• Acheampong, P. K., 1982: Rainfall anomaly along the coast of Ghana—Its nature and causes. Geogr. Ann., 64A , 199211.

  • Barnes, G. M., , and K. Sieckman, 1984: The environment of fast- and slow-moving tropical mesoscale convective cloud lines. Mon. Wea. Rev., 112 , 17821794.

    • Search Google Scholar
    • Export Citation
  • Berry, G. J., , and C. D. Thorncroft, 2005: Case study of an intense African Easterly Wave. Mon. Wea. Rev., 133 , 752766.

  • Buckle, C., 1996: Weather and Climate in Africa. Addison Wesley Longman Limited, 312 pp.

  • Chong, M., , and D. Hauser, 1989: A tropical squall line observed during the COPT 81 experiment in West Africa. Part II: Water budget. Mon. Wea. Rev., 117 , 728744.

    • Search Google Scholar
    • Export Citation
  • Chong, M., , P. Amayenc, , G. Scialom, , and J. Testud, 1987: A tropical squall line observed during the COPT 81 experiment in West Africa. Part I: Kinematic structure inferred from dual-Doppler radar data. Mon. Wea. Rev., 115 , 670694.

    • Search Google Scholar
    • Export Citation
  • Dhonneur, G., 1981: Les amas nuageux mobiles principale composante de la météorologie du Sahel. La Météorologie, 27 , 7582.

  • Duvel, J. P., 1990: Convection over tropical Africa and the Atlantic Ocean during northern summer. Part II: Modulation by easterly waves. Mon. Wea. Rev., 118 , 18551868.

    • Search Google Scholar
    • Export Citation
  • Eldridge, R. H., 1957: A synoptic study of West African disturbance lines. Quart. J. Roy. Meteor. Soc., 83 , 303314.

  • Fink, A. H., , and A. Reiner, 2003: Spatio-temporal variability of the relation between African easterly waves and West African squall lines in 1998 and 1999. J. Geophys. Res., 108 .4332, doi:10.1029/2002JD002816.

    • Search Google Scholar
    • Export Citation
  • Fink, A. H., , D. G. Vincent, , and V. Ermert, 2006: Rainfall types in the West African Sudanian zone during the summer monsoon 2002. Mon. Wea. Rev., 134 , 21432164.

    • Search Google Scholar
    • Export Citation
  • Fortune, M., 1980: Properties of African squall lines from time-lapse satellite imagery. Mon. Wea. Rev., 108 , 153168.

  • 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
  • Houze Jr., R. A., , and A. K. Betts, 1981: Convection in GATE. Rev. Geophys. Space Phys., 19 , 541576.

  • Kamara, I., 1986: The origins and types of rainfall in West Africa. Weather, 41 , 4856.

  • Mathon, V., , A. Diedhiou, , and H. Laurent, 2002a: Relationship between easterly waves and mesoscale convective systems over the Sahel. Geophys. Res. Lett., 29 .1216, doi:10.1029/2001GL014371.

    • Search Google Scholar
    • Export Citation
  • Mathon, V., , H. Laurent, , and T. Lebel, 2002b: Mesoscale convective system rainfall in the Sahel. J. Appl. Meteor., 41 , 10811092.

  • Omotosho, J. B., 1985: The separate contributions of squall lines, thunderstorms and the monsoon to the total rainfall in Nigeria. J. Climatol., 5 , 543552.

    • Search Google Scholar
    • Export Citation
  • Parker, D. E., , L. V. Alexander, , and J. Kennedy, 2004: Global and regional climate in 2003. Weather, 59 , 145152.

  • Payne, S. W., , and M. M. McGarry, 1977: The relationship of satellite inferred convective activity to easterly waves over West Africa and the adjacent ocean during phase III of GATE. Mon. Wea. Rev., 105 , 413420.

    • Search Google Scholar
    • Export Citation
  • Peters, M., , and G. Tetzlaff, 1988: The structure of West African squall lines and their environmental moisture budget. Meteor. Atmos. Phys., 39 , 7484.

    • Search Google Scholar
    • Export Citation
  • Peters, M., , G. Tetzlaff, , and W. Janssen, 1989: Rainfall intensity of West African squall lines. Ann. Geophys., 7 , 227238.

  • Redelsperger, J. L., , A. Diongue, , A. Diedhiou, , J. P. Ceron, , M. Diop, , J. F. Gueremy, , and J. P. Lafore, 2002: Multi-scale description of a Sahelian synoptic weather system representative of the West African monsoon. Quart. J. Roy. Meteor. Soc., 128 , 12291257.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , D. C. Norquist, , and E. E. Recker, 1977: The structure and properties of African wave disturbances observed during Phase III of GATE. Mon. Wea. Rev., 105 , 317333.

    • Search Google Scholar
    • Export Citation
  • Roux, F., 1988: The West African squall line observed on 23 June 1981 during COPT 81: Kinematics and thermodynamics of the convective region. J. Atmos. Sci., 45 , 406426.

    • Search Google Scholar
    • Export Citation
  • Rowell, D. P., , and J. R. Milford, 1993: On the generation of African squall lines. J. Climate, 6 , 11811193.

  • Sommeria, G., , and J. Testud, 1984: COPT 81: A field experiment designed for the study of dynamics and electrical activity of deep convection in continental tropical regions. Bull. Amer. Meteor. Soc., 65 , 410.

    • Search Google Scholar
    • Export Citation
  • Thorncroft, C. D., and Coauthors, 2003: The JET2000 project—Aircraft observations of the African easterly jet and African easterly waves. Bull. Amer. Meteor. Soc., 84 , 337351.

    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., 1977: Mesoscale and convective-scale downdrafts as distinct components of squall-line structure. Mon. Wea. Rev., 105 , 15681589.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Geography of West Africa, with locations of radiosonde stations indicated (Aga: Agadez; Bam: Bamako; Ban: Bangui; Lib: Libreville; Nia: Niamey; Ndj: Ndjamena; Oua: Ouagadougou; Oue: Ouesso; Par: Parakou). Note the location of the Upper Ouémé River valley. The shading shows terrain height above 400, 800, and 1500 m, respectively. (b) Geography of the Upper Ouémé River valley, with dots indicating the locations of pluviograph sites. The box indicates the domain of the Upper Ouémé River valley (i.e., 9°–10.3°N, 1.5°–3°N) used in processing Meteosat data. Topography is indicated in meters.

  • View in gallery

    Meteosat infrared imagery every 3 h from 0° to 15°N for (a) case I, (b) case II, and (c) case III. At times when the 0000 UTC image was not available, it has been replaced by the 2300 UTC image. Cases I, II, and III are highlighted, as are the longitudes of the UOV domain.

