The contribution of fire research to fire management: a critical review of a long-term experiment in the Kruger National Park, South Africa

The present paper reviews a long-term fire experiment in the Kruger National Park, South Africa, established in 1954 to support fire management. The paper’s goals are: (1) to assess learning, with a focus on relevance for fire management; (2) to examine how findings influenced changes in fire management; and (3) to reflect on the experiment’s future. Results show that fire treatments affected vegetation structure and biomass more than species composition. Effects on vegetation were most marked in extreme treatments (annual burning, burning in the summer wet season, or long periods of fire exclusion), and were greater in areas of higher rainfall. Faunal communities and soil physiology were largely unaffected by fire. Since the inception of the experiment, paradigms in savanna ecology have changed to encompass heterogeneity and variability. The design of the experiment, reflecting the understanding of the 1950s, does not cater for variability, and as a result, the experiment had little direct influence on changes in management policy. Notwithstanding this, managers accept that basic research influences the understanding of fundamental ecosystem function, and they recognise that it promotes appropriate adaptive management by contributing to predictive understanding. This has been a major reason for maintaining the experiment for over 50 years.