Species landscape relations at multiple scales: case studies with two owl species, powerful owl (Ninox strenua) and Southern Boobook (Ninox novaeseelandiae)
Access status:
USyd Access
Type
ThesisThesis type
Doctor of PhilosophyAuthor/s
Stone, Brian EdwardAbstract
Models of complex natural systems often demonstrate significant limitations in addressing issues such as the distribution of animal species, particularly in relation to long recognised “problems of scale”. One of the more important theories of scale is Hierarchy Theory (HT). Not ...
See moreModels of complex natural systems often demonstrate significant limitations in addressing issues such as the distribution of animal species, particularly in relation to long recognised “problems of scale”. One of the more important theories of scale is Hierarchy Theory (HT). Not surprisingly this theory provides an important framework for surveying and modelling a range of complex, multi-scale natural systems. In this thesis a range of models are developed that focus on contrasting sized Hawk Owls, the larger Powerful Owl (Ninox strenua) and the smaller Southern Boobook (Ninox novaeseelandiae). The models are used to investigate owl distributions from the micro scale (sampling sites) through the meso scale (250 km2 study area) in forests east of Bathurst NSW, to the macro scale (Bioregion). Two different types of modelling were used. The first is Regression (i.e. Generalised Additive Models (GAMs)). The second type is Spatial Point Process (SPP) modelling. Whilst HT has a valuable role as a conceptual construct for multi scale ecological studies, the results from this study challenge some of the presumptions of HT and its relevance to cost effective, practical management. In this thesis I demonstrate that both regression models and spatial point process models have a role to play in predictive modelling. However, they require an adequate number of data points derived from rigorous spatial sampling design and survey methodology. Furthermore, the results of such models utilised within a specific landscape may lack the capacity to be extrapolated to other landscapes, as well as other locations in a species’ range. Important questions have been addressed relating to the value and role of multi scale spatial species distribution and species environment modelling. For some key species and for some broad scale landscapes, management and species conservation practices may require modelling studies, appropriate spatial sampling design and consequently, costly data requirements.
See less
See moreModels of complex natural systems often demonstrate significant limitations in addressing issues such as the distribution of animal species, particularly in relation to long recognised “problems of scale”. One of the more important theories of scale is Hierarchy Theory (HT). Not surprisingly this theory provides an important framework for surveying and modelling a range of complex, multi-scale natural systems. In this thesis a range of models are developed that focus on contrasting sized Hawk Owls, the larger Powerful Owl (Ninox strenua) and the smaller Southern Boobook (Ninox novaeseelandiae). The models are used to investigate owl distributions from the micro scale (sampling sites) through the meso scale (250 km2 study area) in forests east of Bathurst NSW, to the macro scale (Bioregion). Two different types of modelling were used. The first is Regression (i.e. Generalised Additive Models (GAMs)). The second type is Spatial Point Process (SPP) modelling. Whilst HT has a valuable role as a conceptual construct for multi scale ecological studies, the results from this study challenge some of the presumptions of HT and its relevance to cost effective, practical management. In this thesis I demonstrate that both regression models and spatial point process models have a role to play in predictive modelling. However, they require an adequate number of data points derived from rigorous spatial sampling design and survey methodology. Furthermore, the results of such models utilised within a specific landscape may lack the capacity to be extrapolated to other landscapes, as well as other locations in a species’ range. Important questions have been addressed relating to the value and role of multi scale spatial species distribution and species environment modelling. For some key species and for some broad scale landscapes, management and species conservation practices may require modelling studies, appropriate spatial sampling design and consequently, costly data requirements.
See less
Date
2015-02-01Licence
The author retains copyright of this thesis. It may only be used for the purposes of research and study. It must not be used for any other purposes and may not be transmitted or shared with others without prior permission.Faculty/School
Faculty of Rural ManagementAwarding institution
The University of SydneyShare