• Barros, A. P., , and T. J. Lang, 2003: Monitoring the monsoon in the Himalayas: Observations in central Nepal, June 2001. Mon. Wea. Rev., 131, 14081427.

    • Search Google Scholar
    • Export Citation
  • Bhatt, B. C., , and K. Nakamura, 2005: Characteristics of monsoon rainfall around the Himalayas revealed by TRMM precipitation radar. Mon. Wea. Rev., 133, 149165.

    • Search Google Scholar
    • Export Citation
  • Bhatt, B. C., , and K. Nakamura, 2006: A climatological-dynamical analysis associated with precipitation around the southern part of the Himalayas. J. Geophys. Res., 111, D02115, doi:10.1029/2005JD006197.

    • Search Google Scholar
    • Export Citation
  • Boos, W. R., , and Z. Kuang, 2010: Dominant control of the South Asian monsoon by orographic insulation versus plateau heating. Nature, 463, 218222.

    • Search Google Scholar
    • Export Citation
  • Chung, E.-S., , B.-J. Sohn, , and J. Schmetz, 2008: CloudSat shedding new light on high-reaching tropical deep convection observed with Meteosat. Geophys. Res. Lett., 35, L02814, doi:10.1029/2007GL032516.

    • Search Google Scholar
    • Export Citation
  • CIRA, 2008: CloudSat standard data products handbook. Cooperative Institute for Research in the Atmosphere, Colorado State University, 18 pp. [Available online at http://www.cloudsat.cira.colostate.edu/dataHome.php.]

    • Search Google Scholar
    • Export Citation
  • Dai, A., 2001: Global precipitation and thunderstorm frequencies. Part II: Diurnal variations. J. Climate, 14, 11121128.

  • Dai, A., , F. Giorgi, , and K. E. Trenberth, 1999: Observed and model-simulated diurnal cycles of precipitation over the contiguous United States. J. Geophys. Res., 104, 63776402.

    • Search Google Scholar
    • Export Citation
  • Ding, Y. H., 2007: The variability of the Asian summer monsoon. J. Meteor. Soc. Japan, 85, 2154.

  • Emanuel, K., 1994: Atmospheric Convection. Oxford University Press, 580 pp.

  • Forster, P. M. de F., , and K. P. Shine, 2002: Assessing the climate impact of trends in stratospheric water vapor. Geophys. Res. Lett., 29, 1086, doi:10.1029/2001GL013909.

    • Search Google Scholar
    • Export Citation
  • Fu, R., and Coauthors, 2006: Short circuit of water vapor and polluted air to the global stratosphere by convective transport over the Tibetan Plateau. Proc. Natl. Acad. Sci. USA, 103, 56645669.

    • Search Google Scholar
    • Export Citation
  • Fu, Y., , Q. Liu, , Y. Zi, , S. Feng, , Y. Li, , and G. Liu, 2008: Summer precipitation and latent heating over the Tibetan Plateau based on TRMM measurements (in Chinese). Plateau Mt. Meteor. Res., 28, 818.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , A. E. Dessler, , T. J. Dunkerton, , I. Folkins, , Q. Fu, , and P. W. Mote, 2009: Tropical tropopause layer. Rev. Geophys., 47, RG1004, doi:10.1029/2008RG000267.

    • Search Google Scholar
    • Export Citation
  • Fujinami, H., , and T. Yasunari, 2001: The seasonal and intraseasonal variability of diurnal cloud activity over the Tibetan Plateau. J. Meteor. Soc. Japan, 79, 12071227.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., , and R. W. Jacobson Jr., 1977: Diurnal variation of deep cumulus convection. Mon. Wea. Rev., 105, 11711188.

  • Hendon, H., , and K. Woodberry, 1993: The diurnal cycle of tropical convection. J. Geophys. Res., 98, 162 62316 2637, doi:10.1029/93JD00525.

    • Search Google Scholar
    • Export Citation
  • Houze, R. A., Jr., , D. C. Wilton, , and B. F. Smull, 2007: Monsoon convection in the Himalayan region as seen by the TRMM precipitation radar. Quart. J. Roy. Meteor. Soc., 133, 13891411, doi:10.1002/gq.106.

    • Search Google Scholar
    • Export Citation
  • Im, E., , S. L. Durden, , and C. Wu, 2006: Cloud profiling radar for the CloudSat mission. IEEE Trans. Aerosp. Electron. Syst., 20, 1518.

    • Search Google Scholar
    • Export Citation
  • Jensen, E. J., , A. S. Ackerman, , and J. A. Smith, 2007: Can overshooting convection dehydrate the tropical tropopause layer? J. Geophys. Res., 112, D11209, doi:10.1029/2006JD007943.

    • Search Google Scholar
    • Export Citation
  • Kirk-Davidoff, D. B., , E. J. Hintsa, , J. G. Anderson, , and D. W. Keith, 1999: The effect of climate change on ozone depletion through changes in stratospheric water vapour. Nature, 402, 399401.

    • Search Google Scholar
    • Export Citation
  • Li, W., , F. Li, , Z. Zhao, , F. Liu, , B. Li, , and H. Li, 2009: Technical evaluation report of the construction of L-band meteorological observing systems (in Chinese). China Meteorological Press, 95 pp.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. Zipser, , and S. W. Nesbitt, 2007: Global distribution of tropical deep convection: Different perspectives using infrared and radar as the primary data source. J. Climate, 20, 489503.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. Zipser, , G. Mace, , and S. Benson, 2008: Implications of the differences between daytime and nighttime CloudSat observations over the tropics. J. Geophys. Res., 113, D00A04, doi:10.1029/2008JD009783.

    • Search Google Scholar
    • Export Citation
  • Luo, Z., , and W. B. Rossow, 2004: Characterizing tropical cirrus life cycle, evolution, and interaction with upper-tropospheric water vapor using Lagrangian trajectory analysis of satellite observations. J. Climate, 17, 45414563.

    • Search Google Scholar
    • Export Citation
  • Luo, Z., , G. Y. Liu, , and G. L. Stephens, 2008: CloudSat adding new insight into tropical penetrating convection. Geophys. Res. Lett., 35, L19819, doi:10.1029/2008GL035330.

    • Search Google Scholar
    • Export Citation
  • Mace, G. G., , R. Marchand, , Q. Zhang, , and G. Stephens, 2007: Global hydrometeor occurrence as observed by CloudSat: Initial observations from summer 2006. Geophys. Res. Lett., 34, L09808, doi:10.1029/2006GL029017.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , R. A. Houze Jr., , A. Kumar, , and D. Niyogi, 2010: Summer monsoon convection in the Himalayan region: Terrain and land cover effects. Quart. J. Roy. Meteor. Soc., 136, 593616.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., , and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16, 14561475.

    • Search Google Scholar
    • Export Citation
  • Oltmans, S. J., , and D. J. Hofmann, 2002: Increase in lower-stratospheric water vapour at a mid-latitude Northern Hemisphere site from 1981 to 1994. Nature, 374, 146149, doi:10.1038/374146a0.

    • Search Google Scholar
    • Export Citation
  • Park, M., , W. J. Randel, , A. Gettelman, , S. T. Massie, , and J. H. Jiang, 2007: Transport above the Asian summer monsoon anticyclone inferred from Aura Microwave Limb Sounder tracers. J. Geophys. Res., 112, D16309, doi:10.1029/2006JD008294.

    • Search Google Scholar
    • Export Citation
  • Romatschke, U., , and R. A. Houze Jr., 2011: Characteristics of precipitating convective systems in the South Asian monsoon. J. Hydrometeor., 12, 157180.

    • Search Google Scholar
    • Export Citation
  • Romatschke, U., , S. Medina, , and R. A. Houze Jr., 2010: Regional, seasonal, and diurnal variations of extreme convection in the South Asian region. J. Climate, 23, 419439.

    • Search Google Scholar
    • Export Citation
  • Sawyer, J. S., 1947: The structure of the intertropical front over N.W. India during the S.W. monsoon. Quart. J. Roy. Meteor. Soc., 73, 346369.

    • Search Google Scholar
    • Export Citation
  • Smith, C. A., , R. Toumi, , and J. D. Haigh, 2000: Seasonal trends in stratospheric water vapour. Geophys. Res. Lett., 27, 16871690.

  • Stephens, G. L., and Coauthors, 2002: The CloudSat mission and the A-Train. Bull. Amer. Meteor. Soc., 83, 17711790.

  • Stephens, G. L., and Coauthors, 2008: CloudSat mission: Performance and early science after the first year of operation. J. Geophys. Res., 113, D00A18, doi:10.1029/2008JD009982.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , and LinHo, 2002: Rainy season of the Asian–Pacific summer monsoon. J. Climate, 15, 386398.

