• Blamey, R. C., , and C. J. C. Reason, 2009: Numerical simulation of a mesoscale convective system over the east coast of South Africa. Tellus, 61A , 1734.

    • Search Google Scholar
    • Export Citation
  • Cammas, J-P., , D. Keyser, , G. M. Lackmann, , and J. Molinari, 1987: Diabatic redistribution of potential vorticity accompanying the development of an outflow jet within a strong extratropical cyclone. Preprints, Int. Symp. on the Life Cycles of Extratropical Cyclones, Vol. 2, Bergen, Norway, Amer. Meteor. Soc., 403–409.

    • Search Google Scholar
    • Export Citation
  • Colberg, F., , C. J. C. Reason, , and K. Rodgers, 2004: South Atlantic response to El Niño–Southern Oscillation induced climate variability in an ocean general circulation model. J. Geophys. Res., 109 , C12015. doi:10.1029/2004JC002301.

    • Search Google Scholar
    • Export Citation
  • Cook, C., , C. J. C. Reason, , and B. C. Hewitson, 2004: Wet and dry spells with a particularly wet and dry summers in the South African summer rainfall region. Climate Res., 26 , 1731.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., 2000: The South Indian convergence zone and interannual rainfall variability over southern Africa. J. Climate, 13 , 37893804.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and J. Lindesay, 1993: Water vapour transport over southern Africa during wet and dry early and late summer months. Int. J. Climatol., 13 , 151170.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and P. D. Tyson, 1995: Divergent and non-divergent water vapour transport over southern Africa during wet and dry conditions. Meteor. Atmos. Phys., 55 , 4759.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and P. D. Tyson, 1996: Three-dimensional kinematic trajectory modelling of water vapour transport over Southern Africa. Water S. Afr., 22 , 297306.

    • Search Google Scholar
    • Export Citation
  • De Coning, E., , G. S. Forbes, , and E. P. Poolman, 1998: Heavy precipitation and flooding on 12–14 February 1996 over the summer rainfall regions of South Africa: Synoptic and isentropic analyses. Natl. Wea. Dig., 22 , 2536.

    • Search Google Scholar
    • Export Citation
  • Emmanuel, K. A., , M. Fantini, , and A. J. Thorpe, 1987: Baroclinic instability in an environment of small stability to slantwise moist convection. Part I: Two-dimensional models. J. Atmos. Sci., 44 , 15591573.

    • Search Google Scholar
    • Export Citation
  • Fauchereau, N., , B. Pohl, , C. Reason, , M. Rouault, , and Y. Richard, 2008: Recurrent daily OLR patterns in the Southern Africa/Southwest Indian Ocean region, implications for South African rainfall and teleconnections. Climate Dyn., 32 , 575591.

    • Search Google Scholar
    • Export Citation
  • Harrison, M. S. J., 1984: A generalized classification of South African rain-bearing synoptic systems. Int. J. Climatol., 4 , 547560.

  • Hermes, J. C., , and C. J. C. Reason, 2009: Variability in sea-surface temperature and winds in the tropical south-east Atlantic Ocean and regional rainfall relationships. Int. J. Climatol., 29 , 1121.

    • Search Google Scholar
    • Export Citation
  • Iskenderian, H., 1995: A 10-yr climatology of Northern Hemisphere tropical cloud plumes and their composite flow patterns. J. Climate, 8 , 16301637.

    • Search Google Scholar
    • Export Citation
  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S. K. Yang, , J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP-DOE AMIP II Re-Analysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2003: Tropical–extratropical interactions causing precipitation in northwest Africa: Statistical analysis and seasonal variations. Mon. Wea. Rev., 131 , 30693076.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2005: Tropical–extratropical interactions associated with an Atlantic tropical plume and subtropical jet streak. Mon. Wea. Rev., 133 , 27592776.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2007: Tropical-extratropical interactions related to upper-level troughs at low latitudes. Dyn. Atmos. Oceans, 43 , 3662.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , and J. E. Martin, 2005: Tropical plumes and extreme precipitation in subtropical and tropical West Africa. Quart. J. Roy. Meteor. Soc., 131 , 23372365.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , and J. E. Martin, 2007: The role of dynamic and diabatic processes in the generation of cut-off lows over northwest Africa. Meteor. Atmos. Phys., 96 , 319.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , A. Fink, , A. Reiner, , and P. Speth, 2003: Three late summer/early autumn cases of tropical–extratropical interactions causing precipitation in northwest Africa. Mon. Wea. Rev., 131 , 116135.

    • Search Google Scholar
    • Export Citation
  • Kuhnel, I., 1989: Tropical-extratropical cloudband climatology based on satellite data. Int. J. Climatol., 9 , 441463.

  • Liebmann, B., , and C. A. Smith, 1996: Description of complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77 , 12751277.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , G. N. Kiladis, , J. A. Marengo, , T. Ambrizzi, , and J. D. Glick, 1999: Submonthly convective variability over South America and the South Atlantic convergence zone. J. Climate, 12 , 19771991.

    • Search Google Scholar
    • Export Citation
  • Lindesay, J. A., , and M. R. Jury, 1991: Atmospheric circulation controls and characteristics of a flood event in central South Africa. Int. J. Climatol., 11 , 609627.

    • Search Google Scholar
    • Export Citation
  • Lynch, S., 2003: Development of a RASTER database of annual, monthly and daily rainfall for Southern Africa. Rep. 1156/1/03, WRC, 78 pp.

    • Search Google Scholar
    • Export Citation
  • Lyons, S. W., 1991: Origins of convective variability over equatorial Southern Africa during austral summer. J. Climate, 4 , 2339.

  • McGuirk, J. P., , A. H. Thompson, , and N. R. Smith, 1987: Moisture bursts over the tropical Pacific Ocean. Mon. Wea. Rev., 115 , 787798.

  • Mecikalski, J. R., , and G. J. Tripoli, 1998: Inertial available kinetic energy and the dynamics of tropical plume formation. Mon. Wea. Rev., 126 , 22002216.

    • Search Google Scholar
    • Export Citation
  • Mo, K. C., , and J. Nogues-Paegle, 2001: The Pacific–South American modes and their downstream effects. Int. J. Climatol., 21 , 12111229.

    • Search Google Scholar
    • Export Citation
  • Pohl, B., , N. Fauchereau, , Y. Richard, , M. Rouault, , and C. J. C. Reason, 2009: Interactions between synoptic, intraseasonal and interannual convective variability over Southern Africa. Climate Dyn., 33 , 10331050.

    • Search Google Scholar
    • Export Citation
  • Posselt, D. J., , and J. E. Martin, 2004: The effect of latent heat release on the evolution of a warm occluded thermal structure. Mon. Wea. Rev., 132 , 578599.

    • Search Google Scholar
    • Export Citation
  • Postel, G. A., , and M. H. Hitchman, 1999: Climatology of Rossby wave breaking along the subtropical tropopause. J. Atmos. Sci., 56 , 359373.

    • Search Google Scholar
    • Export Citation
  • Racz, Z., , and R. K. Smith, 1999: The dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 125 , 225252.

  • Reason, C. J. C., , and D. Jagadheesha, 2005: A model investigation of recent ENSO impacts over southern Africa. Meteor. Atmos. Phys., 89 , 181205.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., , W. Landman, , and W. Tennant, 2006: Seasonal to decadal prediction of southern African climate and its links with variability of the Atlantic Ocean. Bull. Amer. Meteor. Soc., 87 , 941955.

    • Search Google Scholar
    • Export Citation
  • Rouault, M., , P. Florenchie, , N. Fauchereau, , and C. J. C. Reason, 2003: South East tropical Atlantic warm events and southern African rainfall. Geophys. Res. Lett., 30 , 8009. doi:10.1029/2003GL014840.

    • Search Google Scholar
    • Export Citation
  • Shen, R., , J. F. Bresch, , and E. R. Reiter, 1986: Vertical interpolation of meteorological variables. Mon. Wea. Rev., 114 , 123134.

  • Todd, M., , and R. Washington, 1999: Circulation anomalies associated with tropical-temperate troughs in Southern Africa and the south west Indian Ocean. Climate Dyn., 15 , 937951.

