• Ba, M. B., , and S. E. Nicholson, 1998: Analysis of convective activity and its relationship to the rainfall over the Rift Valley lakes of East Africa during 1983–90 using the Meteosat infrared channel. J. Appl. Meteor., 37 , 12501264.

    • Search Google Scholar
    • Export Citation
  • Balas, N., , S. E. Nicholson, , and D. Klotter, 2007: The relationship of rainfall variability in West Central Africa to sea-surface temperature fluctuations. Int. J. Climatol., 27 , 13351349.

    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., , and S. A. Listemaa, 2000: An improved scheme for convective/stratiform echo classification using radar reflectivity. J. Appl. Meteor., 39 , 21292150.

    • Search Google Scholar
    • Export Citation
  • Cecil, D. A., , S. J. Goodman, , D. J. Boccippio, , E. J. Zipser, , and S. W. Nesbitt, 2005: Three years of TRMM precipitation features. Part I: Radar, radiometric, and lightning characteristics. Mon. Wea. Rev., 133 , 543566.

    • Search Google Scholar
    • Export Citation
  • Christian, H. J., , R. J. Blakeslee, , and S. J. Goodman, 1992: Lightning Imaging Sensor (LIS) for the Earth Observing System. NASA Tech. Memo. 4350, MSFC, Huntsville, AL, 36 pp.

    • Search Google Scholar
    • Export Citation
  • Cunningham, P., , and D. Keyser, 2000: Analytical and numerical modeling of jet streaks: Barotropic dynamics. Quart. J. Roy. Meteor. Soc., 126 , 31873217.

    • 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
  • Duvel, J-P., 1989: Convection over tropical Africa and the Atlantic Ocean during northern summer. Part I: Interannual and diurnal variations. Mon. Wea. Rev., 117 , 27822799.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , C. Wang, , J. Wilson, , and J. Strom, 2004: Explicit simulations of aerosol physics in a cloud-resolving model: A sensitivity study based on an observed convective cloud. Atmos. Chem. Phys., 4 , 773791.

    • Search Google Scholar
    • Export Citation
  • Flohn, H., , and K. Fraedrich, 1966: Tagesperiodische Zirkulation und Niederschlagsverteilung am Victoria-See (Ostafrika) (The daily periodic circulation and distribution of rainfall over Lake Victoria). Meteor. Rundsch., 19 , 157165.

    • Search Google Scholar
    • Export Citation
  • Fraedrich, K., 1972: A simple climatological model of the dynamics and energetics of the nocturnal circulation at Lake Victoria. Quart. J. Roy. Meteor. Soc., 98 , 322335.

    • Search Google Scholar
    • Export Citation
  • Futyan, J. M., , and A. D. Del Genio, 2007: Deep convective system evolution over Africa and the tropical Atlantic. J. Climate, 20 , 50415060.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., , and T. Dejene, 2005: Regional and diurnal variability of the vertical structure of precipitation systems in Africa based on spaceborne radar data. J. Climate, 18 , 893916.

    • Search Google Scholar
    • Export Citation
  • Grist, J. P., , and S. E. Nicholson, 2001: A study of the dynamic factors influencing the interannual variability of rainfall in the West African Sahel. J. Climate, 14 , 13371359.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., , and C. D. Thorncroft, 1997: Distribution and statistics of African mesoscale convective weather systems based on the ISCCP Meteosat imagery. Mon. Wea. Rev., 125 , 28212837.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Keyser, D., , and M. A. Shapiro, 1986: A review of the structure and dynamics of upper-level frontal zones. Mon. Wea. Rev., 114 , 452499.

    • Search Google Scholar
    • Export Citation
  • Kinter III, J. L., , M. J. Fennessy, , V. Krishnamurthy, , and L. Marx, 2004: An evaluation of the apparent interdecadal shift in the tropical divergent circulation in the NCEP–NCAR reanalysis. J. Climate, 17 , 349361.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., , W. Barnes, , T. Kozu, , J. Shiue, , and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15 , 809817.

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

    • Search Google Scholar
    • Export Citation
  • Laing, A. G., , and J. M. Fritsch, 1993a: Mesoscale convective complexes in Africa. Mon. Wea. Rev., 121 , 22542265.

  • Laing, A. G., , and J. M. Fritsch, 1993b: Mesoscale convective complexes over the Indian monsoon region. J. Climate, 6 , 911919.

  • Laing, A. G., , and J. M. Fritsch, 1997: The global population of mesoscale convective complexes. Quart. J. Roy. Meteor. Soc., 123 , 389405.

    • Search Google Scholar
    • Export Citation
  • Lin, X., , D. A. Randall, , and L. D. Fowler, 2000: Diurnal variability of the hydrologic cycle and radiative fluxes: Comparisons between observations and a GCM. J. Climate, 13 , 41594179.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , and E. J. Zipser, 2005: Global distribution of convection penetrating the tropical tropopause. J. Geophys. Res., 110 , D23104. doi:10.1029/2005JD006063.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. J. Zipser, , and S. W. Nesbitt, 2007: Global distribution of tropical deep convection: Different perspective from TRMM infrared and radar data. J. Climate, 20 , 489503.

    • Search Google Scholar
    • Export Citation
  • McCollum, J. R., , A. Gruber, , and M. B. Ba, 2000: Discrepancy between gauges and satellite estimates of rainfall in equatorial Africa. J. Appl. Meteor., 39 , 666679.

    • Search Google Scholar
    • Export Citation
  • McGarry, M. M., , and R. J. Reed, 1978: Diurnal variations in convective activity and precipitation during phases II and III of GATE. Mon. Wea. Rev., 106 , 101113.

    • Search Google Scholar
    • Export Citation
  • Mekonnen, A., , C. D. Thorncroft, , and A. Aiyyer, 2006: Analysis of convection and its association with African easterly waves. J. Climate, 19 , 54055421.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., 2004: Interannual, monthly, and regional variability in the wet season diurnal cycle of precipitation in sub-Saharan Africa. J. Climate, 17 , 24412453.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and E. J. Zipser, 1996a: Mesoscale convective systems defined by their 85-GHz ice-scattering signature: Size and intensity comparison over tropical oceans and continents. Mon. Wea. Rev., 124 , 24172437.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and E. J. Zipser, 1996b: Defining mesoscale convective systems by their 85-GHz ice-scattering signatures. Bull. Amer. Meteor. Soc., 77 , 11791189.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and C. Thorncroft, 2006: Intense convective systems in West Africa and their relationship to the African easterly jet. Quart. J. Roy. Meteor. Soc., 132 , 163176.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , J. S. Famiglietti, , and E. J. Zipser, 1999: The contribution to tropical rainfall with respect to convective system type, size, and intensity estimated from 85-GHz ice-scattering signature. J. Appl. Meteor., 38 , 596606.

    • 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
  • Nesbitt, S. W., , E. J. Zipser, , and D. J. Cecil, 2000: A census of precipitation features in the Tropics using TRMM: Radar, ice scattering, and ice observations. J. Climate, 13 , 40874106.

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

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 1986: The spatial coherence of African rainfall anomalies: Interhemispheric teleconnections. J. Climate Appl. Meteor., 25 , 13651381.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2008: The intensity, location and structure of the tropical rainbelt over West Africa as factors in interannual variability. Int. J. Climatol., 28 , 17751785.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2009a: A low-level jet along the Benguela coast, an integral part of the Benguela Marine Ecosystem. Climatic Change, in press.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2009b: On the factors modulating the intensity of the tropical rainbelt over West Africa. Int. J. Climatol., 29 , 673689.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and D. Entekhabi, 1987: Rainfall variability in equatorial and southern Africa: Relationships with sea surface temperatures along the southwestern coast of Africa. J. Climate Appl. Meteor., 26 , 561578.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and J. P. Grist, 2003: On the seasonal evolution of atmospheric circulation over West Africa and equatorial Africa. J. Climate, 16 , 10131030.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and P. J. Webster, 2007: A physical basis for the interannual variability of rainfall in the Sahel. Quart. J. Roy. Meteor. Soc., 133 , 20652084.

    • Search Google Scholar
    • Export Citation
  • Oki, T., , and K. Musiake, 1994: Seasonal change of the diurnal cycle of precipitation over Japan and Malaysia. J. Appl. Meteor., 33 , 14451463.

    • Search Google Scholar
    • Export Citation
  • ORSTOM, 1978: Précipitations journalières de l’origine des stations à 1972. Etienne Julien, Paris, Vol. 1, 512 pp.

  • Petersen, W. A., , and S. A. Rutledge, 2001: Regional variability in tropical convection: Observations from TRMM. J. Climate, 14 , 35663586.

    • Search Google Scholar
    • Export Citation
  • Poccard, I., , S. Janicot, , and P. Camberlin, 2000: Comparison of rainfall structures between NCEP/NCAR reanalysis and observed data over tropical Africa. Climate Dyn., 16 , 897915.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , and J. D. Jaffe, 1981: Diurnal variation of summer convection over West Africa and the tropical eastern Atlantic during 1974 and 1978. Mon. Wea. Rev., 109 , 25272534.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., 2002a: Aerosols and ice particle size in tropical cumulonimbus. J. Climate, 15 , 10511063.

  • Sherwood, S. C., 2002b: A microphysical connection among biomass burning, cumulus clouds, and stratospheric moisture. Science, 295 , 12711275.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., , and R. Wahrlich, 1999: Observed evolution of tropical deep convective events and their environment. Mon. Wea. Rev., 127 , 17771795.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., , T. J. Phillips, , and J. S. Wettlaufer, 2006: Small ice crystals and the climatology of lightning. Geophys. Res. Lett., 33 , L05804. doi:10.1029/2005GL025242.

    • Search Google Scholar
    • Export Citation
  • Thorncroft, C. D., , and M. Blackburn, 1999: Maintenance of the African easterly jet. Quart. J. Roy. Meteor. Soc., 125 , 763786.