  • Wang, B., , T. Li, , Y. H. Ding, , R. H. Zhang, , and H. J. Wang, 2005: Eastern Asian–western North Pacific monsoon: A distinctive component of the Asian–Australian monsoon system. The Global Monsoon System: Research and Forecast, WMO/TD-1266, TMRP Rep. 70, C.-P. Chang, Eds., WMO, 72–94.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , J. R. Pelon, , and M. P. McCormick, 2003: The CALIPSO mission: Spaceborne lidar for observation of aerosols and clouds. Lidar Remote Sensing for Industry and Environment Monitoring III, U. N. Singh et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 4893), 1–11.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , W. H. Hunt, , and M. J. McGill, 2007: Initial performance assessment of CALIOP. Geophys. Res. Lett., 34, L19803, doi:10.1029/2007GL030135.

    • Search Google Scholar
    • Export Citation
  • Yu, R., , T. Zhou, , A. Xiong, , Y. Zhu, , and J. Li, 2007: Diurnal variations of summer precipitation over contiguous China. Geophys. Res. Lett., 34, L01704, doi:10.1029/2006GL028129.

    • Search Google Scholar
    • Export Citation
  • Zhou, S., , and R. Zhang, 2005: Decadal variations of temperature and geopotential height over the Tibetan Plateau and their relations with Tibet ozone depletion. Geophys. Res. Lett., 32, L18705, doi:10.1029/2005GL023496.

    • Search Google Scholar
    • Export Citation
  • Zhou, X. J., , C. Luo, , W. Li, , and J. Shi, 1995: Trend in ozone over China and low center over the Tibetan Plateau (in Chinese). Chin. Sci. Bull., 4, 13961398.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Regions defined in this study are highlighted with dashed rectangles. Definitions of the regions are described in detail in the text. Shading represents topography in kilometers of elevation. White dots represent the meteorological stations over the TP.

  • View in gallery

    Vertical cross sections of radar reflectivity (color shading) detected by the CloudSat CPR over the TP–SAMR during (a) the nighttime overpass on 28 Aug 2006 and (b) the daytime overpass on 23 Jun 2009. Black and red dots indicate locations of cloud-layer top and cloud-layer bottom from GEOPROF_LIDAR. Blue lines and green dots near the top of each panel represent horizontal locations of deep convective systems and deep convective cores, respectively. Gray shadings represent elevation of the land surface.

  • View in gallery

    Normalized occurrence frequencies (%) for deep convective systems with various horizontal spans (L) in the 7 regions. For each region, bars represent (left to right) small (L < 120 km), medium (120 < L < 320 km), and large (L > 320 km) convective systems.

  • View in gallery

    Horizontal spans of convective systems averaged (a) over the daytime and the nighttime observations and (b) over the daytime observations (bars without shading) and the nighttime observations (bars with shadings) for the 7 analysis regions. Middle of each bar represents the mean value, and top (bottom) of each bar indicates the mean plus (minus) the standard deviation.

  • View in gallery

    Heights of the tops of deep convective clouds detected by CTH_lidar and CTH_CPR for the 7 regions. Bars defined in Fig. 4.

  • View in gallery

    Heights of tops of 0- and 10-dBZ echo in deep convective cores detected by CloudSat CPR (H_0 dBZ and H_10 dBZ, respectively) for the 7 regions. Bars defined in Fig. 4.

  • View in gallery

    Distances between cloud top detected by CALIPSO lidar and top of 0- and 10-dBZ echo detected by CloudSat CPR (CTH_lidar − H_0dBZ, CTH_lidar − H_10dBZ, respectively) in deep convective cores. Bars defined in Fig. 4.

  • View in gallery

    Schematics of deep convection over the TP, PSS, and in SAMR.

  • View in gallery

    (a) Height of LNB, (b) TPW, and (c) CAPE for the 7 regions. Bars defined in Fig. 4.

  • View in gallery

    Histograms of (a) CTH (CALIPSO lidar), (b) CTH (CloudSat CPR), maximum height of (c) 0 dBZ and (d) 10 dBZ for deep convection. The line color represents each of the 7 regions.

  • View in gallery

    Differences (km) in the means of the physical parameters of deep convective cores between the daytime and nighttime observations (i.e., daytime minus nighttime) for the 7 regions: (a) CTH (CALIPSO lidar), (b) CTH (CloudSat CPR), (c) maximum height of 0 dBZ, (d) maximum height of 10 dBZ, and distance between cloud-top (CALIPSO lidar) and top of (e) 0 dBZ and (f) 10 dBZ.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 97 97 14
PDF Downloads 62 62 6

Intercomparison of Deep Convection over the Tibetan Plateau–Asian Monsoon Region and Subtropical North America in Boreal Summer Using CloudSat/CALIPSO Data

View More View Less
  • 1 State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Society, Beijing, China
  • | 2 Department of Earth and Atmospheric Sciences, and NOAA/CREST Center, City College, City University of New York, New York, New York
  • | 3 School of Mathematical Sciences, Fudan University, Shanghai, China
© Get Permissions
Full access

Abstract

Deep convection in the Tibetan Plateau–southern Asian monsoon region (TP–SAMR) is analyzed using CloudSat and Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) data for the boreal summer season (June–August) from 2006 to 2009. Three subregions are defined—the TP, the southern slope of the plateau (PSS), and the SAMR—and deep convection properties (such as occurrence frequency, internal vertical structure, system size, and local environment) are compared among these subregions. To cast them in a broader context, four additional regions that bear some similarity to the TP–SAMR are also discussed: East Asia (EA), tropical northwestern Pacific (NWP), and western and eastern North America (WNA and ENA, respectively).

The principal findings are as follows: 1) Compared to the other two subregions of the TP–SAMR, deep convection over the TP is shallower, less frequent, and embedded in smaller-size convection systems, but the cloud tops are more densely packed. These characteristics of deep convection over the TP are closely related to the unique local environment, namely, a significantly lower level of neutral buoyancy (LNB) and much drier atmosphere. 2) In a broader context in which all seven regions are brought together, deep convection in the two tropical regions (NWP and SAMR; mostly over ocean) is similar in many regards. A similar conclusion can be drawn among the four subtropical continental regions (TP, EA, WNA, and ENA). However, tropical oceanic and subtropical land regions present some significant contrasts: deep convection in the latter region occurs less frequently, has lower cloud tops but comparable or slightly higher tops of large radar echo (e.g., 0 and 10 dBZ), and is embedded in smaller systems. The cloud tops of the subtropical land regions are generally more densely packed. Hence, the difference between the TP and SAMR is more of a general contrast between subtropical land regions and tropical oceanic regions during the boreal summer. 3) Deep convection over the PSS possesses some uniqueness of its own because of the distinctive terrain (slopes) and moist low-level monsoon flow. 4) Results from a comparison between the daytime (1:30 p.m.) and nighttime (1:30 a.m.) overpasses are largely consistent with researchers’ general understanding of the diurnal variation of tropical and subtropical deep convection.

Corresponding author address: Dr. Yali Luo, State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, 46 Zhong-Guan-Cun South Ave., Beijing 100081, China. E-mail: yali@cams.cma.gov.cn

Abstract

Deep convection in the Tibetan Plateau–southern Asian monsoon region (TP–SAMR) is analyzed using CloudSat and Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) data for the boreal summer season (June–August) from 2006 to 2009. Three subregions are defined—the TP, the southern slope of the plateau (PSS), and the SAMR—and deep convection properties (such as occurrence frequency, internal vertical structure, system size, and local environment) are compared among these subregions. To cast them in a broader context, four additional regions that bear some similarity to the TP–SAMR are also discussed: East Asia (EA), tropical northwestern Pacific (NWP), and western and eastern North America (WNA and ENA, respectively).

The principal findings are as follows: 1) Compared to the other two subregions of the TP–SAMR, deep convection over the TP is shallower, less frequent, and embedded in smaller-size convection systems, but the cloud tops are more densely packed. These characteristics of deep convection over the TP are closely related to the unique local environment, namely, a significantly lower level of neutral buoyancy (LNB) and much drier atmosphere. 2) In a broader context in which all seven regions are brought together, deep convection in the two tropical regions (NWP and SAMR; mostly over ocean) is similar in many regards. A similar conclusion can be drawn among the four subtropical continental regions (TP, EA, WNA, and ENA). However, tropical oceanic and subtropical land regions present some significant contrasts: deep convection in the latter region occurs less frequently, has lower cloud tops but comparable or slightly higher tops of large radar echo (e.g., 0 and 10 dBZ), and is embedded in smaller systems. The cloud tops of the subtropical land regions are generally more densely packed. Hence, the difference between the TP and SAMR is more of a general contrast between subtropical land regions and tropical oceanic regions during the boreal summer. 3) Deep convection over the PSS possesses some uniqueness of its own because of the distinctive terrain (slopes) and moist low-level monsoon flow. 4) Results from a comparison between the daytime (1:30 p.m.) and nighttime (1:30 a.m.) overpasses are largely consistent with researchers’ general understanding of the diurnal variation of tropical and subtropical deep convection.