    • Search Google Scholar
    • Export Citation
  • Todd, M., , R. Washington, , and P. I. Palmer, 2004: Water vapour transport associated with tropical-temperate trough systems over southern Africa and the southwest Indian Ocean. Int. J. Climatol., 24 , 555568.

    • Search Google Scholar
    • Export Citation
  • Vigaud, N., , Y. Richard, , M. Rouault, , and N. Fauchereau, 2009: Moisture transport between the South Atlantic Ocean and southern Africa: Relationships with summer rainfall and associated dynamics. Climate Dyn., 32 , 113123.

    • Search Google Scholar
    • Export Citation
  • Walker, N. D., , and J. A. Lindesay, 1989: Preliminary observations of oceanic influences on the February–March 1988 floods in central South Africa. S. Afr. J. Sci., 85 , 164169.

    • Search Google Scholar
    • Export Citation
  • Ziv, B., 2001: A subtropical rainstorm associated with a tropical plume over Africa and the Middle East. Theor. Appl. Climatol., 69 , 91102.

    • Search Google Scholar
    • Export Citation
  • Ziv, B., , and P. Alpert, 1994: Isobaric to isentropic interpolation errors and implication to potential vorticity analysis. J. Appl. Meteor., 33 , 694703.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Southern African political boundaries, WRC rainfall stations (dots), and key summer season synoptic features.

  • View in gallery

    24-h accumulated rainfall at WRC stations on (a) 1 Jan 1998, (b) 5 Jan 1998, and (c) 6 Jan 1998. (d) The contribution to NDJF season rainfall for the period 1–7 Jan 1998. The upper color scale denotes rainfall in mm for (a),(b), and (c), with the lower color scale denoting contribution % for (d).

  • View in gallery

    (left) Winds (vectors, m s−1) and convergence (contoured, 10−6 s−1) at 850 hPa and OLR (shaded, W m−2) on (a) 31 Dec 1997, (c) 1 Jan 1998, and (e) 2 Jan 1998. (right) Full winds at 250-hPa level (streamlines, m s−1), 250-hPa ageostrophic winds (vectors, >4 m s−1), and 250-hPa divergence (shaded, 10−6 s−1) on (b) 31 Dec 1997, (d) 1 Jan 1998, and (f) 2 Jan 1998. Positions of the Angola low (AL), low-level cyclonic disturbances (L), midlatitude cyclones (MLC), and upper-level troughs (T) are indicated.

  • View in gallery

    Moisture transports (vectors, g kg−1 s−1), moisture convergence (shaded, 10−8 g kg−1 s−1), and pressure (contours, hPa) at 309 K for (a) 31 Dec 1997, (b) 1 Jan 1998, (c) 2 Jan 1998 (case I) and (d) 5 Jan 1998, (e) 6 Jan 1998, and (f) 7 Jan 1998 (case II). Regions of moisture export off Africa and into the midlatitudes are noted as warm conveyor belts (WCBs).

  • View in gallery

    As in Fig. 3, but for (a),(b) 5 Jan 1998; (c),(d) 6 Jan 1998; and (e),(f) 7 Jan 1998 (case II).

  • View in gallery

    As in Fig. 3, but for (a),(b) 15 Dec 2007; (c),(d) 16 Dec 2007; and (e),(f) 17 Dec 2007 (case III).

  • View in gallery

    As in Fig. 4, but at 312 K on (a) 15 Dec 2007, (b) 16 Dec 2007, (c) 17 Dec 2007, and (d) 18 Dec 2007 (case III).

  • View in gallery

    TRMM-estimated accumulated 24-h rainfall (mm) and 700-hPa full omega (contoured every 0.04 Pa s−1) on (a) 16 Dec 2007 and (b) 17 Dec 2007 (case III).

  • View in gallery

    500-hPa NCEP2 omega (contoured every 0.04 Pa s−1) and percentage of NCEP2 omega explained by QG uplift (shaded, %) on (a) 1 Jan 1998, (b) 5 Jan 1998, (c) 6 Jan 1998, and (d) 16 Dec 2007.

  • View in gallery

    345-K PV (thin contours, PVU), PV advection [thick contours starting at +(−)3 PVU day−1 with positive advection dotted (negative advection solid)] and 345-K diabatic PV tendency (shaded, PVU day−1) on (a) 31 Dec 1997, (b) 1 Jan 1998, (c) 2 Jan 1998 (case I) and (d) 4 Jan 1998, (e) 6 Jan 1998, (f) 7 Jan 1998, (g) 8 Jan 1998, and (h) 9 Jan 1998 (case II). Vectors represent 345-K winds >20 m s−1. Note: Shading for (h) depicts OLR (W m−2, color scale as in Fig. 3).

  • View in gallery

    As in Fig. 10, but for (a) 14 Dec 2007, (b) 15 Dec 2007, (c) 16 Dec 2007, (d) 17 Dec 2007, (e) 18 Dec 2007, and (f) 19 Dec 2007 (case III). Note: Shading in (e) depicts OLR, as in Fig. 10h.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 318 318 24
PDF Downloads 275 275 29

Tropical–Extratropical Interactions over Southern Africa: Three Cases of Heavy Summer Season Rainfall

View More View Less
  • 1 Department of Oceanography, University of Cape Town, Cape Town, South Africa
© Get Permissions
Full access

Abstract

The synoptic evolution of three tropical–extratropical (TE) interactions, each responsible for extreme rainfall events over southern Africa, is discussed in detail. Along with the consideration of previously studied events, common features of these heavy rainfall producing tropical temperate troughs (TTTs) over southern Africa are discussed. It is found that 2 days prior to an event, northeasterly moisture transports across Botswana, set up by the Angola low, are diverted farther south into the semiarid region of subtropical southern Africa. The TTTs reach full maturity as a TE cloud band, rooted in the central subcontinent, which is triggered by upper-level divergence along the leading edge of an upper-tropospheric westerly wave trough. Convection and rainfall within the cloud band is supported by poleward moisture transports with subtropical air rising as it leaves the continent and joins the midlatitude westerly flow. It is shown that these systems fit within a theoretical framework describing similar TE interactions found globally.

Uplift forcing for the extreme rainfall of each event is investigated. Unsurprisingly, quasigeostrophic uplift is found to dominate in the midlatitudes with convective processes strongest in the subtropics. Rainfall in the semiarid interior of South Africa appears to be a result of quasigeostrophically triggered convection.

Investigation of TTT formation in the context of planetary waves shows that early development is sometimes associated with previous anticyclonic wave breaking south of the subcontinent, with full maturity of TTTs occurring as a potential vorticity trough approaches the continent from the west. Sensitivity to upstream wave perturbations and effects on anticyclonic wave breaking in the South Indian Ocean are also observed.

Corresponding author address: N. C. G. Hart, Department of Oceanography, University of Cape Town, Rondebosch 7701, Cape Town, South Africa. Email: neil.hart@uct.ac.za

Abstract

The synoptic evolution of three tropical–extratropical (TE) interactions, each responsible for extreme rainfall events over southern Africa, is discussed in detail. Along with the consideration of previously studied events, common features of these heavy rainfall producing tropical temperate troughs (TTTs) over southern Africa are discussed. It is found that 2 days prior to an event, northeasterly moisture transports across Botswana, set up by the Angola low, are diverted farther south into the semiarid region of subtropical southern Africa. The TTTs reach full maturity as a TE cloud band, rooted in the central subcontinent, which is triggered by upper-level divergence along the leading edge of an upper-tropospheric westerly wave trough. Convection and rainfall within the cloud band is supported by poleward moisture transports with subtropical air rising as it leaves the continent and joins the midlatitude westerly flow. It is shown that these systems fit within a theoretical framework describing similar TE interactions found globally.

Uplift forcing for the extreme rainfall of each event is investigated. Unsurprisingly, quasigeostrophic uplift is found to dominate in the midlatitudes with convective processes strongest in the subtropics. Rainfall in the semiarid interior of South Africa appears to be a result of quasigeostrophically triggered convection.

