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

Accueil du site → Doctorat → Afrique du Sud → Plant ecology of the Namaqualand Strandveld : community structure and dynamics in a winter-rainfall desert

University of Cape Town (2000)

Plant ecology of the Namaqualand Strandveld : community structure and dynamics in a winter-rainfall desert

Eccles, Neil Stuart

Titre : Plant ecology of the Namaqualand Strandveld : community structure and dynamics in a winter-rainfall desert

Auteur : Eccles, Neil Stuart

Université de soutenance : University of Cape Town

Grade : Doctor of Philosophy (PhD) 2000

The Strandveld forms the western coastal margin of the arid Namaqualand system. As a whole, Namaqualand has a very unusual diversity of plant life. Firstly it is unusually diverse in the most basic sense - there are lots of species, and many of them are endemic. Secondly, the ratio of species diversity to functional diversity is very high (i.e. there appears to be a great deal of functional overlap). Thirdly, in the specific case of the Strandveld, this diversity exists in a very homogeneous abiotic context. This leads to a challenging theoretical question : how does this diversity persist at a local scale ? Although fundamental in nature, the answer to this question has many applied implications in terms of how we might manage the system, and how we might rehabilitate areas that have been damaged by activities such as mining. In order to begin to address this question, I set out to explore the spatial and temporal structure of selected Strandveld communities. This strategy has proved fruitful in many other systems and was in keeping with prevailing conceptual model for Namaqualand. My earliest observation in these communities was that the majority of perennial plants assemble into cosmopolitan vegetation clumps. This led me to postulate that communities are shaped by a positive feedback between seed dispersal strategies that promote the recruitment into clumps and general physical benefits associated with microhabitat modification and mutual protection. Aspects of this postulate were tested with a combination of manipulative and sampling experiments. The results from a removal experiment suggested that interactions between individuals and their neighbors in clumps are in fact more complex than initially thought. Both negative and positive interactions were evident when different variates were considered, indicating that a balance between positive and negative interactions exists. This result was further supported by the observation that the spatial distribution of below-ground biomass does not mirror the highly clumped above ground pattern, suggesting that competition for water is a likely possibility. Despite this evidence that interactions between individuals may shape the spatial structure of these communities, I could find little evidence that the interactions between individuals of particular species result in any deterministic outcome. There was certainly no evidence for any deterministic competitive exclusion. The species make-up of the vegetation clumps suggests that a large element of chance is involved in the assembly of clumps, and the lack of any strong relationships in populations between the size of individuals and their biotic neighborhoods further supports this suggestion. The implication of this is that these Strandveld communities may be examples of communities with stochastic domination where species coexistence occurs, not in spite of, but because they are functionally equivalent generalists. Together, this understanding of the forces shaping the spatial structure of Strandveld communities, and the suggestion that Strandveld communities may be characterized by stochastic domination constitutes a very valuable contribution to the conceptual framework that has emerged to explain Namaqualand’s unique plant diversity.


Version intégrale

Page publiée le 18 septembre 2017, mise à jour le 10 janvier 2019