Corresponding author address: Dr. Yali Luo, State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, 46 Zhong-Guan-Cun South Ave., Beijing 100081, China. E-mail: yali@cams.cma.gov.cn

1. Introduction

Deep convection and associated precipitation in the Tibetan Plateau–Asian monsoon region is important to the agriculture and well-being of a large portion of the world’s population. Moreover, it interacts with large-scale circulation and may play a critical role in cross-tropopause transport of chemical tracers during boreal summer, which may have contributed to an increasing trend in stratospheric water vapor (Oltmans and Hofmann 2002) during the 1990s (Smith et al. 2000) and a decreasing trend in stratospheric ozone during the 1980s (Zhou et al. 1995; Zhou and Zhang 2005). The increasing trend in stratospheric water vapor probably increased the global greenhouse forcing (Forster and Shine 2002) and enhanced ozone depletion in the Arctic (Kirk-Davidoff et al. 1999). However, investigators have expressed different opinions on the properties of deep convection within this region. For example, Fu et al. (2006) provided evidence to show that tropospheric moist convection driven by elevated surface heating over the Tibetan Plateau is deeper and detrains more water vapor, CO, and ice at the tropopause than over the southern Asian monsoon region. Conversely, Park et al. (2007) argued that the strongest convection occurs over the Asian monsoon region, not over the Tibetan Plateau and its south slope. Since the study of Fu et al. (2006) is largely based on analysis of the Microwave Limb Sounder (MLS) measurements [with the Tropical Rainfall Measuring Mission (TRMM) data being briefly examined] and Park et al. (2007) use outgoing longwave radiation (OLR) as a convective proxy, their identification of deep convection may have included some high-level clouds that are not directly of convective origin [note that Luo and Rossow (2004) showed that about half of the tropical cirrus are not directly related to convective detrainment]. Recent studies by Houze et al. (2007) and Romatschke et al. (2010) involved detailed analysis of the TRMM precipitation radar (PR) data for the Himalayan and South Asian region. Their results indicate weaker convection over the plateau than the south slope of the plateau and the southern Asian monsoon region.

As major components of the A-Train satellite constellation (Stephens et al. 2002), the CloudSat and Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO; Winker et al. 2003) satellites were launched in April 2006 (Stephens et al. 2008), probing nearly the same volumes of the atmosphere within 10–15 s of each other. CloudSat carries with it a 94-GHz cloud profiling radar (CPR; Im et al. 2006) sensitive to both cloud- and precipitation-size particles. The Cloud–Aerosol Lidar with Orthogonal Polarization (Winker et al. 2007) aboard CALIPSO is able to detect optically thin clouds and tenuous cloud tops that could be missed by the CPR. Combined, CloudSat and CALIPSO measurements offer not only the precise location of tops of deep convective clouds but also a glimpse into the internal vertical structures of deep convective clouds, and thus they have provided a new opportunity to gain further insight into deep convection and their distribution globally. It is therefore of interest to use these new measurements to re-examine the question concerning the characteristics of deep convection over the Tibetan Plateau–southern Asian monsoon region.

A few recent studies demonstrated the values of using CloudSat and CALIPSO, in combination with other measurements, to study tropical convection. Chung et al. (2008) used CloudSat data and collocated IR information (from Meteosat-8) to relate warmer water vapor pixels to high-reaching tropical deep convection. Luo et al. (2008) analyzed CloudSat and Moderate Resolution Imaging Spectroradiometer (MODIS; onboard Aqua) data to characterize the internal vertical structure and life cycle of tropical penetrating convection. This current study builds upon the previous work and applies the analysis methods developed to further investigate the characteristics of deep convective clouds in the Tibetan Plateau–Asian monsoon region during boreal summer. First, we contrast deep convection properties among three subregions in the Tibetan Plateau–southern Asian monsoon region (TP–SAMR)—that is, the TP, the southern slope of the plateau (PSS), and the SAMR (shown in Fig. 1)—making an effort to answer the following questions: Does deep convection occur more frequently and penetrate deeper over the SAMR, the PSS, or the TP? How do internal vertical structures of deep convection and the horizontal span of convective systems differ among the three subregions? Then, for a more comprehensive understanding of deep convection in the Tibetan Plateau–Asian monsoon region, we take advantage of the global coverage of the CloudSat/CALIPSO observations and analyze deep convection properties in four other regions—that is, East Asia (EA), the tropical northwestern Pacific (NWP), and western and eastern North America (WNA, ENA) (shown in Fig. 1)—for the boreal summer. The EA and NWP regions are often considered as two other subsystems of the more generally defined Asian summer monsoon (e.g., Wang and LinHo 2002). The WNA and ENA are included because they bear some similarity to the TP and EA regions (i.e., mountainous region versus plains to the east). Houze et al. (2007) and Romatschke et al. (2010) discussed some of the similarities in meteorological conditions and nature of severe convection between the South Asian region and the Great Plains of the United States. Comparisons among the seven analysis regions are made, with a focus on contrasting between the tropical and subtropical regions and between the Asian (TP and EA) and North American regions (WNA and ENA).

Fig. 1.
Fig. 1.

Regions defined in this study are highlighted with dashed rectangles. Definitions of the regions are described in detail in the text. Shading represents topography in kilometers of elevation. White dots represent the meteorological stations over the TP.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

Although CloudSat–CALIPSO makes observations at approximately 1330 and 0130 local time (LT) and as such misses the peak of the diurnal cycle of deep convection over the continental regions (Gray and Jacobson 1977; Hendon and Woodberry 1993; Dai 2001; Nesbitt and Zipser 2003), we will show that contrasts in convection properties depicted by the CloudSat–CALIPSO observations are qualitatively consistent with findings obtained by other researchers using TRMM observations (e.g., Romatschke et al. 2010) that are able to capture the full diurnal cycle of deep convection statistically (but note that CloudSat–CALIPSO sees different parts of convective systems than TRMM). Despite this limitation, we examine the differences in deep convection properties between the CloudSat–CALIPSO daytime (1330 LT) and nighttime (0130 LT) observations as an attempt to at least partially reveal the diurnal differences. In addition to properties of deep convection, atmospheric conditions (e.g., moisture level and air buoyancy) in the regions are also examined to investigate their possible impacts on deep convection properties.

Data and methodology are described in section 2. Properties of deep convection, derived from averaging both the daytime and the nighttime observations, are compared among the TP, PSS, and SAMR in section 3 and among the seven analysis regions in section 4. Major differences between the daytime and nighttime observations are discussed in section 5. Section 6 provides a summary and discussions.

2. Data and methodology

To achieve the objectives, three CloudSat standard data products—publicly available as the level 2 2B Geometrical Profiling Product (GEOPROF), 2B-GEOPROF-lidar, and the European Centre for Medium-Range Weather Forecasts (ECMWF) auxiliary data from a CloudSat product (ECMWF-AUX) are used in the present study. Detailed information about the CloudSat products may be found in the CloudSat Standard Data Products Handbook (CIRA 2008). Specifically, we use vertical distribution of radar reflectivity and cloud mask from 2B-GEOPROF (Mace et al. 2007), heights of cloud-layer top and bottom from 2B-GEOPROF-lidar, and profiles of atmospheric temperature and moisture from ECMWF-AUX during summer [June–August (JJA)] of 2006–09. 2B-GEOPROF-lidar combines CloudSat CPR and CALIPSO lidar cloud masks. ECMWF-AUX contains temperature and moisture profiles from the ECMWF operational analysis interpolated in time and space to the CloudSat track. All the data are collocated at the spatial grid of the CPR with resolutions of 2.5 km along track by 1.4 km across track and 240 m in the vertical.