  • View in gallery

    Time series of various surface meteorological parameters at locations in the UOV (see Fig. 1b) between 2100 UTC 14 and 1200 UTC 15 Sep 2002. (a) Hourly temperature and specific humidity, q, observed at Parakou. (b) The 15-min temperature at Djougou and Doguè (“Dog”) and specific humidity, q, at Djougou only. (c) Maximum wind gusts within 10-min intervals and 10-min-averaged wind barbs, and hourly 24-h pressure tendency at Parakou. The bars indicate 6-min accumulated rainfall. (d), (e) The 15-min averaged wind speed and barbs at Djougou and at Doguè, respectively; the bars indicate 6-min accumulated rainfall.

  • View in gallery

    Relative vorticity at 700-hPa averaged from 7° to 13°N (contour interval is 10 × 10−6 s−1, positive vorticity areas are shaded) and cloud tracks (white circles) for (a) mid-September 2002 and (b) late July and early August 2002. The longitudes of the UOV are indicated. In addition, the approximate propagation of AEW cyclonic vorticity signals are delineated by the bold straight lines in (a). The dashed line in the same panel indicates the weak downstream trough for case II.

  • View in gallery

    Streamlines and equivalent potential temperatures in °C at 850 hPa (a) for 0000 UTC, (b) for 0600 UTC, and (c) for 1200 UTC.

  • View in gallery

    Time–height diagram based on 12-hourly radiosonde ascents at Parakou, Benin, in (a) mid-September and (b) late July and early August 2002. Three extra soundings are also included: 0600 UTC 11 Sep, 0600 UTC 15 Sep, and 0600 UTC 18 Sep. Shading indicates relative humidity.

  • View in gallery

    As in Fig. 6 but for the parcel instability in °C (see text) and without the wind barbs.

  • View in gallery

    As in Fig. 3 but for the period between 1200 UTC 17 and 1200 UTC 18 September 2002.

  • View in gallery

    Time series of various surface meteorological parameters at locations in the UOV (see Fig. 1b) between 1200 UTC 29 and 1200 UTC 30 Jul 2002: (a) 15-min temperature at Djougou and Doguè (“Dog”) and specific humidity, q, at Djougou only; (b) maximum wind gusts within 10-min intervals and 10-min-averaged wind barbs at Parakou; (c) 15-min-averaged wind speed and barbs at Djougou (bars indicate 6-min accumulated rainfall); and (d) 15-min-averaged wind speed and barbs at Doguè.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 21 21 6
PDF Downloads 8 8 0

Three MCS Cases Occurring in Different Synoptic Environments in the Sub-Sahelian Wet Zone during the 2002 West African Monsoon

View More View Less
  • 1 Department of Atmospheric Sciences, Creighton University, Omaha, Nebraska
  • | 2 Institute of Geophysics and Meteorology, University of Cologne, Cologne, Germany
  • | 3 National Meteorological Service of Benin, Cotonou, Benin
© Get Permissions
Full access

Abstract

Three mesoscale convective systems (MCSs) occurring in the sub-Sahelian wet zone of West Africa are examined using observations from the 2002 Integrated Approach to the Efficient Management of Scarce Water Resources in West Africa (IMPETUS) field campaign, the European Centre for Medium-Range Weather Forecasts (ECMWF) operational analyses, and Meteosat infrared imagery. These datasets enable the analysis of the synoptic-scale environment in which the MCSs were embedded, along with a high-resolution monitoring of surface parameters during the systems’ passages. The available data imply that cases I and II were of a squall-type nature. Case I propagated into a moderately sheared and rather moist lower and middle troposphere over the Upper Ouémé Valley (UOV). In contrast, case II was associated with a well-sheared and dry lower troposphere and a large, moist instability. In either case, behind the convective cluster a westward-propagating cyclonic vorticity maximum that was likely captured by the ECMWF analysis as a result of the special upper-air station at Parakou (Benin). In case I, the fast-moving vorticity signal slowed down over the Guinean Highlands where convection dissipated. Farther downstream, it might have played a role in the consolidation of an African easterly waves (AEW) trough over the West African coast and the eastern Atlantic. Case III proved to be a more stationary pattern of convection associated with a vortex in the monsoon flow. It also exhibited a moist and low shear environment.

Corresponding author address: Jon M. Schrage, Dept. of Atmospheric Sciences, Creighton University, 2500 California Plaza, Omaha, NE 68178. Email: jon@creighton.edu

Abstract

Three mesoscale convective systems (MCSs) occurring in the sub-Sahelian wet zone of West Africa are examined using observations from the 2002 Integrated Approach to the Efficient Management of Scarce Water Resources in West Africa (IMPETUS) field campaign, the European Centre for Medium-Range Weather Forecasts (ECMWF) operational analyses, and Meteosat infrared imagery. These datasets enable the analysis of the synoptic-scale environment in which the MCSs were embedded, along with a high-resolution monitoring of surface parameters during the systems’ passages. The available data imply that cases I and II were of a squall-type nature. Case I propagated into a moderately sheared and rather moist lower and middle troposphere over the Upper Ouémé Valley (UOV). In contrast, case II was associated with a well-sheared and dry lower troposphere and a large, moist instability. In either case, behind the convective cluster a westward-propagating cyclonic vorticity maximum that was likely captured by the ECMWF analysis as a result of the special upper-air station at Parakou (Benin). In case I, the fast-moving vorticity signal slowed down over the Guinean Highlands where convection dissipated. Farther downstream, it might have played a role in the consolidation of an African easterly waves (AEW) trough over the West African coast and the eastern Atlantic. Case III proved to be a more stationary pattern of convection associated with a vortex in the monsoon flow. It also exhibited a moist and low shear environment.

Corresponding author address: Jon M. Schrage, Dept. of Atmospheric Sciences, Creighton University, 2500 California Plaza, Omaha, NE 68178. Email: jon@creighton.edu

1. Introduction

Tropical West Africa experiences several different precipitation regimes during the monsoon season. The relative importance of the different precipitation regimes is believed to vary with respect to latitude between the Guinea coastal zone (south of 9°N), the Soudanian climate zone (9°–12°N), and the Sahelian zone (12°–18°N). In the Sahelian zone, the major rain-bearing events are westward-propagating squall line systems that contribute about 80%–90% of the annual rainfall (Dhonneur 1981; Mathon et al. 2002a). Farther south, such systems supply about 50% of the rainfall in the Soudanian zone (Eldridge 1957; Omotosho 1985) and as little as 16%–32% along the Guinea Coast (Acheampong 1982; Omotosho 1985) where monsoonal rains, and local thunderstorms and showers, are increasingly important (Acheampong 1982; Omotosho 1985; Kamara 1986; Buckle 1996). Note that the distribution of different rainfall regimes during the 2002 monsoon season for the Soudanian zone is reviewed by Fink et al. (2006).

