Small mammal responses to Scarp Forest Restoration in the Maputoland-Pondoland-Albany Hotspot, South Africa.

Masters Degree. University of KwaZulu-Natal, Durban.Restoration ecology is a relatively new field. Although a range of attributes have been used to assess restoration success, they have not been standardised across studies. Recently, three main ecological attributes have been identified as key measures to standardise the assessment of restoration success: species diversity, vegetation structure and ecological processes. However few studies have combined more than two of these ecological attributes when assessing restoration success. The aim of this study was to apply these three ecological attributes to determine whether Scarp Forest restoration has been successful from the perspective of small mammals at the Buffelsdraai Landfill Site, Durban, South Africa. I assessed the response of small mammals to Scarp Forest restoration at 2, 4 and 6 year post-restoration periods. I surveyed small mammals every three months for one year in three restoration treatments (2010, 2012 and 2014 restored), as well as in surrounding sugarcane and riverine forest sites. At each site I measured the vegetation structure and small mammal diversity. Additionally, I conducted stable isotope analysis on vegetation and invertebrate samples to compile a baseline database of potential prey, and compared these data with the stable isotope composition of hair and tissue samples collected from rodents and shrews to analyse the trophic structure of the small mammal assemblages. In support for the prediction that vegetation structure should increase in complexity at restored sites, tree species richness, density and height were higher at the 2010 restored than more recently restored sites; and grass height and percentage cover were highest at 2012 restored sites. Except, forb and grass species richness were higher at newly restored sites. Second, rodent abundance was higher at the 2010 restored sites than the 2012 and 2014 restored sites and sugarcane sites. However, shrew abundance and species richness were not significantly different among the study sites. Third, carbon and nitrogen isotopic composition of rodent hairs suggest that these species utilised resources associated with the 2010 restored sites rather than those associated with recently restored sites, sugarcane sites and forests. Further, the stable isotope ratios of carbon and nitrogen in Mastomys natalensis’ tissues showed that these rodents predominantly utilised resources associated with the 2010 restored sites irrespective of the tissue that was analysed. Conversely, carbon and nitrogen isotopic composition of shrew hairs suggest that these species foraged at the sites where they were captured. Taken together, my results suggest that at Buffelsdraai, the restoration efforts have ensured progressive succession in the scarp forest after 10 years, at least from the perspective of most small mammals