  • Thorncroft, C., , and K. Hodges, 2001: African Easterly Wave variability and its relationship to Atlantic tropical cyclone activity. J. Climate, 14 , 11661179.

    • Search Google Scholar
    • Export Citation
  • Toracinta, E. R., , and E. J. Zipser, 2001: Lightning and SSM/I-ice-scattering mesoscale convective systems in the global tropics. J. Appl. Meteor., 40 , 9831002.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and C. J. Guillemot, 1995: Evaluation of the global atmospheric moisture budget as seen from analyses. J. Climate, 8 , 22552272.

    • Search Google Scholar
    • Export Citation
  • Tripoli, G. J., 1986: A numerical investigation of an orogenic mesoscale convective system. Ph.D. dissertation, Colorado State University, 290 pp.

  • Tripoli, G. J., , and W. R. Cotton, 1989a: Numerical study of an observed orogenic mesoscale convective system. Part I: Simulated genesis and comparison with observations. Mon. Wea. Rev., 117 , 273304.

    • Search Google Scholar
    • Export Citation
  • Tripoli, G. J., , and W. R. Cotton, 1989b: Numerical study of an observed orogenic mesoscale convective system. Part II: Analysis of governing dynamics. Mon. Wea. Rev., 117 , 305328.

    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., , and D. R. Johnson, 1979: The coupling of upper and lower tropospheric jet streaks and implications for the development of severe convective storms. Mon. Wea. Rev., 107 , 682703.

    • Search Google Scholar
    • Export Citation
  • Wallace, J. M., 1975: Diurnal variations in precipitation and thunderstorm frequency over the conterminous United States. Mon. Wea. Rev., 103 , 406419.

    • Search Google Scholar
    • Export Citation
  • Yang, G. Y., , and J. Slingo, 2001: The diurnal cycle in the Tropics. Mon. Wea. Rev., 129 , 784801.

  • Yang, S., , and E. A. Smith, 2006: Mechanisms for diurnal variability of global tropical rainfall observed from TRMM. J. Climate, 19 , 51905226.

    • Search Google Scholar
    • Export Citation
  • Yin, X., , and S. E. Nicholson, 1998: The water balance of Lake Victoria. J. Hydrol. Sci., 43 , 789811.

  • Zipser, E. J., , D. J. Cecil, , C. Liu, , S. W. Nesbitt, , and D. P. Yorty, 2006: Where are the most intense thunderstorms on earth? Bull. Amer. Meteor. Soc., 87 , 10571071.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Location of study area and the annual rainfall distribution within four subsectors. For the three southernmost sectors, rainfall distribution is indicated for both the western portion and the eastern portion of the analysis region. (b) Terrain contours and geographic locations mentioned in the text.

  • View in gallery

    Five-year average of (a) relative number of MCSs per year, (b) total volumetric rainfall (km2 mm h−1) from MCSs, (c) volumetric rainfall per MCS, and (d) percentage of convective rainfall. These quantities are averaged over 1° × 1° grid points.

  • View in gallery

    Five-year average of (a) relative number of MCSs per month and per grid point and (b) seasonal rainfall (mm month−1) from MCSs for the four seasons: DJF, MAM, JJA, and SON. These quantities are averaged over 1° × 1° grid points.

  • View in gallery

    Hourly amount of rainfall at selected stations along the shore of Lake Victoria during April and September. The extreme values at Nabuyongo Island, in the center of Lake Victoria, are indicative of the enhance of rainfall by the lake (from Ba and Nicholson 1998).

  • View in gallery

    Monthly means of MCS activity: mean relative number of MCSs within the indicated latitudinal sector, volumetric rainfall per MCS (104 mm km2 h−1), and percentage of convective rainfall within the two northern equatorial zones and two southern equatorial zones. Data on volumetric rainfall and percentage of convective rainfall are omitted for months with relatively few MCSs.

  • View in gallery

    Maps of the diurnal cycle: (a) mean relative number of MCSs within each 2° × 2° lat–lon grid box, (b) the mean volumetric rainfall per MCS (104 mm km2 h−1), and (c) the percentage of convective rainfall at 3-h intervals, starting at 0000–0300 LT.

  • View in gallery

    The diurnal cycle of convection (a) seasonal averages for the relative number of MCSs in 3-h intervals for northern and southern sectors (see Fig. 1a for location). (b) The annual averages of the relative number of MCSs, the mean volumetric rainfall per MCS (104 mm km2 h−1), and the percentage of convective rainfall in 3-h intervals for the northern and southern sectors for the year as a whole.

  • View in gallery

    (a) Five-year mean relative number of lightning flashes (per 1° × 1° latitude–longitude) for the year as a whole, (b) mean percent MCSs with flashes, and (c) and number of flashes per MCS. These are indicators of the overall intensity of convection. All data are averaged for a 1° × 1° grid box, with a three-point smoothing applied. The maxima are highlighted by showing only areas with greater than 20 lightning flashes per year or 10 lightning flashes per season.

  • View in gallery

    Five-year mean relative number of lightning flashes during each 3-month season (total within each 1° × 1° latitude–longitude grid box; three-point smoothing has been applied).

  • View in gallery

    Diurnal cycle of lightning (relative number of flashes per 1° × 1° grid within each 3-h time span). Data are averaged for all months of the 5-yr study period.

  • View in gallery

    Seasonal cycle of flash count (relative flash count per month) within (top) the indicated latitudinal sector and (bottom) average number of flashes per MCS within the sector. Data are shown for the four latitudinal sectors shown in Fig. 1. A three-point smoothing has been applied.

  • View in gallery

    Average relative number of flashes per year in each 1° × 1° grid box, superimposed upon a faux 3D terrain map of Africa. The maximum is highlighted by showing only areas with a relative flash count greater than 10 per year.

  • View in gallery

    Mean winds at 850 and 925 mb during SON.

  • View in gallery

    Mean vertical motion (omega; mb s−1 × 10−2) at 850 mb at 1200 and 1800 UTC (LT is generally 2–3 h later).

  • View in gallery

    Mean precipitable water (kg m−2) over Africa and South America in February and November.

  • View in gallery

    The AEJ-S: (a) mean wind (m s−1) at 600 mb during October; (b) vertical cross section of mean zonal wind at 20°E as a function of latitude (October; m s−1).

  • View in gallery

    Mean temperature (°C) at 925 mb for April and October, based on NCEP–NCAR data.

  • View in gallery

    Mean divergence at 600 and 200 mb and mean vertical motion (omega; mb s−1 × 10−2) at 600 mb during October. The thin crossed lines represent the axes of the AEJ-S.

  • View in gallery

    Mean rainfall (mm) during MAM and SON (from Balas et al. 2007).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 150 150 38
PDF Downloads 142 142 41

Mesoscale Convective Systems over Western Equatorial Africa and Their Relationship to Large-Scale Circulation

View More View Less
  • 1 Department of Meteorology, The Florida State University, Tallahassee, Florida
© Get Permissions
Full access

Abstract

This study examines mesoscale convective systems (MCSs) over western equatorial Africa using data from the Tropical Rainfall Measuring Mission (TRMM) satellite. This region experiences some of the world’s most intense thunderstorms and highest lightning frequency, but has low rainfall relative to other equatorial regions. The analyses of MCS activity include the frequency of occurrence, diurnal and annual cycles, and associated volumetric and convective rainfall. Also evaluated is the lightning activity associated with the MCSs. Emphasis is placed on the diurnal cycle and on the continental-scale motion fields in this region. The diurnal cycle shows a maximum in MCS count around 1500–1800 LT, a morning minimum, and substantial activity during the night; there is little seasonal variation in the diurnal cycle, suggesting stationary influences such as orography. Our analysis shows four maxima in MCS activity, three of which are related to local geography (two orographic and one over Lake Victoria). The fourth coincides with a midtropospheric convergence maximum in the right entrance quadrant of the African easterly jet of the Southern Hemisphere (AEJ-S). This maximum is substantially stronger in the September–November rainy season, when the jet is well developed, than in the March–May rainy season, when the jet is absent. Lightning frequency and flashes per MCS are also greatest during September–November; maxima occur in the right entrance quadrant of the AEJ-S. The lightning maximum is somewhat south of the MCS maximum and coincides with the low-lying areas of central Africa. Overall, the results of this study suggest that large-scale topography plays a critical role in the spatial and diurnal patterns of convection, lightning, and rainfall in this region. More speculative is the role of the AEJ-S, but this preliminary analysis suggests that it does play a role in the anomalous intensity of convection in western equatorial Africa.

Corresponding author address: Sharon E. Nicholson, Department of Meteorology, The Florida State University, Tallahassee, FL 32308. Email: sen@met.fsu.edu

Abstract

This study examines mesoscale convective systems (MCSs) over western equatorial Africa using data from the Tropical Rainfall Measuring Mission (TRMM) satellite. This region experiences some of the world’s most intense thunderstorms and highest lightning frequency, but has low rainfall relative to other equatorial regions. The analyses of MCS activity include the frequency of occurrence, diurnal and annual cycles, and associated volumetric and convective rainfall. Also evaluated is the lightning activity associated with the MCSs. Emphasis is placed on the diurnal cycle and on the continental-scale motion fields in this region. The diurnal cycle shows a maximum in MCS count around 1500–1800 LT, a morning minimum, and substantial activity during the night; there is little seasonal variation in the diurnal cycle, suggesting stationary influences such as orography. Our analysis shows four maxima in MCS activity, three of which are related to local geography (two orographic and one over Lake Victoria). The fourth coincides with a midtropospheric convergence maximum in the right entrance quadrant of the African easterly jet of the Southern Hemisphere (AEJ-S). This maximum is substantially stronger in the September–November rainy season, when the jet is well developed, than in the March–May rainy season, when the jet is absent. Lightning frequency and flashes per MCS are also greatest during September–November; maxima occur in the right entrance quadrant of the AEJ-S. The lightning maximum is somewhat south of the MCS maximum and coincides with the low-lying areas of central Africa. Overall, the results of this study suggest that large-scale topography plays a critical role in the spatial and diurnal patterns of convection, lightning, and rainfall in this region. More speculative is the role of the AEJ-S, but this preliminary analysis suggests that it does play a role in the anomalous intensity of convection in western equatorial Africa.