While a wide variety of mesoscale convective systems (MCSs) are important in the sub-Sahelian wet zone, highly organized squall line systems are perhaps the most extensively studied MCS type. The Global Atmospheric Research Program (GARP) Atlantic Tropical Experiment (GATE) brought about a considerable advancement in our understanding of the mesoscale structure of squall line systems that moved across the West African coast and passed over the GATE ship array (Zipser 1977; Houze and Betts 1981). Fortune (1980) was one of the few GATE convective case studies that described a purely land-based West African squall line system. He described a family of five squall lines, lasting an average of 6–12 h, that were embedded in a longer-lived, westward-propagating parent cloud cluster that lasted more than 48 h. Much of the current understanding about the kinematic, dynamic, and thermodynamic processes that maintain the land-based West African squall line systems is a result of the Convection Profonde Tropicale (COPT 81) field experiment in Ivory Coast (Sommeria and Testud 1984). The GATE and COPT 81 studies, as well as other works (e.g., Barnes and Sieckman 1984; Chong et al. 1987; Roux 1988; Peters and Tetzlaff 1988; Peters et al. 1989; Hodges and Thorncroft 1997; Thorncroft et al. 2003; Parker et al. 2004), postulated a need for dry air at middle levels and low-level wind shear between the lower and middle troposphere to maintain the mesoscale structure of these squall lines. While these environmental conditions are well-known factors favoring squall line genesis at other tropical or even midlatitude locations, West African squall lines seem to be noteworthy for the particularly large fraction of evaporative cooling below the 0°C-isotherm level in their trailing stratiform part (Chong and Hauser 1989). Chong et al. (1987) estimated that the mesoscale downdrafts in the stratiform part of the COPT squall line from 22 June 1981 contributed 60% to the near-surface horizontal density currents.

Westward-propagating, low-level African easterly waves (AEWs) are the dominant synoptic-scale features of the West African monsoon and the tropical Atlantic during boreal summer. Various studies (e.g., Reed et al. 1977; Duvel 1990; Mathon et al. 2002a; Fink and Reiner 2003) have shown that the region ahead of the AEW trough where northerly flow both advects dry air equatorward and increases vertical shear promotes squall line organization. A secondary preferred region of squall line development was ahead of the ridge, where southerly flow brought moist air poleward; this phase of the AEWs was most significant at the northern latitudes where moisture is scarce. Additionally, Fink and Reiner (2003) showed that the relationship between AEWs and squall line genesis was strong (weak) west (east) of the Greenwich meridian. In a recent study, Berry and Thorncroft (2005) showed evidence that the strong convection ahead of the AEW trough augmented the positive potential vorticity (PV) anomaly associated with an adiabatically growing AEW. They further argue that the merger of the AEW-related positive PV anomaly with a stationary PV pattern related to the orographic convection over the Guinean Highlands resulted in a further strengthening of the PV signal that later formed the seedling disturbance of Hurricane Alberto. Fortune (1980) noted a cyclonic vorticity disturbance in the stratiform part of the squall line system K. It was of the order of magnitude as those associated with the AEW trough. However, Fortune noted that the vorticity signal associated with squall line K progressed through the AEW trough region due to the about twice-as-fast propagation speed of the convective system.

While highly organized squall lines dominate the hydrology of the Sahel, the more southerly moist region is clearly characterized by a broader variety of MCSs with varying degrees of apparent organization. Moreover, dynamically rather than thermodynamically forced monsoonal rains, being prolonged periods of steady or intermittent, light to moderate rains from rainstorms with a low electrical activity, seem to be more important in the southerly wet zone. Buckle (1996, p. 169) relate them to vortices or lows in the southwesterly monsoon flow at 850 hPa. Studying these systems has traditionally been difficult due to the lack of high-quality upper air observations in the Soudanian climate zone (cf. Fig. 1). As will be shown in this study, the presence of a single special radiosonde station in Parakou, Benin, during the rainy season 2002 led to a substantial improvement in the representation of the low-level flow and moisture field in the European Centre for Medium-Range Weather Forecasts (ECMWF) analysis. This station was operated by the Integrated Approach to the Efficient Management of Scarce Water Resources in West Africa (IMPETUS) project. Additionally, a number of automated weather stations and a dense surface network of recording rain gauges was maintained in the Upper Ouémé Valley (UOV) in central Benin during this period by the French African Monsoon Multidisciplinary Analyses/Couplage de l’Atmosphère Tropicale et du Cycle Hydrologique (AMMA/CATCH) project and the German IMPETUS project.

Using ECMWF operational analyses and the aforementioned experimental data, the synoptic structure and evolution of three examples of MCSs in the sub-Sahelian latitudes of Benin are examined. A series of Meteosat infrared images for each of the three MCS cases studied are presented in Fig. 2 as an overview. A mesoscale analysis of the structure of these systems is not possible with these datasets. Rather, the present study aspires to demonstrate

  1. the wide variety of environmental synoptic conditions through which cases I–III (see Table 1) propagated in the wet climates of West Africa, focusing on the UOV region;
  2. the association of the squall-type cases I and II with a wake cyclonic vortex;
  3. the relation between the organized convective clusters, their wake vortices, and AEW wave disturbances at the African easterly jet (AEJ) level; and
  4. the characteristics of the vortex-type, nonsqually, case III rainfall during an episode of a nonsheared, moist, and strongly buoyant lower troposphere over the UOV.

Section 2 provides a summary of the data collected during the 2002 IMPETUS field campaign, as well as the description of other datasets and techniques employed in this study. Sections 3, 4, and 5 detail the observations and analyses of MCS cases I, II, and III, respectively. The results are summarized in section 6.

2. Computational and analysis procedures

a. Data collected during the 2002 IMPETUS/CATCH field campaign

The high-resolution vertical profiles of temperature, humidity, pressure, and wind that were collected from the IMPETUS radiosonde campaign at Parakou (9°21′N, 2°37′E) represent a unique dataset to study the thermodynamic vertical structure of the atmosphere before and after rainfall events in the Soudanian zone. These twice-daily (0000 and 1200 UTC) observations at 10-m vertical resolution were taken by the only operational radiosonde station in the Soudanian climate zone (Fig. 1a). For cases I and II, additional soundings at 0600 UTC were available after the MCS passages.

