Thesis (Ph.D.)--University of Wyoming, 2008. / Title from PDF title page (viewed on Dec. 4, 2009). Includes bibliographical references.

Thesis (M.S.)--North Carolina State University, 2004. / Title from PDF t.p. (viewed on Dec. 16, 2005). Includes vita. Includes bibliographical references.

Forest fragmentation is caused by the clearing of patches of indigenous vegetation for agriculture, urban development, and other human land uses. Such action results in patches of remnant natural vegetation being surrounded by altered vegetation. I investigate the effects of forest fragmentation and matrix type on avian diversity and assemblage structure in forest patches of the historically fragmented Karkloof / Balgowan forest range, KwaZulu-Natal, South Africa. This study compares the bird assemblage diversity and composition of indigenous forest patches surrounded by commercial forestry (Gilboa complex) with that surrounded by natural grassland matrix (Balgowan complex). Insularisation of Afromontane Mistbelt forest in KwaZulu-Natal has led to loss of species where forest fragments support fewer bird species than comparably sized patches of mainland forest. Small fragments within natural grassland have fewer bird species per unit area than larger fragments. Forest patch area-dependent density compensation is evident and bird assemblages appear saturated. Bird assemblages are characterised by a non-random species distribution pattern where area-dependent processes are dominant, and the loss of species from fragments follows a deterministic sequence. In forests in the plantation-dominated matrix no island-effect is detectable and it appears that forest patches are converging on the same bird species richness, regardless of forest size. No density compensation is evident and bird assemblages are not saturated. The sequence of species loss from forest patches is not as predictable, where a random yet prominent colonisation process exists. As commercial plantations provide suitable habitat cover for movement of forest birds, colonisation of both distant and small indigenous forest patches has been possible, reducing the effects of area-dependent extinction in the forest patches but also resulting in lower species richness in larger patches. Bird species of the Karkloof / Balgowan forest range appear to be fragmentation adapted, and most species are resilient to further landscape change. Certain species are however more prone to local extinction than others. The major predictors of extinction risk are body size, abundance status, and feeding guild. Patch area is the dominant force governing traits in the natural Balgowan complex where larger species with low natural abundance and an insectivorous diet are most prone to local extinction. In the Gilboa complex the nature of the plantation matrix appears to be masking the species natural responses to fragmentation making it difficult to predict which species are most at risk. In order to preserve maximum bird diversity, including high-risk species, the largest intact forest units (≥302ha) must be conserved. Evidently, the nature of the matrix affects avifaunal diversity and distribution in forest patches, and plantations have the capacity to significantly alter bird assemblage structure and composition in indigenous forest patches. Forest fragments must be considered as integrated parts of a complex landscape mosaic, and this study emphasises the importance of understanding landscape-scale processes. Knowledge of ecological and life history traits proves valuable for predicting community level response to landscape change. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2001.

Grassland is an important part of the ecosystem in the Canadian prairies and its loss and fragmentation affect biodiversity, as well as water and carbon fluxes at local and regional levels. Over the years, native grasslands have been lost to agricultural activities, urban development and oil and gas exploration. This research reports on new methodologies developed for mapping the spatial extent of native grasslands to an unprecedented level of detail and assessing how the grasslands are fragmented. The test site is in the Newell County region of Alberta (NCRA). 72 Landsat and 34 SPOT images from 1985 to 2008 were considered for the analysis. With an airport runway used as a pseudo-invariant feature (PIF), relative radiometric correction was applied to 17 Landsat and 8 SPOT images that included the same airport runway. All the images were classified using the Support Vector Machine (SVM) classification algorithm into grassland, crop, water and road infrastructure classes. The classification results showed an average of 98.2 % overall accuracy for Landsat images and SPOT images. Spatial extents and their temporal change were estimated for all the land cover classes after classifying the images. Fragmentation statistics were obtained using FRAGSTATS 3.3 software that calculated land cover pattern metrics (patch, class and landscape). Based on the available satellite image data, it is found that in Newell County there is almost no significant change found in the grassland and road infrastructure land cover in over two decades. Also, the fragmentation results suggest that fragmentation of grassland was not due to the result of road infrastructure. / x, 105 leaves : ill., ; 29 cm

