Premium
Determinants of savanna vegetation structure: Insights from Colophospermum mopane
Author(s) -
HEMPSON GARETH P.,
FEBRUARY EDMUND C.,
VERBOOM G. ANTHONY
Publication year - 2007
Publication title -
austral ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.688
H-Index - 87
eISSN - 1442-9993
pISSN - 1442-9985
DOI - 10.1111/j.1442-9993.2007.01712.x
Subject(s) - riparian zone , riparian forest , ecology , shrub , woodland , geography , dominance (genetics) , floristics , arid , biology , species richness , habitat , biochemistry , gene
Savannas are structurally heterogeneous at the local, community‐level scale due to fine‐scale floristic heterogeneity as well as the responses of individual species to underlying environmental variation. The structure of mopane woodland, an arid savanna of southern Africa, is dictated largely by local variation in the relative dominance of tall, single‐stemmed and shorter, multi‐stemmed forms of the dominant tree species, Colophospermum mopane (Kirk ex Benth) Léonhard. Here we evaluate the hypothesis that the existence of these alternative forms of C. mopane , as well as the factors that dictate their distribution at a local scale, are driven by fine‐scale environmental variability in available water. We surveyed trees at four sites in the Kruger National Park of South Africa, in each instance surveying both forms of the species, from both riparian and non‐riparian zones. A survey of genetic variation across our sample ( n = 80 individuals), using inter‐simple sequence repeat (ISSR) amplification profiles, indicates that the two forms are not genetically distinct, instead being environmentally determined. While measurements of xylem pressure potentials, determined using a Scholander pressure chamber, show a significant difference between riparian and non‐riparian zones, there is no significant difference between the two growth forms. Although this seems paradoxical in view of the prevalence of tree and shrub form mopane at riparian and non‐riparian sites, respectively, we speculate that such a pattern may emerge through the interaction of moisture stress and top‐down controls, such as those imposed by large mammal browsing and fire.