Corresponding author address: Sharon E. Nicholson, Department of Meteorology, The Florida State University, Tallahassee, FL 32308. Email: sen@met.fsu.edu

1. Introduction

The western equatorial sector of Africa is, from a meteorological standpoint, one of the world’s most interesting, but also the most poorly understood regions. Some of the world’s highest rainfall totals are reported over Mount Cameroon, on the western edge of the region; mean annual rainfall exceeds 10 m at the station Debundscha, Cameroon, according to the Office de la Recherche Scientifique et Technique Outre-Mer (ORSTOM 1978). The coastal sector experiences interannual fluctuations of rainfall in association with Atlantic warmings that rival those produced by El Niño along the South American desert coast (Nicholson and Entekhabi 1987). This region also experiences the world’s most intense thunderstorms and the highest frequency of lightning flashes (Zipser et al. 2006; Toracinta and Zipser 2001; Petersen and Rutledge 2001).

Even within the tropics, western equatorial Africa is a convective anomaly. The only regions with comparable storm intensity, including the United States, Argentina, and parts of the Indian subcontinent, are in the midlatitudes (Mohr and Zipser 1996a,b). Despite the intensity of storms, rainfall in the region is only moderate compared with equatorial regions of the Amazon and Indonesia (Petersen and Rutledge 2001; Zipser et al. 2006). Western equatorial Africa is also the only tropical region with intense convection in all seasons. Storms are anomalously large compared to other tropical regions, with the mean size of all precipitation features exceeding 500 km2 in some parts of the region (Nesbitt et al. 2006). This region also makes a disproportionately large contribution to overshooting convection (i.e., deep convective systems with radar tops above 14 km; Liu and Zipser 2005).

This knowledge of the storm regime in equatorial Africa was made possible by the availability of a decade of observations from the Tropical Rainfall Measuring Mission (TRMM) satellite. Research based on TRMM has described the precipitation regime of western equatorial Africa in great detail. In contrast, virtually no studies of atmospheric dynamics or synoptic situations and features (such as waves) have appeared in the literature. One reason for this is the continued adherence to the traditional climatological explanation for Africa’s equatorial rainfall regime. It has generally been understood to be localized convection enhanced by the twice-yearly passage of the intertropical convergence zone (ITCZ). Consistent with this scenario, in most of the region a bimodal seasonal cycle in rainfall prevails (Fig. 1), with higher amounts during the second rainy season and minimum rainfall during the low-sun season of the respective hemisphere.

Other factors contributing to our lack of meteorological knowledge of western equatorial Africa include the relatively low interannual variability of rainfall and the difficulty in obtaining meteorological data. Unlike the drier regions of Africa, this region has suffered no significant droughts that focused attention on its meteorology. Also, the bulk of the landmass from 5°N to 10°S and eastward to the Rift Valley highlands lies in the Democratic Republic of the Congo (formerly Zaire) or Angola. In both countries, war and economic depression all but closed down the meteorological services for decades.

Fortunately, the availability of information from satellites such as TRMM has altered our picture of equatorial convection. Satellite studies of mesoscale convective systems (MCSs) have dramatically underscored the fallacy of the local convection scenario (e.g., Laing and Fritsch 1993a,b, 1997; Mohr and Zipser 1996a,b; Mohr et al. 1999; Nesbitt et al. 2006). MCSs, systems exceeding 2000 km2 in raining area, produce more than 70% of the rainfall in western equatorial Africa (Nesbitt et al. 2006). Contributions on this order, 50%–90%, are typical for heavy rain regions of the global tropics, including the Sahel, the south-central United States, and the west coast of Central America (Nesbitt et al. 2006).

Other recent work relates to the factors governing interannual variability and the seasonal cycle of the general atmospheric circulation in this region. Balas et al. (2007) demonstrated remarkable complexity in the spatiotemporal pattern of the interannual variability of rainfall. The region is highly heterogeneous with respect to interannual variability, especially compared with West Africa, where a single time series provides a first approximation of rainfall variations throughout the region. In western equatorial Africa, the regions of coherent variability are about an order of magnitude smaller and the relationship to sea surface temperature varies greatly within the region. Control on the variability shifts seasonally between the Atlantic, Pacific, and Indian Oceans, as do the interregional teleconnections. Nicholson and Grist (2003) showed that a midlevel easterly jet stream, analogous to the African easterly jet of northern Africa, is present during much of the year. The rainbelt lies between the cores of these two midlevel jets: the African easterly jet of the Southern Hemisphere (AEJ-S) and the African easterly jet of the Northern Hemisphere (AEJ-N; Grist and Nicholson 2001). Both jets are best developed around 650 mb and migrate seasonally with the rainbelt. The AEJ-N is generally within the latitudes of 5°–15°N; the AEJ-S is generally within the latitudes of 5°–10°S and is evident mainly from August to November. A recently discerned low-level coastal jet along the Atlantic coast of Angola and Namibia (Nicholson 2009a) may also be a factor in the region’s meteorology.

The overall goal of our work is to increase our understanding both of the atmospheric controls on the seasonal cycle and the interannual variability of rainfall in western equatorial Africa. This study represents one contribution to that understanding. In particular, its primary goal is to examine the seasonal and diurnal cycles of convection throughout this region. A second goal is to examine the anomalous characteristics of convection in this region, seeking an explanation in the regional atmospheric circulation. We further hope that the work will help us to develop a better understanding of the high degree of spatial heterogeneity apparent in the rainfall regime (Balas et al. 2007).

This article begins with an overview of MCS activity in equatorial Africa and associated rainfall in section 3a. This is followed in sections 3b and 3c by an examination of the seasonal and diurnal cycles, respectively. Lightning is considered in section 3d. The relationship of MCS activity, rainfall, lightning, and the diurnal cycle to topography and atmospheric circulation is discussed in section 4. Overall, the results of the study suggest that large-scale topography is probably the most important factor in the convective regime prevailing in western equatorial Africa and that the AEJ-S may play some role as well.

2. Data and methodology

a. Satellite, blended, and conventional datasets to be utilized

The TRMM satellite, launched in November 1997, carried the first quantitative spaceborne precipitation radar. It also carried a suite of complementary passive sensors, including the Microwave Imager (TMI), the Visible and Infrared Scanner (VIRS), and the Lightning Imaging Sensor (LIS; Kummerow et al. 1998, 2000). Collectively these allowed a number of characteristics of tropical rainfall to be derived on a global basis for the first time. For example, coincident information on the radar reflectivity field and passive microwave brightness temperature allowed for a detailed look at the structure of precipitating systems (Nesbitt and Zipser 2003). TRMM’s low-altitude, 35°-inclination, sun-synchronous orbit allowed for coverage from 36°N to 36°S and for sampling throughout the diurnal cycle (Zipser et al. 2006).

This study uses the TRMM database compiled at the University of Utah, described by Nesbitt et al. (2000) and Cecil et al. (2005). The precipitation feature (PF) level 3 products utilized here include the frequency of occurrence, diurnal cycle, and volumetric rainfall associated with three classes of systems. The database also includes information on stratiform versus convective rainfall and lightning. The stratiform–convective distinction is based on the precipitation radar (PR) algorithm 2A23 (see Biggerstaff and Listemaa 2000; details of the algorithm may be found online at http://trmm.gsfc.nasa.gov/2a23.html). Lightning data are taken from the LIS on board TRMM (Christian et al. 1992).

This database archives information on three types of PFs: PFs without ice scattering, PFs with ice scattering, and MCSs. This study uses only the last category, defined by Nesbitt and Zipser (2003) as a precipitation feature with “at least 2000 km2 of contiguous area with 85-GHz polarization corrected temperature (PCT) ≤ 250 K and 185 km2 ≤ 225 K.” Such features are ensured to be large convective systems (Nesbitt and Zipser 2003). The spatial resolution of this database is 1° × 1°.

It is important to point out that the numbers we derive for lightning flashes and MCSs are based on features detected during the passage of the TRMM satellite. Thus, the numbers are relative because of the low sampling frequency of TRMM and the narrow swath (215 km) of TRMM’s precipitation radar. Near the equator the sampling is on the order of 0.5 times day−1. This results in significant geographical undersampling. The use of several years of data allows us to produce numbers that are comparable geographically and temporally, but the actual number of “events” (flashes or MCSs) is only qualitative. The terms “relative number” and “relative count” are used to underscore this point. A detailed discussion of the sampling issue is found in Nesbitt and Zipser 2003).

The National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset is used in analyses of atmospheric circulation (Kalnay et al. 1996). We have used it in numerous studies of atmospheric dynamics over Africa (e.g., Nicholson and Grist 2003; Nicholson 2008; Nicholson and Webster 2007) and have verified select results with West African pibal and rawinsonde reports (Grist and Nicholson 2001). However, the conclusions based on the NCEP–NCAR data should be treated cautiously, because there are uncertainties in some variables and regions. NCEP–NCAR estimates of wind fields are considered to be relatively reliable, but there are difficulties with tropical divergent circulations and rainfall (Poccard et al. 2000; Kinter et al. 2004). There are also biases related to steep orography (Trenberth and Guillemot 1995), such as the mountain ranges over central Africa.

