Variations in annual rainfall may have profound effects on the performance of reedbuck, waterbuck, lechwe and puku.

Rainfall is the principal determinant of numbers of common ungulates in southern African savannas (Mills et al 1995).

There is a strong relationship between the survival of tsessebe and the accumulated surplus or deficit in rainfall over a long period (Dunham & Robertson 2001, Dunham et al 2003). A strong correlation between adult and juvenile mortality and the rainfall in the late dry season was found. Rainfall in late dry season appears to be critical, affecting the animals' condition, survival rate, late stage of pregnancy and early stage of lactation.

The long term surplus or deficit in rainfall may be the main determinant of the condition of the floodplain grasslands in the Caprivi, which is the main habitat for all of these species. A prolonged drying out process would affect water tables and the catenas in vegetation from river banks to the upper reaches of catchments. Reedbuck, waterbuck, lechwe and puku would find their preferred habitats shrinking to narrow bands close to rivers. Martin (2004) examined the extent to which the given regime of adult female and juvenile mortality, which was derived for average conditions, would have to change in order to throw the 'generic' population into a decline (Table 3).

Reedbuck

Joubert and Mostert (1975) 'felt' that there were no more than 50 reedbuck in the country in 1975. This was at a time when the cumulative rainfall deviations were in a deficit mode. Whilst placing no weight on the survey estimates for reedbuck (Figure 17), the species was recorded in modest numbers from 1978 to 1994. During this time the cumulative surplus/deficit in rainfall was positive. From 1995-2002 the long term rainfall regime swung into a deficit mode coinciding with concerns expressed about the status of reedbuck.

Red Lechwe

Estimates of lechwe numbers (Figure 18) were consistently higher than 4,000 up until 1995, coinciding with a peak in the cumulative rainfall surpluses. The drop from 13,000 estimated in 1980 to 4,000 estimated in 1986 appears to track the long term rainfall regime fairly closely. It seems that after 1986, the lechwe numbers were unable to bounce back immediately to follow the secondary peak in cumulative rainfall surpluses which occurred around 1990. After 1996, the long term rainfall moved into a deficit mode and the lechwe population appears to have crashed. In 1998, the estimate was only 119 animals on a comprehensive survey of the Caprivi.

n Figure 19 lechwe numbers are shown in relation to the cumulative surplus/deficit rainfall record for Andara beginning in 1915. It is of interest to note that in 81 years only one complete cycle of surpluses and deficits has occurred (1914-1977). A second cycle is underway at present but it would appear to be less than half-way through its time. Early narrative accounts of lechwe (Shortridge 1934) suggest that lechwe populations were high in the 1930s - which coincided with a positive half-cycle in the rainfall lasting from 1914-1944. Child and von Richter (1969) remark that in the late 1960s lechwe were in decline in northern Botswana and the Caprivi. Joubert and Mostert (1975) estimated that there were fewer than 100 lechwe in Namibia in 1975. This was during a negative rainfall cycle lasting from 1945-1977. The early part of the survey record for lechwe shown in Fig.18 coincided with a rainfall surplus period from 1978-1984 and, during this time, lechwe were fairly abundant. Having now entered a deficit mode, it would seem that lechwe populations are once more in a parlous state. If, as it seems, this long term rainfall regime is significant in regulating lechwe populations, then there is little that can be done to assist the species in its natural range. The best strategy would be to protect a small nucleus of the animals so that when the rainfall moves once more into a surplus mode, the population can begin to increase.