Savannas are ecosystems dominated by trees and grasses. A relatively recent hypothesis to explain tree-grass coexistence is the patch-dynamics theory, which proposes that savannas can be interpreted as patch-dynamic systems where landscapes are composed of many patches in different states of cyclical transition between woody and grassy dominance. Transition is spatially asynchronous within patches and is driven by self-thinning and localized tree recruitment events. Self-thinning arises from the density-dependent mortality of woody plants and drives the transition of a tree-dominated to a grass-dominated patch. In arid and  semi-arid savannas, repeatedly overlapping rain events may induce mass germination of a tree cohort and promote the transition of a grassy patch to woody dominance.

If savanna dynamics is driven by self-thinning and localized tree recruitment, this should be visible in the spatio-temporal tree distribution. We individually followed ca. 2900 Acacia erioloba trees in space and time based on multi-temporal images of a savanna site in South Africa. Based on emergence and vanishing of trees between two consecutive images, we identified new and dead trees. We statistically analyzed the spatial distribution of new and dead trees relative to the distribution of mature trees using scale-dependent point pattern analyses.

Results/Conclusions

Our results show that most dead trees are located in areas with an overall high density of trees. This means that the probability of mortality increases with tree-density and that A. erioloba undergoes self-thinning. The spatial distribution of new trees is clumped and spatially segregated from mature trees. Thus, there is patchy regeneration in agreement with localized recruitment driven by rainfall.

These results give further support to the paradigm of patch dynamic savannas driven by self-thinning and patchy regeneration.