Precipitation events in the UOV region were observed using 49 recording rain gauges distributed unevenly over the catchment area of about 20 000 km2 (Fig. 1b). These surface precipitation stations were used to determine the local rate of propagation for the convection systems by producing isochrones of the time of onset of precipitation for each event. Fifteen-minute-averaged temperature, relative humidity, 2-m wind speed and direction, and rainfall data from two automated weather stations at Doguè and Djougou (Fig. 1b) were used to delineate the changes in these surface parameters during the MCS events. For the same purpose, hourly rainfall and 24-h pressure tendency observations were digitized from the records of the synoptic station at Parakou (Fig. 1b), where maximum gusts within 10-min intervals, as well as 10-min-averaged wind speed and direction, were also logged from an anemograph at an altitude of 9 m.

b. ECMWF data

Large-scale kinematic processes are examined in this study using analyses from the ECMWF operational model. The six-hourly ECMWF isobaric analyses of the meridional and zonal winds and the relative vorticity were employed at 1° × 1° resolution for the summer of 2002. While the fields were available at the 10 standard pressure levels up to 100 hPa, only the results for 700 hPa are represented in this paper, as they best illustrate the configuration of the waves in the AEJ. The data from the experimental radiosonde station at Parakou were transmitted in real time to the ECMWF. The ECMWF provided the authors with a comparison between the four-dimensional variational data assimilation (4DVAR) analysis and the observations, as well as analysis increments for each sounding. Thus, it was possible to assess the impact of each observation associated with cases I, II, and III on the analyzed wind, temperature, and humidity fields.

3. Case I (14/15 September 2002)

a. Surface observations

The convection in case I was particularly widespread, and statistics for precipitation in the UOV and infrared brightness temperature data are summarized for this and the other two cases in Table 1. During this period, all 45 operating stations in the UOV reported at least 1 mm of rainfall, with an average of 32.4 mm and a peak rainfall of 50.3 mm observed. The rainfall event lasted an average of 3.8 h, with one station reporting precipitation for approximately 5.3 h. Analysis of the times of onset of precipitation at the field stations indicates that the squall line had a local propagation speed of about 14.4 m s−1 as it crossed the UOV.

Changes in the high-resolution surface data were consistent with a squall-line nature of this convective system (Fig. 3). Temperatures in the UOV had fallen and stabilized near the dewpoint temperature (not shown) before the passage of the squall line. As the system passed each station, a marked temperature drop was observed—about 4°C at Parakou and about 3°C at Djougou and Doguè. The observed temperature drops are somewhat moderate because the squall line crossed the UOV at night. Behind the gust front (Figs. 3c,d,e), specific humidity values were markedly lower (Figs. 3a,b), despite the ongoing stratiform rain, consistent with the depiction presented by Chong and Hauser (1989). A time series of hourly 24-h pressure tendencies at Parakou (Fig. 3c) reveals the presence of a 1-hPa mesoscale high with the passage of the squall line. Other classic characteristics of squall line passage are readily noted in Fig. 3, including the rapid onset of heavy precipitation accompanied by gusty winds, a shift of surface winds out of the east as easterly momentum from the AEJ is transported to the surface by downdrafts, and a protracted period of stratiform precipitation behind the squall line (Figs. 3c,d,e). An independent verification of the squall line nature of the system was obtained by a Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) overpass downstream of the UOV over Ghana at 0745 UTC 15 September 2002; the arc-shaped leading edge was clearly visible in the precipitation intensity (not shown).

b. Synoptic environment

Case I originated over the Jos Plateau in Nigeria 1230 UTC 14 September 2002. As shown in Fig. 2, this convective system grew rapidly as it propagated westward for the next 18 h. West of the Greenwich meridian, the region of coldest cloud tops began to shrink as the convection propagated in a more west-southwesterly direction, ultimately dissipating over the Guinea Highlands near the coast at 0330 UTC 16 September 2002.

The relationship between the squall line and the troughs and ridges in the AEJ is suggested in a Hovmoeller diagram of relative vorticity averaged between 7° and 13°N (Fig. 4a), which indicates robust propagation of positive and negative vorticity features westward, especially downstream of the UOV. These vorticity signals, delineated by bold lines in Fig. 4a, travel westward at the typical speed of an AEW, between 6° and 8° longitude per day (Reed et al. 1977). Notice, however, that the phase speeds of the centers of gravity of the 233-K infrared brightness temperatures are considerably faster. The cloud tracking technique is described in Fink et al. (2006). Case I is seen to be initiated well to the east of a very broad trough in the AEJ. Since the phase speed of the clouds is considerably greater than that of the trough, the convection soon caught up with the peak relative vorticity values. The cyclonic vorticity maximum in the wake of the squall line (Fig. 4a) propagates faster than AEW vortices and suggests that it may be related to the production of midlevel cyclonic vorticity mainly in the region of stratiform rainfall (cf. Berry and Thorncroft 2005). Figure 4a further implies that, as soon as the convective cluster dissipates over the Guinean Highlands just west of 10°W, the associated vorticity signal seems to slow down and form the proper AEW signal over the eastern Atlantic Ocean.

The question arises to what extent the ECMWF analysis is able to capture the formation of a mesovortex in the wake of a squall line. The study of Berry and Thorncroft (2005) indicated the usefulness of the analyses, especially west of the central Nigerian Highlands, but also pointed to analysis deficiencies in the data-sparse regions of central North Africa (cf. Fig. 1a). We argue here that even the presence of a single radiosonde station in the Soudanian zone of West Africa has had a beneficial impact on the low-level wind, temperature, and moisture analyses during case I and II. Figure 5 displays streamlines and equivalent potential temperature at 850 hPa for 0000, 0600, and 1200 UTC 15 September 2002. Note that the case I system passed over the station just after the midnight sounding and that soundings were performed in the wake of the MCSs at 0600 and 1200 UTC. It is obvious from Figs. 5b and 5c that the post-MCS drying of the lower troposphere, as evident in the skew T–log p diagrams (not shown), caused a larger-scale strip of low values of equivalent potential temperature values behind the convective system. Moreover, the 4DVAR winds were corrected by the radiosonde winds to yield a northerly flow and cyclonic curvature to the east of the UOV (Fig. 5a) that crossed the region in the following 12 h (Figs. 5b and 5c).

c. Vertical environment

Ten days of radiosonde observations taken at Parakou are shown in Fig. 6. Prior to the onset of case I, quite high relative humidity values prevailed below 8 km due to three rainfall events on 12 and 13 September 2002 (not shown) and southerly winds on 14 September 2002. Moreover, the low-level wind shear was quite low due to the absence of a strong AEJ (Fig. 6a) and low-level values of equivalent potential temperature were not particularly high (not shown). This is noteworthy as strong squall lines are typically associated with dry air at midlevels and strong low-level wind shear (Barnes and Sieckman 1984; Roux 1988). Even though the midday sounding at 1200 UTC 14 September 2002 indicated a tendency toward higher AEJ wind speeds and midlevel drying, the conditions were less than ideal for the squall line as it crossed the UOV. In contrast, the system likely experienced dry conditions at 700 hPa during its genesis over the Jos Plateau, as inferred from the ECMWF operational analyses (not shown).