Investigation of TTT formation in the context of planetary waves shows that early development is sometimes associated with previous anticyclonic wave breaking south of the subcontinent, with full maturity of TTTs occurring as a potential vorticity trough approaches the continent from the west. Sensitivity to upstream wave perturbations and effects on anticyclonic wave breaking in the South Indian Ocean are also observed.

Corresponding author address: N. C. G. Hart, Department of Oceanography, University of Cape Town, Rondebosch 7701, Cape Town, South Africa. Email: neil.hart@uct.ac.za

1. Introduction

Cloud bands resulting from tropical–extratropical (TE) interactions are common features in many regions, typically presenting as an elongated region of cloudiness rooted in the tropics and extending poleward and eastward into the midlatitudes (e.g., Kuhnel 1989; Iskenderian 1995). These events typically involve the export of moisture and heat from the tropics to the midlatitudes, which may replace deep Hadley cell overturning during periods of weakened ITCZ activity (McGuirk et al. 1987).

On seasonal time scales, subtropical moisture convergence zones (CZs) can support persistent TE cloud band features and thus are often excluded from synoptic studies (e.g., McGuirk et al. 1987). The South Atlantic convergence zone (SACZ) and South Indian Ocean convergence zone (SICZ), however, exhibit high synoptic variability (e.g., Lyons 1991; Liebmann et al. 1999; Pohl et al. 2009). Work by Cook (2000) on the SICZ, which forms in austral summer over southern Africa and the southwest Indian Ocean, produced a conceptual model applicable to land-based convergence zones and their formation but did not address synoptic variability.

Over the southern African region of interest here, cloud bands are the major synoptic rainfall-producing weather system (Harrison 1984) during the summer, yet their dynamics and variability are not yet well understood. Regionally, these cloud bands are known as tropical temperate troughs (TTTs) and will be referred to as such for the rest of this study.

Globally, a wealth of studies of similar TE interactions exists. These regions include Australia, South America, the North Pacific, the North Atlantic, and North Africa, with many recent dynamical studies focusing on the latter two regions (Knippertz 2003; Knippertz et al. 2003; Knippertz 2005; Knippertz and Martin 2005). A full summary of this literature is not provided here but for an extensive overview of previous TE interaction case studies the reader is referred to Ziv (2001) and Knippertz et al. (2003).

Knippertz (2007, hereafter K07) synthesized the current body of knowledge regarding these interactions and developed a theoretical framework for their formation and evolution. The key feature of these phenomena is the presence of a low-latitude upper-tropospheric trough, which facilitates the TE interaction and resulting cloud band. Despite very similar events occurring over southern Africa, TTTs receive little mention in general TE cloud band studies. The present study seeks to address this gap in the literature, further motivated by the considerable importance of TTTs for rainfall in the semiarid central southern Africa.

The primary goal of this paper is to consider the dynamics common to the development, maturation, and decay of TTTs over southern Africa. Additionally, the study examines whether TTTs fit within the theoretical framework set out in K07. To help achieve these goals upper- and lower-tropospheric circulation and associated moisture transports are presented for three extreme, warm season rainfall events caused by TTTs over southern Africa.

Forcing factors for uplift that produces heavy rainfall are presented to gain insight into the dynamics driving the high precipitation rates associated with TE interactions. Three primary mechanisms have been identified as drivers of uplift: positive vorticity advection, adiabatic uplift, and diabatic processes (see section 5 and 6 in K07). Following Knippertz and Martin (2005), the quasigeostrophic (QG) omega is calculated to include the contributions of positive vorticity advection and adiabatic uplift to the heavy rainfall of the events presented here.

Finally, a brief investigation of upper-tropospheric potential vorticity (PV) evolution for each event is presented to get a perspective on the association of TTTs to planetary wave activity over the South Atlantic and South Indian Oceans, an aspect of TE interactions over southern Africa, which has not been previously explored.

Three TTT events that resulted in heavy rainfall over large parts of South Africa on 1 January 1998, 5–6 January 1998, and 16 December 2007 are examined. The data and methods are described in section 2. The synoptic evolution of these three events is presented in section 3 while section 4 addresses the forcing for uplift on extreme precipitation days. The large-scale context of TTT development is discussed in section 5 with common synoptic features and aspects addressing TTTs in light of the K07 theoretical framework mentioned in section 6. Climatological implications are considered in section 7 and conclusions are summarized in section 8.

2. Data and methods

Three extreme rainfall events were chosen through two selection procedures outlined below: case I: 1 January 1998; case II: 5–6 January 1998, and case III: 16 December 2007.

The South African Water Research Commission (WRC) daily rainfall dataset (Lynch 2003) comprising 7665 stations within South Africa (Fig. 1), only available for the period 1979–98, was used for cases I and II with Tropical Rainfall Monitoring Mission (TRMM) satellite daily accumulated rainfall covering the period 1998 to the present at 0.5° × 0.5° resolution used for case III. The November–February (NDJF) period, representative of the core wet summer season, was extracted to calculate extreme rainfall days in both rainfall datasets.

Extreme rainfall criteria to be met were the following: 1) daily rainfall be above the 90th percentile rainfall value for all rain days at that location (grid point or station) and 2) this criterion is valid for at least 10% of locations to ensure that the extreme rainfall had a large areal extent.

In Fauchereau et al. (2008), it was shown that the daily outgoing longwave radiation (OLR) variability over southern Africa is well defined by seven statistically distinct clusters, one of which represented TTT events producing continental rainfall. As a final filter for events of interest, only extreme rain days that matched days assigned to this cluster were used. Thus, a candidate pool of TTT events associated with heavy precipitation was created from which these three events were chosen.

The synoptic evolution of the three cases is investigated using the 2.5° × 2.5° National Centers for Environmental Prediction–Department of Energy (NCEP–DOE) Reanalysis II (NCEP2) dataset (Kanamitsu et al. 2002) with 17 pressure levels. These data are 6-hourly and were averaged to daily means to match the time resolution of the rainfall data. Daily OLR values from the dataset described in Liebmann and Smith (1996) were used as a proxy for deep convection.

Both upper-troposphere PV and low-level moisture transports are plotted on isentropic surfaces. Interpolation of the NCEP2 data to isentropic levels was performed in a two-step process. First, pressure was interpolated to 19 isentropic surfaces assuming a constant lapse rate between the pressure surfaces. This assumption used by Shen et al. (1986) was shown in Ziv and Alpert (1994) to be the method of choice when performing such an interpolation, especially when applied to obtain PV on an isentropic surface. Pressure on the isentropic surfaces was then used to interpolate all other atmospheric variables using polynomial interpolation, assuming variable ∝ (lnp)2 (for details see Shen et al. 1986).

The assumption that the potential temperature (θ) is conserved during motion of the air parcels implies that flow on isentropic surfaces is a truer representation of atmospheric motion than flow on pressure surfaces (see section 4.3 in Ziv 2001). This allows some inferences regarding air parcel trajectories to be made, especially when synoptic situations exhibit some stationarity for a few days.

3. Synoptics of TTT development

For ease of reference, key circulation features of the southern African summer pertinent to TE cloud bands are highlighted by Fig. 1. The Angola low is a semipermanent feature of tropical southern African circulation during summer (Reason et al. 2006), dominating the lower- to midtroposphere circulation. Often a weak surface high over southern Mozambique and the neighboring ocean helps enhance the pressure gradient across Botswana and Zimbabwe. This pressure gradient sets up strong low-level northeasterly flow across the central subcontinent, which then promotes the flow of tropical easterlies north of Madagascar deep into the subcontinent. Continental heat lows (Racz and Smith 1999) often form over the central Kalahari Desert helping to induce a weak cyclonic circulation, which can then divert the low-level northeasterlies farther south. Eastward ridging of the South Atlantic high (SAH) often induces onshore flow and coastal showers along the south coast and sometimes up the east coast. Northwestward extension of the South Indian high (SIH), coupled with a surface depression in the Mozambique channel can then produce an easterly wave flow into Mozambique and eastern South Africa.

In this scenario, TTTs may then occur across southern Africa and extend from the Angola low, southeast to a location over the South Indian Ocean that is typically just east of the ridging SAH pressure system and associated with a midlatitude westerly wave depression.

