Causes Of Droughts In South Africa

Having described the pattern of rainfall-producing systems in the country, attention now focuses on some of the causes of periods of insufficient rainfall and droughts. The forcing mechanisms for droughts are well documented for the southern African region (see, e.g., Tyson, 1986; Mason and Jury, 1997; Lindesay, 1998; Tyson and Preston-Whyte, 2000). Much of the recent research in the country has focused on the prolonged droughts of the last two decades (namely the 1980s and 1990s) Tyson and Dyer, 1978; Dent et al., 1987; Jury and Levey, 1993; Alexander, 1995; Mason and Jury, 1997; Lindesay, 1998; Landman and Mason, 1999).

Rainfall variability for the country has been associated with atmospheric circulation configurations and interchanges in easterly and westerly circulations, the interactions between tropical and temperature systems, and the variation in pressure patterns over Marion and Gough Island (summarized in Tyson, 1986; Lindsay, 1998). Prolonged heat waves and droughts are linked, in most cases to the predominance of prevailing anticycIonic circulation over the country. Longer periods, such as extended wet spells, for example, are usually caused by an invigoration of tropically induced circulation disturbances forced by tropical easterlies whereas the extended dry spells usually occur with an expansion and increase of westerly disturbances. In the latter case, summers become drier in the summer rainfall region (Tyson and Lindesay, 1992). Aspects of this circulation patterning during wet and

4 CAUSES OF DROUGHTS IN SOUTH AFRICA 837

dry spells have been shown over long time scales, possibly during late Interglacial and mid-Holocene warming phases (Partridge, 1997).

Much of the work on seeking causal mechanisms for rainfall variability has focused on the relationship between rainfall and the circulation parameters within the region, such as changes in the intensity and positioning of pressure systems (Tyson, 1981, 1984; Miron and Lindesay, 1983; Taljaard, 1986, 1989). Adjustment in factors such as cloud bands (important conduits of energy and momentum into the subcontinent) and ocean temperature (Walker and Lindesay, 1989; Jury, 1995) have been shown to also induce rainfall-related changes over the subcontinent and more specifically over South Africa (Harrison, 1984a, b, c; Lindesay, 1988; Mason, 1990).

More recent work has also increasingly focused on the global-scale forcing mechanisms of rainfall and linkages to atmospheric circulation changes across the Southern Hemisphere. The Southern Oscillation (the climatic oscillation between warm and cold periods in the tropical Pacific) and sea surface temperature (SST) in the South Atlantic and South Indian Oceans, have become key foci of climate research in recent years (e.g., Tyson, 1986; Lindesay, 1988, 1998; Harrison, 1986; Jury et al, 1996; Mason and Jury, 1997; Landman and Mason, 1999). Aspects of these, relating specifically to drought, are highlighted below.

The Southern Oscillation is a major forcing mode of the interannual circulation variations over southern Africa (Lindesay, 1988; Allan et al, 1996). One mechanism by which the Southern Oscillation signal is transmitted to southern Africa is via the Indian Ocean Walker Circulation. High- and low-phase years of this circulation (known as El Nino Southern Oscillation, ENSO) have been shown, lurthermore, to influence rainfall over southern Africa (e.g., Lindesay, 1988, 1998). Briefly, low-phase years of the El Nino Southern Oscillation are accompanied by reductions in heat release and convection over tropical southern Africa caused by adjustments in the Walker Circulation and a retarded number of tropical-temperate troughs across South Africa. Fewer cloud bands occur and rainfall is diminished (Harrison, 1986). The reverse essentially occurs for high-phase or wetter years (Fig. 3). This general association between the Southern Oscillation phase changes and South African rainfall has been shown to exist in both present and preinstrumental rainfall records in South Africa from at least 1820 (Lindesay and Vogel, 1990).

An integral part of the Southern Oscillation is the role that sea surface temperatures play in modulating the occurrence of low- and high-phase changes and associated atmospheric circulation interactions. Although not explicitly direct, relationships between sea surface temperature anomalies, atmospheric circulation, and rainfall have been shown to exist in several instances (Glantz et al, 1987; Ogallo et al, 1988; Nicholson and Entekhabi, 1986; Mason, 1990; Walker, 1990) including relationships between above-average sea surface temperatures in the Benguela area and dry years such as 1982-1983 (Walker et al, 1984; Philander, 1986).

ENSO warm events have been associated with drought, resulting in a variety of impacts over much of southern Africa (e.g., Ogallo, 1987; Enfield, 1989; Cane et al, 1994). The 1982-1983 ENSO event, for example, served to exacerbate the prevailing dry conditions in much of the subcontinent (Bhalotra, 1985; Dent et al, 1987; Taljaard, 1989). The rainfall-producing systems of the subcontinent were displaced

Stronger South Atlantic Anticyclone and Gough Island/west-ooast index; positive pressure anomal y t—j-^ . f. ^ . Southward shift of storm tracks —^ '—^ stranger storms, south-western Cape winters drier

Weaker South Atlantic Anticyclone and Gough Islan devest-coast 'index; negative \ 1

Cloud bands locate preferentially over Madagascar and Indian Ocean

Cloud bands locate preferentially over Madagascar and Indian Ocean

Weaker South Atlantic Anticyclone and Gough Islan devest-coast 'index; negative \ 1

weaker storms; south-western Cape winters wetter

Figure 3 Model of the circulation changes over southern Africa during wet and dry spells (after Tyson and Preston-Why te, 2Û00, with permission, Oxford University Press, Cape Town).

5 CLASSIFYING DROUGHTS 839

eastward during this time (Fig. 3) (Harrison, 1983; Tyson, 1986; Lindesay, 1988; Muller and Tyson, 1988).

The recent ENSO event of 1997-1998 was a very powerful one with anomalously high SSTs and, thus, indications for a possible drought (of similar magnitude to that of 1982-1983); interventions were planned, both locally and elsewhere (see Thomson et al„ 1998; Mason et al, 1999; and NOAA-OGP, 1999). The El Nino impact was reduced in some parts of southern Africa and appears to have been modulated by temperatures in the Indian and Arabian Sea (Landman and Mason, 1999). Despite the strong linkage between ENSO and rainfall in parts of southern Africa, it must be remembered that not all droughts are associated with ENSO. There thus remain areas of uncertainty and predictive skill in seasonal forecasting for the country because of the complex interactions between the oceans, atmosphere, and land (Mason et al., 1994, 1996; Landman and Mason, 1999).

In summary, current research on the causes of droughts in South Africa indicates that rainfall over the country is influenced by a number of interactive mechanisms, details of which are still being examined (Mason and Jury, 1997; Landman and Mason, 1999; Mason and Tyson, 2000). The interaction between tropically and extratropically sourced weather systems, when combined with upper-air tropospheric dynamics and convectively unstable air masses, usually results in high possibilities of rainfall. Extreme drought events in southern Africa are the result of a number of atmospheric circulation interactions (Tyson, 1986; Lindesay, 1998) with some droughts and periods of reduced rainfall, for example, connected with ENSO events.

0 0

Post a comment