Unfortunately, with the paucity of upper-air soundings, blended datasets such as NCEP–NCAR are the best available tool for examining atmospheric circulation. We have designed our analyses in ways that are intended to reduce the effect of the two shortcomings noted above. For one, we use primarily what are termed “A variables,” those strongly influenced by observational data and, hence, the most reliable (Kalnay et al. 1996). These include, for example, wind and pressure fields. Less reliable are the B variables, the derivation of which is about equally dependent on observations and modeling. The only B variable we utilize is omega. Divergence is also a B variable. It was calculated offline from NCEP–NCAR winds, because the NCEP–NCAR analysis provides this variable at only two sigma levels. The impact of surface topography is perhaps more difficult to remove and in much of the analysis region, we cannot rule out the possibility that the NCEP–NCAR results reflect primarily model physics. However, we attempted to reduce the impact by examining the 925- and 850-mb (hPa) levels, which are above the surface over most of the analysis sector.

b. Analyses

We will examine the characteristics of MCSs over equatorial Africa, including frequency of occurrence, diurnal cycle and annual cycles, associated volumetric and convective rainfall, and interannual variability. Only land areas are considered. Also evaluated is the lightning activity associated with the MCSs. This is a proxy for convective intensity (Zipser et al. 2006), in that lightning requires both strong updrafts and a mixed phase microphysical environment (Toracinta and Zipser 2001). Our analyses will consider the entire continent from 20°N to 20°S, but will focus on four sectors in western equatorial Africa. Shown in Fig. 1, these extend across the continent to 25°E and each covers a 5° latitudinal band, collectively spanning the equatorial latitudes from 10°S to 10°N.

Of special interest is the diurnal cycle of convective activity. Mechanisms of the diurnal cycle in the tropics include afternoon boundary layer destabilization by radiation (Wallace 1975; Dai et al. 1999; Dai 2001), local effects of sea breezes and complex terrain (Oki and Musiake 1994; Yang and Slingo 2001; Yang and Smith 2006), and the long, nocturnal life cycle of MCSs (Sherwood and Wahrlich 1999; Nesbitt and Zipser 2003). All of these factors are present in varying degrees over western equatorial Africa. At the same time, one of the unusual climatological features of the region is the remarkable complexity in the spatiotemporal pattern of the interannual variability of rainfall (Balas et al. 2007). The region is very heterogeneous with respect to this interannual variability, especially compared with West Africa. Factors controlling the variability vary seasonally and over short distances. A better understanding of the diurnal cycle may help us to unravel and explain the complexity of the factors controlling interannual variability in this region.

The TRMM instrument that allows for analysis of the diurnal cycle is the PR. Despite the many advantages of TRMM, the narrow swath of the PR leads to a geographic undersampling on a daily basis. The satellite’s orbital characteristics result in a sampling time that varies from 0.5 times per day at the equator to nearly 2 times per day at 35° latitude. Also, the TRMM satellite takes 46 days to return to a given position at a given local time. As a result, local sampling of the complete diurnal cycle with any degree of confidence requires considerable compositing. Lin et al. (2000) suggested that at least 3 months of PR data must be combined to adequately sample a 4° × 5° grid at 1-h resolution. Thus, in this study we examine the diurnal cycle on a multiyear, multimonth basis to temper sampling errors.

3. Results

a. Climatology of MCSs and associated rainfall

Figure 2 shows the 5-yr mean of the relative number of MCSs over western equatorial Africa. The number generally increases with decreasing latitude and increases toward the interior of the continent. Relatively few occur east of the Rift Valley. Four maxima are clearly apparent, three of which are related to local geography. Two orographic maxima are centered on Mount Cameroon and the Ethiopian highlands. A third, locally induced maximum lies over Lake Victoria.

The Lake Victoria maximum is related to the nocturnal enhancement of convection by a combination of lake breezes and mountain–valley breezes (Flohn and Fraedrich 1966; Fraedrich 1972). As a result, rainfall over the lake is nearly 50% greater than in the surrounding catchment (1791 mm yr−1 versus roughly 1300 mm yr−1 in the catchment, 800–1200 mm in most of the surrounding region; Yin and Nicholson 1998).

A larger maximum extends from roughly 5°N to 5°S, and from near the Atlantic coast to the western edge of the Rift Valley highlands. Its core lies over eastern Congo/Zaire, at roughly 1°S, 27°E. It is not associated with local topography or lakes.

Figure 2 also shows for tropical Africa the mean climatology of volumetric rainfall from MCSs, rainfall per MCS, and the percentage of convective rainfall. The spatial pattern of volumetric rainfall is very similar to that corresponding to the number of MCSs. Both show a strong maximum in the central equatorial sector, to the west of Lake Victoria and the Rift Valley highlands. There is somewhat more spatial detail in the pattern of rainfall per MCS. In general, the maximum is somewhat farther north and west of the maxima in MCS activity and rainfall. A second maximum in rainfall per MCS is evident across a latitudinal band near 10°N, an area coinciding with the southern track of African easterly waves (AEWs) associated with the AEJ-N (e.g., Thorncroft and Hodges 2001). In contrast, the percentage of convective rainfall is quite uniform throughout most of the region and generally on the order of 60%–70%. A cursory comparison with local topography shows that the few scattered maxima (where convective rainfall exceeds 70%) correspond to low-lying regions.

Figure 3a shows the seasonal patterns of MCS activity. The extreme seasons [December–February (DJF) and June–August (JJA)] contrast sharply in terms of the spatial location of MCSs. During DJF, MCS activity is principally south of the equator, while it is mostly north of the equator in JJA. This is commensurate with the movement of the sun and the ITCZ from their most extreme positions in the Southern and Northern Hemispheres. The JJA pattern reflects the track of the AEWs. In DJF the area of MCS activity is considerably broader than during JJA. The two topographic maxima (Mount Cameroon and the Ethiopian highlands) show up well in JJA but not in DJF, while the Lake Victoria maximum is evident in DJF but not in JJA.

The contrast between the September–November (SON) and March–May (MAM) seasons is more surprising because both of these equatorial rainy seasons are traditionally explained by the north–south passage of the ITCZ. The local maxima over Mount Cameroon and Lake Victoria are evident in both seasons, but that over the Ethiopian highlands is evident only in SON. The Lake Victoria maximum is most pronounced during MAM, when the local enhancement of convection by the lake is at a maximum (Ba and Nicholson 1998). In contrast, the Mount Cameroon maximum is stronger in SON.

Overall, MCS activity is notably stronger in SON than in MAM. This is evident from the difference between these two seasons in terms of the size of the area in which the relative monthly frequency of MCSs is greater than 0.15 and from the large areas during SON in which the relative number of systems is greater than 0.3 month−1. More striking is the spatial pattern of activity. During MAM the central maximum in activity (i.e., that not associated with local geographic factors) runs roughly east and west. During SON the central maximum lies along a diagonal from northeast to southwest. A very interesting feature, a pronounced local maximum in MCS activity, appears near the equator and from about 25° to 28°E. Geographical considerations, such as topographic gradients, provide no immediately obvious explanation for this maximum.

Figure 3b shows the seasonal rainfall associated with MCS activity. Not surprisingly, the patterns evident in the number of MCSs are for the most part apparent in the rainfall fields. The rainbelt is latitudinally most expansive in SON and most restricted in JJA. The two local topographic maxima are most pronounced in JJA and, to a lesser extent, SON. The MCS rainfall maximum over Lake Victoria is evident throughout the year, but the effect is dramatic in MAM. Surface gauges (Fig. 4) likewise show the maximum during this season.

b. The seasonal cycle

Figure 5 shows the seasonal cycle of MCS activity and various MCS characteristics for the two northern equatorial sectors and the two southern equatorial sectors. The parameters shown include the mean relative number of MCSs, volumetric rainffigall per MCS, and the percentage of convective rainfall.

In the northernmost sector (5°–10°N) the number of MCSs peaks in August, but is relatively high from April to October. The remaining sectors all have a bimodal rather than unimodal distribution. Just north of the equator (0°–5°N) the number is relatively uniform from March to November, but peak months are April/May and October/November. In the Southern Hemisphere sectors 0°–5°S and 5°–10°S, peak months are April and December and March and November/December, respectively. In all four sectors, the number of MCSs is at a minimum during the winter months of the respective hemisphere.

The seasonal cycle of MCS count (Fig. 5) closely resembles that of rainfall (Fig. 1), although the maxima may differ by a month or two. Both variables show a single peak in the northernmost sector and a double peak elsewhere, with the intensity and length of the winter dry season increasing from the equator southward.

The seasonal cycle of volumetric rain per MCS is notably different. This variable shows much less seasonal variation. Volumetric rainfall per MCS is relatively constant from May to September in the sector 5°–10°N. It peaks in April in the sector 0°–5°N. Maxima occur in September and February in the sector 0°–5°S and in September in the sector 5°–10°S.

The percentage of convective rainfall (Fig. 5c) is likewise fairly steady, but shows some tendency for an inverse relationship with the number of MCSs (Fig. 5c). In both the Northern and Southern Hemisphere sectors, it varies between roughly 60% and 80%. The contribution of convective rainfall falls to roughly 60% during months with a large number of MCSs. This is consistent with the results of Nesbitt et al. (2006), indicating that for the tropics as a whole 48% of the precipitation over land is stratiform and 50% of the precipitation associated with MCSs over land is stratiform.

c. The diurnal cycle

A handful of studies have examined the diurnal cycle of precipitation and convection in the global tropics, focusing mainly on ocean versus land contrasts (e.g., Lin et al. 2000; Dai 2001; Yang and Slingo 2001; Yang and Smith 2006). A comprehensive study of Nesbitt and Zipser (2003) provided a very detailed view, using the advanced capabilities of the TRMM satellite. For the African continent as a whole, information on the diurnal cycle is generally limited to what can be gleaned from this global perspective. The exceptions are a study by Duvel (1989), based on Meteosat data, and a recent papers by Mohr (2004) and Futyan and Del Genio (2007), based on Meteosat and TRMM. McGarry and Reed (1978) and Reed and Jaffe (1981) also provide a detailed look at the diurnal cycle, but these studies were limited to West Africa.