Figure 7 displays the evolution of the buoyancy in terms of positive and negative temperature differences of a pseudoadiabatically ascending parcel with respect to the ambient air. The starting temperature and humidity values of the parcel were averaged between the surface (i.e., the screen-level observation) and 925 hPa. The averaging over the lowest 400 m (or approximately 50 hPa) has been used in other studies (Barnes and Sieckman 1984; Roux 1988), ensuring more representative values. The pseudoadiabatic buoyancy development clearly shows that the parcels only exhibit a small temperature excess above the level of free convection (LFC) at about an altitude of 3 km in the 36 h leading to the squall line event (Fig. 7a). Below the LFC, an ascending parcel would have a considerable negative buoyancy. For example, the convective inhibition (CIN) is about 130 J kg−1 at 1200 UTC 14 September 2002. The post-squall sounding of 0600 UTC 15 September 2002 clearly displays the expected destruction of positive buoyancy by the moist convection. The rather low values of potential moist instability are related to the moist southeasterly flow (Fig. 6a) and daily rainfall events in the preceding days.

4. Case II (18 September 2002)

a. Surface observations

Since case II occurred merely 2–3 days after case I, this case is readily apparent in the time series and Hovmoeller diagram plotted in the previous section. The three days between case I and case II were almost dry in the UOV domain (not shown). A total of 43 of the 45 stations in the UOV reported precipitation during case II. An average of 19.8 mm fell over a mean duration of 2.4 h; a peak rainfall of 79.7 mm and a peak duration of 5.2 h were also observed (Table 1). Analysis of the times of the onset of precipitation at the field stations indicated that the squall line had a local propagation speed of about 17.2 m s−1 as it crossed the UOV.

The ground-truth observations from case II (Fig. 8) clearly suggest that the system was a well-defined West African squall line. Prior to the arrival of the squall line, local thunderstorm cells occurred over the UOV; however, these storms apparently did not moisten the environment sufficiently to preclude the possibility of a classic, West African squall line. At Parakou, a rapid temperature drop of 4°–5°C in the 2-h period between 0100 and 0300 UTC—which is quite significant for nocturnal observations—is observed (Fig. 8a). At Doguè, the temperature dropped in two steps since the rain from the local convection just after midnight was quite intensive (Fig. 8b). However, the strongest gusts in combination with a wind shift to the east were only observed at the time of the arrival of the second heavy convective rains at Parakou and Doguè (Figs. 8c and 8d). At these stations, the high-resolution observations clearly capture the fact that case II was followed by a larger region of stratiform precipitation. Additionally, a considerable pressure rise of about 2 hPa was observed in Parakou between about 0000 and 0600 UTC 18 September 2002 (Figs. 8c); this mesohigh is believed to develop under the stratiform region of a West African squall line due to a combination of hydrostatic and nonhydrostatic processes and is a separate feature from the nonhydrostatic mesohigh that typically is found associated with the gust front but was not captured in these observations. The surface observations alone seem to suggest that case II was a more vigorous squall line than case I. Case II was faster propagating, had stronger mean winds and gusts at Parakou and field sites (Figs. 8c,d,e), and had a stronger mesohigh at the synoptic station at Parakou.

b. Synoptic environment

As shown in Fig. 2, case II had a considerably smaller spatial extent and was shorter lived than case I. Like the first case, case II originated over the Jos Plateau. Starting at 1200 UTC 17 September 2002, the area of convection propagated almost due west and dissipated just west of the Greenwich meridian at 1400 UTC 18 September 2002. The 700-hPa streamline maps (not shown) and the Hovmoeller diagram of relative vorticity at the same level (Fig. 4a) reveal that this convection initially developed behind a small-amplitude trough in the easterlies. This trough is indicated by the dashed line in Fig. 4a. Whereas this trough weakens as the squall line approaches the UOV region (Fig. 4a), another trough and corresponding cyclonic vorticity maximum develops behind the convective system. Figure 4a again suggests that the vorticity signal in the wake of the squall line propagates farther downstream after termination of the convection and seems to interact with the formation of a broad vorticity signal that was later identified as an AEW (Fig. 4a).

c. Vertical environment

Referring back to Fig. 6a, the monsoon layer (as defined by a southerly wind component and high relative humidity) is found to be quite shallow before the onset of precipitation in case II, although within the monsoon layer equivalent potential temperatures were increasing (not shown). All tropospheric levels above 4 km had northerly winds starting at 1200 UTC 16 September 2002, resulting in a considerable drying of these layers and a well-developed low-level wind shear. The wind shift from east-northeast before to east-southeast after the convection is consistent with the passage of the observed small-amplitude trough behind the convection (Fig. 6a).

Contrary to case I, Fig. 7a shows that during the dry days preceding the MCS event, a build-up of strong positive midlevel buoyancy has been observed as a result of the aforementioned increase in the low-level equivalent potential temperature. At the same time, the low-level heat fluxes reduced the CIN near the surface. As the squall line arrived shortly after midnight on 18 September 2002, it found a highly unstable environment that also allowed for some individual cells to develop in the UOV before midnight of 19 September 2002. This notion is supported by the pseudoadiabatic convective available potential energy (CAPE) value of 1585 J kg−1, inferred for a parcel ascending from the surface to 925-hPa layer to the level of zero buoyancy (LZB) from the 0000 UTC 18 September 2002 sounding. Again, the post-squall sounding at 0600 UTC 18 September 2002 reveals the strong stabilization of the troposphere. Together with the surface observations and the faster local propagation speed, this strongly supports the idea that case II was a better-organized squall system than case I. Moreover, the arc-shaped leading edge in the Meteosat infrared satellite imagery is more distinct in comparison to case I.

5. Case III (29/30 July 2002)

a. Surface observations

Precipitation was considerably less widespread during case III, with only about 75% of the stations reporting precipitation (Table 1). However, case III had the highest observed rainfall totals of any of the three cases, with an average rainfall of 33.8 mm (at 32 stations that received measurable precipitation) and a peak rainfall of 116.7 mm. Average duration of the precipitation was 3.2 h, but the slow propagation of the system resulted in much longer durations of up to 7 h. The more stationary nature of this system also meant that it was impossible to determine a well-defined local propagation speed based on the network of pluviometers in the UOV.

