School of Agriculture, Food and Ecosystem Sciences - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    Multi-scale drivers of vertebrate communities in flammable landscapes
    Dorph, Annalie Joan ( 2021)
    For successful biodiversity conservation, knowledge of the patterns and processes acting in the landscape to influence faunal species diversity is essential. One hypothesis developed to explain how spatial variation affects species diversity is the environmental heterogeneity hypothesis. This hypothesis postulates that increased spatial variability in resources will lead to greater niche availability and subsequent increases in faunal diversity. Environmental heterogeneity is created by two key sources (1) biophysical factors (e.g. topography, climate) and (2) stochastic or anthropogenic disturbances (e.g. fire, land clearing). Species and communities respond to this environmental heterogeneity at both the local- (<1 ha) and landscape-levels (10-1000s ha). Understanding the multi-level effects of different sources of environmental heterogeneity on faunal communities will help land managers to successfully implement management programs and conservation activities. In this thesis, I examined how environmental heterogeneity measured at multiple levels influenced faunal communities in flammable landscapes. I used two case studies from separate landscapes to understand the effects of biophysical factors, disturbances and habitat structure on species and community responses. In the first case study, I examined a mammal community in a topographically-complex, unfragmented landscape. In the second case study, I examined a reptile community in a highly-modified landscape. In both cases I measured environmental drivers at multiple scales using novel approaches, such as gradient modelling to account for spatial pattern created by continuous environmental variables. Local-level measures were better for predicting mammal species richness, while reptile richness was not influenced by any of the local-level measures tested. Specifically, habitat structural complexity and biophysical factors were the strongest drivers of mammal richness at the local-level. Disturbance from fire had no local-level effect on the mammal or reptile community. Landscape-level measures of environmental heterogeneity were stronger predictors of species richness for disturbance from fire and changes to land use in both the mammal and reptile community. Mammal richness was most strongly driven by landscape-level measures of vegetation productivity, time since fire, elevation and habitat complexity. Reptile richness was most strongly driven by landscape-level measures of land use and habitat complexity. Time since fire had no effect on reptile richness at the landscape-level, however individual species were driven by spatial fire patterns at the landscape-level. Studies of the response of species and communities to local- and landscape-level measures of their environment are still limited in many areas of ecology. In this thesis, I advance our understanding of how faunal communities respond to two key sources of environmental heterogeneity in landscapes affected by different processes. From a management perspective, I found that the impact of disturbances on fauna are better assessed at the landscape-level rather than local-level. Additionally, I re-emphasise the importance of retaining native vegetation in modified landscapes for maintaining species diversity. In future, more work is needed to understand how species resource selection and faunal community response are determined by the relative influence of local- and landscape-level measures of different processes acting in both fragmented and unfragmented landscapes