The term TTT is used to describe the full complement of synoptic features that relate to these TE interactions over southern Africa. These synoptic features however, will be referred to as TE features (e.g., TE cloud band).

a. Case I: 31 December 1997–2 January 1998

The 24-h accumulated rainfall for 1 January 1998 is presented in Fig. 2a. Upper (right panels) and lower-level (left panels) synoptic evolution for case I are presented in Fig. 3. Low-level moisture transports for this case appear in Figs. 4a–c, represented on the 309-K isentropic surface.

On 31 December 1997, key features of the developing TTT were well manifested. Low-level northeasterlies across Zimbabwe and Botswana were found on the southeastern flank of the well-formed Angola low, as evidenced by the 850-hPa divergence field (Fig. 3a).

The low OLR signature over the central subcontinent indicates that convection had already begun in the semiarid region, aided by upper-level divergence in the entrance zone of a moderate southwesterly jet region. The cause of this jet was likely deep convection over tropical Africa during previous days. This convection enhanced the upper-level anticyclone over the central subcontinent, leading to a stronger gradient wind over its southeastern flank, as suggested by the ageostrophic wind (Fig. 3b).

The low-level northeasterlies resulted in strong transports of tropical moisture converging in the semiarid subtropics, supporting the convection (Fig. 4a). The first indication of a TE link about to form was the weak warm moist conveyor transporting moisture off the southeastern African coast ahead of the approaching midlatitude wave.

By 1 January 1998, the TTT was fully developed with a band of low OLR values extending from tropical southern Africa into the southwest Indian Ocean (SWIO), terminating in a midlatitude cyclone near 45°S (Fig. 3c). The Angola low convergence maxima had intensified to the southeast, weakening the northeasterlies. Although not clear on the 850-hPa surface, the isentropic flow field clearly indicated a warm conveyor feeding moisture from subtropical Africa into the midlatitude depression over the SWIO (Fig. 4b).

The amplification of the westerly wave as warm subtropical air moved poleward and upward, increased the gradient wind effect, promoting a band of upper-level divergence along the leading edge of the wave. This amplification was further encouraged by the divergent response to the deep convection in the TE cloud band (Fig. 3d). These factors promoted the intensification of the jet core above the midlatitude cyclone to 65 m s−1.

Substantial precipitation had begun on 31 January, but it was the amplifying upper-level wave, strong moisture supply, and enhanced low-level convergence of 1 January 1998 that supported the heaviest rainfalls. Many stations throughout central northeastern South Africa recorded daily totals exceeding 50 mm (Fig. 2a). The heaviest precipitation was found in the mountainous region near Lesotho (28°S, 28°E) with some stations recording up to 80 mm.

On 2 January 1998 light rainfall (not shown) was still recorded for central South Africa. The TE cloud band (Fig. 3e) was still present, but contiguous flow from subtropical Africa into the midlatitudes was shut off by the eastward ridging of the SAH under the trailing edge of the upper-level wave (Fig. 3f).

Intensification of the Angola low shown in this case was evident in both the coherent low-level cyclonic circulation and strong convergence as well as in the strengthened northeasterly moisture transports across much of central subtropical southern Africa (Fig. 4c). This pattern was important for the next event, which occurred a few days later as discussed below.

b. Case II: 5–7 January 1998

Rainfall on 5 and 6 January 1998 is shown in Figs. 2b,c with the synoptic evolution of the event presented in Fig. 5. Near-surface moisture transports are displayed in Figs. 4d–f.

The development of this second event was promoted by a strong Angola low facilitating transport of moisture into the semiarid subtropical southern Africa during 2–4 January 1998. The stationarity of the synoptic situation during this period allows a tentative inference to be made from the isentropic plots that the moisture was primarily sourced from the tropical Indian Ocean and transported in a low-level easterly jet across the northern Madagascar (Fig. 4c). The tropical southeast Atlantic was a secondary source with moist onshore flow over Angola (Figs. 4c–e) during this build-up period.

On 5 January 1998, strong low-level northeasterlies were present on the southeastern flank of the intense Angola low as indicated by strong 850-hPa convergence (Fig. 5a). OLR indicated convection had already begun to occur in the southeastern interior of South Africa (Fig. 5a). This convection was promoted by upper-level divergence ahead of an upper-tropospheric trough, approaching from the west (Fig. 5b).

Moisture transport deep into subtropical southern Africa was likely due to the formation of a low-level west coast trough as suggested by both 850-hPa surface and 309-K surface circulation. The favorable large-scale synoptic situation and continued moisture supply resulted in the dry central to western South Africa experiencing station rainfall totals of up to 50 mm for 5 January 1998 (Fig. 2b).

By 6 January, a mature TTT had formed linking deep convection over much of southern Africa to the frontal cloud of a midlatitude cyclone in the SWIO (Fig. 5c). Although not visible at 850 hPa, substantial export of moisture into the midlevel extratropics was occurring along the axis of this cloud band (Fig. 4e).

Baroclinic wave growth, due to the warm, poleward flow in the lower troposphere rising into the temperate latitudes, had amplified the upper-level trough as it moved over southern Africa (Fig. 5d), increasing curvature in the upper-level flow. This wave growth intensified the upper-level divergence favoring convection and was further reinforced by the upper-level divergent response to convective activity already occurring in the cloud band. Heavy showers, supported by the intensified large-scale forcing, continued to fall over much of South Africa on 6 January, although generally farther east than the previous day (Fig. 2c).

Substantial precipitation was recorded on 7 January 1998 over the eastern subcontinent with intensified convection indicated by very low OLR east of 25°E (Fig. 5e). The TE cloud band was not as prominent as the previous day because of rapid westward propagation of the midlatitude frontal cloud; however, strong TE flow at lower-levels continued to support significant export of moisture from the tropics (Fig. 4f). The upper-tropospheric flow over southern Africa had become more zonal with upper-level divergence now confined to the subtropical subcontinent and the midlatitude cyclone in the SWIO.

Interestingly, a second OLR minimum (Figs. 5a,c,e; 25°S, 50°E), associated with a sharp upper-level trough over Madagascar, formed during 2–6 January 1998. This feature linked up to the midlatitude cyclone on 7 January 1998 to form a secondary TE interaction site southwest of Madagascar.

During the following 2 days (8–9 January 1998), a second higher-amplitude midlatitude trough approached from the west (not shown) reinvigorating the export of tropical moisture to the extratropics. As a result, heavy rainfall occurred over Mozambique and the SWIO. A particularly striking feature of this renewed TTT is that there was a well-defined cloud band present over both South America and the central South Indian Ocean. The comanifestation of South American and southern African cloud bands is addressed further in section 5.

c. Case III: 15–17 December 2007

Upper- and lower-level synoptic evolution is shown in Fig. 6 for 15, 16, and 17 December 2007 with satellite rainfall estimates for 16 and 17 December 1998 (displayed in Fig. 8). Unfortunately, WRC station data are not available after 1999. Moisture transports on the 312-K surface are presented in Fig. 7 for the period 15–18 December 2007.

On 15 December 2007, cyclonic circulation around the well-expressed Angola low resulted in the northwesterlies over Angola, converging with northeasterlies across Botswana. The latter extended into South Africa, where weak convective activity was beginning to occur (Fig. 6a). Divergence ahead of an approaching upper-level trough (Fig. 6b) appeared to have encouraged a cyclonic disturbance to form in the lower troposphere (30°S, 17°E) and promote onshore flow into the more southerly 850-hPa convergence maximum.

Moisture transport into the subtropical subcontinent was facilitated by the Botswanan northeasterlies, with at least some moisture entering from the east, below the 312-K level (not shown). A substantial contribution also seemed to come from central tropical Africa (Fig. 7a). A similar moisture transport field persisted into 16 December, allowing time for air parcels to follow trajectories implied by these daily mean moisture fields. Strong moisture convergence in the arid Kalahari Desert was collocated with a continental heat low, implied by a sharp drop in isentropes in a vertical cross section (not shown) of the feature.

