils. J. Atmos. Sci., 45 , 16061623.

  • Back, L. E., , and C. S. Bretherton, 2006: Geographic variability in the export of moist static energy and vertical motion profiles in the tropical Pacific. Geophys. Res. Lett., 33 .L17810, doi:10.1029/2006GL026672.

    • Search Google Scholar
    • Export Citation
  • Bjerknes, J., 1966: A possible response of the atmospheric Hadley circulation to the equatorial anomalies of ocean temperature. Tellus, 18 , 820829.

    • Search Google Scholar
    • Export Citation
  • Bjerknes, J., 1969: Atmospheric teleconnections from the equatorial Pacific. Mon. Wea. Rev., 97 , 163172.

  • Boehm, M. T., , and S. Lee, 2003: The implications of tropical Rossby waves for tropical tropopause cirrus formation and for the equatorial upwelling of the Brewer–Dobson circulation. J. Atmos. Sci., 60 , 247261.

    • Search Google Scholar
    • Export Citation
  • Cau, P., , J. Methven, , and B. J. Hoskins, 2005: Representation of dry tropical layers and their origins in ERA-40 data. J. Geophys. Res., 110 .D06110, doi:10.1029/2004JD004928.

    • Search Google Scholar
    • Export Citation
  • Ciesielski, P. E., , R. H. Johnson, , P. T. Haertel, , and J. Wang, 2003: Corrected TOGA COARE sounding humidity data: Impact on diagnosed properties of convection and climate over the warm pool. J. Climate, 16 , 23702384.

    • Search Google Scholar
    • Export Citation
  • Corti, T., , B. P. Luo, , Q. Fu, , H. Vomel, , and T. Peter, 2006: The impact of cirrus clouds on tropical troposphere-to-stratosphere transport. Atmos. Chem. Phys. Discuss., 6 , 17251747.

    • Search Google Scholar
    • Export Citation
  • Danielsen, E. F., 1993: In situ evidence of rapid, vertical, irreversible transport of lower tropospheric air into the lower tropical stratosphere by convective cloud turrets and by larger-scale upwelling in tropical cyclones. J. Geophys. Res., 98 , 86658681.

    • Search Google Scholar
    • Export Citation
  • Dima, I. M., , J. M. Wallace, , and I. Kraucunas, 2005: Tropical zonal momentum balance in the NCEP reanalyses. J. Atmos. Sci., 62 , 24992513.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , and R. V. Martin, 2005: The vertical structure of tropical convection and its impact on the budgets of water vapor and ozone. J. Atmos. Sci., 62 , 15601573.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , M. Loewenstein, , J. Podolske, , S. J. Oltmans, , and M. Proffitt, 1999: A barrier to vertical mixing at 14 km in the tropics: Evidence from ozonesondes and aircraft measurements. J. Geophys. Res., 104 , D18. 2209522102.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , K. K. Kelly, , and E. M. Weinstock, 2002: A simple explanation for the increase in relative humidity between 11 and 14 km in the tropics. J. Geophys. Res., 107 .D23. 4736, doi:10.1029/2002JD002185.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , H. Wernli, , and T. Peter, 2004: Tropical troposphere-to-stratosphere transport inferred from trajectory calculations. J. Geophys. Res., 109 .D03108, doi:10.1029/2003JD004069.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , M. Bonazzola, , P. H. Haynes, , and T. Peter, 2005: Stratospheric water vapor predicted from the Lagrangian temperature history of air entering the stratosphere in the tropics. J. Geophys. Res., 110 .D08107, doi:10.1029/2004JD005516.

    • Search Google Scholar
    • Export Citation
  • Gettelman, A., , and P. M. de F. Forster, 2002: A climatology of the tropical tropopause layer. J. Meteor. Soc. Japan, 80 , 911924.

  • Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106 , 447462.

  • Highwood, E. J., , and B. J. Hoskins, 1998: The tropical tropopause. Quart. J. Roy. Meteor. Soc., 124 , 15791604.

  • Hoskins, B. J., , A. J. Simmons, , and D. G. Andrews, 1977: Energy dispersion in a barotropic atmosphere. Quart. J. Roy. Meteor. Soc., 103 , 553567.

    • Search Google Scholar
    • Export Citation
  • Jackson, D. R., , J. Methven, , and V. Pope, 2001: Transport in the low latitude tropopause zone diagnosed using particle trajectories. J. Atmos. Sci., 58 , 173192.

    • Search Google Scholar
    • Export Citation
  • Kerr-Munslow, A. M., , and W. A. Norton, 2006: Tropical wave driving of the annual cycle in tropical tropopause temperatures. Part I: ECMWF analyses. J. Atmos. Sci., 63 , 14101419.

    • Search Google Scholar
    • Export Citation
  • Kley, D., , A. L. Schmeltekopf, , K. Kelly, , R. H. Winkler, , T. L. Thompson, , and M. McFarland, 1982: Transport of water through the tropical tropopause. Geophys. Res. Lett., 9 , 617620.

    • Search Google Scholar
    • Export Citation
  • Knollenberg, R. G., , A. J. Dascher, , and D. Huffman, 1982: Measurements of the aerosol and ice crystal populations in tropical stratospheric cumulonimbus anvils. Geophys. Res. Lett., 9 , 613616.

    • Search Google Scholar
    • Export Citation
  • Kraucunas, I., , and D. L. Hartmann, 2007: Tropical stationary waves in a nonlinear shallow-water model with realistic basic states. J. Atmos. Sci., 64 , 25402557.

    • Search Google Scholar
    • Export Citation
  • Krishnamurti, T. N., 1971: Tropical east–west circulations during the northern summer. J. Atmos. Sci., 28 , 13421347.

  • Krishnamurti, T. N., , M. Kanamitsu, , W. J. Koss, , and J. D. Lee, 1973: Tropical east–west circulations during the northern winter. J. Atmos. Sci., 30 , 780787.

    • Search Google Scholar
    • Export Citation
  • Kuang, Z., , and C. S. Bretherton, 2004: Convective influence of the heat balance of the tropical tropopause layer: A cloud-resolving model study. J. Atmos. Sci., 61 , 29192927.

    • Search Google Scholar
    • Export Citation
  • Lee, S., 1999: Why are the climatological zonal winds easterly in the equatorial upper troposphere. J. Atmos. Sci., 56 , 13531363.

  • Lilly, D. K., 1988: Cirrus outflow dynamics. J. Atmos. Sci., 45 , 15941605.

  • Matsuno, T., 1966: Quasi-geostrophic motions in the equatorial area. J. Meteor. Soc. Japan, 44 , 2542.

  • McFarquhar, G. M., , A. J. Heymsfield, , J. Spinhirne, , and B. Hart, 2000: Thin and subvisual tropopause tropical cirrus: Observations and radiative impacts. J. Atmos. Sci., 57 , 18411853.

    • Search Google Scholar
    • Export Citation
  • Mote, P. W., and Coauthors, 1996: An atmospheric tape recorder: The imprint of tropical tropopause temperatures on stratospheric water vapor. J. Geophys. Res., 101 , 39894006.

    • Search Google Scholar
    • Export Citation
  • Newell, R. E., , and S. Gould-Stewart, 1981: A stratospheric fountain? J. Atmos. Sci., 38 , 27892796.

  • Newell, R. E., , J. W. Kidson, , D. G. Vincent, , and G. J. Boer, 1974: The General Circulation of the Tropical Atmosphere and Interactions with Extratropical Latitudes. Vol. 2, MIT Press, 371 pp.

    • Search Google Scholar
    • Export Citation
  • Norton, W. A., 2001: Longwave heating of the tropical lower stratosphere. Geophys. Res. Lett., 28 , 36533656.

  • Norton, W. A., 2006: Tropical wave driving of the annual cycle in tropical tropopause temperatures. Part II: Model results. J. Atmos. Sci., 63 , 14201431.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., , and F. Wu, 2005: Kelvin wave variability near the equatorial tropopause observed in GPS radio occultation measurements. J. Geophys. Res., 110 .D03102, doi:10.1029/2004JD005006.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., , R. R. Garcia, , and F. Wu, 2002: Time-dependent upwelling in the tropical lower stratosphere estimated from the zonal-mean momentum budget. J. Atmos. Sci., 59 , 21412152.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , and C. L. Vlcek, 1969: The annual temperature variation in the lower tropical stratosphere. J. Atmos. Sci., 26 , 163167.

  • Sadler, J. C., cited. 2006: The upper tropospheric circulation over the global tropics. [Available online at http://www.soest.hawaii.edu/Library/Sadler.html#database.].

  • Sherwood, S. C., 2000: A stratospheric “drain” over the Maritime Continent. Geophys. Res. Lett., 27 , 677680.

  • Simmons, A. J., 1982: The forcing of stationary wave motion by tropical diabatic heating. Quart. J. Roy. Meteor. Soc., 108 , 503534.

  • Simmons, A. J., , A. Untch, , C. Jakob, , P. Kallberg, , and P. Unden, 1999: Stratospheric water vapor and tropical tropopause temperatures in ECMWF analyses and multi-year simulations. Quart. J. Roy. Meteor. Soc., 125 , 353386.

    • Search Google Scholar
    • Export Citation
  • Thompson Jr., R. M., , S. W. Payne, , E. E. Recker, , and R. J. Reed, 1979: Structure and properties of synoptic-scale wave disturbances in the intertropical convergence zone of the eastern Atlantic. J. Atmos. Sci., 36 , 5372.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , D. P. Stepaniak, , and J. M. Caron, 2000: The global monsoon as seen throught the divergent atmospheric circulation. J. Climate, 13 , 39693993.

