Patterns of animal endemism in the Maputaland-Pondoland-Albany biodiversity hotspot.

Thesis (Ph.D.)-University of KwaZulu-Natal, Durban, 2013.

The Maputaland-Pondoland-Albany (MPA) hotspot, as is the case of all such global biodiversity hotspots, has primarily been recognised based on its high floristic endemism and delimited intuitively. Boundaries of global biodiversity hotspots have seldom been empirically tested in terms of species distribution patterns and only a few have been examined for patterns of animal endemism. This thesis presents the results of a zoogeographical study of all five major vertebrate groups and selected invertebrate groups in south-eastern Africa, refining the delimitation of the MPA hotspot and identifying areas and centres of endemism within and around it. It also provides zoogeographical regionalisation schemata for the whole of south-eastern Africa. The study employed methods of, (a) preliminary qualitative identification of “Endemic Vertebrate Distributions”, (b) phenetic clustering of operational geographical units based on species incidence matrices, using the Jaccard’s coefficient of similarity and the Unweighted Pair-Group Method using Arithmetic means (UPGMA) clustering algorithm, (c) Parsimony Analysis of Endemicity, and (d) ArcGIS-based mapping of various measures of endemism (e.g. narrow endemism and weighted endemism). The results reveal that the MPA hotspot, though defined so due to its exceptional floristic endemism, is a hotspot for the endemism of animals too, especially for the herpetofauna and invertebrate groups like the velvet worms, land snails and many others. But the current boundary of the hotspot is arbitrarily defined and not exactly matching the patterns of animal endemism (and, likely, neither those in plants). Hence, a greater Maputaland-Pondoland-Albany (GMPA) region of animal endemism is proposed as a broad priority region of conservation concern, while centres of endemism within the GMPA are identified and patterns of quantitative measures of endemism are mapped. The study also proposes a zoogeographical regionalisation placing the GMPA and Highveld regions at the province rank in the global zoogeographical hierarchy, within the south east African dominion, also describing zoogeographical districts and assemblages nested within each. Results from the vertebrate and invertebrate analyses reveals the possibility of a common zoogeographical regionalisation for south-eastern Africa. The study emphasises the importance of quantitative biogeographical methodologies in conservation biogeography, in addition to their uses in the theoretical/descriptive biogeography.