Habitat destruction and degradation are the most important threatening processes for the majority of species of conservation concern. As landscapes become more fragmented their spatial structure becomes increasingly important for the viability of wildlife populations. Therefore, understanding the dynamics of how populations interact with landscape structure is crucial for developing effective wildlife management and monitoring strategies. To achieve robust management, such an understanding should be incorporated into decision theory approaches for the design of management and monitoring strategies. Monitoring is a crucial component for assessing the outcome of management, because this can then be used to inform future management actions in an adaptive management framework. However, for many species, their spatial population dynamics are not well understood and objective management frameworks not well developed. In this thesis I use modelling approaches to develop the components of an adaptive management framework for species of conservation concern that occur in fragmented landscapes. I use the koala (Phascolarctos cinereus) as a study organism and specifically address three key components that are important for adaptive management. These components are: (1) the development of novel approaches for modelling and understanding spatial population processes, (2) the application of modelling approaches to develop general principles and methods for managing wildlife populations in fragmented landscapes and (3) the application of modelling approaches to identify the best spatio-temporal sampling strategies for monitoring population trends in fragmented landscapes. This research advances our understanding of the dynamics of wildlife populations in fragmented landscapes, but importantly also develops tools and general principles for the adaptive management of these populations. To improve our understanding of the dynamics of populations in fragmented landscapes I develop several novel modelling approaches. Firstly, I develop an approach for disassociating the effect of natural habitat and anthropogenic influences on population distributions using static occupancy models. This approach allows inferences to be made about how important the distribution of habitat is compared to other human-related factors. Secondly, I develop modelling approaches for understanding habitat selection and dispersal processes. In particular, I address common problems for modelling complex habitat selection processes and parameterising dispersal simulation models. The habitat selection models are novel because they account for habitat preferences that depend upon the spatial location of habitat. The novel emphasis for the dispersal models, is showing that pattern-oriented approaches are useful for parameterising dispersal simulation models, even when we have little data. These models make an important contribution to our understanding of and methods for understanding spatial population processes. These and other modelling approaches are then applied to some key management questions, in order to develop general principles and methods for wildlife management decision-making. These studies relate to identifying the best spatial allocation of different management strategies, the planning of road networks to minimise impacts on wildlife and the impact of different mortality rates on the ranking of habitat protection strategies. First, I show that the relative spatial distribution of habitat and anthropogenic influences has important implications for the spatial allocation of management strategies. Second, I show that increasing traffic volume on existing roads is generally preferential to building new roads in terms of minimising the impact on wildlife mortality. Third, I use a multi-criteria decision analysis to show that the spatial arrangement of protected habitat is much less important for populations subject to high mortality rates than those subject to low mortality rates. Finally, spatially-explicit modelling approaches are applied to questions related to how sampling effort should best be allocated spatially and spatio-temporally to monitor population trends. These studies reveal that the best sampling strategy depends predictably upon the monitoring objectives and the dynamics of the species. Using the principles developed, improved monitoring strategies and ultimately improved management strategies will be possible. This thesis makes an important contribution to the development of adaptive management strategies for wildlife populations in fragmented landscapes. To implement an adaptive management approach, the modelling, management and monitoring components must be explicitly linked. By developing each of these components, this thesis provides a strong basis from which an adaptive management framework can be constructed. How this may be achieved and key areas for future research are discussed.

Thesis (doctoral)--Swedish University of Agricultural Sciences, 2007. / Thesis documentation sheet inserted. Includes appendix of four papers and manuscripts co-authored with Bo Söderström. Includes bibliographical references. Also issued electronically via World Wide Web in PDF format; online version lacks appendix.

Thesis (Ph.D.)--University of Adelaide, Dept. of Geographical and Environmental Studies, 2001. / Includes bibliographical references (p. 301-319).

Thesis (M.App.Sc.)--University of Adelaide, Dept. of Animal Science, 2002. / "September 2002" Bibliography: leaves 77-85.

Thesis (Ph.D.)--University of Wollongong, 2005. / Typescript. Includes bibliographical references: leaf 153-172.

Thesis (Ph. D.)--Oregon State University, 2009. / Printout. Includes bibliographical references (leaves 97-103). Also available on the World Wide Web.