By 16 December 2007, a band of cloud extended from tropical Africa through the subtropics into the frontal cloud of a midlatitude depression (Fig. 6c). Cyclonic flow around the Angola low was being diverted farther south into the strengthened cyclonic disturbance over South Africa.

Baroclinic growth, due to poleward flow of warm continental air in lower levels, deepened the upper-tropospheric trough and strengthened the upper-level divergence on its leading edge (Fig. 6d). Deep convection likely helped reinforce this divergence, further strengthening the near-surface depression.

Tropically sourced moisture was transported along the cloud band axis into a region of strong moisture convergence where flow was rising poleward, into midlevels, as it left the continent (Fig. 7b). This substantial supply of moist air supported the widespread heavy rainfall over much of southern Africa (Fig. 8a). Interestingly, there was a split in regions of heavy precipitation over South Africa; namely, a northeastern region with maximum rainfall in Mozambique, and a southern region maintained by strong vertical velocities below the upper-level divergence here.

By 17 December 2007, the cloud band gained the typical northwest–southeast orientation of cases I and II as the southerly cyclonic disturbance was incorporated into the westward-moving midlatitude cyclone (Fig. 6e). The Angola low retained its expression continuing to encourage flow of tropical air in the subtropics and supporting the deep convection over Zimbabwe.

The upper-level trough became positively tilted, with subgeostrophic flow near its axis and convection in the cloud band promoting a TE band of upper-level divergence (Fig. 6f). This large-scale situation favored continued heavy precipitation over Mozambique and the SWIO (Fig. 8b).

Tropical moisture supporting this rainfall pattern was still directed off the continent, rising into the midlevel troposphere above the SWIO, along a well-defined band of moisture convergence (Fig. 7c). By the following day, the moisture flow off the continent ceased as the midlatitude wave continued eastward, breaking the band of moisture convergence (Fig. 7d). Despite this situation, a weak expression of the TE cloud band remained visible on 18 December 2007.

Similar to case II, full maturation of this event during 17–18 December 2007 was also synchronous with weak TE cloud band development rooted in South America.

4. Forcing factors for precipitation

In this section, processes producing the uplift responsible for the heavy precipitation of these three events are explored. A simple calculation of topographic uplift using NCEP2 winds produced no useful result since the regional topography is poorly represented at such a coarse resolution. Similarly, lack of mesoscale data limited investigations into uplift due to convection. Thus, only large-scale synoptic forcing is assessed here.

In the QG framework, uplift can be produced by adiabatic temperature advection, which is generally stronger in the lower troposphere, and positive vorticity advection, which is primarily performed by upper-tropospheric troughs. The QG omega was calculated using the traditional omega equation [Eq. (1)] for a limited domain (7.5°–45°S, 7.5°–60°E) on the standard 17 pressure levels of the NCEP2 reanalysis:
i1520-0493-138-7-2608-e1

In solving Eq. (1), the QG omega ω was assumed to equal full omega w at the domain boundaries and f0 was calculated for 30°S. The equation was solved using the freely available Octave software’s built-in linear equation solver, which relies on the LAPACK libraries.

The Rossby number was calculated at each grid point for each day with R0 < 0.15 for the whole domain in days prior to and including the extreme rainfall days. Values only exceeded 0.3 in days following the events as the baroclinic waves amplified and wind speeds increased. Good spatial coherence of QG omega field with the full uplift field was found, which together with the low Rossby numbers, gives confidence to the following analysis.

The percentage of total 500-hPa upward motion attributable to QG forcing is displayed in Fig. 9. The full omega field is overlaid for reference. Since this investigation focuses on uplift forcing only, this percentage was only calculated where w < −0.04 Pa s−1.

On 1 January 1998, 50% of uplift in the region of strongest vertical velocities (Fig. 9a; 30°S, 32.5°E) appeared to be due to QG forcing. The QG forcing, however, was clearly remote from the core precipitation region with convection likely the dominant rain-producing process across the subcontinent. As expected from theory, QG forcing becomes more dominant in the midlatitudes, toward the poleward end of the cloud band.

The 5 January 1998 case (Fig. 9b) showed half the total uplift explained by ω in the main precipitation zone over the southwestern interior of South Africa. This result suggests that QG forcing may have acted as a trigger for convection in a region that had been primed with tropically sourced moisture (see section 3b).

On 6 January 1998 (Fig. 9c) much of the uplift was explained by QG processes, especially along the east of the core uplift region and extending poleward. Indeed much of the widespread precipitation on this day was to the east of this core uplift region, collocated with the dominant QG forcing.

On 16 December 2007, TRMM rainfall estimates had strong spatial coherence with the core uplift region (Fig. 8a). Results in Fig. 9d for the south coast zone of rainfall (25°E, Fig. 8) suggest the QG forcing was the primary precipitation driver, which matches well with the strong upslope flow on the 312-K surface in the region (Fig. 7b). This would be captured by the temperature advection term in Eq. (1) (second term, right-hand side). Rainfall in the north of the domain appeared to be more convectively driven with QG percentages less than 40%, implying a weakening role of QG forcing toward the tropics.

5. Planetary waves

The large-scale planetary wave aspects of TTT formation are discussed with the aid of PV maps, an analysis that, to the authors’ knowledge, has never been performed for southern African TE interaction events. Diabatic destruction of PV is calculated following the method described in Posselt and Martin (2004) based on latent heating rates calculated after Emmanuel et al. (1987) and applied to the PV context by Cammas et al. (1987). Positive values imply PV reduction by latent heating and, together with PV advection, provide useful information regarding the evolution of the PV field. To avoid confusion due to Southern Hemisphere PV being negative, strongly negative values of this property will still be referred to as high PV and weakly negative values as low PV, consistent with this property increasing poleward, as in the Northern Hemisphere. Similarly, positive values of diabatic PV tendency and advection are regarded as reducing potential vorticity.

a. Case I

The PV evolution for 31 December 1997 and 1, 2, and 4 January 1998 is presented in Figs. 10a–d.

On 31 December 1997, a PV trough passing south of Madagascar left a low-latitude streamer of marginally higher PV air (1 PVU) across subtropical southern Africa (Fig. 10a). This streamer likely encouraged a surface depression, initiating moisture transport deeper into the subtropics. A weak PV trough had begun to form near 20°E, amplifying into 1 January 1998. It was on the leading edge of this PV trough that the TE cloud band of case I occurred. The strongest (>2 PVU day−1) diabatic PV destruction (Fig. 10b, 75°E) of any of the cases presented here took place downstream of this trough in a PV ridge that rapidly amplified into the extratropics the following day (Fig. 10c).

Meanwhile, upstream of the trough, moderate diabatic PV destruction near 10°E (Fig. 10b) and southeastward advection of low PV air initiated an anticyclonic wave breaking event on 2 January 1998 (Fig. 10c) that ultimately resulted in the PV field observed on 4 January 1998 (Fig. 10d). The presence of the high PV air over the subcontinent in Fig. 10d would have favored a surface depression facilitating advection of tropical moisture into the subtropics.

b. Case II

The PV maps for 6–9 January 1998 are shown in Figs. 10e–h. The PV situation on 5 January 1998 was very similar to 4 January 1998 and is therefore not plotted.

A broad, long-wave PV trough to the east of southern Africa on 4 January 1998 (Fig. 10d) finally neared southern Africa on 6 January 1998 (Fig. 10e). The case II TTT event thus did have some association with the leading edge of a weak trough.

Upstream, in the southeast Atlantic, substantial PV advection (>3 PVU day−1) had begun to deform a lower-latitude PV trough (axis at 30°W), downstream of a confluence of a subtropical jet and polar jet over South America. This structure intensified into 7 January 1998 (Fig. 10f) with the PV gradients in the jet confluence (50°W) tightening and downstream diabatic PV destruction contributing to rapid ridging of low PV air into high latitudes as seen 8 January 1998 (Fig. 10g). The continued ridging contributed to strong equatorward advection of high PV air into a PV trough.

By 9 January 1998, the trough had reached southern Africa and the strong poleward flow on its leading edge reinvigorated the decaying TTT in the region (Fig. 10h). Downstream of Africa it appeared that the trough encouraged an anticyclonic wave breaking event over the South Indian Ocean in the following days (not shown).