Rains were widespread in the northern and northwestern part of the UOV during case III; therefore, the only automated weather station reporting rain was Djougou (cf. Fig. 1b). The surface observations at this station shown in Fig. 9 are not consistent with the classic depiction of a squall line passage. The time series of surface temperatures suggest only the diurnal cycle and not a temperature drop associated with the onset of precipitation, which was at about 2000 UTC for Djougou (Fig. 9a). Wind speeds remain quite low throughout the event, and surface winds are out of the west and south, rather than the east (Fig. 9c). Moreover, no sudden drop in specific humidity at Djougou was observed (Fig. 9a); rather, the specific humidity values slowly decreased with the duration of the rainfall. Very low wind speeds were also measured at Parakou and Doguè, where no rains were observed (Figs. 9b and 9d). After the rain stopped at Djougou, winds at Parakou turned to the northwest (Fig. 9b), indicating some cold air outflow from the convection to the northwest of the station. Similarly, the outflow from the rain to the north may have caused the lull in the monsoonal southerlies at Doguè (Fig. 9d).

b. Synoptic environment

Case III covered the smallest fraction of the UOV of the three cases examined in this study. As suggested by Fig. 2, this event also demonstrated the slowest propagation speed. An analysis of 200-hPa wind vectors (not shown) indicated that the tropical easterly jet (TEJ) was quite strong (i.e., greater than 30 m s−1) during case III. The core of the jet was found at about 4°N; therefore, the UOV was in a region of anticyclonic shear in the upper troposphere. The overall pattern was consistent with that presented for August 1992 by Redelsperger et al. (2002).

Case III was one of several nearly stationary or slow MCSs that crossed the UOV in late July and early August 2002. At that time, the 700-hPa relative vorticity field (Fig. 4b) averaged between 7° and 13°N shows little suggestion of systematic westward propagation of AEWs. Rather, the Soudanian climate zone of West Africa was dominated by a broad pattern of strong but stationary positive vorticity, indicating a standing cyclonic vortex. Such a quasi-stationary low- to midlevel flow pattern was typical for the vortex-type rainfall events in the UOV as classified in Fink et al. (2006).

c. Vertical environment

Ten days of radiosonde data from Parakou during late July and early August 2002 are shown in Fig. 6b. The most salient feature of the analysis is the long-lived and vertically extensive moist layer at this time. Relative humidity values greater than 60% are the norm below a height of 10 km throughout this period, and equivalent potential temperatures at 850 hPa rose substantially just before the onset of case III (not shown). Case III occurred in an episode of several days during which unusually deep westerly flow was temporarily observed in the UOV region (cf. Fink et al. 2006), the latter being related to the presence of a cyclonic vortex to the north of the UOV. Also apparent is the strong TEJ, whereas the AEJ is not seen on this plot; examination of the ECMWF analyses suggests that the AEJ was farther north than Parakou at this time.

The stability pattern in Parakou during case III (Fig. 7b) appears to have been altogether different than that calculated for either case I or case II. In marked contrast to the events that occurred in September 2002, case III was characterized by a deep layer of moderately strong positive buoyancy from an altitude of about 4 km to the tropopause. Simultaneously, the negative buoyancy of the lower atmosphere was significantly less than was computed for presquall environments in cases I and II. Notice, too, that the convection of case III as well as rainfall events during previous days does not appear to have consumed the convective instability to the same extent that cases I and II did. It is speculated that, in addition to the reduced discharge of convective energy, the recharge of moisture in the case III environment is faster due to a stronger moisture convergence (not shown).

It would appear that a vortex in the low-level monsoon flow was able to exploit the unusually low level of free convection (LFC) and large potential instability to produce copious rainfall. Aloft, the convection occurred in the exit region of a TEJ maximum, as was also observed by Redelsperger et al. (2002). Profiles and plan views of the divergence in the ECMWF operational analyses (not shown) failed to present a consistent depiction of the synoptic forcings created by this configuration of the jet. The dynamic and kinematic mechanisms supporting this favored region for convection remain unclear and require further research and improved datasets.

6. Summary and conclusions

Our analyses based on surface and sounding data suggest that cases I and II were both squall lines, the most prominent rainfall systems of the Soudano–Sahelian zone (Omotosho 1985; Mathon et al. 2002b; Fink et al. 2006). Both cloud systems developed over the central Nigerian Highlands, an area known to be a major genesis region of West African squall lines (Omotosho 1985; Rowell and Milford 1993; Fink and Reiner 2003). Case I experienced an atmosphere that was only moderately unstable with insufficiently dry conditions in the lower and middle troposphere. In contrast, case II propagated across the UOV in an environment that was considerably less stable and more conducive to production of strong downdrafts and robust outflow due to dry conditions at 700 hPa. Moreover, the low-level shear was also more pronounced during case II. Consistent with the local environmental conditions, case II was faster and exhibited stronger wind gusts and mesohighs. Case II, which was drier in the midtroposphere than case I, was found to propagate more rapidly across the UOV. This is consistent with the idea that dryness at middle levels of the atmosphere leads to more cooling by evaporation of raindrops and, therefore, stronger density currents near the surface. However, dryness at middle levels of the troposphere is one of many factors that hypothetically could influence the speed of propagation of West African squall lines.

The investigation of the relation between the squall lines of case I and case II and the 700-hPa flow revealed the existence of cyclonic vorticity maxima to the east of the convective cluster. These vorticity signals traveled at the speed of the squall lines (i.e., about 15 m s−1). After the termination of the squall systems these vorticity disturbances might have played a role in the consolidation of an AEW trough over the West African coast and the eastern Atlantic. In a recent study, Berry and Thorncroft (2005) investigated the interaction between a strong AEW and convection at the peak of the rainy season of 2002. They showed that moist convection was enhanced in the southern wet zone by southward advection of high low-level equivalent potential temperature related to the northerly AEW vortex. In turn, the production of positive PV anomalies by the deep convection enhanced the southerly AEW vortex. However, if squall lines and AEWs propagated westward at approximately the same speed, this configuration shown by Berry and Thorncroft (2005, their Fig. 10b) would likely be stable for very long periods of time. Various studies (e.g., Fortune 1980; Payne and McGarry 1977; Fink and Reiner 2003) have shown that typical squall lines propagate westward at speeds of nearly double those of AEWs. Thus, it seems reasonable to assume that the longer-living MCSs propagate through the AEW trough, limiting the time at which the convectively produced positive PV anomaly superimposes on the AEW PV anomaly. This is an avenue for further research.