Using Meteosat data with a resolution of 2.5° of latitude and longitude, Duvel showed that deep convection over equatorial Africa has a maximum around 1800 LT and a minimum around 0900 LT. This is in general agreement with the results of Nesbitt and Zipser (2003) and others. However, the latter study, at a resolution of 10° of latitude and longitude, showed some zonal variation in the timing of rainfall associated with MCSs. In the western sectors of equatorial Africa, the maximum was generally in the early morning hours. In the more central regions, rainfall associated with MCSs tended to peak during the night.

Using a higher-resolution TRMM dataset compiled by Nesbitt and Zipser, we examined the diurnal cycle over western equatorial Africa in greater detail. Figure 6 presents a spatial view of the diurnal cycle of MCS activity, volumetric rainfall per MCS and percentage of convective rain. Because the diurnal cycle does not show substantial seasonal variation, these maps are typical for the year as a whole. The maximum in MCS frequency occurs around 1500–1800 LT. This appears to be coincident with a minimum in volumetric rain per MCS and a maximum in convective rainfall. Volumetric rain per MCS peaks in the night and early morning hours, when convective rainfall is reaching a minimum.

Figure 7, which presents regional averages of the data in Fig. 6, confirms this pattern. In all seasons there is a morning minimum in MCS count and a late afternoon to early evening maximum (Fig. 7a). In general, there is substantial MCS activity during the night. Some contrasts between the seasons are evident. The most obvious is the minimal MCS count during JJA in the southern sector and DJF in the northern sector, the dry seasons in the respective hemispheres. The maxima are SON in the Northern Hemisphere and MAM in the Southern Hemisphere.

Figure 7b compares the diurnal cycle of MCS activity, volumetric rain per MCS and the percentage of convective rainfall. Data are averaged for the year as a whole for the northern and southern equatorial sectors. Diurnal cycles are similar in the two sectors. Notably, at night the volumetric rainfall per MCS is greater in the Southern Hemisphere sectors.

The results in both Figs. 6 and 7 underscore the contrast in the diurnal cycles of MCS count and rainfall per MCS. The latter has a nocturnal or morning maximum and a late afternoon/early evening minimum, while MCS count peaks around 1500–2100 LT. This is consistent with the findings of Nesbitt and Zipser (2003) and other studies of tropical rainfall (e.g., Laing and Fritsch 1997). The cycle of convective rainfall is similar to that for MCSs: a maximum in the afternoon/early evening hours and a nocturnal to early morning minimum, usually between the hours of 0300 and 0900 LT. Thus, the minimum in MCS count and convective rainfall is coincident with the maximum in rainfall per MCS. This presumably reflects the substantial contribution of stratiform rain to the total volume of rain associated with MCSs (Nesbitt et al. 2006). The stratiform rain is dominant during the night, when the systems expand and reach their maximum areal extent.

d. Lightning activity

Figure 8a shows the 5-yr mean relative number of lightning flashes, based on data from the TRMM-LIS. A pronounced maximum extends from roughly 5°N to 5°S across the continent from the Rift Valley to the Atlantic coast. Within this sector, the number increases toward the interior, so that the highest flash frequency is centered over the equator and at roughly 20°E. This is associated with a maximum in MCS activity (Fig. 2) and a maximum in the percent of MCSs with flashes and the number of flashes per MCS (Figs. 8b,c). A broad belt with several maxima is centered at roughly 10°N. It is evident also in the percent of MCSs with flashes and the number of flashes per MCS. This belt of strong convective activity generally corresponds to the southerly track of AEWs (Thorncroft and Hodges 2001). The maxima within this zone (areas with greater than 40 relative counts per year) are located near topographic features: Mount Cameroon, Darfur, and the Jos highlands. Isolated maxima in flash count also appear over Lake Victoria and on the northern slopes of the Ethiopian highlands.

Figure 9 shows the relative seasonal counts of MCSs with lightning flashes. The most notable feature of the distributions is the relative constancy of the equatorial maximum in flash count. Throughout the year there is a maximum centered around the equator and extending 5° or 10° of latitude on either side. The location of the maximum in MCS count (Fig. 3a) is much more geographically variable. This suggests more stationary influences on the lightning activity.

Figure 10 shows the diurnal cycle in the flash count for the year as a whole. The phase of the cycle does not vary much by season (not shown), suggesting that fixed geographical factors are important determinants. Lightning is both frequent and widespread between 1200 and 0000 LT. The diurnal maximum occurs between 1500 and 1800 LT and the minimum occurs between 0600 and 0900 LT.

Figure 11a shows the monthly mean relative flash count in the four equatorial sectors. In the Southern Hemisphere there are more flashes in the sector closest to the equator during most months. Notable exceptions are the months of the SON transition season, when flash frequency is strongly maximized in the zone 5°–10°S. In the Northern Hemisphere more flashes per MCS occur in the zone closer to the equator during most months. However, there is little difference in the two zones during August–October and flashes are most frequent farther north from May to July.

Figure 11b shows the monthly mean relative number of flashes per MCS in these same sectors. In the Southern Hemisphere the number of flashes per MCS is greater in the sector closest to the equator during most months. Again, notable exceptions are the transition-season months SON, when flash frequency per MCS is maximized in the zone 5°–10°S. In the Northern Hemisphere there is much less difference between the two sectors. A notable exception is the month of March, when considerably more flashes per MCS occur in the zone closer to the equator.

4. Discussion

A number of authors have commented on the paradox of the Congo/Zaire basin having the most intense storms but relatively low precipitation compared to other equatorial regions. Zipser et al. (2006) explicitly state that a fundamental question is “why…rainfall in equatorial Africa is less than that in Indonesia and equatorial South America, but why African storms are so often more intense.” Petersen and Rutledge (2001) confirmed the anomalous intensity of convection in this region. Of 22 regions evaluated, equatorial Africa had the highest ice-water content at the 7–9-km level, a prime factor in the production of lightning. Small effective diameter ice crystals are also associated with climatological maxima in lightning activity (Sherwood et al. 2006). Various studies (e.g., Sherwood 2002a,b; Ekman et al. 2004) have suggested that atmospheric aerosols reduce this diameter; hence, extensive biomass burning over equatorial Africa may play some role in creating the lightning maximum (Sherwood et al. 2006).

The cause of the low rainfall, despite the high convective intensity, is still unclear. However, a few papers have provided possible thermodynamic and cloud physics explanations for this paradox. Factors reducing the efficiency in rainfall production in this region include thermodynamic stability profiles that inhibit convection and, relative to those observed in South America, a smaller effective droplet radius, less water vapor in the atmospheric column, higher cloud bases, relatively dry air in the lower troposphere, and a relatively larger proportion of nonraining clouds (Petersen and Rutledge 2001; McCollum et al. 2000; Liu et al. 2007). A study by Geerts and Dejene (2005) also underscores various contrasts between Central Africa and the Amazon, but the results have to be seen as tentative, since the authors evaluated the DJF dry season in central Africa (see Fig. 1a). Their study did show differences in climatological relative humidity, convective available potential energy (CAPE), and low-level wind shear that might contribute to differences in rainfall amount.

Much less is understood about how the regional atmospheric circulation may contribute to the relatively low rainfall or anomalous intensity of convection. The only relevant study is that of McCollum et al. (2000), who show weak moisture flux at low levels and conclude that the Rift Valley highlands block transport from the Indian Ocean. It is relevant to point out that the net result of the circulation is that two dry seasons occur over western equatorial Africa, while only one moderately dry season occurs over the Amazon. This is certainly a contributing factor.

Here we evaluate select kinematic aspects of the regional circulation: divergence, midtropospheric winds, and vertical motion. These do not directly assess the storm intensity, which is essentially updraft strength. However, these factors do produce a mean basic state that may favor or inhibit the development of strong updrafts. We also examined atmospheric moisture during the equatorial rainy seasons. Our analyses are considered jointly with regional geographic factors and our climatological analyses of convective activity in an attempt to further understand the unique attributes of the convective regime of western equatorial Africa. In view of the uncertainties of the NCEP analysis in the region and the broad nature of the analysis, we can merely speculate on associations and suggest potential avenues for further research.

a. The lightning maximum over the Congo/Zaire basin

Figure 12 compares a high-resolution terrain map of the Congo/Zaire basin and the mean number of lightning flashes. The latter is for the year as a whole and is based on 5 yr: 1998, 1999, 2000, 2002, and 2003. The lightning maximum is clearly coincident with the regions of lowest elevation, in the center of the basin, and the number of flashes generally decreases progressively toward the surrounding regions of higher terrain.

The association between lightning frequency and low-level terrain is also apparent on a smaller scale. The maximum associated with the Jos plateau of northern Nigeria (10°N, 8°E) lies in the lower regions to the north of the plateau. An extremely strong but localized maximum in lightning flashes (Fig. 12), flashes per MCS (Fig. 8c), and rainfall per MCS lies to the west and south of the highlands of Darfur in the Sudan (14°N, 24°E).

The spatial coincidence of the lightning maximum and the low-level terrain suggests that orographic effects play a role in the production of this maximum. The seasonal constancy of the location of the lightning maximum and the seasonally constant diurnal cycle of lightning further implicate stationary geographic factors and, hence, also support the role of topography in creating the lightning maximum.

Assuming that the terrain plays a role, the mechanism at play may be akin to that described by Tripoli (1986) and Tripoli and Cotton (1989a,b) to explain the diurnal cycle of convection over the Great Plains. It applies to the case of geostrophic flow perpendicular to the terrain. Interaction of this large-scale flow with upslope flow in the afternoon creates intense convection in the lee but compensatory subsidence farther leeward over the plain. The net spatial and temporal development of intense convection results from the interaction of the geostrophic flow, slope winds, and boundary layer and radiative processes (Yang and Smith 2006).

The spatial resolution of NCEP–NCAR data is inadequate to confirm or reject this hypothesis. However, some interesting features are apparent that are consistent with the hypothesis. Figure 13 shows the mean winds at 925 and 850 mb during SON, the season of most intense convective activity and highest flash count, based on NCEP–NCAR reanalysis data. During that season the spatial distribution of lightning is similar to that for the year as a whole (Fig. 8), with maximum frequency in the Congo/Zaire basin and centered around the equator. In the sector between 5° and 10°S, on average over 50% of all lightning flashes (Fig. 11) and nearly 40% percent of all MCSs (Fig. 5) occur during this season.