Over South America, the subtropical jet had been displaced equatorward, leading to a TE cloud band linking tropical Brazil to the subtropical–polar jet confluence deep in the high-latitude South Atlantic.

c. Case III

The PV evolution for the period 14–19 December 2007 is displayed in Fig. 11. A low-amplitude anticyclonic wave breaking event occurred near southern Africa at about 40°S on 15 December 2007 (Fig. 11b) as the PV streamer present on 14 December 2007 (Fig. 11a) was separated from the high PV reservoir. The cutoff appeared to become incorporated into a PV trough on 16 December 2007 (Fig. 11c), the arrival of which signified the start of heavy rainfall over southern Africa.

This trough appeared to be the remnant of a trough involved in minor anticyclonic wave breaking near 20°W (Fig. 11a) on 14 December 2007, which propagated quickly east and began to reamplify as baroclinic growth occurred due to warm advection off the subcontinent.

By 17 December 2007 (Fig. 11d), the PV trough had gained a positive tilt as advection continued to deform the PV field south of Africa. The continued growth of the PV trough encouraged strong poleward flow of the TTT now positioned over the SWIO. Tilting of the trough continued into 18 December as anticyclonic wave breaking clearly began to occur (Fig. 11e). A clear TE cloud band on the trough’s leading edge was now concomitant with a similar structure emanating from South America.

On 19 December 2007 (Fig. 11f) the wave breaking event left a PV streamer extending from the south coast of southern Africa into the SWIO, potentially prolonging showers in the wake of the TTT.

6. Discussion

In this section, common features of the three case studies are discussed and compared to previous TTT studies and TE interaction globally.

The Angola low, a well-documented feature of southern African summer circulation (Reason et al. 2006), was well formed during the early stages of all of the events and appeared to weaken somewhat during their evolution. Its primary role in these cases was to facilitate moisture transport into subtropical southern Africa. A relatively small contribution of moisture from the eastern tropical South Atlantic Ocean, a source noted in (D’Abreton and Lindesay 1993; D’Abreton and Tyson 1995; Cook et al. 2004), was brought onshore by westerlies to the north of the Angola low in cases I and III. However, the primary moisture supply occurred in the low-level northeasterly flow on the southwest flank of the low. This flow conveyed moisture from the tropical Indian Ocean, a primary moisture source for southern African rainfall (Walker and Lindesay 1989; D’Abreton and Tyson 1995; Todd et al. 2004).

A secondary, low-level low pressure feature was present near the Namibia–South Africa border 2 days before extreme rainfall days in cases II and III, acting to divert moisture deeper into the subtropics. Upper-level divergence downstream of an upper-tropospheric trough was likely responsible for the development of the low-level cyclonic disturbance. In case III, a similar process occurred farther east. Isentropic surface depressions in early stages of all cases but especially case III indicated that continental heat lows were collocated with regions of intense moisture convergence in the semiarid Kalahari Desert.

In all three cases, TE cloud band formation was finally triggered by the arrival of an upper-level trough over southern Africa associated with band of divergence east of its leading edge. These cases corroborate the conclusion of previous studies regarding the importance of upper-tropospheric troughs to TTT development (e.g., Harrison 1984; Lindesay and Jury 1991; Lyons 1991). The poleward extremity of all three TTT events was associated with a midlatitude depression beneath the jet core, downstream of the trough.

Once the TTTs were fully developed, moisture was transported off the continent along the axis of a band of moisture convergence, rising from near 800 hPa over the landmass into midlevel flow (700–650 hPa) in the midlatitudes. This transport is consistent with trajectory analysis of a TTT event on 22 January 1981 (Fig. 4 in D’Abreton and Tyson 1996) and work on TE interactions over North Africa by Knippertz and Martin (2005).

All three cases exhibit the key ingredients common to the well studied tropical–extratropical interactions found in other regions of both the Northern Hemisphere and Southern Hemisphere, broadly termed tropical plumes (summarized in Fig. 11; K07). We argue that TTTs generally fall within the theoretical framework put forward in K07. A distinction that TTTs have from similar events in the North Pacific and North Atlantic is their formation over a landmass. Many oceanic “tropical plumes” discussed in K07 only have mid- to upper-level cloud through the subtropics. In the southern African context, however, continental heating can support deep convection and heavy rainfall along the full extent of the cloud band.

The results of the uplift forcing analysis compare well with the results of Knippertz and Martin (2005) that QG forcing plays a significant role in promoting the uplift observed in TE interactions. Nevertheless, the heaviest rainfall of these three events was recorded in case I, where QG forcing was weak, implying the role of convective overturning. Spatial inhomogeneities in the magnitude of the rainfall for all cases (e.g., the extreme precipitation over Mozambique during case III) suggests phenomenon such as mesoscale convective systems (e.g., Blamey and Reason 2009) may be embedded in TE cloud bands, at least over South Africa. Indeed, De Coning et al. (1998) found the tropical root of a TTT on 12 February 1996 to be a mesoscale convective vortex centered near the Angola low.

Attempts were made to investigate the role of inertial (and symmetric) instability in aiding development of the TTTs (Mecikalski and Tripoli 1998; Knippertz and Martin 2005); however, calculating these parameters from the 2.5° × 2.5° NCEP2 reanalysis produce no regions of instability, which is not unexpected, given the coarse resolution of the data and the mesoscale nature of instability regions.

A final aspect of these TE interactions over southern Africa is that they are clearly related to planetary wave structures that, although alluded to in Todd et al. (2004), have not been demonstrated before. Anticyclonic wave breaking a few days before each TTT event left a weak (1–2 PVU) PV streamer across the southern subcontinent, with this low PV air aloft, encouraging a surface low to initiated poleward moisture transport into subtropical southern Africa. Convection and cloud band formation were triggered when the leading edge of an amplifying PV trough neared the subcontinent’s western margin, promoted both by baroclinic wave growth near southern Africa and upstream deformation as low PV air was advected poleward. The wave breaking events in the South Atlantic agree well with research by Postel and Hitchman (1999).

In cases II and III, similar cloud band features evolved concurrently over South America, emphasizing these systems are results of large-scale planetary wave growth and decay. Indeed the TE interactions presented here appear sensitive to upstream perturbations in both the subtropical and the polar jet over South America. Downstream of southern Africa ridging of low PV air into the extratropics (case II) and anticyclonic wave breaking (cases I and II) suggested TTTs contribute to significantly South Indian Ocean planetary wave activity.

7. Climatological implications

The key role of the Angola low in TTT development in these three case studies provides support to previous climatological studies linking the strength of the Angola low to rainfall variability over southern Africa on the seasonal to interannual time scale (Rouault et al. 2003; Cook et al. 2004; Reason and Jagadheesha 2005; Reason et al. 2006). The weak moisture transports from the South Atlantic associated with the Angola low also corroborates recent work by Hermes and Reason (2009) and Vigaud et al. (2009).

An interesting feature of cases II and III was their development over the central subcontinent, but full maturation over Mozambique and the southwest Indian Ocean. Composite studies (Todd and Washington 1999) and climatology (Cook 2000) do not capture this behavior and indicate only that the dominant position of the systems lies over the southwest Indian Ocean, understandable with this being the region in which the systems manifest most strongly. The current study reveals the importance of resolving these systems on the synoptic scale to capture the development stages of TTTs over the continent.

The development of these three cases over the subcontinent and their eastward propagation agrees well with the results of Fauchereau et al. (2008). Their third cluster looks very similar to the early stages of the three cases presented here and, although only 15% of the days following it are assigned to a cluster describing TTTs, we hypothesize that this 15% corresponds specifically to extreme rainfall events over semiarid central South Africa. Further investigations would be needed to confirm this hypothesis.

Although TTTs are recognized as the primary rain producing system in the South African summer season, little work has been done to link them directly to wet spells. The quick succession of cases I and II produced a 7-day wet spell during which more than 40% of the 1997/98 NDJF season’s rainfall was recorded over much of South Africa (Fig. 2d). This contribution is particularly notable since South Africa received near-average rainfall during this season, despite a strong El Niño event, which often causes drought.