    • Search Google Scholar
    • Export Citation
  • Troup, A. J., 1965: The Southern Oscillation. Quart. J. Roy. Meteor. Soc., 91 , 490506.

  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Reanalysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Wang, P-H., , P. Minnis, , M. P. McCormick, , G. S. Kent, , and K. M. Skeens, 1996: A 6-year climatology of cloud occurrence frequency from Stratospheric Aerosol and Gas Experiment II observations (1985–1990). J. Geophys. Res., 101 , D23. 2940729430.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., 1972: Response of the tropical atmosphere to local steady forcing. Mon. Wea. Rev., 100 , 518541.

  • Webster, P. J., 1981: Mechanisms determining the atmospheric response to sea surface temperature anomalies. J. Atmos. Sci., 38 , 554571.

    • Search Google Scholar
    • Export Citation
  • Woodruff, S. D., , R. J. Slutz, , R. L. Jenne, , and P. M. Steurer, 1987: A comprehensive ocean–atmosphere data set. Bull. Amer. Meteor. Soc., 68 , 12391250.

    • 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.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Annual-mean sea surface temperature (contour interval = 1°C) and 300-hPa vertical velocity (colored shading, Pa s−1). The 25°C sea surface temperature contour is shown in red, for reference.

  • View in gallery

    Annual-mean 300-hPa vertical velocity (colored shading, Pa s−1) superimposed on (top) 150-hPa geopotential height (contours, m) and wind (arrows, m s−1); and (bottom) sea level pressure (contour interval = 1.5 hPa, min/max values shown 1009/1021 hPa) and boundary layer (1000–925 hPa) surface winds (arrows). The wind arrows are plotted every 7.5° latitude × 15° longitude, up to 23° latitude. The contour interval for geopotential height is 100 m (gray lines); additional contours at 10-m intervals (black) are inserted in the tropical belt. Contour succession for 150-hPa height: (. . . 14 100, 14 200, 14 210, 14 220, . . .) m, with the first black contour at the separation between gray and black contours corresponding to 14 210 m.

  • View in gallery

    Annual-mean temperature of various layers as indicated, as estimated from the respective thicknesses (contour interval = 0.5°C for black lines and 1°C for gray lines). Superimposed are the wind (arrows, m s−1) and vertical velocity (colored shading, Pa s−1) estimated at (a) 100, (b) 300, (c) 600, and (d) 925 hPa. The first black contour at the separation between gray and black lines represents (a) −72°, (b) −36°, (c) −3°, and (d) 22°C. Note that in the top panel the polarity of the temperature pattern is reversed and the vertical velocities are much smaller than in the other panels.

  • View in gallery

    Longitude–height cross section of the eddy component of geopotential height Z* (colored shading, m) and zonal and vertical wind components (arrows, m s−1 for zonal wind and cm s−1 for vertical velocity) for the latitudinally averaged belt 10°S–5°N. The longitudinal sector 0°–60°E is repeated on the rhs for visual continuity. The vertical velocities are stretched relative to the zonal velocities to make them more clearly visible.

  • View in gallery

    As in Fig. 4 but for the eddy component of the temperature field T*, in units of °C. The vertical velocity arrows have been rescaled by dividing them by the horizontally averaged static stability (Γd − Γ) at each level to make them proportional to heating rate (°C day−1).

  • View in gallery

    As in Fig. 5 but for the total relative humidity field, in units of %.

  • View in gallery

    Vertical profiles of rms amplitude of the annual-mean stationary waves in the (a) eddy geopotential height (m), (b) temperature (°C), (c) vertical velocity (Pa s−1) fields, and (d) covariance between vertical velocity and temperature (Pa °C s−1).

  • View in gallery

    Vertical profiles of vertical velocity (Pa s−1) in pressure coordinates in different regions, as indicated by the red rectangles in the bottom plot showing the annual-mean 300-hPa vertical velocity. Color bar is the same as in Fig. 1.

  • View in gallery

    Annual-mean meridional cross section for the sector extending across the Pacific, from 180° to 90°W, showing temperatures (departures from the 20°N–20°S means) and the meridional wind component and vertical velocity averaged over this sector. Plotting and scaling conventions are the same as in Fig. 5.

  • View in gallery

    Time–longitude section of the eddy component of the 150-hPa geopotential height field indicated by shading, color bar in meters, and the total (zonal mean plus eddy) fields of 150-hPa zonal wind in m s−1 and 300-hPa vertical velocity in cm s−1, as indicated by the arrows. Data are meridionally averaged from 10°S to 5°N.

  • View in gallery

    As in Fig. 10 but for 100-hPa temperature, zonal wind, and vertical velocity.

  • View in gallery

    Maps of eddy fields: (top) regression on standardized CTI with sign reversed (contour interval = 2.5 m); (bottom) annual mean EPW, for the 150-hPa geopotential height (contour interval is 10 m). Dashed contours indicate negative values, and wind (arrows, m s−1) and 300-hPa vertical velocity (colored shading, Pa s−1) are also shown.

  • View in gallery

    Composite maps based on extreme values of the CTI, an indicator of the status of the ENSO cycle. The fields represented are 150-hPa geopotential height (contours), wind (arrows), and 300-hPa vertical velocity (color). Plotting conventions are the same as in Fig. 2a.

  • View in gallery

    Schematic of the vertical structure of the EPW over the western Pacific, as represented in the ERA-40 reanalyses. The layering scheme shown at the left is based on the profiles of temperature, vertical velocity, and the release of available potential energy shown in Fig. 7. Most of the detrainment from the hypothetical deep convective cloud is depicted as occurring below 175 hPa, the level of peak amplitude of the EPW. Convectively forced descent is indicated by the downward-pointing arrows to the right of the cloud, and overshooting cloud tops by the small turret protruding from the top of the cloud. The plume of planetary-scale ascent over the western Pacific is represented by the upward-pointing arrows just above the convective cloud. Extensive subvisible cirrus cloud layers, represented by the wavy lines, are assumed to be present within the plume. The plume is depicted as spreading out at or just above the cold point, ventilating and lifting the entire tropical lower stratosphere. The vertical profile at the right represents the planetary-scale divergence over the western Pacific. The planetary-scale divergence is depicted as peaking at or just above the level of strongest detrainment from the convective clouds.

  • View in gallery

    Schematic view of the impact of the EPW on the vertical velocity and temperature fields. The region with convective clouds in the center of the figure represents the equatorial Pacific and the full width of the diagram represents the entire tropical belt. Arrows indicate the strength of the upwelling, and the blue shading indicates the intensity of the adiabatic cooling. Above the cold point, the cooling is stronger because of the higher static stability.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 33 33 11
PDF Downloads 26 26 5

Structure of the Annual-Mean Equatorial Planetary Waves in the ERA-40 Reanalyses

View More View Less
  • 1 Department of Atmospheric Sciences, University of Washington, Seattle, Washington
© Get Permissions
Full access

Abstract

The three-dimensional structure of the annual-mean equatorial planetary waves in the 40-yr ECMWF Re-Analysis (ERA-40) is documented. The features in the free atmosphere are predominantly equatorially symmetric, driven by east–west heating gradients. The geopotential height and wind perturbations are strongest at or just below the 150-hPa level. Below the level of maximum amplitude, the circulations in the waves are thermally direct with latent heat release in deep convective clouds and radiative cooling in the intervening cloud-free regions. Within the overlying capping layer, the wave-related circulations are thermally indirect, with rising of the coldest air and sinking of air that is less cold. At the cold point, just above the 100-hPa (17 km) level, the ERA-40 annual-mean vertical velocity in the equatorial belt ranges up to 3 mm s−1 over the equatorial western Pacific during the boreal winter, implying diabatic heating rates of up to 3°C day−1, an order of magnitude larger than typical clear-sky values. Strong heating is consistent with evidence of widespread thin and subvisible cirrus cloud layers over this region. It is hypothesized that the air mass as a whole is rising (as opposed to just the air in the updrafts of convective clouds) and that this plume of ascending air spreads out horizontally at or just above the cold point, ventilating and lifting the entire lower stratosphere.

El Niño years are characterized by anomalously weak equatorial planetary waves in the Indo-Pacific sector and slightly enhanced waves over the Atlantic sector and cold years of the El Niño–Southern Oscillation (ENSO) cycle by the opposite conditions. Equatorial Pacific sea surface temperature is as well correlated with the strength of the equatorial planetary waves in the upper troposphere over the Indo-Pacific sector as it is with the conventional Southern Oscillation index based on sea level pressure.

Corresponding author address: John M. Wallace, Department of Atmospheric Sciences, Department of Atmospheric Sciences, 408 ATG Bldg., Box 351640, Seattle, WA 98195-1640. Email: wallace@atmos.washington.edu

Abstract

The three-dimensional structure of the annual-mean equatorial planetary waves in the 40-yr ECMWF Re-Analysis (ERA-40) is documented. The features in the free atmosphere are predominantly equatorially symmetric, driven by east–west heating gradients. The geopotential height and wind perturbations are strongest at or just below the 150-hPa level. Below the level of maximum amplitude, the circulations in the waves are thermally direct with latent heat release in deep convective clouds and radiative cooling in the intervening cloud-free regions. Within the overlying capping layer, the wave-related circulations are thermally indirect, with rising of the coldest air and sinking of air that is less cold. At the cold point, just above the 100-hPa (17 km) level, the ERA-40 annual-mean vertical velocity in the equatorial belt ranges up to 3 mm s−1 over the equatorial western Pacific during the boreal winter, implying diabatic heating rates of up to 3°C day−1, an order of magnitude larger than typical clear-sky values. Strong heating is consistent with evidence of widespread thin and subvisible cirrus cloud layers over this region. It is hypothesized that the air mass as a whole is rising (as opposed to just the air in the updrafts of convective clouds) and that this plume of ascending air spreads out horizontally at or just above the cold point, ventilating and lifting the entire lower stratosphere.

