Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2003 → Modeling the Impact of the African Elephant, Loxodonta africana, on Woody Vegetation in Semi-Arid Savannas


Modeling the Impact of the African Elephant, Loxodonta africana, on Woody Vegetation in Semi-Arid Savannas

Baxter Peter William Joseph

Titre : Modeling the Impact of the African Elephant, Loxodonta africana, on Woody Vegetation in Semi-Arid Savannas

Auteur : Peter William Joseph Baxter


Grade : Doctor of Philosophy in Environmental Science, Policy and Management 2003

Concerns over elephant impacts to woody plants in African savannas have highlighted shifts in vegetation community composition with implications for possible reductions in biodiversity. I developed a grid-based savanna model that differs from previous elephant vegetation models by accounting for tree demographics, tree-grass interactions, stochastic environmental variables (fire and rainfall) and spatial contagion of fire and tree recruitment. The vegetation component of the model produces long-term tree-grass coexistence and realistic fire frequencies. The tree-grass balance of the model is more sensitive to changes in rainfall conditions and tree growth rates while less sensitive to fire regime. Introducing elephants into this model savanna has the expected effect of reducing tree cover, although at an elephant density of 1.0 per square kilometer, woody plants still persist for over a century. I tested the effect of plant responses to elephant impact : faster growth was a more successful strategy than elephant-enhanced germination or adult resilience to impact. I elaborated the model by including a second, more “r-selected” tree species to investigate the effects of elephant impacts on species composition within the tree community. The model produces similar dynamics when run with either tree species alone ; when both species are included it replicates ecological succession, with competitive exclusion of the early-successional species by the later-successional species on a timescale of centuries. Increases in growth, fecundity or survival of the earlysuccessional species increase the likelihood of its persistence over 500 years. Inclusion of the faster-growing tree species in the model enables both species to survive greater elephant densities. Spatial heterogeneity of the woody plant component increases with elephant density. I examined the interaction of the two tree strategies – adult resilience and elephant-enhanced germination – with elephant preference for either species. Adult tree resilience was the more successful strategy and may act synergistically between tree species. Fire suppression also moderates the effects of elephant damage. I conclude that while elephants may cause woodland to decline, they may also enhance biodiversity at lower densities, and increase spatial heterogeneity. Conservation workers should be conscious of the array of species types and their interactions when planning to manage savannas and/or elephant populations for biodiversity.

Version intégrale

Page publiée le 9 janvier 2017