An examination of records provided by the ECMWF showed that the data assimilated from the special radiosonde station at Parakou greatly improved the low-level wind field. The 4DVAR scheme generally accepted the winds and thermodynamic observations in the presquall environments. In the postsquall environments, some wind observations near the surface were rejected (presumably as not being representative of the large-scale flow); however, the moisture and temperature observations were assimilated into the analyses, improving the representation of the onion-shaped sounding often noted in postsquall observations (Zipser 1977). During the upcoming AMMA field campaign in 2006, the planned radiosonde stations in Tamale, Parakou, and Abuja—along with the new operational station at Cotonou—might help to identify the role of the low-level flow in triggering convective events in the southern wet zone and the evolution of southerly AEW vortices.

In contrast, case III was clearly not a squall line in the classic sense and occurred in a different kinematic and thermodynamic environment. An analysis of the thermodynamic and dynamic environment in which case III developed showed that the environment was not conducive to squall line development; the AEJ was too far north to provide the low-level shear, and the moist layer was deep. Despite this, case III actually produced the most precipitation (at stations that observed precipitation) over the UOV of the three cases investigated, probably due to the very high values of CAPE and the very slow propagation speed for the system. Moreover, the strong TEJ may have contributed to good outflow conditions aloft. Rainfall during such a weather pattern has been termed vortex-type rainfall by Fink et al. (2006) since the occurrences of intermittent deep westerlies were the consequence of a nearly stationary midlevel vortex to the north of the UOV. These vortices and their associated convection are frequent toward the Guinean coast at the height of the rainy season (Buckle 1996, p. 169). The synoptic structure and evolution of such vortices should be better understood after the AMMA field campaign in the summer of 2006, when an array of radiosonde stations will be operational across much of West Africa. The enhanced network was specifically designed to yield four-dimensional depictions of the low-level monsoon flow and the associated mass and moisture fluxes.

Moreover, like in the COPT 81 experiment, the present paper has not investigated the thermodynamic environments during the genesis and lysis periods of westward-propagating MCSs. This is partly related to the lack of upper-air data over central Nigeria and downstream of the UOV. In the course of the AMMA field campaign in 2006, additional upper-air stations will be installed, which will allow for a better analysis of the wind and moisture field in the wetter Soudano-Guinean climate zones of West Africa. In combination with other ground-based remote sensing instruments (e.g., radars, lidars, and lightning detection) and aircraft data, further case studies will be able to shed light on the environmental conditions that control MCS genesis, propagation speed, and maintenance, as well as modulate their rainfall efficiency. In addition, the interaction between convectively produced vorticity maxima and the growth and propagation of AEWs west of the Greenwich Meridian can be further studied. Finally, results from the AMMA 2006 field campaign can be compared with the IMPETUS 2002 experiment as the UOV is one of the three AMMA mesoscale sites in which enhanced observations will be taken in the West African rainy season of 2006.

Acknowledgments

This research is funded by the Federal German Ministry of Education and Research (BMBF) under Grant 07 GWK 02 and by the Ministry of Science and Research (MWF) of the federal state of North Rhine-Westphalia under Grant 223-21200200. The two grants support the IMPETUS project. We also acknowledge the support of the Graduate Dean at Creighton University for the support of the first author’s travel to Cologne. We are indebted to Susan Pohle and Peggy Reiner for graphics production. We are grateful to Christian Depraetere and Jean-Michel Bouchez from the Institute de Recherche pour le Development for providing us with the rainfall data and the climate observations from Djougou. Simone Giertz from the Institute of Geography of the University of Bonn kindly provided the climate data from the automated weather station near the village of Doguè. A special thanks is due to all members of the IMPETUS radiosonde team; namely, Michael Christoph, Andreas Weimer, and Elisabeth van den Akker. Without their engagement the successful campaign would not have been possible. The authors would also like to acknowledge the contributions from the staff of the synoptic station at Parakou who wrote down the hourly observations in a special spreadsheet. Two anonymous reviewers helped to greatly improve the manuscript.

REFERENCES

  • Acheampong, P. K., 1982: Rainfall anomaly along the coast of Ghana—Its nature and causes. Geogr. Ann., 64A , 199211.

  • Barnes, G. M., , and K. Sieckman, 1984: The environment of fast- and slow-moving tropical mesoscale convective cloud lines. Mon. Wea. Rev., 112 , 17821794.

    • Search Google Scholar
    • Export Citation
  • Berry, G. J., , and C. D. Thorncroft, 2005: Case study of an intense African Easterly Wave. Mon. Wea. Rev., 133 , 752766.

  • Buckle, C., 1996: Weather and Climate in Africa. Addison Wesley Longman Limited, 312 pp.

  • Chong, M., , and D. Hauser, 1989: A tropical squall line observed during the COPT 81 experiment in West Africa. Part II: Water budget. Mon. Wea. Rev., 117 , 728744.

    • Search Google Scholar
    • Export Citation
  • Chong, M., , P. Amayenc, , G. Scialom, , and J. Testud, 1987: A tropical squall line observed during the COPT 81 experiment in West Africa. Part I: Kinematic structure inferred from dual-Doppler radar data. Mon. Wea. Rev., 115 , 670694.

    • Search Google Scholar
    • Export Citation
  • Dhonneur, G., 1981: Les amas nuageux mobiles principale composante de la météorologie du Sahel. La Météorologie, 27 , 7582.

  • Duvel, J. P., 1990: Convection over tropical Africa and the Atlantic Ocean during northern summer. Part II: Modulation by easterly waves. Mon. Wea. Rev., 118 , 18551868.

    • Search Google Scholar
    • Export Citation
  • Eldridge, R. H., 1957: A synoptic study of West African disturbance lines. Quart. J. Roy. Meteor. Soc., 83 , 303314.

  • Fink, A. H., , and A. Reiner, 2003: Spatio-temporal variability of the relation between African easterly waves and West African squall lines in 1998 and 1999. J. Geophys. Res., 108 .4332, doi:10.1029/2002JD002816.

    • Search Google Scholar
    • Export Citation
  • Fink, A. H., , D. G. Vincent, , and V. Ermert, 2006: Rainfall types in the West African Sudanian zone during the summer monsoon 2002. Mon. Wea. Rev., 134 , 21432164.

    • Search Google Scholar
    • Export Citation
  • Fortune, M., 1980: Properties of African squall lines from time-lapse satellite imagery. Mon. Wea. Rev., 108 , 153168.

  • 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
  • Houze Jr., R. A., , and A. K. Betts, 1981: Convection in GATE. Rev. Geophys. Space Phys., 19 , 541576.

  • Kamara, I., 1986: The origins and types of rainfall in West Africa. Weather, 41 , 4856.

  • Mathon, V., , A. Diedhiou, , and H. Laurent, 2002a: Relationship between easterly waves and mesoscale convective systems over the Sahel. Geophys. Res. Lett., 29 .1216, doi:10.1029/2001GL014371.