El Niño years are characterized by anomalously weak equatorial planetary waves in the Indo-Pacific sector and slightly enhanced waves over the Atlantic sector and cold years of the El Niño–Southern Oscillation (ENSO) cycle by the opposite conditions. Equatorial Pacific sea surface temperature is as well correlated with the strength of the equatorial planetary waves in the upper troposphere over the Indo-Pacific sector as it is with the conventional Southern Oscillation index based on sea level pressure.

Corresponding author address: John M. Wallace, Department of Atmospheric Sciences, Department of Atmospheric Sciences, 408 ATG Bldg., Box 351640, Seattle, WA 98195-1640. Email: wallace@atmos.washington.edu

1. Introduction

The equatorial planetary waves (EPW), forced by zonal asymmetries in the distribution of diabatic heating, are an important element in the global general circulation. They force the zonal flow by inducing equatorward eddy fluxes of westerly momentum at the jet stream level. In the annual mean, the eddy flux convergence would produce a strong equatorial superrotation were it not balanced by a flux divergence of westerly momentum by the zonally symmetric component of the seasonally varying monsoon circulations (Lee 1999; Dima et al. 2005). The EPW also induce mean meridional circulations that play a role in the heat balance of the equatorial belt. Boehm and Lee (2003), Kerr-Munslow and Norton (2006), and Norton (2006) have argued that EPW-induced mean-meridional circulations contribute substantially to the coldness of the equatorial tropopause.

The structure of the EPW in the upper-tropospheric wind field is dominated by the equatorially symmetric component (Newell et al. 1974; Sadler 2006), which is present year round and linked to a distinctive signature in the geopotential height field (Dima et al. 2005). A prominent feature of the circulation in the equatorial plane is the so-called Walker cell over the equatorial Pacific, with low-level easterly trade winds in the lower branch, ascent over the western Pacific warm pool, westerly flow in the upper branch, and descent over the eastern Pacific (Bjerknes 1969; Krishnamurti et al. 1973).

Many of the features in the upper tropospheric wind field of the EPW are replicated in the linear response to an equatorial mass source in a barotropic model (Matsuno 1966) or a shallow-water wave equation model (Hoskins et al. 1977), or as the response to an equatorial heat source in a baroclinic model (Webster 1972; Gill 1980; Webster 1981; Simmons 1982; Highwood and Hoskins 1998). In these simulations, the response to the prescribed mass source or heating consists of a superposition of equatorially trapped Kelvin and Rossby waves. The observed waves are greater in meridional extent than those predicted by the linear theory, and the equatorial symmetry is more robust with respect to changes in the latitude of the heat source (Dima et al. 2005). Kraucunas and Hartmann (2007) have shown that the prominence of the equatorially symmetric component in the upper-tropospheric flow, even during the solsticial seasons, can be explained in terms of dynamical processes that can be represented in a shallow-water wave equation model. Under solsticial conditions, the stronger driving of the EPW in the summer hemisphere is largely counterbalanced by the stronger response per unit forcing in the winter hemisphere that occurs by virtue of the stronger westerly background flow.

Important questions remain concerning the vertical velocity field in the EPW, its relation to deep cumulus convection and subvisible cirrus clouds, and its role in troposphere–stratosphere exchange. Still at issue is the vertical extent of the upward mass flux over the region of heavy rainfall over the Indo-Pacific region on the planetary scale and how it relates to the vertical structure of deep cumulus convection. Is the planetary-scale upwelling in this region merely an average of the vertical velocity in the convective-scale updrafts and the subsidence that they induce, or is it influenced by planetary-scale dynamics as well? Newell and Gould-Stewart (1981) postulated that most of the air entering the stratosphere passes through a planetary-scale “fountain” of adiabatically cooled ascending air over the western Indo-Pacific warm pool. Boehm and Lee (2003), Kerr-Munslow and Norton (2006), and Norton (2006) have emphasized the role of EPW-driven mean-meridional circulation, with ascent in the equatorial belt. Kley et al. (1982), Knollenberg et al. (1982), Ackerman et al. (1988), Danielsen (1993), and numerous subsequent studies have emphasized the importance of overshooting updrafts in deep convective clouds over the Indo-Pacific warm pool in ventilating the lower stratosphere. Sherwood (2000) has postulated the existence of a planetary-scale stratospheric mass sink over this region associated with convectively driven subsidence. Sharply contrasting interpretations abound because confidence in observations relating to the vertical velocity field is insufficient to rule any of them out.

Data derived from models has been used only sparingly for documenting the structure of the EPW because of concerns that it might be unduly influenced by dubious radiative and convective parameterizations. Fields derived from data assimilation schemes are much less subject to this limitation than those derived from climate models because they are constrained to some degree by observations. The multivariate data assimilation schemes used in modern reanalyses protocols impose a dynamical consistency between the fields of different variables that is not present in the earlier univariate analyses.

Trenberth et al. (2000) analyzed the seasonally varying climatology in the European Centre for Medium-Range Forecasts (ECMWF) operational model and the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalyses by performing complex EOF analysis of the divergent circulation. Although that study did not focus exclusively on the EPW, it provided a number of valuable insights into their three-dimensional structure. The present study is based on the 40-yr ECMWF Re-Analysis (ERA-40) for the period of record 1979–2001, as discussed in section 2. Sections 3 and 4 document the three-dimensional structure of the EPW, with emphasis on the vertical velocity field and its relation to the fields of geopotential height, wind, and temperature in various layers of the equatorial atmosphere. Section 5 describes the meridional circulation associated with the ITCZ. Sections 6 and 7 show how the structure of the EPW varies with season and with the phase of the ENSO cycle. Section 8 presents a summary and discussion of the results and their implications for troposphere–stratosphere exchange and the role of subvisible cirrus cloud layers, and the final section offers a few concluding remarks.

2. Data

The datasets used in this study are the wind, temperature, geopotential height, relative humidity, and vertical velocity fields from the ERA-40 described in Uppala et al. (2005). ERA-40 is based on data from a variety of sources, including radiosonde observations, satellite radiances, and winds derived from inertial navigation systems on commercial aircraft and from the tracks of clouds in time-lapse satellite imagery. The data are mapped on a 2.5° × 2.5° global latitude–longitude grid. This study is based on data for the 1000-, 925-, 850-, 700-, 775-, 600-, 500-, 400-, 300-, 250-, 200-, 150-, 100-, 70-, and 50-hPa pressure levels, for the period of record extending from 1979, the first year of coverage for many of the satellite-based datasets, through 2001, the last year included in ERA-40.

For constructing vertical profiles, the data were interpolated onto a finer vertical grid with data points at nearly evenly spaced logarithmic pressure intervals (1000, 963, 925, 889, 850, 812, 775, 740, 700, 652, 600, 549, 500, 452, 400, 345, 300, 272, 250, 226, 200, 176, 150, 123, 100, 83, 70, 60, 50 hPa). The interpolation was performed using a built-in Matlab function called “interp,” which resamples the data using a symmetric filter.

3. Horizontal structure

The spatial pattern of annual-mean 300-hPa vertical velocity, shown in Fig. 1, bears a strong relation to the temperature of the underlying surface. Analogous patterns for the ECMWF operational analyses and the NCEP reanalyses as shown in Fig. 7 of Trenberth et al. (2000) are quite similar. Ascent in the equatorial belt (the rising branch of the Hadley cell) is concentrated over the continents, where the diurnal cycle in surface heating creates conditions favorable for the development of deep convection, and over the Indo-Pacific warm pool and the narrower belts of warm water beneath the ITCZs. Descent within this belt is weak and largely confined to the relatively cool equatorial dry zones in the eastern Pacific and Atlantic.

Figure 2 shows the horizontal structure of the annual-mean EPW in the sea level pressure (SLP) and boundary layer wind fields (lower panel) and in the 150-hPa geopotential height and wind fields (upper panel), both superimposed on the 300-hPa vertical velocity field. The boundary layer wind field is represented by the average for the 1000- and 925-hPa levels. The SLP pattern is dominated by the contrasts between the subtropical oceanic highs and the equatorial trough centered over the western Pacific warm pool ∼165°E. The boundary layer wind field exhibits a systematic downgradient component, with easterlies across most of the equatorial Pacific and Atlantic, and northward cross-equatorial flow toward the ITCZ. The 150-hPa height patterns (Fig. 2, top panel) exhibit a more equatorially symmetric, wavelike structure, with an off-equatorial Rossby wave couplet centered at 120°E connected by a ridge of high geopotential height that crosses the equator at 165°E. Equatorial westerlies prevail to the east of the ridge and easterlies to the west of it. The ridge lies directly over the SLP minimum in the equatorial trough and at the boundary between the eastern Pacific equatorial dry zone and the large region of ascent over the Indo-Pacific warm pool.

Northward cross-equatorial flow into the Pacific and Atlantic ITCZs is clearly apparent in the boundary layer wind field and southward return flow is discernible in the 150-hPa field in the same longitudinal sector. The equatorially symmetric planetary wave features are more prominent in the upper troposphere and the equatorially asymmetric ITCZ-related mean meridional circulation is more prominent in the boundary layer field.