Deep convective cores are defined using vertical distribution of radar reflectivity and cloud mask from 2B-GEOPROF, combined with information on cloud layer from 2B-GEOPROF-lidar. Cloud mask value ≥30 is used to identify clouds and to locate cloud tops. Based on previous studies using CloudSat (e.g., Luo et al. 2008), a deep convective core (DCC) is defined as a CloudSat profile that satisfies the following two criteria: 1) it contains continuous radar echo from cloud top to near the surface (i.e., the distance between the cloud-layer bottom and the ground or sea surface is <3 km) and 2) the top of the cloud layer (which corresponds to the top of the echo of about −30 dBZ) and the maximum height of the 0 dBZ echo and the 10 dBZ echo detected by the CloudSat CPR are located above 12, 11, and 9 km, respectively. The basic idea behind this definition of DCC is that, deep convection normally extends from the planetary boundary layer to the upper troposphere with strong upward motion inside to sustain hydrometeors (both small cloud particles and larger precipitation particles) at high altitudes. Deep convective systems (DCSs) are then defined as a contiguous region (i.e., vertical cross section) consisting of cloudy radar profiles that contain at least one DCC with the noncore cloudy profiles either having cloud tops above 10 km or maximum reflectivities greater than 10 dBZ. Using this definition, cirrus and anvil clouds as well as shallow precipitating clouds that are connected to the deep convective core are considered as inherent parts of a DCS. This adds to the previous depiction of DCS by TRMM PR since both precipitating and nonprecipitating segments of the convective systems are now included. Figure 2 shows two typical examples of CloudSat–CALIPSO overpasses across the TP–SAMR that include DCSs. The blue lines and green dots near the top of each panel mark the selection of deep convective systems and the embedded deep convective cores, respectively. Some differences in the vertical and horizontal extent of the DCSs can be easily identified among the three subregions. For example, DCSs have a larger size over the SAMR than the TP. Statistical analysis will be conducted to further corroborate this view (section 3).

Fig. 2.
Fig. 2.

Vertical cross sections of radar reflectivity (color shading) detected by the CloudSat CPR over the TP–SAMR during (a) the nighttime overpass on 28 Aug 2006 and (b) the daytime overpass on 23 Jun 2009. Black and red dots indicate locations of cloud-layer top and cloud-layer bottom from GEOPROF_LIDAR. Blue lines and green dots near the top of each panel represent horizontal locations of deep convective systems and deep convective cores, respectively. Gray shadings represent elevation of the land surface.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

To examine the internal vertical structure of deep convection, we analyze statistics of the following parameters: 1) cloud-top heights (CTHs) seen by CALIPSO lidar (CTH_lidar) and CloudSat CPR (CTH_CPR), 2) echo-top heights (ETHs; the maximum heights reached by the 0- and 10-dBZ CPR echoes, referred to as H_0 dBZ and H_10 dBZ, respectively), and 3) the distance between CTH and ETH, which indicates the fuzziness of cloud top (i.e., a larger distance means a more tenuous cloud top, and a smaller distance means a more packed cloud top).

Amounts of moisture and convective available potential energy (CAPE) in the atmosphere are critical environmental factors determining the formation and growth of convection (Emanuel 1994); hence, they are expected to have significant impacts on not only the occurrence frequency and intensity of deep convection but also the horizontal size of convective systems. The value of CAPE is affected by surface elevation, which varies substantially among the analysis regions. Therefore, we also use the height of the level of neutral buoyancy (LNB), which is more directly related to CTH, for a more fair comparison. The total precipitable water (TPW), CAPE, and height of LNB in the seven analysis regions are calculated using the temperature and moisture profiles from ECMWF-AUX. Moreover, to validate the environmental conditions derived from ECMWF-AUX over the TP region, temperature and moisture profiles obtained by the L-band digital radiosonde systems (Li et al. 2009) at nine meteorological stations over the TP (locations shown in Fig. 1) are also utilized. These measurements have a resolution of ∼300 m in the vertical and were taken at 0000 and 1200 UTC (i.e., 0600–0700 LT and 1800–1900 LT) during JJA of 2008/09. For a more accurate comparison, the atmospheric conditions are determined from the ECMWF-AUX data within 1° × 1° boxes centered at each of the meteorological stations during JJA of 2008/09 and compared to those estimated with the radiosonde observations.

A total of seven regions are defined in this study (Fig. 1), including the TP, PSS, SAMR, which are the primary focus of this paper; the EA and NWP, two other subsystems of the more generally defined Asian summer monsoon; and the WNA and ENA on the other side of the world. The TP, PSS, and SAMR are located within the TP–SAMR and are defined the same as in Fu et al. (2006): as the area with elevation >3 km within 25°–40°N, 70°–105°E; the area with elevation <3 km within 25°–35°N, 70°–105°E; and the area within 10°–25°N, 70°–105°E covering both land (India and Myanmar) and ocean (Arabian Sea and Bay of Bengal), respectively. Three of the other four regions (EA, WNA, and ENA) are located in the subtropics (25°–40°N) and the last one (NWP) in the tropics (10°–25°N). EA, WNA, and ENA are defined as the land areas within 25°–40°N, 108°–122°E; within 25°–40°N, 100°–125°W with elevation >1 km; and within 25°–40°N, 70°–100°W, respectively. The NWP is defined as the area within 10°–25°N, 105°–140°E mainly consisting of ocean. One can see that the TP, EA, WNA, and ENA are subtropical (25°–40°N) continental regions with various levels of elevation, while the NWP and SAMR belong to the tropical section (10°–25°N) of the Asian summer monsoon (Wang et al. 2005; Ding 2007). The PSS is a subtropical continental region that is sandwiched between the TP and SAMR with elevation varying from the sea level to <3 km.

3. Properties of deep convection at the TP–SAMR

In this section, we investigate the following aspects of deep convection for the Tibetan Plateau–southern Asian monsoon region (i.e., TP, PSS, and SAMR): 1) occurrence frequency; 2) horizontal span of the deep convective systems; 3) internal vertical structure; and 4) local environment, such as the LNB height and TPW. Although previous studies have used the TRMM PR data to characterize the monsoonal deep convection along similar lines (e.g., Houze et al. 2007; Romatschke et al. 2010), CloudSat and CALIPSO with a different view of convection (i.e., high sensitivity to cloud particles) will provide a complementary depiction of deep convection properties and thus will add a new dimension to these previous findings.

a. Occurrence frequency of deep convection

One definition of the occurrence frequency of deep convection is the number of DCC profiles divided by the number of all profiles (including clear and cloudy ones), which we call DCC_freq. We can also define the occurrence frequency of deep convection as the number of DCS profiles (i.e., DCC plus the attached anvils) divided by the number of all profiles. This we call DCS_freq. DCC_freq gives a description of how frequently convective cores or strong updrafts occur in a given region, whereas DCS_freq characterizes the coverage of the whole deep convective systems. Table 1 shows the results from both definitions. DCC_freq is the lowest over the TP: 1.19%, which is only about 60% of that over the PSS and in the SAMR (∼1.9%). DCS_freq shows the highest value in the SAMR (24.15%), the second over the PSS (16.12%), and remains the lowest over the TP (11.50%).

Table 1.

Number of profiles and occurrence frequencies (%, see text for details) of deep convection for the 7 regions.

Table 1.

b. Horizontal span of the deep convective systems

We estimate the size of the DCSs through the horizontal span of the CloudSat profiles (see Fig. 2 for examples; the blue lines near the top of the panels show the estimate of the DCS size). Note that this measure of DCS size is somewhat different from that by visible or IR imagers, which give a plan view of the cloud system. Nevertheless, this is probably not a major concern because we are mostly interested in the contrasts among the regions as opposed to the absolute values. Also, four years of statistics (2006–09) will help reduce the random noises that may result from different “cuts” through the DCSs.

We define three size categories for deep convective systems: small (<120 km), medium (between 120 and 320 km), and large (>320 km), based on the occurrence frequency of DCSs over the TP, PSS, and SAMR, respectively, as a function of their horizontal span. Figure 3 shows the contribution from the three different size categories to the total population of deep convective systems. The TP has the most occurrences (46.3%) of the small convective systems and least occurrences (15.4%) of the large systems. The opposite occurs to the SAMR. The PSS comes between them. Figure 4a shows the mean size of the DCSs: it is 396 km in the SAMR, 221 km over the PSS, and 182 km over the TP. Romatschke and Houze (2011) investigated the characteristics of precipitating convective systems for this region using TRMM PR data during June–September for 8 yr (1999–2006). They also found more occurrences of large systems, presumably mesoscale convective systems (MCSs), over the SAMR (especially over the Bay of Bengal) relative to the PSS and TP. The latter two regions generally have smaller systems, such as isolated convective towers. It should, however, be noted that TRMM PR only detects the precipitating part of DCSs, whereas CloudSat–CALIPSO also sees the nonprecipitating part, such as cirrus anvils. But they arrive at qualitatively similar conclusions on size differences of the DCSs over the TP–SAMR.

Fig. 3.
Fig. 3.

Normalized occurrence frequencies (%) for deep convective systems with various horizontal spans (L) in the 7 regions. For each region, bars represent (left to right) small (L < 120 km), medium (120 < L < 320 km), and large (L > 320 km) convective systems.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

Fig. 4.
Fig. 4.