Figure 13 shows that the large-scale geostrophic flow during SON is such that the Congo/Zaire basin lies in the lee of high terrain to the north, east, and west, as required for the mechanism proposed by Tripoli and Cotton (1989a,b). Early in the afternoon (1200 UTC, i.e., 1400 LT) ascent prevails over the highlands. At 1800 UTC, a peak time of convection and lightning, the mean vertical motion field shows ascent over the high terrain at 850 mb and weak subsidence downslope over the plains (Fig. 14).

b. The low rainfall over equatorial Africa

McCollum et al. (2000) demonstrate the relative dryness of the atmosphere above equatorial Africa, compared to other equatorial regions and South America, in particular. Equatorial Africa is especially dry in the layer surface–700 mb, where most of the moisture resides. Typical values of precipitable water in this layer are 25–40 kg m−2 along most of the equator, but less than 25 over equatorial Africa. For the year as a whole, total column water vapor averages 40–50 kg m−2 over South America, compared to 30–40 kg m−2 over equatorial Africa (McCollum et al. 2000). Figure 15 illustrates the contrast with South America for November and February, two months that are common to the rainy seasons in the equatorial regions of both continents. Values over equatorial South America are on the order of 40–50 kg m−2. Over Africa values exceed 40 kg m−2 in only a very limited area of the western equatorial region. East of the highlands, values between 25 and 30 kg m−2 prevail.

McCollum et al. noted the “aridity” east of the highlands and suggested that this feature and the restriction of moisture transport from the Indian Ocean by the highlands are major factors in the relatively low amounts of rainfall over equatorial Africa. We suggest that additional aspects of the regional circulation play a role, as well.

For example, several points of evidence suggest that the Atlantic is also an important moisture source for this region. The spatial distribution of moisture (Fig. 15) and the low-level flow (Fig. 13) (Nicholson 2009a) suggest that the primary transport is from the Atlantic. During the SON season of intense convective activity, rainfall in the western extreme of the region is strongly correlated with Atlantic sea surface temperatures, but not with sea surface temperatures in the Indian Ocean (Balas et al. 2007). As with flow from the east, the transport of moisture is blocked at low levels by the high terrain.

We speculate that two other situations may further contribute to the anomalously low rainfall. One is the advection of relatively dry air from North Africa into the northern rim of the Congo/Zaire basin (Fig. 13). The other is the mesoscale situation described by Tripoli and Cotton (1989a,b) to explain convection in the lee of high terrain. In areas of subsidence over the plains in the lee (i.e., within the Congo/Zaire basin), dry area is mixed with the surface air and adiabatic heating further reduces the relative humidity.

c. The Darfur region

An area south and west of the highlands of Darfur, in the southern Sudan (10°N, 22°E), stands out as a bull’s-eye in many of our analyses related to convective intensity. Examples are rainfall per MCS (Fig. 2), lightning flash count (Fig. 8a), and flashes per MCS (Fig. 8c). Because of the extreme values noted in the analysis, the data for the region were individually checked and the maxima over this region were confirmed.

The intensity of convection in this region is noteworthy because several studies (e.g., Hodges and Thorncroft 1997; Mekonnen et al. 2006) have suggested that convection over Darfur triggers many of the AEWs that move westward during the summer rainy season. The anomalous intensity of convection in this regions lends support to this idea.

d. The intensity of convective activity.

The orographic effects described in section 4a could be a factor in the extreme intensity of convective activity over the Congo/Zaire basin. An examination of regional circulation suggests one other possible factor. Although intense convective activity is ubiquitous over the basin, it clearly has a spatial maximum in the easternmost region in the latitudes from the equator to 10°S and a seasonal maximum during the SON season. Here, we examine that season in further detail, using NCEP–NCAR reanalysis data and rainfall from the African archive of the second author. An examination of all three months shows similar patterns, so that October is used to represent the season. Monthly means are utilized. Although the time scale of individual MCSs is much shorter, we feel the monthly mean provides a reasonable cursory look. For one, convective activity is present during most days in October. Also, an examination of winds on individual days shows that the monthly pattern is clearly evident on individual days.

The most striking feature of the October wind field is a midlevel easterly jet stream (the AEJ-S) with a core at 600 mb and 8°S (Fig. 16). The core lies west of the highlands of eastern Africa both in the mean and on the individual days when it is well developed. This feature is present throughout the SON rainy season, but completely absent during the MAM rainy season (Nicholson and Grist 2003). Its core speed averages 12 m s−1, but speeds can well exceed 15 m s−1 on individual days. The AEJ-S appears to be a result of low-level surface temperature gradients (Fig. 17), which are strong following the dry season of the Southern Hemisphere subtropics (i.e., in SON) but extremely weak following the Southern Hemisphere rainy season (i.e., in MAM). The same thermal pattern is evident also at 850 mb, so it is unlikely to be a result of the NCEP–NCAR model physics and is readily confirmed by numerous satellite analyses of surface temperature.

We hypothesize that the presence of the AEJ-S may enhance convection during the seasons when it is present and speculate that the physical link is a jet streak circulation similar to that known from midlatitude studies. Upper-level jets are characterized by considerable zonal asymmetry, with distinct entrance and exit regions and ensuing vertical circulations with distinct areas of convergence and divergence. Uccellini and Johnson (1979) developed a four-quadrant conceptual model of a “jet streak” (core region of maximum velocity) to describe this asymmetry. The model prescribes for a Northern Hemisphere jet divergence in the right entrance region, convergence in the left entrance region, and the reverse patterns in the exit region. The converse pattern prevails for a Southern Hemisphere jet.

The conditions for the development of such a jet streak are stringent (Keyser and Shapiro 1986), with the most basic being geostrophy. Two considerations suggest that the flow in the vicinity of the AEJ-S is approximately geostrophic. One is that the winds are roughly parallel to the geopotential height contours (not shown). The other is the Rossby number (Ug/Lf, where Ug is the geostrophic speed, L is a typical length scale of the jet, and f is the Coriolis parameter at 10°S). A calculation with typical speeds and length scale give a value of 0.3, indicating that quasigeostrophic theory, including jet streak concepts, can be appropriately utilized in evaluating the AEJ-S (Cunningham and Keyser 2000).

Figure 18 shows the mean divergence during October at 600 and 200 mb, calculated offline from NCEP–NCAR winds, and the mean vertical motion at 600 mb. A four-quadrant checkerboard pattern in divergence is evident at 600 mb. The axis of the AEJ-S (Fig. 16) separates the right and left quadrants and a large area of convergence corresponds to the right entrance region of the jet. A core of vertical motion on the order of 0.3–0.5 mb s−1 × 10−2 is coincident with this area of convergence. Divergence at 200 mb overrides it. The jet’s axis closely follows the edge of the divergence maximum at 200 mb and the maximum in convective intensity during SON. Thus, the kinematic characteristics of the motion field are consistent with a jet streak circulation.

Requirements for both intense lightning and strong convection include strong updrafts (Petersen and Rutledge 2001; Toracinta and Zipser 2001). The aforementioned convergence maximum coincides with the lightning maximum around 25°–30°E (Figs. 11 and 12). Although convection commences near the surface, its intensification and deepening would be enhanced by the combination of midlevel convergence and strong outflow in the upper troposphere. Notably, a similar situation prevails during the rainy season over West Africa (Nicholson and Webster 2007; Nicholson 2008, 2009b). Within the core of the rainbelt, convergence associated with the AEJ-N prevails at midlevels, divergence associated with the tropical easterly jet prevails at 200 mb, and a core of strong vertical motion lies in between.

Our interpretation is that the presence of the AEJ-S produces the midlevel convergence prevailing over the Congo/Zaire basin and, hence, plays a role in the anomalous intensity of convection in this region. Nonetheless, the question arises as to whether the jet itself may be a product of the latent heat release associated with intense convection. Thorncroft and Blackburn (1999) suggested that latent heat release is a factor in the development of the AEJ-N over North Africa.

The reality of the situation for equatorial Africa cannot be determined from the cursory analysis performed here. However, there are several points of supporting evidence for our conjecture that the AEJ-S may enhance convection, rather than be a result of it. One is the absence of the convective maximum west of Lakes Victoria and Tanganyika during the MAM rainy season (Fig. 3). Although total rainfall (Fig. 19) and, hence, latent heat release in the equatorial latitudes is comparable during the MAM and SON seasons, the AEJ-S does not develop during the MAM season (Nicholson and Grist 2003). Second, the development of the jet during SON and its absence during MAM are consistent with the gradients of surface heating during these seasons (Fig. 17). The strong surface gradient during SON results from the prolonged dry season over southern Africa. Finally, the convection is equatorward of the jet. During SON, thermal wind considerations indicate that latent heat release at midlevels would tend to reduce rather than enhance easterly winds. Above 600 mb, the temperature gradient is reversed because of convective heating equatorward of the jet, resulting in the location of the core at 600 mb.

The AEJ-S might also contribute to the convective intensity in other ways. Zipser et al. (2006) identify commonalities of the other regions with large numbers of huge MCSs, the central United States and southeast South America. Both regions are characterized by strong low-level wind shear, a low-level jet that brings in very moist air, highlands that trigger disturbances that lift the low-level air and release convective instability. Two of these features, highlands and low-level wind shear, are present over the Congo/Zaire basin. Zipser et al. (2006) further note the possible importance of midlevel jets, such as the AEJ-N, in producing the requisite shear. Mohr and Thorncroft (2006) confirm the link between intense convection over West Africa and the AEJ-N, implicating shear as the important factor. The AEJ-S likewise markedly enhances the vertical and horizontal shear in western equatorial Africa.