Figure 3 shows the mean temperature fields in four layers as estimated from geopotential height data, making use of the hypsometric equation. Within the boundary layer (1000 to 850 hPa; Fig. 3d), the temperature pattern over the oceans conforms closely to the underlying sea surface temperature pattern (Fig. 1) and the continents tend to be warmer than the surrounding oceans. The oceanic ITCZs are clearly apparent. In the 850-to-500-hPa layer (Fig. 3c), the temperature pattern is weaker and more equatorially symmetric than the boundary layer pattern, with little, if any, indication of an ITCZ signature. A similar pattern is evident in the 400-to-200-hPa layer (Fig. 3b), but with higher amplitude, and in the 150-to-70-hPa layer (Fig. 3a) with reversed polarity. Despite the prominent ITCZ signature in the 300-hPa vertical velocity pattern, the equatorially symmetric planetary wave pattern dominates the temperature field in the free atmosphere.

The vertical velocity fields within the four layers are qualitatively similar, except that in the topmost panel (100 hPa), the ITCZ is not discernible and ascent prevails throughout virtually the entire equatorial belt. At all but the lowest level, the strongest ascent is located over the western Pacific. In contrast to the circulations at lower levels, the planetary-scale circulation in the topmost layer in Fig. 3 is thermally indirect, with the coldest air, which is located over the western Pacific, rising the most rapidly. The 100-hPa vertical velocity pattern over the Maritime Continent is patchy, with small regions of subsidence, interspersed with regions of ascent, reminiscent of the pattern in Fig. 2b of Norton (2001) based on ECMWF operational analyses. Sherwood (2000) also found evidence of a patch of subsidence over the Maritime Continent based on the analysis of the divergence field derived from radiosonde observations.

4. Vertical structure of the equatorial planetary waves

This section begins with a series of vertical cross sections for atmospheric variables averaged over the equatorial belt, which, for purposes of this study, is defined as extending from 10°S to 5°N. The belt is centered slightly to the south of the equator in order to exclude the ITCZ and to accentuate the east–west contrasts in the sea surface temperature and sea level pressure fields, which are centered a few degrees south of the equator. In all longitudinal cross sections, the winds are longitudinally averaged using a 5-point running mean, and the wind arrows are plotted every fourth horizontal grid point to better represent the planetary-scale structure.

Figure 4 shows zonal and vertical velocities in a vectorial format, superimposed upon the eddy component (i.e., the departure from the zonal mean) of the geopotential height field. The height perturbations are largest near the 150-hPa level, where the pattern is dominated by the high centered ∼165°E over the western Pacific and the lows over the eastern Pacific and eastern Atlantic. At these levels, the zonal flow tends to be directed down the zonal pressure gradient, out of the highs and into the lows. A notable exception is over the eastern Pacific, where the strongest westerlies are collocated with the minimum in geopotential height rather than to the west of it. Consistent with the 150-hPa wind field in Fig. 2a, the dominant features in the section are the easterlies across Indonesia and the westerlies across the central Pacific, which extend from 400 to 70 hPa (7 to 19 km).

The geopotential height fluctuations at the earth’s surface in Fig. 4 tend to be out of phase with those in the upper troposphere, with minima over the western Pacific and the continents and maxima over the relatively cool eastern Pacific and eastern Atlantic and over the western Indian Ocean near 40°E. The amplitude of the low-level features decreases monotonically with height up to the node ∼450 hPa. The lower-tropospheric zonal flow is dominated by Atlantic and Pacific easterly trade winds, which blow down the zonal pressure gradient. Ascent prevails throughout most of the section, which lies within the rising branch of the Hadley cell. Ascent occurs over the western Pacific and the continents and descent is strongest over the eastern Pacific.

The overall pattern in the wind field in this section resembles the widely used schematic of Krishnamurti et al. (1973, their Fig. 2) except that the region of ascent in the Walker circulation is shifted eastward by ∼40° degrees of longitude, from the Maritime Continent into the western Pacific.

The corresponding equatorial cross section for the temperature field is shown in Fig. 5. To make the vertical velocity perturbations associated with the stationary waves more clearly visible in the lower stratosphere, we have scaled them by multiplying them by the factor Γd − Γ, the difference between the observed and the dry adiabatic lapse rate. Assuming steady-state conditions and neglecting horizontal temperature advection by the time–mean flow and the transients, the adiabatic cooling (or heating) due to ascent (or descent) is balanced by the diabatic heating; that is,
i1520-0469-64-8-2862-e1
where w is the vertical velocity and Q is the diabatic heating rate. Hence, the rescaled vertical velocity, with sign reversed, is indicative of the diabatic heating rate. At the cold point (∼100 hPa), where Γ = 0, Q in °C day−1 is approximately equal to w in mm s−1. The rescaled vertical velocities show a planetary-scale plume of ascent over the western Pacific extending all the way up to at least 100 hPa. Above 150 hPa, vertical velocities elsewhere in the section are weak and mostly upward, consistent with the 100-hPa vertical velocity field shown in the top panel of Fig. 3.

In accordance with the hypsometric equation, each of the upper-tropospheric features in the geopotential height field in Fig. 4 is attended by a warm/cold dipole in Fig. 5: warm below/cold above height maxima and cold below/warm above height minima. In the lower centers in these dipoles, near the 300-hPa level, the overturning circulations in the waves are thermally direct, with warmer air rising and colder air sinking, releasing potential energy, whereas in the upper centers, near the 100-hPa level, the circulation is thermally indirect, presumably driven by the upward flux of planetary-scale wave activity from below. Given that ascending air in the vicinity of the tropopause tends to be depleted in ozone, the reversal in the polarity of the temperature perturbations around the 150-hPa (∼14 km) level explains why ozone concentration and temperature tend to be negatively correlated below that level and positively correlated above (Folkins et al. 1999).

It should be emphasized that Lagrangian air parcel trajectories are much flatter than suggested by the visual appearance of the arrows in Fig. 5, particularly at the uppermost levels. The bundles of upward arrows over the western Pacific warm pool and the continents are not indicative of coherent plumes of rising air parcels. They simply reflect a tendency for air parcels to ascend as they pass through these sectors on varied, gently sloping trajectories as described in Jackson et al. (2001), Fueglistaler et al. (2004), and Cau et al. (2005).

Figure 6 shows the same wind cross section superimposed on the field of relative humidity (RH). Within the midtroposphere, values range from as low as 20% in regions of descent to as high as 60% in regions of ascent. In the boundary layer and in the 150–100-hPa layer, RH is generally above 60% and ranges as high as 80% in the regions of strongest ascent. The increase in RH with height in the upper troposphere is consistent with sounding data presented in Fig. 1 of Folkins et al. (2002) and in Ciesielski et al. (2003). The prevalence of high RH in the 150-to-100-hPa layer is consistent with the notion that ascent is occurring in this layer, not only within the updrafts in the towers of deep convective clouds, but also along sloping three-dimensional trajectories in the stably stratified outside air. However, it must be acknowledged that the relative humidity field at these levels says more about the model than about the real atmosphere, since the humidity field is only weakly constrained by direct measurements.

Figure 7 shows vertical profiles of root-mean-squared amplitude of the EPW in the height, temperature, vertical velocity in pressure coordinates (ω), a measure of the vertical mass flux, and the covariance between vertical velocity and temperature (ωT). Each profile was computed by 1) averaging the data from 10°S to 5°N, 2) squaring the eddy (i.e., the total field minus the zonal mean) component at each longitude, 3) averaging over all longitudes, and 4) in the case of the variances, taking the square root to obtain the root-mean-squared (rms) amplitude.

The EPW in the geopotential height field (Fig. 7a) and the wind field (not shown) exhibit well-defined maxima in the rms amplitude, at the earth’s surface and at 150 hPa. Wave amplitude exhibits a distinct minimum at ∼450 hPa where the geopotential height and wind perturbations undergo a sign reversal. The waves in the temperature field (Fig. 7b) exhibit a more complicated vertical structure that (in accordance with the hypsometric equation) resembles the absolute value of the vertical derivative of the geopotential height profile. The vertical profile of rms temperature amplitude exhibits a low-level maximum in the boundary layer, a midlevel maximum near 300 hPa, and an upper-level maximum near 100 hPa. In the low-level maximum, the temperature perturbations are coupled to the underlying surface, while in the mid- and upper-level maxima they reflect the more equatorially symmetric planetary wave pattern. The mid- and upper-level maxima in the temperature profile are separated by a well-defined node, coincident with the 150-hPa height maximum. The vertical profile of rms wave amplitude in vertical velocity (Fig. 7c) exhibits a broad, flat maximum extending from 750 to 450 hPa, in which the flow appears to be nondivergent.

The release of planetary-scale available potential energy in the EPW is given by ωT d(ln p) integrated over the equatorial belt. Hence, in Fig. 7d, the lightly shaded areas are representative of a release of available potential energy in the planetary-scale overturning in the waves, and areas of dark shading represent layers in which the circulation is thermally indirect. Most of the release of available potential energy occurs within the boundary layer and near the midlevel maximum in the rms temperature profile (Fig. 7b). The release of planetary-scale available potential energy is small in the 800-to-450-hPa layer, where the release of potential energy in deep convection is presumably the largest. The energy conversion is small in this layer because the temperature perturbations are weak (Fig. 7b) and poorly defined (Fig. 5). The positive covariances above 150 hPa are indicative of thermally indirect planetary-scale circulations that are doing work against the pull of gravity by lifting colder, denser air, relative to its surroundings.