Horizontal spans of convective systems averaged (a) over the daytime and the nighttime observations and (b) over the daytime observations (bars without shading) and the nighttime observations (bars with shadings) for the 7 analysis regions. Middle of each bar represents the mean value, and top (bottom) of each bar indicates the mean plus (minus) the standard deviation.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

c. Internal vertical structure of deep convective core

We examine the following parameters pertaining to the internal vertical structure of deep convection: CTH, ETH, and distance between CTH and ETH. Figures 57 provide the mean values and standard deviations, and Fig. 8 gives a schematic sketch showing the major features more clearly. All the heights are above mean sea level. Heights of the DCC tops are comparable between the PSS and SAMR with the means of 16.13 and 16.32 km (CTH_lidar) and 15.18 and 15.06 km (CTH_CPR), respectively. They are noticeably higher than those over the TP, where the mean values of CTH_lidar and CTH_CPR are 14.74 and 14.26 km, respectively. All of these CTHs are above the base of the so-called tropical tropopause layer (TTL), located at an average of 14 km. Deep convection penetrating into the TTL has a better chance of actively participating in the stratosphere–troposphere exchange of water substance because of the gentle rising motion within the TTL owing to radiative heating (Fueglistaler et al. 2009). For the comparison of ETHs (i.e., H_0 dBZ and H_10 dBZ), the PSS stands out as the highest (13.46 and 11.47 km), the SAMR is slightly lower (13.08 and 11.01 km), and the TP has the lowest values (12.62 and 10.84 km). The lowest CTHs and ETHs over the TP are consistent with the findings using the TRMM PR data by Romatschke et al. (2010), who found that the total depth of the monsoonal deep convective cores over the TP is smaller than those observed over lower terrains on the PSS and the SAMR. This agreement suggests that the strength of deep convection is manifested in the vertical extent of both cloud-size particles (CloudSat–CALIPSO) and precipitation-size particles (TRMM PR).

Fig. 5.
Fig. 5.

Heights of the tops of deep convective clouds detected by CTH_lidar and CTH_CPR for the 7 regions. Bars defined in Fig. 4.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

Fig. 6.
Fig. 6.

Heights of tops of 0- and 10-dBZ echo in deep convective cores detected by CloudSat CPR (H_0 dBZ and H_10 dBZ, respectively) for the 7 regions. Bars defined in Fig. 4.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

Fig. 7.
Fig. 7.

Distances between cloud top detected by CALIPSO lidar and top of 0- and 10-dBZ echo detected by CloudSat CPR (CTH_lidar − H_0dBZ, CTH_lidar − H_10dBZ, respectively) in deep convective cores. Bars defined in Fig. 4.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

Fig. 8.
Fig. 8.

Schematics of deep convection over the TP, PSS, and in SAMR.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

The distances between CTH and ETH, which indicates the fuzziness of convective top, are the smallest over the TP (i.e., the most packed cloud top) and the largest in the SAMR. The mean distance between CTH_lidar and H_10 dBZ is 3.90 km over the TP, which is 0.76 and 1.41 km smaller than those at the PSS and in the SAMR, respectively. As will be discussed later, the small distance between the CTH and ETH of deep convection over the TP is also observed at subtropical North America and East Asia. In contrast, the tops of deep convective cores in the NWP are nearly as fuzzy as those in the SAMR. This is a typical contrast between continental and oceanic deep convection, as also shown by TRMM PR data (Liu et al. 2007). The fuzziness of deep convection tops is an indication of the overall size of the cloud particles that are transported up to the upper troposphere. The smaller CTH-to-ETH distance means a more packed convective top, which is dominated by larger particles. It has been suggested whether penetrating convection hydrates or dehydrates the TTL depends on the size of the ice crystals lofted (e.g., Jensen et al. 2007). Most of the deep convection over the TP–SAMR indeed penetrates into the TTL.

d. LNB height, CAPE, and TPW

The LNB height, CAPE, and TPW are similar between the PSS and SAMR, but they are significantly lower over the TP (Fig. 9): The mean LNB height over the TP is lower by 2.4 and 2.8 km than the PSS and SAMR, respectively. The CAPE and TPW are only approximately 14% and 20%, respectively, of those over the PSS and in the SAMR. Note that higher moisture level and taller LNB (and larger CAPE) are intrinsically related: given everything else fixed, larger water vapor amount (concentrated mostly near the lower troposphere) means higher near-surface equivalent potential temperature (θe), which leads to greater convective buoyancy during moist ascent and consequently higher LNB (and larger CAPE). The higher LNB and larger CAPE over the PSS and the SAMR, compared to the TP, are consistent with the higher CTH and ETH there (Figs. 5 and 6) because the higher LNB and larger CAPE allow greater depth for deep convection to grow. The moister conditions for the former two regions are conducive to higher occurrence frequencies of deep convection (Table 1) and are probably related to the larger convective system as well (Figs. 3 and 4a).

Fig. 9.
Fig. 9.

(a) Height of LNB, (b) TPW, and (c) CAPE for the 7 regions. Bars defined in Fig. 4.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

The mean values of the TPW, LNB height, and CAPE at the 9 meteorological stations over the TP (locations shown in Fig. 1) during JJA of 2008–2009 are compared between the L-band radiosonde data and ECMWF-AUX. The results suggest more TPW (14.78 versus 11.72 kg m−2), higher LNB (11.35 versus 10.35 km), and larger CAPE (273 versus 75 J kg−1) based on the L-band radiosonde data than ECMWF-AUX. Multiple factors could contribute to these differences, including uncertainties associated with both data sources and the time difference of the two datasets (0130 and 1330 LT of ECMWF versus 0600–0700 and 1800–1900 LT of the radiosonde observations). It is important to note that the differences between the two datasets are small compared to the differences between the TP and the other regions derived from ECMWF-AUX (Fig. 9). Therefore, the conclusion of much drier and less buoyant atmosphere over the TP than the other analysis regions (the PSS and SAMR in particular) remains unchanged.

e. Brief summary

Put together, CloudSat/CALIPSO observations suggest that deep convection occurs more frequently over the PSS and in the SAMR than over the TP. CTH and ETH are also higher in the former two regions than the latter. The less frequent and shallower DCC over the TP suggests generally weaker convective activity there. The weaker convection is closely related to the significantly lower LNB, smaller CAPE, and much drier atmosphere over the plateau. The convective system over the TP is also smaller in horizontal size than the other two regions. These results are broadly in agreement with previous studies using TRMM PR (e.g., Houze et al. 2007; Romatschke et al. 2010).

Recently, a new paradigm has been proposed by Boos and Kuang (2010) that has largely changed our traditional understanding of the South Asia summer monsoon: the authors found, through observation analysis and model simulations, that the Himalaya Mountains (which prevent the high moist entropy air from the south from being diluted by low moist entropy air from the north), instead of the heat source of the TP, that determine the strength of the South Asia summer monsoon and related precipitation. They showed that the surface of the TP contains much lower θe than the PSS and SAMR due to the much lower water vapor amounts (so the TP is not so much of a heat source if we use θe as a measure for heat). Lower surface θe cannot support strong convection over the plateau. Hence, Boos and Kuang (2010)’s study is consistent with our findings using CloudSat–CALIPSO and collocated ECMWF analyses.

Caution, however, needs to be exercised to interpret our results in the proper context. One caveat of using CloudSat–CALIPSO to study convection is a lack of full sampling of the diurnal cycle due to the sunsynchronous orbit followed by CloudSat and CALIPSO (approximately 1:30 a.m. and 1:30 p.m. equator local crossing time). Convection over the TP tends to have a strong diurnal cycle with the peak in late afternoon (e.g., Fujinami and Yasunari 2001; Fu et al. 2008). Nevertheless, the overall agreement with previous studies using TRMM PR data (which has the full diurnal sampling) suggests that our conclusions are still qualitatively valid. In addition, our current study adds to the TRMM analysis by showing the structures and patterns of clouds that are associated with the deep convection.

4. Comparing deep convection among seven analysis regions

In order to cast the comparison of deep convection properties over the Tibetan Plateau–southern Asian monsoon region into a broader context, we utilize four other regions—EA, NWP, WNA, and ENA (locations shown in Fig. 1)—and compare these with the TP–SAMR subregions following what has been done in section 3; that is, we investigate the following parameters related to deep convection: 1) the occurrence frequency of DCC and DCS (Table 1), 2) the horizontal span of deep convective system (Figs. 3 and 4a), 3) the physical parameters reflecting the internal vertical structure of deep convection (Figs. 57), and 4) the TPW, LNB height, and CAPE (Fig. 9). As will be described in more detail below, general similarities are found among the four subtropical continental regions (TP, EA, WNA, and ENA) and within the two tropical regions (SAMR and NWP). The PSS possesses some unique features of its own.