    • Search Google Scholar
    • Export Citation
  • Mathon, V., , H. Laurent, , and T. Lebel, 2002b: Mesoscale convective system rainfall in the Sahel. J. Appl. Meteor., 41 , 10811092.

  • Omotosho, J. B., 1985: The separate contributions of squall lines, thunderstorms and the monsoon to the total rainfall in Nigeria. J. Climatol., 5 , 543552.

    • Search Google Scholar
    • Export Citation
  • Parker, D. E., , L. V. Alexander, , and J. Kennedy, 2004: Global and regional climate in 2003. Weather, 59 , 145152.

  • Payne, S. W., , and M. M. McGarry, 1977: The relationship of satellite inferred convective activity to easterly waves over West Africa and the adjacent ocean during phase III of GATE. Mon. Wea. Rev., 105 , 413420.

    • Search Google Scholar
    • Export Citation
  • Peters, M., , and G. Tetzlaff, 1988: The structure of West African squall lines and their environmental moisture budget. Meteor. Atmos. Phys., 39 , 7484.

    • Search Google Scholar
    • Export Citation
  • Peters, M., , G. Tetzlaff, , and W. Janssen, 1989: Rainfall intensity of West African squall lines. Ann. Geophys., 7 , 227238.

  • Redelsperger, J. L., , A. Diongue, , A. Diedhiou, , J. P. Ceron, , M. Diop, , J. F. Gueremy, , and J. P. Lafore, 2002: Multi-scale description of a Sahelian synoptic weather system representative of the West African monsoon. Quart. J. Roy. Meteor. Soc., 128 , 12291257.

    • Search Google Scholar
    • Export Citation
  • Reed, R. J., , D. C. Norquist, , and E. E. Recker, 1977: The structure and properties of African wave disturbances observed during Phase III of GATE. Mon. Wea. Rev., 105 , 317333.

    • Search Google Scholar
    • Export Citation
  • Roux, F., 1988: The West African squall line observed on 23 June 1981 during COPT 81: Kinematics and thermodynamics of the convective region. J. Atmos. Sci., 45 , 406426.

    • Search Google Scholar
    • Export Citation
  • Rowell, D. P., , and J. R. Milford, 1993: On the generation of African squall lines. J. Climate, 6 , 11811193.

  • Sommeria, G., , and J. Testud, 1984: COPT 81: A field experiment designed for the study of dynamics and electrical activity of deep convection in continental tropical regions. Bull. Amer. Meteor. Soc., 65 , 410.

    • Search Google Scholar
    • Export Citation
  • Thorncroft, C. D., and Coauthors, 2003: The JET2000 project—Aircraft observations of the African easterly jet and African easterly waves. Bull. Amer. Meteor. Soc., 84 , 337351.

    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., 1977: Mesoscale and convective-scale downdrafts as distinct components of squall-line structure. Mon. Wea. Rev., 105 , 15681589.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

(a) Geography of West Africa, with locations of radiosonde stations indicated (Aga: Agadez; Bam: Bamako; Ban: Bangui; Lib: Libreville; Nia: Niamey; Ndj: Ndjamena; Oua: Ouagadougou; Oue: Ouesso; Par: Parakou). Note the location of the Upper Ouémé River valley. The shading shows terrain height above 400, 800, and 1500 m, respectively. (b) Geography of the Upper Ouémé River valley, with dots indicating the locations of pluviograph sites. The box indicates the domain of the Upper Ouémé River valley (i.e., 9°–10.3°N, 1.5°–3°N) used in processing Meteosat data. Topography is indicated in meters.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 2.
Fig. 2.

Meteosat infrared imagery every 3 h from 0° to 15°N for (a) case I, (b) case II, and (c) case III. At times when the 0000 UTC image was not available, it has been replaced by the 2300 UTC image. Cases I, II, and III are highlighted, as are the longitudes of the UOV domain.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 3.
Fig. 3.

Time series of various surface meteorological parameters at locations in the UOV (see Fig. 1b) between 2100 UTC 14 and 1200 UTC 15 Sep 2002. (a) Hourly temperature and specific humidity, q, observed at Parakou. (b) The 15-min temperature at Djougou and Doguè (“Dog”) and specific humidity, q, at Djougou only. (c) Maximum wind gusts within 10-min intervals and 10-min-averaged wind barbs, and hourly 24-h pressure tendency at Parakou. The bars indicate 6-min accumulated rainfall. (d), (e) The 15-min averaged wind speed and barbs at Djougou and at Doguè, respectively; the bars indicate 6-min accumulated rainfall.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 4.
Fig. 4.

Relative vorticity at 700-hPa averaged from 7° to 13°N (contour interval is 10 × 10−6 s−1, positive vorticity areas are shaded) and cloud tracks (white circles) for (a) mid-September 2002 and (b) late July and early August 2002. The longitudes of the UOV are indicated. In addition, the approximate propagation of AEW cyclonic vorticity signals are delineated by the bold straight lines in (a). The dashed line in the same panel indicates the weak downstream trough for case II.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 5.
Fig. 5.

Streamlines and equivalent potential temperatures in °C at 850 hPa (a) for 0000 UTC, (b) for 0600 UTC, and (c) for 1200 UTC.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 6.
Fig. 6.

Time–height diagram based on 12-hourly radiosonde ascents at Parakou, Benin, in (a) mid-September and (b) late July and early August 2002. Three extra soundings are also included: 0600 UTC 11 Sep, 0600 UTC 15 Sep, and 0600 UTC 18 Sep. Shading indicates relative humidity.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 7.
Fig. 7.

As in Fig. 6 but for the parcel instability in °C (see text) and without the wind barbs.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 8.
Fig. 8.

As in Fig. 3 but for the period between 1200 UTC 17 and 1200 UTC 18 September 2002.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Fig. 9.
Fig. 9.

Time series of various surface meteorological parameters at locations in the UOV (see Fig. 1b) between 1200 UTC 29 and 1200 UTC 30 Jul 2002: (a) 15-min temperature at Djougou and Doguè (“Dog”) and specific humidity, q, at Djougou only; (b) maximum wind gusts within 10-min intervals and 10-min-averaged wind barbs at Parakou; (c) 15-min-averaged wind speed and barbs at Djougou (bars indicate 6-min accumulated rainfall); and (d) 15-min-averaged wind speed and barbs at Doguè.

Citation: Journal of the Atmospheric Sciences 63, 9; 10.1175/JAS3757.1

Table 1.

Selected statistics describing rainfall and cloud coverage in the UOV for each of the three convective systems.

Table 1.
Save