The influence of abiotic processes, competition and predation on the community structure of rodents and shrews.

Abstract

Predation and abiotic processes rather than competition should influence the community structure of rodents and shrews with life histories characterised by high fecundity, short longevity and unstable populations. I investigated the influence of abiotic processes, predation and competition on three parameters of community structure (species composition, phenotypic and phylogenetic niches) of rodents and shrews at Mkhuze and Kube Yini, two game reserves in KwaZulu-Natal, South Africa, using null models and multivariate analyses. Rodents and shrews were sampled between 2007 and 2009. Sample-based rarefaction curves indicated that rodent species richness was higher at Mkhuze than at Kube Yini, while shrew species richness was identical at both reserves. Species richness estimators indicated that estimates of species richness were fairly accurate, hence strengthening the results from my null model analyses. I found evidence that immigration and extinction operating at a regional scale influenced rodent species composition. Moreover, habitat filtering operating at a local scale influenced rodent and shrew species composition. These processes produced nested assemblages: species present at species-poor sites were subsets of species present at species-rich sites. Habitat filtering also influenced the phenotypic niche of rodents and shrews: sympatric species showed similar phenotypic adaptations (phenotypic niches were underdispersed), probably in response to similar food requirements. Furthermore, shrew phenotypic traits showed a convergent evolution, and local assemblages comprised distantly related species (phylogenetic evenness), suggesting the influence of habitat filtering on the phylogenetic niche structure of shrews. Predation influenced shrew phenotypes. Bullae and ears were underdispersed and larger than expected by chance, probably to reduce predation risk through increased hearing sensitivity. In contrast, I found no evidence that predation influenced the rodent phenotypic niche. Competition influenced the phenotypic niches of rodents and shrews in species-rich assemblages (phenotypic niches were overdispersed). In these assemblages, the coexistence of species was facilitated by dietary and microhabitat partitioning. Competition also influenced the phylogenetic niche of rodents: phenotypic traits showed a convergent evolution, and local assemblages comprised closely related species (phylogenetic clustering). In conclusion, both abiotic and biotic processes influenced different parameters of the community structure of rodents and shrews. However, despite similar life-history traits, the community structure of local assemblages differed between rodents and shrews. Comparing patterns and processes of community structure across taxa would help find general trends of community organisation.