According to Table 1, the tropical regions (NWP and SAMR) and the PSS clearly favor the presence of deep convection relative to the subtropical regions, with the DCC occurrence frequencies at the NWP, SAMR, and PSS (∼1.9%) being nearly double those at the WNA, ENA, and EA (∼1.0%). The TP has slightly higher DCC occurrence frequency (1.2%) than the other three subtropical regions. Also obvious is that convective systems grow to a larger size in the tropical section of the Asian monsoon region (Figs. 3 and 4a). Nearly half of the convective systems have horizontal spans >320 km at the SAMR and NWP. The fractions reduce to about one-fourth at the PSS and EA and further decrease to less than one-fifth over the TP, and in WNA and ENA. The small-size convective systems (with a horizontal span <120 km) account for approximately one-fourth of the total population of convective systems over the SAMR and in the NWP, while their percentages increase to 36% at the PSS and to more than 40% over the TP, and in WNA, ENA, and EA. The mean values of the horizontal span of deep convective systems (Fig. 4a) are 396 and 404 km in the SAMR and NWP, respectively—substantially larger than the PSS (221 km), EA (227 km), ENA (193 km), WNA (185 km), and TP (182 km). TRMM PR observations also generally suggest smaller-size convective systems over subtropical North America (C. Liu 2010, personal communication) and the TP (Romatschke and Houze 2011).

The distributions of CTH and ETH are nearly Gaussian with only a single mode (Fig. 10). The distributions of CTH (Figs. 10a and 10b) differ significantly among the regions, with the modes distributed over a wide range (e.g., from 13.5 km for the WNA to 17.5 km for the PSS in Fig. 10a). The PSS, SAMR, and NWP have the highest CTHs, while the WNA has the lowest one. In an ascending order, the CTH distributions of ENA, TP, and EA shift toward higher altitudes. In contrast, the ETH distributions (Figs. 10c and 10d) are more alike among different regions. Compared to other regions, the ETH distributions at the PSS slightly shift toward higher altitudes with less pronounced modes and flatter distributions than the other regions, suggesting more (less) occurrences of the highest (lowest) echo tops at PSS. Student’s t test confirmed that the means of H_0 dBZ and H_10 dBZ at the PSS differ from those in the other regions at a 0.01 significant level. Given the general similar ETH distributions among the regions, the higher CTHs in the NWP and SAMR lead to larger distances between CTH and ETH (i.e., more fuzzy cloud tops) compared to the subtropical regions (Fig. 7). As mentioned earlier, this reflects the general difference between continental and oceanic convection since the subtropical regions selected in this study are all over land, while the two tropical regions are mostly over ocean.

Fig. 10.
Fig. 10.

Histograms of (a) CTH (CALIPSO lidar), (b) CTH (CloudSat CPR), maximum height of (c) 0 dBZ and (d) 10 dBZ for deep convection. The line color represents each of the 7 regions.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

The means and standard deviations of the TPW, LNB height, and CAPE when DCC was observed by CloudSat are shown in Fig. 9. The ranks of the means of the TPW among the regions are the same as those of LNB height, suggesting a close connection between the TPW and LNB height. Unsurprisingly, moisture is notably more abundant over the two tropical regions (SAMR and NWP) and PSS, where the LNB is also higher. These regions all favor deeper convection and larger-size convective systems. The smaller TWP and CAPE and the lower LNB at the TP and WNA are directly related to the higher elevation in the two regions.

In summary, the differences in deep convection characteristics between the TP and SAMR as discussed in the previous section are shown to be more of a general contrast between the tropical monsoonal convection (with much of the region covered by ocean) and the subtropical continental convection during boreal summer. The PSS possesses some of its own uniqueness because of the distinctive terrain (slope), moist low-level monsoon flow, and synoptic dynamics (e.g., Sawyer 1947; Houze et al. 2007; Medina et al. 2010; Romatschke et al. 2010; Romatschke and Houze 2011).

5. Comparison between the daytime and nighttime overpasses

One limitation of using the A-Train data to study convection is the inability to sample the full diurnal cycle because of the sun synchronous orbit (approximately 1:30 a.m. and 1:30 p.m. equator local crossing time) (Liu et al. 2008). Nevertheless, contrasting observations from the daytime (1:30 p.m.) and nighttime (1:30 a.m.) overpasses may provide a means to at least partially reveal the diurnal differences.

We first examine the partition of DCC between the daytime and nighttime overpasses. Contributions from the nighttime observations to the total occurrences of DCC vary among the regions (Table 2). There is more deep convection during the daytime (i.e., early afternoon) than nighttime (i.e., soon after midnight) over subtropical North America (WNA and ENA) and the TP, with daytime contributions of 90%, 69%, and 68%, respectively. In contrast, more convection occurs during the nighttime (64%) than daytime over the PSS. The daytime and nighttime contributions are about equal over the tropical regions (SAMR and NWP) and in EA. These results are generally consistent with the findings from previous studies (e.g., Romatschke et al. 2010; Dai et al. 1999). The substantially more occurrences of deep convection in the daytime in the subtropical land regions (WNA, ENA and TP) result from the diurnal variation of surface solar heating, which has significant impacts on the diurnal variation of low-level atmospheric stability. The mechanisms that lead to the diurnal variation over the PSS are daytime upslope wind and nighttime downslope wind due to daytime heating and nighttime cooling over elevated terrain (Romatschke and Houze 2011). The nighttime downslope wind converges with the moist monsoon flow in lower elevations, which leads to the triggering of convection over the foothills during the nighttime (Barros and Lang 2003; Bhatt and Nakamura 2005, 2006; Romatschke et al. 2010). The nearly equal occurrences of deep convection during the daytime and nighttime over the EA may result from substantial subregional variations of the diurnal cycle (Yu et al. 2007), so when averaged over the entire region, no clear preference over daytime or nighttime is obtained. Finally, it is well established that over the tropical oceans, the diurnal cycle of deep convection is rather weak (Hendon and Woodberry 1993), which may contribute to the similar occurrences at daytime and nighttime over the SAMR and NWP where oceans cover a large portion of the regions. Again, we stress that CloudSat does not sample the entire diurnal cycle, so this section is not meant to be a comprehensive survey of the diurnal variation of deep convection. Nevertheless, the contrasts in convection occurrences between 1:30 p.m. and 1:30 a.m. are consistent with our general understanding of the diurnal cycle of tropical and subtropical convection.

Table 2.

Number of deep convective core profiles that are observed by the CloudSat nighttime overpasses and its contribution (%) to the total number of deep convective core profiles that are observed by both the daytime and nighttime overpasses.

Table 2.

We also investigate the size of the convective systems at the two times (Fig. 4b). Four regions exhibit large differences—TP, EA, ENA and PSS—and all of them see larger systems during the nighttime than the daytime. The mean size of the convective systems over the TP is more than twice as large during the night (304 km) than the day (140 km); the EA, ENA and PSS have less of a contrast but still differ by ∼50% night versus day. Other regions have very small diurnal contrast in convection size. The reasons for these differences are not immediately clear to us, but we speculate that this may be attributable to the different formation mechanisms for continental convection during different times of the day. During the summer season, a significant fraction of the daytime convection is driven by local solar heating of the land surface. Such convection is usually isolated and small. But solar heating is lacking during night, so other mechanisms, most likely mechanisms of larger scale (e.g., mesoscale or synoptic disturbances), will play a more important role. This would explain the diurnal difference in convective system sizes over these regions. However, since we do not see this day–night contrast over the WNA, other factors may also be important. Further study is needed to unravel the nature of the underlying mechanisms. We once again caution the generalization of the results discussed here because no full coverage of the diurnal cycle is examined.

Finally, we examine the difference in internal vertical structures of deep convection between the daytime and nighttime overpasses (Fig. 11). Since the vertical resolution of CloudSat is 480 m (oversampled to 240 m), we focus on the differences that are at least larger than this value. Figure 11 shows that the most striking difference occurs over the TP: CTH_lidar is higher during the nighttime than the daytime by 0.49 km; however, the H_0 dBZ and H_10 dBZ are higher during the daytime than nighttime (by 0.70 and 0.58 km, respectively), resulting in a substantially smaller distance between CTH_lidar and the ETHs during the daytime (the means differ by 1.19 and 1.07 km, respectively). In other words, the daytime deep convective clouds have more condensed or packed tops relative to the nighttime ones. In fact, the distances between CTH_lidar and the ETHs are almost always smaller during the daytime for all the regions selected for this study (with a few exceptions, such as the NWP, where ocean coverage dominates). This is probably related to the strong surface heating during the day, which leads to strong updrafts lifting large particles to near the cloud top.

Fig. 11.
Fig. 11.