The influence of large-scale modes that produce significant mean Rossby wave perturbations, such as the Pacific–South America pattern (Mo and Nogues-Paegle 2001; Colberg et al. 2004) and the El Niño–Southern Oscillation extratropical response over the southern African region remain poorly understood. The clear link between synoptic-scale PV wave activity and TTTs demonstrated here is therefore an important point to consider in studies looking for teleconnections between subtropical southern Africa and the Pacific Ocean. It further highlights the need for scale-interaction studies linking synoptic-scale rainfall to season total precipitation.

8. Conclusions

Investigation of the three heavy precipitation producing TTT events over southern Africa was performed using the NCEP2 gridded dataset (Kanamitsu et al. 2002), allowing a comprehensive study of the synoptic evolution of each system. Comparison of these three events revealed clear similarities with one another and with previous TTT case studies. We conclude that TTTs, despite some minor differences, are similar to TE interactions observed elsewhere and do fit the theoretical framework developed by K07.

Uplift forcing analysis of the vertical velocities and associated heavy rainfall of the three events presented here indicates adiabatic temperature advection and positive vorticity advection play a significant role in producing uplift in TTTs. It is clear, however, that local land-based convection is as important, especially in the subtropical portion of the cloud band. A high-resolution mesoscale model would be needed, however, to more fully assess this contribution and capture the details of the precipitation field. The fact that heavy rainfall is produced along the full length of these TTT cases highlights a key difference between TE interactions occurring over landmasses or oceans.

The importance of planetary waves for TTT development is demonstrated, motivating for the necessity of a more detailed investigation of this aspect of TTTs. The key finding of this study is that TTTs are embedded within planetary wave structures, are influenced by upstream perturbations, and in turn, influence wave activity in the South Indian Ocean. This result highlights the South Atlantic and South Indian Ocean as regions where vigorous planetary wave activity influences lower-latitude cloudiness associated with TE interactions, similar to processes occurring in the North Atlantic context as discussed in Knippertz and Martin (2007).

In conclusion, this study has identified the key components necessary for TTT development, providing insight to further studies investigating regional rainfall variability related to TTT formation over southern Africa. Additionally, the study has discussed TTTs in context to TE interactions that occur globally.

Acknowledgments

The authors thank Michael Mehari for preprocessing TRMM satellite data and Marlan Perumal for assistance with the calculation of QG omega. Ross Blamey and two anonymous reviewers are thanked for comments that greatly improved this study. This paper is an extension of the first author’s masters work funded by SANAP. Interpolated OLR and NCEP2 data were provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado (see online at http://www.cdc.noaa.gov/).

REFERENCES

  • Blamey, R. C., , and C. J. C. Reason, 2009: Numerical simulation of a mesoscale convective system over the east coast of South Africa. Tellus, 61A , 1734.

    • Search Google Scholar
    • Export Citation
  • Cammas, J-P., , D. Keyser, , G. M. Lackmann, , and J. Molinari, 1987: Diabatic redistribution of potential vorticity accompanying the development of an outflow jet within a strong extratropical cyclone. Preprints, Int. Symp. on the Life Cycles of Extratropical Cyclones, Vol. 2, Bergen, Norway, Amer. Meteor. Soc., 403–409.

    • Search Google Scholar
    • Export Citation
  • Colberg, F., , C. J. C. Reason, , and K. Rodgers, 2004: South Atlantic response to El Niño–Southern Oscillation induced climate variability in an ocean general circulation model. J. Geophys. Res., 109 , C12015. doi:10.1029/2004JC002301.

    • Search Google Scholar
    • Export Citation
  • Cook, C., , C. J. C. Reason, , and B. C. Hewitson, 2004: Wet and dry spells with a particularly wet and dry summers in the South African summer rainfall region. Climate Res., 26 , 1731.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., 2000: The South Indian convergence zone and interannual rainfall variability over southern Africa. J. Climate, 13 , 37893804.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and J. Lindesay, 1993: Water vapour transport over southern Africa during wet and dry early and late summer months. Int. J. Climatol., 13 , 151170.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and P. D. Tyson, 1995: Divergent and non-divergent water vapour transport over southern Africa during wet and dry conditions. Meteor. Atmos. Phys., 55 , 4759.

    • Search Google Scholar
    • Export Citation
  • D’Abreton, P. C., , and P. D. Tyson, 1996: Three-dimensional kinematic trajectory modelling of water vapour transport over Southern Africa. Water S. Afr., 22 , 297306.

    • Search Google Scholar
    • Export Citation
  • De Coning, E., , G. S. Forbes, , and E. P. Poolman, 1998: Heavy precipitation and flooding on 12–14 February 1996 over the summer rainfall regions of South Africa: Synoptic and isentropic analyses. Natl. Wea. Dig., 22 , 2536.

    • Search Google Scholar
    • Export Citation
  • Emmanuel, K. A., , M. Fantini, , and A. J. Thorpe, 1987: Baroclinic instability in an environment of small stability to slantwise moist convection. Part I: Two-dimensional models. J. Atmos. Sci., 44 , 15591573.

    • Search Google Scholar
    • Export Citation
  • Fauchereau, N., , B. Pohl, , C. Reason, , M. Rouault, , and Y. Richard, 2008: Recurrent daily OLR patterns in the Southern Africa/Southwest Indian Ocean region, implications for South African rainfall and teleconnections. Climate Dyn., 32 , 575591.

    • Search Google Scholar
    • Export Citation
  • Harrison, M. S. J., 1984: A generalized classification of South African rain-bearing synoptic systems. Int. J. Climatol., 4 , 547560.

  • Hermes, J. C., , and C. J. C. Reason, 2009: Variability in sea-surface temperature and winds in the tropical south-east Atlantic Ocean and regional rainfall relationships. Int. J. Climatol., 29 , 1121.

    • Search Google Scholar
    • Export Citation
  • Iskenderian, H., 1995: A 10-yr climatology of Northern Hemisphere tropical cloud plumes and their composite flow patterns. J. Climate, 8 , 16301637.

    • Search Google Scholar
    • Export Citation
  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S. K. Yang, , J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP-DOE AMIP II Re-Analysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2003: Tropical–extratropical interactions causing precipitation in northwest Africa: Statistical analysis and seasonal variations. Mon. Wea. Rev., 131 , 30693076.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2005: Tropical–extratropical interactions associated with an Atlantic tropical plume and subtropical jet streak. Mon. Wea. Rev., 133 , 27592776.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., 2007: Tropical-extratropical interactions related to upper-level troughs at low latitudes. Dyn. Atmos. Oceans, 43 , 3662.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , and J. E. Martin, 2005: Tropical plumes and extreme precipitation in subtropical and tropical West Africa. Quart. J. Roy. Meteor. Soc., 131 , 23372365.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , and J. E. Martin, 2007: The role of dynamic and diabatic processes in the generation of cut-off lows over northwest Africa. Meteor. Atmos. Phys., 96 , 319.

    • Search Google Scholar
    • Export Citation
  • Knippertz, P., , A. Fink, , A. Reiner, , and P. Speth, 2003: Three late summer/early autumn cases of tropical–extratropical interactions causing precipitation in northwest Africa. Mon. Wea. Rev., 131 , 116135.

    • Search Google Scholar
    • Export Citation
  • Kuhnel, I., 1989: Tropical-extratropical cloudband climatology based on satellite data. Int. J. Climatol., 9 , 441463.

  • Liebmann, B., , and C. A. Smith, 1996: Description of complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77 , 12751277.

    • Search Google Scholar
    • Export Citation
  • Liebmann, B., , G. N. Kiladis, , J. A. Marengo, , T. Ambrizzi, , and J. D. Glick, 1999: Submonthly convective variability over South America and the South Atlantic convergence zone. J. Climate, 12 , 19771991.

    • Search Google Scholar
    • Export Citation
  • Lindesay, J. A., , and M. R. Jury, 1991: Atmospheric circulation controls and characteristics of a flood event in central South Africa. Int. J. Climatol., 11 , 609627.

