The mitigation of the effects of drought on pastoral populations in sub-Saharan Africa has received much attention in the literature (Wade, 1974; Charney et al, 1975; Hogg, 1980; Campbell, 1984; Sinclair and Fryxell, 1985; Moris, 1988). With few exceptions (e.g. McCabe and Ellis, 1987), the commonly held view is that traditional pastoral societies are increasingly unable to cope with drought, as indicated by large losses of herd capital, widening poverty and frequent famine. Traditional pastoral systems are thus thought by many to be in the process of gradual destruction through the combined effects of internal and external forces as exacerbated by drought. The primary objective of this chapter is to review information collected during and after the 1983-84 drought on the Borana Plateau in terms of livestock and human response and to compare the findings with those from other pastoral systems. A secondary objective is to highlight more cursory observations of the effects of the 1990-91 drought to determine whether cattle population dynamics in the Borana system during 1981 -91 resemble equilibrial or non-equilibrial patterns (Ellis and Swift, 1988; Behnke and Scoones, 1991). An integrated perspective will be introduced in Section 7.2: A theory of local system dynamics in which the general effects of subsequent droughts on the Borana system can be predicted.
Before the 1980s, the last drought on the Borana Plateau occurred in the mid-1970s (Donaldson, 1986: p 28). In contrast to an isolated dry year, a drought is defined here as two or more consecutive years when rainfall [or length of growing period (LGP)] is less than 75% of the long-term average (see Section 2.4.1.4: Climate, primary production and carrying capacity).
The 1983-84 drought began with the long rains (April to June) in 1983 when precipitation was roughly 50% of the long-term average. This was followed by three consecutive rainy seasons with deficient rainfall: (1) the short rains in October and November 1983; (2) the long rains in 1984; and (3) the short rains in 1984. These wet seasons commonly had enough rainfall to green-up the rangelands for a short period of time, but new forage was quickly consumed (Donaldson, 1986: p 38).
In terms of rainfall, the drought ended during the long rains of 1985. The total period for which rainfall was below average was thus two years. Forage production was high and crops of maize were produced during the long rains of 1985 as an opportunistic means of recovery for the Boran (Cossins and Upton, 1988b: p 272). However, response lags in the food chain dictated that milk production would not significantly resume until April 1986 and that recovery of near pre-drought levels of milk offtake by people was probably delayed until 1988 (see Section 7.2: A theory of local system dynamics). Duration of acute drought impact on the production system was thus on the order of five years.