Differences (km) in the means of the physical parameters of deep convective cores between the daytime and nighttime observations (i.e., daytime minus nighttime) for the 7 regions: (a) CTH (CALIPSO lidar), (b) CTH (CloudSat CPR), (c) maximum height of 0 dBZ, (d) maximum height of 10 dBZ, and distance between cloud-top (CALIPSO lidar) and top of (e) 0 dBZ and (f) 10 dBZ.

Citation: Journal of Climate 24, 8; 10.1175/2010JCLI4032.1

6. Summary and discussions

Deep convection over the Tibetan Plateau–southern Asian monsoon region (TP–SAMR) is analyzed in this study using CloudSat and CALIPSO data for the boreal summer season (June – August) from 2006 to 2009. A number of previous studies have examined deep convection properties over this region during the monsoon season (e.g., Houze et al. 2007; Romatschke et al. 2010) and studied their impact on the cross-tropopause transport of trace gases (e.g., Fu et al. 2006; Park et al. 2007) using different spaceborne instruments (MLS, TRMM, etc.). However, some disagreement still lingers. The newly launched CloudSat and CALIPSO, which fly within 10–15 s of each other, add a new dimension to these existing measurements by providing not only the precise location of the tops of the deep convective clouds but also a glimpse of their internal vertical structure. It is thus of interest to use these new and unique measurements to reexamine the questions concerning deep convection over the TP–SAMR.

Three subregions in the TP–SAMR are defined (Fig. 1): the Tibetan Plateau (TP), the southern slope of the plateau (PSS), and the southern Asian monsoon region (SAMR), following an earlier study by Fu et al. (2006). To cast them in a broader context, we also bring in four additional regions that have some similarity to the TP–SAMR: East Asia (EA), the tropical northwestern Pacific (NWP), and western and eastern North America (WNA and ENA, respectively). Based on the characteristics of the CloudSat radar reflectivity, we define deep convective cores (DCCs) as profiles with cloud top and tops of large radar echo (reflectivity of 0 and 10 dBZ, respectively) extending to high altitudes. We further define a deep convective system (DCS) as a cross section of contiguous cloudy profiles that contain at least one DCC. This includes the associated nonprecipitating cirrus anvils that are not observed by TRMM PR. The analysis in this study is largely based on the statistics collected from these selected DCCs and DCSs, including occurrence frequency, horizontal span, and internal vertical structures. In addition, we calculated the environmental parameters associated with the deep convection, such as the height of the level of neutral buoyancy (LNB) and total precipitable water (TPW). The principal findings of the paper are as follows:

  1. Compared to the PSS and SAMR, deep convection occurs less frequently over the TP and has lower cloud-top height (CTH) and radar echo-top height (ETH) but a smaller distance between the CTH and ETH (i.e., more packed convective tops). The shallower vertical extents of both cloud-size particles (as revealed by CTH) and the precipitation-size particles (as revealed by ETH) suggest that deep convection is generally weaker over the plateau. The more packed convective top is a common feature for subtropical, continental convection during the summertime. DCSs over the TP have smaller horizontal spans compared to the other two subregions. These deep convection characteristics are closely related to the unique local environment of the TP, that is, significantly lower LNB height, smaller CAPE, and much drier atmosphere. This is in agreement with a new paradigm that has been proposed by Boos and Kuang (2010) to explain the control of the South Asian summer monsoon.
  2. Deep convection shows a number of similarities between the PSS and SAMR in terms of occurrence frequency and intensity (as indicated by CTH and ETH), although the PSS seems to have slightly higher ETH. The main difference between the two subregions is the horizontal span of DCSs: the mean size of DCSs at the SAMR is almost twice that at the PSS, suggesting that the deep convection tends to be more organized into, for example, large MCSs at the SAMR. Another difference between the PSS and SAMR is that the convective tops are more packed in the PSS. This reflects the difference between continental and oceanic convection, as a large portion of the SAMR is covered by oceans (Arabian Sea and the Bay of Bengal).
  3. In a broader context where all seven regions are brought together, deep convection in the two tropical regions (NWP and SAMR) is similar in many regards, including the occurrence frequency, CTH–ETH, system size, and local thermodynamic environment. A similar conclusion can be drawn for the four subtropical regions (except that TPW and LNB height usually scale with the elevation over the land areas). Tropical and subtropical regions, however, present a significant contrast in convective properties: convection in the latter region generally occurs less frequently, has lower CTH but comparable or slightly higher ETH (e.g., H_10 dBZ), and is embedded in smaller systems. Hence, the difference in deep convection between the TP and SAMR is more of a general contrast between subtropical continental regions and tropical oceanic regions during the boreal summer.
  4. Deep convection over the PSS shows some uniqueness because of its distinctive terrain (slopes) and moist low-level monsoon flow. Deep convection occurs almost as frequently as in the tropical regions (although embedded in smaller DCSs) with comparable CTH and slightly higher ETH. The presence of extremely intense convection over the PSS has been attributable to a number of factors by previous studies (e.g., Sawyer 1947; Houze et al. 2007), including: 1) strong instability near the Himalayan foothills, caused by warm and moist low-level monsoon flow from the Arabian Sea capped by dry westerly or northwesterly midlevel flow coming down from higher terrain; and 2) orographic lifting of the low-level flow when the low-level monsoon flow impinges on the foothills of the mountain ranges.
  5. Although CloudSat–CALIPSO observations lack  a full sample of the diurnal cycle (local equator crossing time is approximately 1:30 a.m.–1:30 p.m.), we nevertheless made an attempt to contrast the daytime (1:30 p.m.) and nighttime (1:30 a.m.) overpasses to at least partially reveal the day/night differences in deep convection. Results are largely consistent with our general understanding of the diurnal variation of tropical and subtropical convection as well as findings from recent studies based on TRMM observations (e.g., Romatschke and Houze 2011). Moreover, this study suggests that the daytime deep convective clouds have more packed tops relative to nighttime ones in nearly all of the analysis regions.

In summary, analysis of the CloudSat/CALIPSO observations at the Tibetan Plateau–southern Asian monsoon region during the boreal summer generally corroborates findings from previous studies that used the TRMM PR. However, it is important to note that CloudSat–CALIPSO, with its unique sensitivity to both cloud- and precipitation-size particles, presents a different view of the convective systems and the overall consistency with previous work adds to our knowledge of the deep convection in this region.

Acknowledgments

This study was jointly supported by the National Natural Science Foundation of China (Projects 40875064 and 40921003), the Basic Research Fund of the Chinese Academy of Meteorological Sciences (2007R001), and the Special Fund for Research in Meteorology (GYHY200806020). Dr. Z. Luo acknowledges the support from the NASA CloudSat/CALIPSO Science Team under Grant NNX10AM31G and NASA MAP Grant NNX09AJ46G. NASA’s CloudSat project products are provided online (www.cloudsat.cira.colostate.edu).

REFERENCES

  • Barros, A. P., , and T. J. Lang, 2003: Monitoring the monsoon in the Himalayas: Observations in central Nepal, June 2001. Mon. Wea. Rev., 131, 14081427.

    • Search Google Scholar
    • Export Citation
  • Bhatt, B. C., , and K. Nakamura, 2005: Characteristics of monsoon rainfall around the Himalayas revealed by TRMM precipitation radar. Mon. Wea. Rev., 133, 149165.

    • Search Google Scholar
    • Export Citation
  • Bhatt, B. C., , and K. Nakamura, 2006: A climatological-dynamical analysis associated with precipitation around the southern part of the Himalayas. J. Geophys. Res., 111, D02115, doi:10.1029/2005JD006197.

    • Search Google Scholar
    • Export Citation
  • Boos, W. R., , and Z. Kuang, 2010: Dominant control of the South Asian monsoon by orographic insulation versus plateau heating. Nature, 463, 218222.

    • Search Google Scholar
    • Export Citation
  • Chung, E.-S., , B.-J. Sohn, , and J. Schmetz, 2008: CloudSat shedding new light on high-reaching tropical deep convection observed with Meteosat. Geophys. Res. Lett., 35, L02814, doi:10.1029/2007GL032516.

    • Search Google Scholar
    • Export Citation
  • CIRA, 2008: CloudSat standard data products handbook. Cooperative Institute for Research in the Atmosphere, Colorado State University, 18 pp. [Available online at http://www.cloudsat.cira.colostate.edu/dataHome.php.]

    • Search Google Scholar
    • Export Citation
  • Dai, A., 2001: Global precipitation and thunderstorm frequencies. Part II: Diurnal variations. J. Climate, 14, 11121128.

  • Dai, A., , F. Giorgi, , and K. E. Trenberth, 1999: Observed and model-simulated diurnal cycles of precipitation over the contiguous United States. J. Geophys. Res., 104, 63776402.