5. Structure of the ITCZ

The foregoing sections have emphasized the structure of the equatorially symmetric component of the EPW in which the motions are dominated by overturning in the zonal plane, as epitomized by the Walker circulation. In contrast, the circulation cells in the meridional plane associated with the Pacific and Atlantic ITCZs are predominantly zonally asymmetric and much smaller in horizontal scale. The rainbelts associated with the ITCZs are also much narrower than those that drive the east–west circulations in the EPW. This section contrasts the vertical structure of the ITCZ circulations with that of the EPW in the longitude–height plane.

Vertical velocity profiles for the eastern Pacific and Atlantic ITCZs are compared with those for the rain areas over the continents and the Indo-Pacific warm pool region in Fig. 8. The continental/warm-pool profile (upper right panel in Fig. 8) exhibits a broad midtropospheric maximum analogous to the stationary-wave amplitude profile in Fig. 7c. In contrast, the ITCZ profile (upper left panel in Fig. 8) exhibits a major peak at the 850-hPa level, a minimum near 550 hPa and a secondary maximum near 300 hPa. The contrasting profiles are analogous to those shown in Fig. 6 of Thompson et al. (1979), which were derived from radiosonde data in the Atlantic ITCZ region and the western Pacific. A thermodynamical interpretation of the distinctions between ERA-40 eastern and western Pacific vertical velocity profiles is offered by Back and Bretherton (2006).

Meridional cross sections of temperature, vertical velocity, and the meridional wind component in the vicinity of the equatorial Pacific ITCZ are shown in Fig. 9. The dominant feature in the vertical velocity field is the strong ascent over the ITCZ and the weaker subsidence over the equator. In agreement with Fig. 8, the low-level flow into the ITCZ is much more concentrated in the boundary layer than the flow into the continental and Indo-Pacific rain areas in Figs. 4 and 5. Consistent with results of Zhang et al. (2004), the meridional flow across the equator reverses near the top of the boundary layer and weak southward flow prevails within the 850-to-550-hPa layer. However, the meridional circulation above the boundary layer is much weaker than the circulation in the longitude–height plane in Figs. 4 –6. The temperature field in this section bears little relation to the mean meridional circulation associated with the ITCZ. Consistent with the temperature maps shown in Fig. 3, the pattern is equatorially symmetric, with relatively warm air over the equator. The ITCZ does not possess a well-defined temperature signature above the 700-hPa level or a well-defined sinking branch. The ITCZ-related vertical velocities at the 100-hPa level in this section are not as strong as those in Fig. 5, consistent with the absence of a well-defined ITCZ signature in the 100-hPa vertical velocity field (Fig. 3).

6. Seasonal variations

Recent studies of Dima et al. (2005) and Kerr-Munslow and Norton (2006) have emphasized different aspects of the seasonality of the EPW, the former pointing out the semiannual cycle in eddy kinetic energy of the equatorially symmetric component, with maxima during the monsoon seasons, and the latter drawing attention to the annual cycle in the upward Eliassen–Palm fluxes induced by the EPW, with a maximum during the boreal winter. Here we attempt to put these two different kinds of seasonality into context.

Figure 10 shows longitude–time (calendar month) sections of 150-hPa eddy geopotential height and zonal wind and 300-hPa vertical velocity, averaged from 10°S to 5°N. The July maximum in eddy kinetic energy reported in Dima et al. (2005) is evidently a reflection of the tropical easterly jet centered over the Indian Ocean sector during the boreal summer (Krishnamurti 1971). This feature, in combination with the weaker January maximum over the same sector associated with the austral summer monsoon, imparts a semiannual rhythm to the eddy kinetic energy in the equatorial belt. A semiannual rhythm is also apparent in the eddy component of the 150-hPa geopotential height over the western Pacific, with the January maximum being more prominent. The rate of ascent over the western Pacific is strongest in February and weakest in September–October: it shows little, if any, indication of a semiannual component.

Analogous sections, but for the 100-hPa temperature, zonal wind, and vertical velocity are shown in Fig. 11. At this level, the annual cycle in upwelling is much more prominent. The strong, seasonally dependent upwelling is concentrated in the western Pacific sector: averaged over the belt from 10°S to 5°N, 120°E to 160°W, and from December through February the vertical velocity is ∼3 mm s−1, consistent with results of Simmons et al. (1999) based on the prior ERA-15 reanalyses. It is only over the western Pacific that the equatorial upwelling at the 100-hPa level exhibits a strong annual cycle. In contrast, the 100-hPa temperature exhibits a strong annual cycle at all longitudes, with a minimum during the boreal winter. Throughout the year, the lowest equatorial 100-hPa temperatures are observed over the central and western Pacific.

7. ENSO-related variability

The remarkable similarity between the structure of the ENSO-related variability and the EPW in the 150-hPa geopotential height and wind fields is illustrated in Fig. 12 (the top and bottom panel, respectively). The ENSO signature (with sign reversed) is derived by regressing the eddy component of the 150-hPa field on the cold tongue index (CTI; sea surface temperature anomalies averaged over the equatorial Pacific cold tongue region 6°S–6°N, 180°–90°W, minus the globally averaged SST anomaly). The sea surface temperature data used in computing the index are from the Comprehensive Atmosphere–Ocean Dataset (COADS; Woodruff et al. 1987).

The respective centers of action of the ENSO pattern over the Pacific sector are shifted west of the corresponding centers in the EPW, but the two patterns nonetheless project strongly upon one another. The centers of action in ENSO and the EPW are of opposing sign over the Indian and Pacific sectors and of like sign over the Atlantic sector. Hence, during the warm phase of the ENSO cycle, the EPW are suppressed over the Indian and Pacific sectors and enhanced over the Atlantic sector.

Wind and geopotential height patterns at the 150-hPa level during extreme warm and cold phases of the ENSO cycle are compared in Fig. 13. As documented in Table 1, the warm phase composite is based on data for 12 months: the warmest January, the warmest February, . . . , the warmest December in the 23-yr record, as defined by the CTI and the cold composite is defined in an analogous manner. The cold composite shown in the lower panel is marked by enhanced EPW amplitude over the Indo-Pacific sector, enhanced subsidence in the western part of the equatorial dry zone, and relatively weak troughs over the Atlantic sector, and the warm composite shown in the upper panel is marked by the opposite conditions.

The ENSO cycle has long been recognized as an episodic strengthening and weakening of the southeasterly trade winds in the Indo-Pacific sector (e.g., see Troup 1965; Bjerknes 1966, 1969), as reflected in indices of Southern Oscillation. Here we have shown that it bears an equally strong relation to the structure of the EPW in the upper troposphere. The strength of the waves in the Indo-Pacific sector, as indicated by the 150-hPa zonal wind (0°, 150°W) minus the zonal wind at (0°, 90°E) is correlated with the CTI at a level of −0.90, based on 12-month running mean data from 1979–2001, which is comparable to the strength of the correlations between the CTI and indices of the Southern Oscillation.

8. Discussion

a. Horizontal structure of the waves

The annual-mean EPW are made up of two patterns: 1) shallow features in the SLP and boundary layer temperature and wind fields such as the ITCZ that are closely coupled to the temperature field at the earth’s surface, and 2) a deep, equatorially symmetric pattern composed of a superposition of equatorial Rossby and Kelvin waves, with an equatorial ridge over the western Pacific. The shallow features pattern mirror the patterns in the underlying surface temperature field and their signature is evident in the distribution of deep convection, as represented by the 300-hPa vertical velocity field. The temperature and pressure fields in the free atmosphere are dominated by the larger-scale equatorially symmetric wave pattern (2), which broadly resembles the idealized pattern obtained in the numerical simulation presented in section 5 of Highwood and Hoskins (1998).

b. Vertical structure of the waves

In terms of the dynamical features shown in the vertical cross sections (Figs. 3 –7), the tropical atmosphere, as represented in the ERA-40 reanalyses, can be divided into four layers, as illustrated schematically in Fig. 14.

  1. The boundary layer or lower baroclinic layer is marked by strong divergence/convergence, baroclinicity, and release of available potential energy in planetary-scale overturning circulations.
  2. The barotropic layer extending from the top of the boundary layer to ∼450 hPa is distinguished by its weak baroclinicity and the absence of a systematic release of available potential energy on the planetary scale. Within this layer, the planetary-scale vertical velocity is strong and nearly independent of height and therefore, the horizontal wind field tends to be nondivergent. The wind and temperature fields weaken with height.
  3. The middle baroclinic layer extending from around 400 to 175 hPa is marked by a well-defined maximum in the amplitude of the temperature perturbations and a strong release of potential energy in planetary-scale thermally direct circulations. The wind and temperature perturbations in the EPW increase with height in this layer.
  4. The capping layer or upper baroclinic layer extending from the level of maximum amplitude in the geopotential height and wind field near 150 hPa up to the upper limit of the waves in the ERA-40 reanalyses, somewhere between 100 and 70 hPa. The distinguishing feature of the capping layer is that the temperature perturbations in the waves are of reversed polarity of those at lower levels so that the planetary-scale circulations are thermally indirect, presumably forced from below. The vertical velocity field in the capping layer is dominated by a plume of planetary-scale ascent of cold air over the western Pacific: smaller-scale features such as the ITCZ are less prominent than they are in the underlying layers.

