To Cull or Not to Cull: Lessons from a Southern African Drought

(1) The ecological consequences of a severe drought on four wildlife conservation areas in southern Africa (Tuli, Klaserie, Kruger, Umfolozi) are compared. (2) Herbivore mortality occurred during the second year of the drought, when failure of the spring rains coincided with depleted food reserves. It was severe in Klaserie, with up to 90% mortality. In Tuli, mortality was high but restricted mostly to wildebeest and zebra; impala (a mixed grazer-browser) and browsers were little affected. Mortality was up to 35% in Kruger; it included browsers, but wildebeest and zebra were little affected. Mortality was low in Umfolozi, even in an unculled block, although pre-emptive culling exceeded the average mortalities at Kruger. Ungulate populations recovered to predrought levels within 2 years in Kruger but, apart from impala, will take longer in the more severely affected areas. (3) Mortality of grass tufts was extensive in Klaserie, Umfolozi and Tuli, and was limited to areas near water in Kruger. Initial recovery was rapid in three areas after one season of above-average rainfall, but was retarded in Tuli, and later in Klaserie because of continuing drought and (possibly) continued impala grazing. Little tree mortality was recorded. (4) Ungulate mortality was high in the most arid areas (Tuli and Klaserie) where grass mortality was particularly high and, most importantly, where extensive surface water permitted a high pre-drought biomass and even distribution of ungulates. (5) It is concluded that culling is ecologically unnecessary where sufficient spatial heterogeneity exists to provide reserve forage. Some drought-related mortality is natural and probably beneficial to both animal and plant populations. Wild ungulate populations should be allowed to fluctuate within limits set by management objectives and culling is likely to be necessary only where the provision of water points has eliminated reserve forage areas.