How important is patch size relative to patch isolation in an arid inselberg landscape?

Abstract

Patch size (area) and isolation (inverse distance to source) are key variables of island biogeography, but their application in conservation planning has recently come under scrutiny. Based on animal distribution data in fragmented landscapes from around the world it was postulated that occupancy was poorly predicted by patch size and isolation. Do these findings also apply to arid areas and to plants?This question was investigated in four inselberg landscapes in the Namib Desert stretching over a 770 km climatic north–south gradient. Inselbergs (isolated mountains) are special habitats and can be considered ‘islands’ of more mesic conditions in an arid matrix. Species richness of inselberg specialists was used as an indicator and regressed against area, distance to nearest habitat and distance to ‘mainland’ (or source pool, which are proxies for level of isolation). Additionally, the influence of landscape geographic position, land use and rock types were studied. Responses of different life forms and individual plant species as well as possible dispersal limitations were also investigated.Whether patch size is more important than isolation in determining species richness on arid inselbergs could not be answered at a landscape level. However, when investigated on a functional and species level, isolation was of greater importance than patch size for trees and also for selected species such as Commiphora saxicola and Heliotropium steudneri. The effect of distance to source in this study is likely a function of spatial distance as well as a coast-inland moisture gradient. Inselberg specialist-richness was not affected by land use or the bioclimatic position. Yet rock type, which is a reflection of soil physical and chemical properties did affect species richness with granites and basalts supporting more species. Life forms of inselberg specialists were influenced by the investigated environmental variables, but dispersal limitations were not. Despite inconclusive results at a landscape level, functional groups and individual species provided more conclusive results. This points towards the necessity of additional investigations at a finer scale to gain an understanding of the factors driving species distributions. In future more detail regarding the ‘matrix’ surrounding ‘patches’, as well as evolutionary and vegetation historic factors should be explored. Phylogenetic studies would be one tool to help unravelling some of the presently unknown factors affecting species distributions on Namib inselbergs. This study also highlighted the importance of inselbergs in arid landscapes as ‘hotspots’ of species richness and species of conservation importance.