It is instructive to compare these four layers defined on the basis of the EPW structure with the layering scheme proposed by Folkins and Martin (2005, their Fig. 2), which is based on static stability considerations as applied to convective clouds. Our boundary layers are quite comparable and our barotropic layer corresponds roughly to their shallow outflow layer, which is also quasi nondivergent by virtue of the near balance between the mass flux divergence out of the updrafts shallow cumulus clouds and the convergence into evaporatively forced downdrafts associated with deep convective clouds. Within this layer, the stratification is slightly unstable. Our middle baroclinic layer encompasses their pseudoadiabatic layer (5–10 km) as well as most of their deep outflow (10–17 km) layer. In ERA-40, the planetary-scale upper-level divergence out of regions of heavy rainfall extends from about 8 to 18 km as compared to 10 to 17 km in the Folkins and Martin scheme.

The capping layer in our dynamically based layering scheme encompasses the so-called tropopause transition layer (TTL), broadly defined as the transition region between the convectively dominated troposphere and the radiatively controlled stratosphere (Gettelman and de F. Forster 2002) and the lower stratosphere, extending up to the highest level at which the waves are detectable. The upper part of the capping layer is much drier, and it is more stably stratified by roughly a factor of 3 than the lower part. With levels at 150-, 100-, and 70-hPa, the version of ERA-40 used in this study does not have sufficient vertical resolution to clearly define the vertical structure of the temperature perturbations in the EPW in the capping layer. Climatological mean January temperature maps at the actual levels of the assimilating model, kindly provided by A. J. Simmons of ECMWF (not shown), indicate that the temperature perturbations in the EPW are about twice as large at levels 24 (80 hPa) and 25 (96 hPa) as they are at level 26 (113 hPa). There are also indications that the temperature exhibits much more temporal variability in the tropical lower stratosphere than in the TTL. For example, Reed and Vlcek (1969) found a tenfold amplification of the annual cycle and Randel and Wu (2005) found evidence of a sharp upward increase in the amplitude of the temperature perturbations associated with vertically propagating gravity waves in a thin layer centered on the cold point.

c. On the role of thin and subvisible cirrus clouds

In ERA-40, the rate of ascent at the 100-hPa level ranges up to 3 mm s−1 over the western Pacific during the boreal winter. The corresponding radiative heating rates, as inferred from Eq. (1), range up to several °C day−1, which is up to an order of magnitude higher than the range of values estimated for clear-sky radiation at these levels. Lilly (1988), McFarquhar et al. (2000), Boehm and Lee (2003), and others have speculated that the presence of subvisible cloud layers at these levels could substantially enhance the radiative heating rates. Subvisible cirrus cloud layers have been shown to be present in the 150–100-hPa layer, with the highest frequency of occurrence over the equatorial western Pacific during the boreal winter (Wang et al. 1996).

On the basis of radiative transfer calculations using existing observations of the optical thickness of subvisible cloud layers, Corti et al. (2006) have estimated radiative heating rates at these levels to be on the order of 1°C day−1, about 3 times as large as the clear-sky values, but only about ⅓ as large as required to balance the adiabatic cooling implied by the rate of ascent in the ERA-40 reanalyses over the western Pacific during the boreal winter. The condensation of water vapor onto ice crystals in these clouds could also significantly dry the air as it rises toward the cold point in the temperature profile, provided that the crystals grow large enough to have fall speeds comparable to the ascent rate of the air (mm s−1). Enhanced radiative heating of subvisible cirrus cloud layers would cancel much of the adiabatic cooling in the upwelling regions of planetary-scale circulations, reducing the amplitude of the dynamically induced negative temperature perturbations, and thereby enabling the planetary-scale plumes of ascent in EPW to penetrate deeper into the capping layer than they could under clear-sky conditions.

d. Implications for troposphere–stratosphere exchange

In the absence of the EPW, convective-scale fluxes of mass and heat would be the dominant mechanism for troposphere–stratosphere exchange and for determining the temperature of the cold point. Under these conditions, adiabatic cooling induced by convective-scale downward heat fluxes near the cloud-top level is balanced by radiative heating, as in the numerical simulations of Kuang and Bretherton (2004) with a cloud-resolving model. In ERA-40, dynamically forced, planetary-scale ascent over the western Pacific induces adiabatic cooling (and radiative heating) rates ranging up to an order of magnitude stronger than those in the simulations of Kuang and Bretherton (2004).

Table 2 compares ERA-40 annual-mean vertical velocities and vertical mass fluxes over the equatorial western Pacific (125°E–160°W), the equatorial belt (10°S-10°N), and the entire Tropics (20°S–20°N) at selected levels in the upper troposphere and lower stratosphere. The rate of ascent at the 70-hPa level is spatially quite uniform with a value of 0.24 mm s−1, consistent with the rate of ascent of moist and dry layers formed by the annual cycle in lower stratospheric temperature, the so-called tropical tape recorder (Mote et al. 1996). The mass flux through the 70-hPa surface integrated over the entire Tropics (0.56 × 1010 kg s−1) is comparable to the mass flux through the 100-hPa surface over the equatorial western Pacific alone (0.55 × 1010 kg s−1). The important role of the western Pacific in the Eulerian upward mass flux at the 100-hPa level is qualitatively consistent with results of Jackson et al. (2001) and Fueglistaler et al. (2004, 2005, which indicate that many of the three-dimensional trajectories of air entering the stratosphere pass over the western Pacific. However, it should be acknowledged that these results are also model-dependent, the first two studies being based on the ECMWF operational analyses and the third on ERA-40.

The statistics presented in Table 2 are not inconsistent with the notion that the concentrated plume or fountain of planetary-scale ascent over the equatorial western Pacific at the 100-hPa level spreads out horizontally within the upper part of the capping layer to feed the much more spatially uniform upwelling extending over the entire Tropics at the 70-hPa level, as postulated by Newell and Gould-Stewart (1981). Air parcels ascending through the cold point and emerging from the tops of the subvisible cloud layers experience roughly a threefold increase in static stability (Γd − Γ) in combination with an abrupt decrease in radiative heating rate, both of which inhibit further upward penetration of the plume, which should cause it to spread out horizontally, as depicted in Fig. 15. Indeed, climatological-mean January vertical velocity fields in the full resolution ERA-40 data show little if any indication of the plume at level 24 (80 hPa), just above the cold point (not shown; A. J. Simmons 2006, personal communication). If the spreading takes place within a thin layer at or just above the cold point, then the outflow from the plume of planetary-scale ascent over the western Pacific may be viewed as hydrostatically lifting and ventilating the entire tropical lower stratosphere. Given radiative relaxation times in the lower stratosphere possibly as long as 50–100 days (Randel et al. 2002), ascent rates of a few tenths of a mm s−1, as indicated in Table 2, would be capable of maintaining tropical temperatures of 5°–10°C below their radiative equilibrium values just above the cold point.

Whether the mass flux through the cold-point tropopause in the plume of planetary-scale ascent over the western Pacific dominates the ventilation of stratosphere depends upon how it compares in magnitude with the mass flux through this level in the mean meridional circulation induced by the upward flux of wave activity in the EPW, as proposed by Norton (2006), and with the rate of ascent in the Brewer–Dobson circulation forced by the equatorward flux of wave activity in the lower stratosphere. A quantitative comparison of these three contributions to the ventilation of the stratosphere is beyond the scope of this paper.

9. Concluding remarks

We have made a cursory examination of the structure of the EPW in the NCEP–NCAR reanalyses, which also allows for the possibility of radiatively active subvisible cloud layers. While most of the features discussed in this study are also apparent in the ERA-40 and NCEP–NCAR datasets, there are important differences in the representation of the capping layer. The plume over the western Pacific warm pool does not extend as high into the capping layer in the NCEP–NCAR reanalyses as it does in ERA-40. There is no indication of the plume at the 100-hPa level and the vertical velocities at that level are of mixed sign, as opposed to being predominantly upward in ERA-40. Results of Trenberth et al. (2000) suggest that the planetary waves are slightly deeper in the ECMWF operational analyses than in the NCEP–NCAR reanalyses. These differences serve as a reminder that the reanalysis products, despite the observational constraints imposed on them, are still to some degree model-dependent. Two acknowledged shortcomings of ERA-40 that would tend to contribute to an overestimate of the 100-hPa vertical velocities are the excessive strength of the stratospheric Brewer–Dobson circulation and the excessively heavy rainfall over the western Pacific, both of which are documented in Uppala et al. (2005). Hence, it should be emphasized that the statistics presented in Table 2 are not necessarily representative of the real atmosphere.

Even if the actual vertical velocities over the western Pacific at the 100-hPa level were only ⅓ as strong as those in ERA-40, the implied heating rates would still be much larger than the clear-sky radiative heating rates, and the adiabatic cooling would still be stronger than the cooling produced by the overshooting tops of convective clouds in the model of Kuang and Bretherton (2004). To invalidate our conclusion that the mass and heat balances at this level are strongly influenced by the ascent of stably stratified air outside of convective clouds, either the vertical velocities would have to be overestimated by more than an order of magnitude in ERA-40, and/or the convective mass fluxes would need to be much stronger than those obtained in the simulation of Kuang and Bretherton.

In the ERA-40 scheme, thin and subvisible cirrus cloud layers are quite extensive around the cold-point tropopause over the western Pacific during boreal winter. That the winter-mean analysis increment (i.e., the mean correction that needs to be added to the 6-h forecasts to make them consistent with the observations) reaches a maximum warming of about 1°C day−1 at the 100-hPa level over the western Pacific suggests that either the simulated radiative heating rates in the clouds are too small or the upward vertical motion are too large (A. J. Simmons 2006, personal communication).