5. Summary and conclusions

To further our understanding of the atmospheric dynamics regulating precipitation in western equatorial Africa, this study examined the seasonal and diurnal cycles of convection throughout this region. This study also examined potential factors in the anomalous intensity of convection in this region and the anomalously low rainfall.

The most salient results of our study are 1) the seasonally invariant diurnal cycles of convective activity and lightning, 2) the seasonally varying spatial distribution of MCSs, 3) the relatively seasonally invariant location of the lightning maximum, 4) the development of maximum convective intensity during the second rainy season (SON), 5) the afternoon maximum in MCS occurrence and the percentage of convective rainfall, and 6) the morning maximum in total rainfall from MCSs.

The results of this study suggest that large-scale topography is a critical factor in the spatial and diurnal patterns of convection, lightning, and rainfall in western equatorial Africa. The interaction of topographic effects and regional circulation may play some role in the relatively low rainfall, but do not fully explain this feature. Our results further suggest that the AEJ-S may play some role in the anomalous intensity of convection in this region. The complexity of these factors provides at least a partial explanation for the complexity and spatial heterogeneity of the rainfall regime, as noted by Balas et al. (2007).

In the western equatorial latitudes of Africa there are four maxima in MCS activity, three of which are related to geographical factors. Pronounced maxima in the frequency of occurrence of MCSs and the volumetric rainfall associated with them correspond to the high terrain of Mount Cameroon and the surrounding highlands, the Ethiopian highlands, and Lake Victoria. The fourth maximum, near 0°, 25°–28°E, shows no association with geographical factors. It coincides with the right entrance region of the AEJ-S. From this study alone, it is impossible to determine if the jet is a factor in or a result of the intense convection, or is merely coincidentally collocated with the convective maximum. However, several observations suggest that the AEJ-S may contribute to the intensity of convection.

The conclusions we draw here are tenuous, as the study is limited by the limitations of the NCEP–NCAR reanalysis dataset and by use of monthly means in the analysis. Nevertheless, clear temporal and spatial associations have been demonstrated between convective activity and regional topography, and circumstantial evidence is presented for an association between the anomalous intensity of convection and the AEJ-S. Further research should concentrate on determining the mesoscale motions in the region and their relationship to terrain and convection and on evaluating possible links between the AEJ-S and convection on synoptic time scales.

Acknowledgments

This study was supported by NSF Grant ATM-0004479. Much of the work in this article is taken from the M.S. thesis of the first author. We express our thanks to Steve Nesbitt, Ed Zipser, Chuntao Liu, and others who carried out the tedious job of compiling the precipitation feature database and making it available to others. Without this data base, our study could not have been carried out. We also thank Ed Zipser and two anonymous reviewers for their thorough reading of the manuscript and valuable suggestions for improvement.

REFERENCES

  • Ba, M. B., , and S. E. Nicholson, 1998: Analysis of convective activity and its relationship to the rainfall over the Rift Valley lakes of East Africa during 1983–90 using the Meteosat infrared channel. J. Appl. Meteor., 37 , 12501264.

    • Search Google Scholar
    • Export Citation
  • Balas, N., , S. E. Nicholson, , and D. Klotter, 2007: The relationship of rainfall variability in West Central Africa to sea-surface temperature fluctuations. Int. J. Climatol., 27 , 13351349.

    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., , and S. A. Listemaa, 2000: An improved scheme for convective/stratiform echo classification using radar reflectivity. J. Appl. Meteor., 39 , 21292150.

    • Search Google Scholar
    • Export Citation
  • Cecil, D. A., , S. J. Goodman, , D. J. Boccippio, , E. J. Zipser, , and S. W. Nesbitt, 2005: Three years of TRMM precipitation features. Part I: Radar, radiometric, and lightning characteristics. Mon. Wea. Rev., 133 , 543566.

    • Search Google Scholar
    • Export Citation
  • Christian, H. J., , R. J. Blakeslee, , and S. J. Goodman, 1992: Lightning Imaging Sensor (LIS) for the Earth Observing System. NASA Tech. Memo. 4350, MSFC, Huntsville, AL, 36 pp.

    • Search Google Scholar
    • Export Citation
  • Cunningham, P., , and D. Keyser, 2000: Analytical and numerical modeling of jet streaks: Barotropic dynamics. Quart. J. Roy. Meteor. Soc., 126 , 31873217.

    • 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
  • Duvel, J-P., 1989: Convection over tropical Africa and the Atlantic Ocean during northern summer. Part I: Interannual and diurnal variations. Mon. Wea. Rev., 117 , 27822799.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , C. Wang, , J. Wilson, , and J. Strom, 2004: Explicit simulations of aerosol physics in a cloud-resolving model: A sensitivity study based on an observed convective cloud. Atmos. Chem. Phys., 4 , 773791.

    • Search Google Scholar
    • Export Citation
  • Flohn, H., , and K. Fraedrich, 1966: Tagesperiodische Zirkulation und Niederschlagsverteilung am Victoria-See (Ostafrika) (The daily periodic circulation and distribution of rainfall over Lake Victoria). Meteor. Rundsch., 19 , 157165.

    • Search Google Scholar
    • Export Citation
  • Fraedrich, K., 1972: A simple climatological model of the dynamics and energetics of the nocturnal circulation at Lake Victoria. Quart. J. Roy. Meteor. Soc., 98 , 322335.

    • Search Google Scholar
    • Export Citation
  • Futyan, J. M., , and A. D. Del Genio, 2007: Deep convective system evolution over Africa and the tropical Atlantic. J. Climate, 20 , 50415060.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., , and T. Dejene, 2005: Regional and diurnal variability of the vertical structure of precipitation systems in Africa based on spaceborne radar data. J. Climate, 18 , 893916.

    • Search Google Scholar
    • Export Citation
  • Grist, J. P., , and S. E. Nicholson, 2001: A study of the dynamic factors influencing the interannual variability of rainfall in the West African Sahel. J. Climate, 14 , 13371359.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., , and C. D. Thorncroft, 1997: Distribution and statistics of African mesoscale convective weather systems based on the ISCCP Meteosat imagery. Mon. Wea. Rev., 125 , 28212837.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Keyser, D., , and M. A. Shapiro, 1986: A review of the structure and dynamics of upper-level frontal zones. Mon. Wea. Rev., 114 , 452499.

    • Search Google Scholar
    • Export Citation
  • Kinter III, J. L., , M. J. Fennessy, , V. Krishnamurthy, , and L. Marx, 2004: An evaluation of the apparent interdecadal shift in the tropical divergent circulation in the NCEP–NCAR reanalysis. J. Climate, 17 , 349361.

    • Search Google Scholar
    • Export Citation
  • Kummerow, C., , W. Barnes, , T. Kozu, , J. Shiue, , and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol., 15 , 809817.

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

    • Search Google Scholar
    • Export Citation
  • Laing, A. G., , and J. M. Fritsch, 1993a: Mesoscale convective complexes in Africa. Mon. Wea. Rev., 121 , 22542265.

  • Laing, A. G., , and J. M. Fritsch, 1993b: Mesoscale convective complexes over the Indian monsoon region. J. Climate, 6 , 911919.

  • Laing, A. G., , and J. M. Fritsch, 1997: The global population of mesoscale convective complexes. Quart. J. Roy. Meteor. Soc., 123 , 389405.

    • Search Google Scholar
    • Export Citation
  • Lin, X., , D. A. Randall, , and L. D. Fowler, 2000: Diurnal variability of the hydrologic cycle and radiative fluxes: Comparisons between observations and a GCM. J. Climate, 13 , 41594179.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , and E. J. Zipser, 2005: Global distribution of convection penetrating the tropical tropopause. J. Geophys. Res., 110 , D23104. doi:10.1029/2005JD006063.

    • Search Google Scholar
    • Export Citation
  • Liu, C., , E. J. Zipser, , and S. W. Nesbitt, 2007: Global distribution of tropical deep convection: Different perspective from TRMM infrared and radar data. J. Climate, 20 , 489503.

    • Search Google Scholar
    • Export Citation
  • McCollum, J. R., , A. Gruber, , and M. B. Ba, 2000: Discrepancy between gauges and satellite estimates of rainfall in equatorial Africa. J. Appl. Meteor., 39 , 666679.

    • Search Google Scholar
    • Export Citation
  • McGarry, M. M., , and R. J. Reed, 1978: Diurnal variations in convective activity and precipitation during phases II and III of GATE. Mon. Wea. Rev., 106 , 101113.

    • Search Google Scholar
    • Export Citation
  • Mekonnen, A., , C. D. Thorncroft, , and A. Aiyyer, 2006: Analysis of convection and its association with African easterly waves. J. Climate, 19 , 54055421.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., 2004: Interannual, monthly, and regional variability in the wet season diurnal cycle of precipitation in sub-Saharan Africa. J. Climate, 17 , 24412453.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and E. J. Zipser, 1996a: Mesoscale convective systems defined by their 85-GHz ice-scattering signature: Size and intensity comparison over tropical oceans and continents. Mon. Wea. Rev., 124 , 24172437.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and E. J. Zipser, 1996b: Defining mesoscale convective systems by their 85-GHz ice-scattering signatures. Bull. Amer. Meteor. Soc., 77 , 11791189.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , and C. Thorncroft, 2006: Intense convective systems in West Africa and their relationship to the African easterly jet. Quart. J. Roy. Meteor. Soc., 132 , 163176.

    • Search Google Scholar
    • Export Citation
  • Mohr, K. I., , J. S. Famiglietti, , and E. J. Zipser, 1999: The contribution to tropical rainfall with respect to convective system type, size, and intensity estimated from 85-GHz ice-scattering signature. J. Appl. Meteor., 38 , 596606.

    • 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
  • Nesbitt, S. W., , E. J. Zipser, , and D. J. Cecil, 2000: A census of precipitation features in the Tropics using TRMM: Radar, ice scattering, and ice observations. J. Climate, 13 , 40874106.