    • Search Google Scholar
    • Export Citation
  • Lynch, S., 2003: Development of a RASTER database of annual, monthly and daily rainfall for Southern Africa. Rep. 1156/1/03, WRC, 78 pp.

    • Search Google Scholar
    • Export Citation
  • Lyons, S. W., 1991: Origins of convective variability over equatorial Southern Africa during austral summer. J. Climate, 4 , 2339.

  • McGuirk, J. P., , A. H. Thompson, , and N. R. Smith, 1987: Moisture bursts over the tropical Pacific Ocean. Mon. Wea. Rev., 115 , 787798.

  • Mecikalski, J. R., , and G. J. Tripoli, 1998: Inertial available kinetic energy and the dynamics of tropical plume formation. Mon. Wea. Rev., 126 , 22002216.

    • Search Google Scholar
    • Export Citation
  • Mo, K. C., , and J. Nogues-Paegle, 2001: The Pacific–South American modes and their downstream effects. Int. J. Climatol., 21 , 12111229.

    • Search Google Scholar
    • Export Citation
  • Pohl, B., , N. Fauchereau, , Y. Richard, , M. Rouault, , and C. J. C. Reason, 2009: Interactions between synoptic, intraseasonal and interannual convective variability over Southern Africa. Climate Dyn., 33 , 10331050.

    • Search Google Scholar
    • Export Citation
  • Posselt, D. J., , and J. E. Martin, 2004: The effect of latent heat release on the evolution of a warm occluded thermal structure. Mon. Wea. Rev., 132 , 578599.

    • Search Google Scholar
    • Export Citation
  • Postel, G. A., , and M. H. Hitchman, 1999: Climatology of Rossby wave breaking along the subtropical tropopause. J. Atmos. Sci., 56 , 359373.

    • Search Google Scholar
    • Export Citation
  • Racz, Z., , and R. K. Smith, 1999: The dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 125 , 225252.

  • Reason, C. J. C., , and D. Jagadheesha, 2005: A model investigation of recent ENSO impacts over southern Africa. Meteor. Atmos. Phys., 89 , 181205.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., , W. Landman, , and W. Tennant, 2006: Seasonal to decadal prediction of southern African climate and its links with variability of the Atlantic Ocean. Bull. Amer. Meteor. Soc., 87 , 941955.

    • Search Google Scholar
    • Export Citation
  • Rouault, M., , P. Florenchie, , N. Fauchereau, , and C. J. C. Reason, 2003: South East tropical Atlantic warm events and southern African rainfall. Geophys. Res. Lett., 30 , 8009. doi:10.1029/2003GL014840.

    • Search Google Scholar
    • Export Citation
  • Shen, R., , J. F. Bresch, , and E. R. Reiter, 1986: Vertical interpolation of meteorological variables. Mon. Wea. Rev., 114 , 123134.

  • Todd, M., , and R. Washington, 1999: Circulation anomalies associated with tropical-temperate troughs in Southern Africa and the south west Indian Ocean. Climate Dyn., 15 , 937951.

    • Search Google Scholar
    • Export Citation
  • Todd, M., , R. Washington, , and P. I. Palmer, 2004: Water vapour transport associated with tropical-temperate trough systems over southern Africa and the southwest Indian Ocean. Int. J. Climatol., 24 , 555568.

    • Search Google Scholar
    • Export Citation
  • Vigaud, N., , Y. Richard, , M. Rouault, , and N. Fauchereau, 2009: Moisture transport between the South Atlantic Ocean and southern Africa: Relationships with summer rainfall and associated dynamics. Climate Dyn., 32 , 113123.

    • Search Google Scholar
    • Export Citation
  • Walker, N. D., , and J. A. Lindesay, 1989: Preliminary observations of oceanic influences on the February–March 1988 floods in central South Africa. S. Afr. J. Sci., 85 , 164169.

    • Search Google Scholar
    • Export Citation
  • Ziv, B., 2001: A subtropical rainstorm associated with a tropical plume over Africa and the Middle East. Theor. Appl. Climatol., 69 , 91102.

    • Search Google Scholar
    • Export Citation
  • Ziv, B., , and P. Alpert, 1994: Isobaric to isentropic interpolation errors and implication to potential vorticity analysis. J. Appl. Meteor., 33 , 694703.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Southern African political boundaries, WRC rainfall stations (dots), and key summer season synoptic features.

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 2.
Fig. 2.

24-h accumulated rainfall at WRC stations on (a) 1 Jan 1998, (b) 5 Jan 1998, and (c) 6 Jan 1998. (d) The contribution to NDJF season rainfall for the period 1–7 Jan 1998. The upper color scale denotes rainfall in mm for (a),(b), and (c), with the lower color scale denoting contribution % for (d).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 3.
Fig. 3.

(left) Winds (vectors, m s−1) and convergence (contoured, 10−6 s−1) at 850 hPa and OLR (shaded, W m−2) on (a) 31 Dec 1997, (c) 1 Jan 1998, and (e) 2 Jan 1998. (right) Full winds at 250-hPa level (streamlines, m s−1), 250-hPa ageostrophic winds (vectors, >4 m s−1), and 250-hPa divergence (shaded, 10−6 s−1) on (b) 31 Dec 1997, (d) 1 Jan 1998, and (f) 2 Jan 1998. Positions of the Angola low (AL), low-level cyclonic disturbances (L), midlatitude cyclones (MLC), and upper-level troughs (T) are indicated.

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 4.
Fig. 4.

Moisture transports (vectors, g kg−1 s−1), moisture convergence (shaded, 10−8 g kg−1 s−1), and pressure (contours, hPa) at 309 K for (a) 31 Dec 1997, (b) 1 Jan 1998, (c) 2 Jan 1998 (case I) and (d) 5 Jan 1998, (e) 6 Jan 1998, and (f) 7 Jan 1998 (case II). Regions of moisture export off Africa and into the midlatitudes are noted as warm conveyor belts (WCBs).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 5.
Fig. 5.

As in Fig. 3, but for (a),(b) 5 Jan 1998; (c),(d) 6 Jan 1998; and (e),(f) 7 Jan 1998 (case II).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 6.
Fig. 6.

As in Fig. 3, but for (a),(b) 15 Dec 2007; (c),(d) 16 Dec 2007; and (e),(f) 17 Dec 2007 (case III).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 7.
Fig. 7.

As in Fig. 4, but at 312 K on (a) 15 Dec 2007, (b) 16 Dec 2007, (c) 17 Dec 2007, and (d) 18 Dec 2007 (case III).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 8.
Fig. 8.

TRMM-estimated accumulated 24-h rainfall (mm) and 700-hPa full omega (contoured every 0.04 Pa s−1) on (a) 16 Dec 2007 and (b) 17 Dec 2007 (case III).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 9.
Fig. 9.

500-hPa NCEP2 omega (contoured every 0.04 Pa s−1) and percentage of NCEP2 omega explained by QG uplift (shaded, %) on (a) 1 Jan 1998, (b) 5 Jan 1998, (c) 6 Jan 1998, and (d) 16 Dec 2007.

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 10.
Fig. 10.

345-K PV (thin contours, PVU), PV advection [thick contours starting at +(−)3 PVU day−1 with positive advection dotted (negative advection solid)] and 345-K diabatic PV tendency (shaded, PVU day−1) on (a) 31 Dec 1997, (b) 1 Jan 1998, (c) 2 Jan 1998 (case I) and (d) 4 Jan 1998, (e) 6 Jan 1998, (f) 7 Jan 1998, (g) 8 Jan 1998, and (h) 9 Jan 1998 (case II). Vectors represent 345-K winds >20 m s−1. Note: Shading for (h) depicts OLR (W m−2, color scale as in Fig. 3).

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Fig. 11.
Fig. 11.

As in Fig. 10, but for (a) 14 Dec 2007, (b) 15 Dec 2007, (c) 16 Dec 2007, (d) 17 Dec 2007, (e) 18 Dec 2007, and (f) 19 Dec 2007 (case III). Note: Shading in (e) depicts OLR, as in Fig. 10h.

Citation: Monthly Weather Review 138, 7; 10.1175/2010MWR3070.1

Save