The three-dimensional trajectories of air parcels entering the stratosphere are much more complicated and diverse than the time-averaged Eulerian-mean vertical velocity fields presented in Figs. 4 –6 would suggest, as clearly illustrated in the studies of Jackson et al. (2001), Fueglistaler et al. (2004), Cau et al. (2005), and Fueglistaler et al. (2005). Air parcels ascend gently as they pass through the regions of Eulerian time–mean upward motion, but the air parcels that reach the cold point are not necessarily the ones that ascended to the base of the capping layer in deep convection over the western Pacific.

In interpreting the vertical structure of the EPW in ERA-40 we have drawn a distinction between the planetary-scale and convective-scale vertical velocity fields, postulating the existence of a plume of planetary-scale ascent over the western Pacific that penetrates deep into the capping layer, beyond the tops of all but the tallest convective clouds, as depicted schematically in Fig. 14. Such a distinction is more clearly evident in the Asian summer monsoon, in which most of the deep convection occurs over India, southeast Asia, and the adjacent oceans while the associated plume of cold, planetary-scale ascent in the capping layer is located farther to the north, in the northerly flow on the eastern flank of the Tibetan anticyclone (W. J. Randel, NCAR, 2006, personal communication).

The analysis presented in this paper is largely based on the thermodynamic energy balance and the inferences concerning the energetics that can be derived from it. The large remaining uncertainties concerning the strength of the ascent over the western Pacific at the 100-hPa level might be narrowed by considering the vorticity balance in the EPW.

Acknowledgments

We thank Stefan Fueglistaler and Qiang Fu for stimulating conversations that were instrumental in shaping the ideas in this paper; Adrian Simmons for the detailed information that he provided concerning ERA-40; William Randel, Brian Hoskins, and Ian Kraucunas for their constructive reviews; and Christopher Bretherton and Robert Wood for helpful comments and suggestions. This work was supported by the National Science Foundation under Grant ATM 0318675.

REFERENCES

  • Ackerman, T. P., , K-N. Liou, , F. P. J. Valero, , and L. Pfister, 1988: Heating rates in tropical anvils. J. Atmos. Sci., 45 , 16061623.

  • Back, L. E., , and C. S. Bretherton, 2006: Geographic variability in the export of moist static energy and vertical motion profiles in the tropical Pacific. Geophys. Res. Lett., 33 .L17810, doi:10.1029/2006GL026672.

    • Search Google Scholar
    • Export Citation
  • Bjerknes, J., 1966: A possible response of the atmospheric Hadley circulation to the equatorial anomalies of ocean temperature. Tellus, 18 , 820829.

    • Search Google Scholar
    • Export Citation
  • Bjerknes, J., 1969: Atmospheric teleconnections from the equatorial Pacific. Mon. Wea. Rev., 97 , 163172.

  • Boehm, M. T., , and S. Lee, 2003: The implications of tropical Rossby waves for tropical tropopause cirrus formation and for the equatorial upwelling of the Brewer–Dobson circulation. J. Atmos. Sci., 60 , 247261.

    • Search Google Scholar
    • Export Citation
  • Cau, P., , J. Methven, , and B. J. Hoskins, 2005: Representation of dry tropical layers and their origins in ERA-40 data. J. Geophys. Res., 110 .D06110, doi:10.1029/2004JD004928.

    • Search Google Scholar
    • Export Citation
  • Ciesielski, P. E., , R. H. Johnson, , P. T. Haertel, , and J. Wang, 2003: Corrected TOGA COARE sounding humidity data: Impact on diagnosed properties of convection and climate over the warm pool. J. Climate, 16 , 23702384.

    • Search Google Scholar
    • Export Citation
  • Corti, T., , B. P. Luo, , Q. Fu, , H. Vomel, , and T. Peter, 2006: The impact of cirrus clouds on tropical troposphere-to-stratosphere transport. Atmos. Chem. Phys. Discuss., 6 , 17251747.

    • Search Google Scholar
    • Export Citation
  • Danielsen, E. F., 1993: In situ evidence of rapid, vertical, irreversible transport of lower tropospheric air into the lower tropical stratosphere by convective cloud turrets and by larger-scale upwelling in tropical cyclones. J. Geophys. Res., 98 , 86658681.

    • Search Google Scholar
    • Export Citation
  • Dima, I. M., , J. M. Wallace, , and I. Kraucunas, 2005: Tropical zonal momentum balance in the NCEP reanalyses. J. Atmos. Sci., 62 , 24992513.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , and R. V. Martin, 2005: The vertical structure of tropical convection and its impact on the budgets of water vapor and ozone. J. Atmos. Sci., 62 , 15601573.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , M. Loewenstein, , J. Podolske, , S. J. Oltmans, , and M. Proffitt, 1999: A barrier to vertical mixing at 14 km in the tropics: Evidence from ozonesondes and aircraft measurements. J. Geophys. Res., 104 , D18. 2209522102.

    • Search Google Scholar
    • Export Citation
  • Folkins, I., , K. K. Kelly, , and E. M. Weinstock, 2002: A simple explanation for the increase in relative humidity between 11 and 14 km in the tropics. J. Geophys. Res., 107 .D23. 4736, doi:10.1029/2002JD002185.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , H. Wernli, , and T. Peter, 2004: Tropical troposphere-to-stratosphere transport inferred from trajectory calculations. J. Geophys. Res., 109 .D03108, doi:10.1029/2003JD004069.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , M. Bonazzola, , P. H. Haynes, , and T. Peter, 2005: Stratospheric water vapor predicted from the Lagrangian temperature history of air entering the stratosphere in the tropics. J. Geophys. Res., 110 .D08107, doi:10.1029/2004JD005516.

    • Search Google Scholar
    • Export Citation
  • Gettelman, A., , and P. M. de F. Forster, 2002: A climatology of the tropical tropopause layer. J. Meteor. Soc. Japan, 80 , 911924.

  • Gill, A. E., 1980: Some simple solutions for heat-induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106 , 447462.

  • Highwood, E. J., , and B. J. Hoskins, 1998: The tropical tropopause. Quart. J. Roy. Meteor. Soc., 124 , 15791604.

  • Hoskins, B. J., , A. J. Simmons, , and D. G. Andrews, 1977: Energy dispersion in a barotropic atmosphere. Quart. J. Roy. Meteor. Soc., 103 , 553567.

    • Search Google Scholar
    • Export Citation
  • Jackson, D. R., , J. Methven, , and V. Pope, 2001: Transport in the low latitude tropopause zone diagnosed using particle trajectories. J. Atmos. Sci., 58 , 173192.

    • Search Google Scholar
    • Export Citation
  • Kerr-Munslow, A. M., , and W. A. Norton, 2006: Tropical wave driving of the annual cycle in tropical tropopause temperatures. Part I: ECMWF analyses. J. Atmos. Sci., 63 , 14101419.

    • Search Google Scholar
    • Export Citation
  • Kley, D., , A. L. Schmeltekopf, , K. Kelly, , R. H. Winkler, , T. L. Thompson, , and M. McFarland, 1982: Transport of water through the tropical tropopause. Geophys. Res. Lett., 9 , 617620.

    • Search Google Scholar
    • Export Citation
  • Knollenberg, R. G., , A. J. Dascher, , and D. Huffman, 1982: Measurements of the aerosol and ice crystal populations in tropical stratospheric cumulonimbus anvils. Geophys. Res. Lett., 9 , 613616.

    • Search Google Scholar
    • Export Citation
  • Kraucunas, I., , and D. L. Hartmann, 2007: Tropical stationary waves in a nonlinear shallow-water model with realistic basic states. J. Atmos. Sci., 64 , 25402557.

    • Search Google Scholar
    • Export Citation
  • Krishnamurti, T. N., 1971: Tropical east–west circulations during the northern summer. J. Atmos. Sci., 28 , 13421347.

  • Krishnamurti, T. N., , M. Kanamitsu, , W. J. Koss, , and J. D. Lee, 1973: Tropical east–west circulations during the northern winter. J. Atmos. Sci., 30 , 780787.

    • Search Google Scholar
    • Export Citation
  • Kuang, Z., , and C. S. Bretherton, 2004: Convective influence of the heat balance of the tropical tropopause layer: A cloud-resolving model study. J. Atmos. Sci., 61 , 29192927.

    • Search Google Scholar
    • Export Citation
  • Lee, S., 1999: Why are the climatological zonal winds easterly in the equatorial upper troposphere. J. Atmos. Sci., 56 , 13531363.

  • Lilly, D. K., 1988: Cirrus outflow dynamics. J. Atmos. Sci., 45 , 15941605.

  • Matsuno, T., 1966: Quasi-geostrophic motions in the equatorial area. J. Meteor. Soc. Japan, 44 , 2542.

  • McFarquhar, G. M., , A. J. Heymsfield, , J. Spinhirne, , and B. Hart, 2000: Thin and subvisual tropopause tropical cirrus: Observations and radiative impacts. J. Atmos. Sci., 57 , 18411853.

    • Search Google Scholar
    • Export Citation
  • Mote, P. W., and Coauthors, 1996: An atmospheric tape recorder: The imprint of tropical tropopause temperatures on stratospheric water vapor. J. Geophys. Res., 101 , 39894006.

    • Search Google Scholar
    • Export Citation
  • Newell, R. E., , and S. Gould-Stewart, 1981: A stratospheric fountain? J. Atmos. Sci., 38 , 27892796.