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

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 1986: The spatial coherence of African rainfall anomalies: Interhemispheric teleconnections. J. Climate Appl. Meteor., 25 , 13651381.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2008: The intensity, location and structure of the tropical rainbelt over West Africa as factors in interannual variability. Int. J. Climatol., 28 , 17751785.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2009a: A low-level jet along the Benguela coast, an integral part of the Benguela Marine Ecosystem. Climatic Change, in press.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., 2009b: On the factors modulating the intensity of the tropical rainbelt over West Africa. Int. J. Climatol., 29 , 673689.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and D. Entekhabi, 1987: Rainfall variability in equatorial and southern Africa: Relationships with sea surface temperatures along the southwestern coast of Africa. J. Climate Appl. Meteor., 26 , 561578.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and J. P. Grist, 2003: On the seasonal evolution of atmospheric circulation over West Africa and equatorial Africa. J. Climate, 16 , 10131030.

    • Search Google Scholar
    • Export Citation
  • Nicholson, S. E., , and P. J. Webster, 2007: A physical basis for the interannual variability of rainfall in the Sahel. Quart. J. Roy. Meteor. Soc., 133 , 20652084.

    • Search Google Scholar
    • Export Citation
  • Oki, T., , and K. Musiake, 1994: Seasonal change of the diurnal cycle of precipitation over Japan and Malaysia. J. Appl. Meteor., 33 , 14451463.

    • Search Google Scholar
    • Export Citation
  • ORSTOM, 1978: Précipitations journalières de l’origine des stations à 1972. Etienne Julien, Paris, Vol. 1, 512 pp.

  • Petersen, W. A., , and S. A. Rutledge, 2001: Regional variability in tropical convection: Observations from TRMM. J. Climate, 14 , 35663586.

    • Search Google Scholar
    • Export Citation
  • Poccard, I., , S. Janicot, , and P. Camberlin, 2000: Comparison of rainfall structures between NCEP/NCAR reanalysis and observed data over tropical Africa. Climate Dyn., 16 , 897915.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , and J. D. Jaffe, 1981: Diurnal variation of summer convection over West Africa and the tropical eastern Atlantic during 1974 and 1978. Mon. Wea. Rev., 109 , 25272534.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., 2002a: Aerosols and ice particle size in tropical cumulonimbus. J. Climate, 15 , 10511063.

  • Sherwood, S. C., 2002b: A microphysical connection among biomass burning, cumulus clouds, and stratospheric moisture. Science, 295 , 12711275.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., , and R. Wahrlich, 1999: Observed evolution of tropical deep convective events and their environment. Mon. Wea. Rev., 127 , 17771795.

    • Search Google Scholar
    • Export Citation
  • Sherwood, S. C., , T. J. Phillips, , and J. S. Wettlaufer, 2006: Small ice crystals and the climatology of lightning. Geophys. Res. Lett., 33 , L05804. doi:10.1029/2005GL025242.

    • Search Google Scholar
    • Export Citation
  • Thorncroft, C. D., , and M. Blackburn, 1999: Maintenance of the African easterly jet. Quart. J. Roy. Meteor. Soc., 125 , 763786.

  • Thorncroft, C., , and K. Hodges, 2001: African Easterly Wave variability and its relationship to Atlantic tropical cyclone activity. J. Climate, 14 , 11661179.

    • Search Google Scholar
    • Export Citation
  • Toracinta, E. R., , and E. J. Zipser, 2001: Lightning and SSM/I-ice-scattering mesoscale convective systems in the global tropics. J. Appl. Meteor., 40 , 9831002.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and C. J. Guillemot, 1995: Evaluation of the global atmospheric moisture budget as seen from analyses. J. Climate, 8 , 22552272.

    • Search Google Scholar
    • Export Citation
  • Tripoli, G. J., 1986: A numerical investigation of an orogenic mesoscale convective system. Ph.D. dissertation, Colorado State University, 290 pp.

  • Tripoli, G. J., , and W. R. Cotton, 1989a: Numerical study of an observed orogenic mesoscale convective system. Part I: Simulated genesis and comparison with observations. Mon. Wea. Rev., 117 , 273304.

    • Search Google Scholar
    • Export Citation
  • Tripoli, G. J., , and W. R. Cotton, 1989b: Numerical study of an observed orogenic mesoscale convective system. Part II: Analysis of governing dynamics. Mon. Wea. Rev., 117 , 305328.

    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., , and D. R. Johnson, 1979: The coupling of upper and lower tropospheric jet streaks and implications for the development of severe convective storms. Mon. Wea. Rev., 107 , 682703.

    • Search Google Scholar
    • Export Citation
  • Wallace, J. M., 1975: Diurnal variations in precipitation and thunderstorm frequency over the conterminous United States. Mon. Wea. Rev., 103 , 406419.

    • Search Google Scholar
    • Export Citation
  • Yang, G. Y., , and J. Slingo, 2001: The diurnal cycle in the Tropics. Mon. Wea. Rev., 129 , 784801.

  • Yang, S., , and E. A. Smith, 2006: Mechanisms for diurnal variability of global tropical rainfall observed from TRMM. J. Climate, 19 , 51905226.

    • Search Google Scholar
    • Export Citation
  • Yin, X., , and S. E. Nicholson, 1998: The water balance of Lake Victoria. J. Hydrol. Sci., 43 , 789811.

  • Zipser, E. J., , D. J. Cecil, , C. Liu, , S. W. Nesbitt, , and D. P. Yorty, 2006: Where are the most intense thunderstorms on earth? Bull. Amer. Meteor. Soc., 87 , 10571071.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

(a) Location of study area and the annual rainfall distribution within four subsectors. For the three southernmost sectors, rainfall distribution is indicated for both the western portion and the eastern portion of the analysis region. (b) Terrain contours and geographic locations mentioned in the text.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 2.
Fig. 2.

Five-year average of (a) relative number of MCSs per year, (b) total volumetric rainfall (km2 mm h−1) from MCSs, (c) volumetric rainfall per MCS, and (d) percentage of convective rainfall. These quantities are averaged over 1° × 1° grid points.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 3.
Fig. 3.

Five-year average of (a) relative number of MCSs per month and per grid point and (b) seasonal rainfall (mm month−1) from MCSs for the four seasons: DJF, MAM, JJA, and SON. These quantities are averaged over 1° × 1° grid points.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 4.
Fig. 4.

Hourly amount of rainfall at selected stations along the shore of Lake Victoria during April and September. The extreme values at Nabuyongo Island, in the center of Lake Victoria, are indicative of the enhance of rainfall by the lake (from Ba and Nicholson 1998).

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 5.
Fig. 5.

Monthly means of MCS activity: mean relative number of MCSs within the indicated latitudinal sector, volumetric rainfall per MCS (104 mm km2 h−1), and percentage of convective rainfall within the two northern equatorial zones and two southern equatorial zones. Data on volumetric rainfall and percentage of convective rainfall are omitted for months with relatively few MCSs.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 6.
Fig. 6.

Maps of the diurnal cycle: (a) mean relative number of MCSs within each 2° × 2° lat–lon grid box, (b) the mean volumetric rainfall per MCS (104 mm km2 h−1), and (c) the percentage of convective rainfall at 3-h intervals, starting at 0000–0300 LT.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 7.
Fig. 7.

The diurnal cycle of convection (a) seasonal averages for the relative number of MCSs in 3-h intervals for northern and southern sectors (see Fig. 1a for location). (b) The annual averages of the relative number of MCSs, the mean volumetric rainfall per MCS (104 mm km2 h−1), and the percentage of convective rainfall in 3-h intervals for the northern and southern sectors for the year as a whole.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 8.
Fig. 8.

(a) Five-year mean relative number of lightning flashes (per 1° × 1° latitude–longitude) for the year as a whole, (b) mean percent MCSs with flashes, and (c) and number of flashes per MCS. These are indicators of the overall intensity of convection. All data are averaged for a 1° × 1° grid box, with a three-point smoothing applied. The maxima are highlighted by showing only areas with greater than 20 lightning flashes per year or 10 lightning flashes per season.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 9.
Fig. 9.

Five-year mean relative number of lightning flashes during each 3-month season (total within each 1° × 1° latitude–longitude grid box; three-point smoothing has been applied).

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 10.
Fig. 10.

Diurnal cycle of lightning (relative number of flashes per 1° × 1° grid within each 3-h time span). Data are averaged for all months of the 5-yr study period.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 11.
Fig. 11.

Seasonal cycle of flash count (relative flash count per month) within (top) the indicated latitudinal sector and (bottom) average number of flashes per MCS within the sector. Data are shown for the four latitudinal sectors shown in Fig. 1. A three-point smoothing has been applied.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 12.
Fig. 12.

Average relative number of flashes per year in each 1° × 1° grid box, superimposed upon a faux 3D terrain map of Africa. The maximum is highlighted by showing only areas with a relative flash count greater than 10 per year.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 13.
Fig. 13.

Mean winds at 850 and 925 mb during SON.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 14.
Fig. 14.

Mean vertical motion (omega; mb s−1 × 10−2) at 850 mb at 1200 and 1800 UTC (LT is generally 2–3 h later).

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 15.
Fig. 15.

Mean precipitable water (kg m−2) over Africa and South America in February and November.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 16.
Fig. 16.

The AEJ-S: (a) mean wind (m s−1) at 600 mb during October; (b) vertical cross section of mean zonal wind at 20°E as a function of latitude (October; m s−1).

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 17.
Fig. 17.

Mean temperature (°C) at 925 mb for April and October, based on NCEP–NCAR data.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 18.
Fig. 18.

Mean divergence at 600 and 200 mb and mean vertical motion (omega; mb s−1 × 10−2) at 600 mb during October. The thin crossed lines represent the axes of the AEJ-S.

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Fig. 19.
Fig. 19.

Mean rainfall (mm) during MAM and SON (from Balas et al. 2007).

Citation: Monthly Weather Review 137, 4; 10.1175/2008MWR2525.1

Save