    • Search Google Scholar
    • Export Citation
  • Ding, Y. H., 2007: The variability of the Asian summer monsoon. J. Meteor. Soc. Japan, 85, 2154.

  • Emanuel, K., 1994: Atmospheric Convection. Oxford University Press, 580 pp.

  • Forster, P. M. de F., , and K. P. Shine, 2002: Assessing the climate impact of trends in stratospheric water vapor. Geophys. Res. Lett., 29, 1086, doi:10.1029/2001GL013909.

    • Search Google Scholar
    • Export Citation
  • Fu, R., and Coauthors, 2006: Short circuit of water vapor and polluted air to the global stratosphere by convective transport over the Tibetan Plateau. Proc. Natl. Acad. Sci. USA, 103, 56645669.

    • Search Google Scholar
    • Export Citation
  • Fu, Y., , Q. Liu, , Y. Zi, , S. Feng, , Y. Li, , and G. Liu, 2008: Summer precipitation and latent heating over the Tibetan Plateau based on TRMM measurements (in Chinese). Plateau Mt. Meteor. Res., 28, 818.

    • Search Google Scholar
    • Export Citation
  • Fueglistaler, S., , A. E. Dessler, , T. J. Dunkerton, , I. Folkins, , Q. Fu, , and P. W. Mote, 2009: Tropical tropopause layer. Rev. Geophys., 47, RG1004, doi:10.1029/2008RG000267.

    • Search Google Scholar
    • Export Citation
  • Fujinami, H., , and T. Yasunari, 2001: The seasonal and intraseasonal variability of diurnal cloud activity over the Tibetan Plateau. J. Meteor. Soc. Japan, 79, 12071227.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., , and R. W. Jacobson Jr., 1977: Diurnal variation of deep cumulus convection. Mon. Wea. Rev., 105, 11711188.

  • Hendon, H., , and K. Woodberry, 1993: The diurnal cycle of tropical convection. J. Geophys. Res., 98, 162 62316 2637, doi:10.1029/93JD00525.

    • Search Google Scholar
    • Export Citation
  • Houze, R. A., Jr., , D. C. Wilton, , and B. F. Smull, 2007: Monsoon convection in the Himalayan region as seen by the TRMM precipitation radar. Quart. J. Roy. Meteor. Soc., 133, 13891411, doi:10.1002/gq.106.

    • Search Google Scholar
    • Export Citation
  • Im, E., , S. L. Durden, , and C. Wu, 2006: Cloud profiling radar for the CloudSat mission. IEEE Trans. Aerosp. Electron. Syst., 20, 1518.

    • Search Google Scholar
    • Export Citation
  • Jensen, E. J., , A. S. Ackerman, , and J. A. Smith, 2007: Can overshooting convection dehydrate the tropical tropopause layer? J. Geophys. Res., 112, D11209, doi:10.1029/2006JD007943.

    • Search Google Scholar
    • Export Citation
  • Kirk-Davidoff, D. B., , E. J. Hintsa, , J. G. Anderson, , and D. W. Keith, 1999: The effect of climate change on ozone depletion through changes in stratospheric water vapour. Nature, 402, 399401.

    • Search Google Scholar
    • Export Citation
  • Li, W., , F. Li, , Z. Zhao, , F. Liu, , B. Li, , and H. Li, 2009: Technical evaluation report of the construction of L-band meteorological observing systems (in Chinese). China Meteorological Press, 95 pp.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. Zipser, , and S. W. Nesbitt, 2007: Global distribution of tropical deep convection: Different perspectives using infrared and radar as the primary data source. J. Climate, 20, 489503.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. Zipser, , G. Mace, , and S. Benson, 2008: Implications of the differences between daytime and nighttime CloudSat observations over the tropics. J. Geophys. Res., 113, D00A04, doi:10.1029/2008JD009783.

    • Search Google Scholar
    • Export Citation
  • Luo, Z., , and W. B. Rossow, 2004: Characterizing tropical cirrus life cycle, evolution, and interaction with upper-tropospheric water vapor using Lagrangian trajectory analysis of satellite observations. J. Climate, 17, 45414563.

    • Search Google Scholar
    • Export Citation
  • Luo, Z., , G. Y. Liu, , and G. L. Stephens, 2008: CloudSat adding new insight into tropical penetrating convection. Geophys. Res. Lett., 35, L19819, doi:10.1029/2008GL035330.

    • Search Google Scholar
    • Export Citation
  • Mace, G. G., , R. Marchand, , Q. Zhang, , and G. Stephens, 2007: Global hydrometeor occurrence as observed by CloudSat: Initial observations from summer 2006. Geophys. Res. Lett., 34, L09808, doi:10.1029/2006GL029017.

    • Search Google Scholar
    • Export Citation
  • Medina, S., , R. A. Houze Jr., , A. Kumar, , and D. Niyogi, 2010: Summer monsoon convection in the Himalayan region: Terrain and land cover effects. Quart. J. Roy. Meteor. Soc., 136, 593616.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., , and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16, 14561475.

    • Search Google Scholar
    • Export Citation
  • Oltmans, S. J., , and D. J. Hofmann, 2002: Increase in lower-stratospheric water vapour at a mid-latitude Northern Hemisphere site from 1981 to 1994. Nature, 374, 146149, doi:10.1038/374146a0.

    • Search Google Scholar
    • Export Citation
  • Park, M., , W. J. Randel, , A. Gettelman, , S. T. Massie, , and J. H. Jiang, 2007: Transport above the Asian summer monsoon anticyclone inferred from Aura Microwave Limb Sounder tracers. J. Geophys. Res., 112, D16309, doi:10.1029/2006JD008294.

    • Search Google Scholar
    • Export Citation
  • Romatschke, U., , and R. A. Houze Jr., 2011: Characteristics of precipitating convective systems in the South Asian monsoon. J. Hydrometeor., 12, 157180.

    • Search Google Scholar
    • Export Citation
  • Romatschke, U., , S. Medina, , and R. A. Houze Jr., 2010: Regional, seasonal, and diurnal variations of extreme convection in the South Asian region. J. Climate, 23, 419439.

    • Search Google Scholar
    • Export Citation
  • Sawyer, J. S., 1947: The structure of the intertropical front over N.W. India during the S.W. monsoon. Quart. J. Roy. Meteor. Soc., 73, 346369.

    • Search Google Scholar
    • Export Citation
  • Smith, C. A., , R. Toumi, , and J. D. Haigh, 2000: Seasonal trends in stratospheric water vapour. Geophys. Res. Lett., 27, 16871690.

  • Stephens, G. L., and Coauthors, 2002: The CloudSat mission and the A-Train. Bull. Amer. Meteor. Soc., 83, 17711790.

  • Stephens, G. L., and Coauthors, 2008: CloudSat mission: Performance and early science after the first year of operation. J. Geophys. Res., 113, D00A18, doi:10.1029/2008JD009982.

    • Search Google Scholar
    • Export Citation
  • Wang, B., , and LinHo, 2002: Rainy season of the Asian–Pacific summer monsoon. J. Climate, 15, 386398.

  • Wang, B., , T. Li, , Y. H. Ding, , R. H. Zhang, , and H. J. Wang, 2005: Eastern Asian–western North Pacific monsoon: A distinctive component of the Asian–Australian monsoon system. The Global Monsoon System: Research and Forecast, WMO/TD-1266, TMRP Rep. 70, C.-P. Chang, Eds., WMO, 72–94.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , J. R. Pelon, , and M. P. McCormick, 2003: The CALIPSO mission: Spaceborne lidar for observation of aerosols and clouds. Lidar Remote Sensing for Industry and Environment Monitoring III, U. N. Singh et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 4893), 1–11.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , W. H. Hunt, , and M. J. McGill, 2007: Initial performance assessment of CALIOP. Geophys. Res. Lett., 34, L19803, doi:10.1029/2007GL030135.

    • Search Google Scholar
    • Export Citation
  • Yu, R., , T. Zhou, , A. Xiong, , Y. Zhu, , and J. Li, 2007: Diurnal variations of summer precipitation over contiguous China. Geophys. Res. Lett., 34, L01704, doi:10.1029/2006GL028129.

    • Search Google Scholar
    • Export Citation
  • Zhou, S., , and R. Zhang, 2005: Decadal variations of temperature and geopotential height over the Tibetan Plateau and their relations with Tibet ozone depletion. Geophys. Res. Lett., 32, L18705, doi:10.1029/2005GL023496.

    • Search Google Scholar
    • Export Citation
  • Zhou, X. J., , C. Luo, , W. Li, , and J. Shi, 1995: Trend in ozone over China and low center over the Tibetan Plateau (in Chinese). Chin. Sci. Bull., 4, 13961398.

    • Search Google Scholar
    • Export Citation
Save