  • Newell, R. E., , J. W. Kidson, , D. G. Vincent, , and G. J. Boer, 1974: The General Circulation of the Tropical Atmosphere and Interactions with Extratropical Latitudes. Vol. 2, MIT Press, 371 pp.

    • Search Google Scholar
    • Export Citation
  • Norton, W. A., 2001: Longwave heating of the tropical lower stratosphere. Geophys. Res. Lett., 28 , 36533656.

  • Norton, W. A., 2006: Tropical wave driving of the annual cycle in tropical tropopause temperatures. Part II: Model results. J. Atmos. Sci., 63 , 14201431.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., , and F. Wu, 2005: Kelvin wave variability near the equatorial tropopause observed in GPS radio occultation measurements. J. Geophys. Res., 110 .D03102, doi:10.1029/2004JD005006.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., , R. R. Garcia, , and F. Wu, 2002: Time-dependent upwelling in the tropical lower stratosphere estimated from the zonal-mean momentum budget. J. Atmos. Sci., 59 , 21412152.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , and C. L. Vlcek, 1969: The annual temperature variation in the lower tropical stratosphere. J. Atmos. Sci., 26 , 163167.

  • Sadler, J. C., cited. 2006: The upper tropospheric circulation over the global tropics. [Available online at http://www.soest.hawaii.edu/Library/Sadler.html#database.].

  • Sherwood, S. C., 2000: A stratospheric “drain” over the Maritime Continent. Geophys. Res. Lett., 27 , 677680.

  • Simmons, A. J., 1982: The forcing of stationary wave motion by tropical diabatic heating. Quart. J. Roy. Meteor. Soc., 108 , 503534.

  • Simmons, A. J., , A. Untch, , C. Jakob, , P. Kallberg, , and P. Unden, 1999: Stratospheric water vapor and tropical tropopause temperatures in ECMWF analyses and multi-year simulations. Quart. J. Roy. Meteor. Soc., 125 , 353386.

    • Search Google Scholar
    • Export Citation
  • Thompson Jr., R. M., , S. W. Payne, , E. E. Recker, , and R. J. Reed, 1979: Structure and properties of synoptic-scale wave disturbances in the intertropical convergence zone of the eastern Atlantic. J. Atmos. Sci., 36 , 5372.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , D. P. Stepaniak, , and J. M. Caron, 2000: The global monsoon as seen throught the divergent atmospheric circulation. J. Climate, 13 , 39693993.

    • Search Google Scholar
    • Export Citation
  • Troup, A. J., 1965: The Southern Oscillation. Quart. J. Roy. Meteor. Soc., 91 , 490506.

  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Reanalysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Wang, P-H., , P. Minnis, , M. P. McCormick, , G. S. Kent, , and K. M. Skeens, 1996: A 6-year climatology of cloud occurrence frequency from Stratospheric Aerosol and Gas Experiment II observations (1985–1990). J. Geophys. Res., 101 , D23. 2940729430.

    • Search Google Scholar
    • Export Citation
  • Webster, P. J., 1972: Response of the tropical atmosphere to local steady forcing. Mon. Wea. Rev., 100 , 518541.

  • Webster, P. J., 1981: Mechanisms determining the atmospheric response to sea surface temperature anomalies. J. Atmos. Sci., 38 , 554571.

    • Search Google Scholar
    • Export Citation
  • Woodruff, S. D., , R. J. Slutz, , R. L. Jenne, , and P. M. Steurer, 1987: A comprehensive ocean–atmosphere data set. Bull. Amer. Meteor. Soc., 68 , 12391250.

    • 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.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Annual-mean sea surface temperature (contour interval = 1°C) and 300-hPa vertical velocity (colored shading, Pa s−1). The 25°C sea surface temperature contour is shown in red, for reference.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 2.
Fig. 2.

Annual-mean 300-hPa vertical velocity (colored shading, Pa s−1) superimposed on (top) 150-hPa geopotential height (contours, m) and wind (arrows, m s−1); and (bottom) sea level pressure (contour interval = 1.5 hPa, min/max values shown 1009/1021 hPa) and boundary layer (1000–925 hPa) surface winds (arrows). The wind arrows are plotted every 7.5° latitude × 15° longitude, up to 23° latitude. The contour interval for geopotential height is 100 m (gray lines); additional contours at 10-m intervals (black) are inserted in the tropical belt. Contour succession for 150-hPa height: (. . . 14 100, 14 200, 14 210, 14 220, . . .) m, with the first black contour at the separation between gray and black contours corresponding to 14 210 m.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 3.
Fig. 3.

Annual-mean temperature of various layers as indicated, as estimated from the respective thicknesses (contour interval = 0.5°C for black lines and 1°C for gray lines). Superimposed are the wind (arrows, m s−1) and vertical velocity (colored shading, Pa s−1) estimated at (a) 100, (b) 300, (c) 600, and (d) 925 hPa. The first black contour at the separation between gray and black lines represents (a) −72°, (b) −36°, (c) −3°, and (d) 22°C. Note that in the top panel the polarity of the temperature pattern is reversed and the vertical velocities are much smaller than in the other panels.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 4.
Fig. 4.

Longitude–height cross section of the eddy component of geopotential height Z* (colored shading, m) and zonal and vertical wind components (arrows, m s−1 for zonal wind and cm s−1 for vertical velocity) for the latitudinally averaged belt 10°S–5°N. The longitudinal sector 0°–60°E is repeated on the rhs for visual continuity. The vertical velocities are stretched relative to the zonal velocities to make them more clearly visible.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 5.
Fig. 5.

As in Fig. 4 but for the eddy component of the temperature field T*, in units of °C. The vertical velocity arrows have been rescaled by dividing them by the horizontally averaged static stability (Γd − Γ) at each level to make them proportional to heating rate (°C day−1).

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 6.
Fig. 6.

As in Fig. 5 but for the total relative humidity field, in units of %.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 7.
Fig. 7.

Vertical profiles of rms amplitude of the annual-mean stationary waves in the (a) eddy geopotential height (m), (b) temperature (°C), (c) vertical velocity (Pa s−1) fields, and (d) covariance between vertical velocity and temperature (Pa °C s−1).

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 8.
Fig. 8.

Vertical profiles of vertical velocity (Pa s−1) in pressure coordinates in different regions, as indicated by the red rectangles in the bottom plot showing the annual-mean 300-hPa vertical velocity. Color bar is the same as in Fig. 1.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 9.
Fig. 9.

Annual-mean meridional cross section for the sector extending across the Pacific, from 180° to 90°W, showing temperatures (departures from the 20°N–20°S means) and the meridional wind component and vertical velocity averaged over this sector. Plotting and scaling conventions are the same as in Fig. 5.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 10.
Fig. 10.

Time–longitude section of the eddy component of the 150-hPa geopotential height field indicated by shading, color bar in meters, and the total (zonal mean plus eddy) fields of 150-hPa zonal wind in m s−1 and 300-hPa vertical velocity in cm s−1, as indicated by the arrows. Data are meridionally averaged from 10°S to 5°N.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 11.
Fig. 11.

As in Fig. 10 but for 100-hPa temperature, zonal wind, and vertical velocity.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 12.
Fig. 12.

Maps of eddy fields: (top) regression on standardized CTI with sign reversed (contour interval = 2.5 m); (bottom) annual mean EPW, for the 150-hPa geopotential height (contour interval is 10 m). Dashed contours indicate negative values, and wind (arrows, m s−1) and 300-hPa vertical velocity (colored shading, Pa s−1) are also shown.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 13.
Fig. 13.

Composite maps based on extreme values of the CTI, an indicator of the status of the ENSO cycle. The fields represented are 150-hPa geopotential height (contours), wind (arrows), and 300-hPa vertical velocity (color). Plotting conventions are the same as in Fig. 2a.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 14.
Fig. 14.

Schematic of the vertical structure of the EPW over the western Pacific, as represented in the ERA-40 reanalyses. The layering scheme shown at the left is based on the profiles of temperature, vertical velocity, and the release of available potential energy shown in Fig. 7. Most of the detrainment from the hypothetical deep convective cloud is depicted as occurring below 175 hPa, the level of peak amplitude of the EPW. Convectively forced descent is indicated by the downward-pointing arrows to the right of the cloud, and overshooting cloud tops by the small turret protruding from the top of the cloud. The plume of planetary-scale ascent over the western Pacific is represented by the upward-pointing arrows just above the convective cloud. Extensive subvisible cirrus cloud layers, represented by the wavy lines, are assumed to be present within the plume. The plume is depicted as spreading out at or just above the cold point, ventilating and lifting the entire tropical lower stratosphere. The vertical profile at the right represents the planetary-scale divergence over the western Pacific. The planetary-scale divergence is depicted as peaking at or just above the level of strongest detrainment from the convective clouds.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Fig. 15.
Fig. 15.

Schematic view of the impact of the EPW on the vertical velocity and temperature fields. The region with convective clouds in the center of the figure represents the equatorial Pacific and the full width of the diagram represents the entire tropical belt. Arrows indicate the strength of the upwelling, and the blue shading indicates the intensity of the adiabatic cooling. Above the cold point, the cooling is stronger because of the higher static stability.

Citation: Journal of the Atmospheric Sciences 64, 8; 10.1175/JAS3985.1

Table 1.

Warm and cold ENSO years used in composites in Fig. 13.

Table 1.
Table 2.

Numerical values of annual-mean vertical velocity and vertical mass flux at selected levels: WP = western Pacific (125°E–160°W), ZAvg = zonal average.

Table 2.
Save