Thesis (MS)--Montana State University--Bozeman, 2009. / Typescript. Chairperson, Graduate Committee: Duncan T. Patten. Includes bibliographical references (leaves 122-131).

Thesis (Ph.D.)--University of Washington, 1995. / Vita. Includes bibliographical references (leaves [110]-144).

Thesis (M.S.)--Montana State University--Bozeman, 2007. / Typescript. Chairperson, Graduate Committee: Brian L. McGlynn. Includes bibliographical references (leaves 129-134).

no abstract provided by author / Master of Science

Thesis (PhD)--Montana State University--Bozeman, 2008. / Typescript. Chairperson, Graduate Committee: Brian L. McGlynn. Includes bibliographical references.

The revegetation of severely, disturbed areas using indigenous vegetation forms the topic of this thesis. Mountain Fynbos was chosen for the study because of its importance as a unique vegetation-type worthy of conservation. Unfortunately, this vegetation faces long-term threats associated with escalating development pressures, which include the demand for improved infrastructures. These factors are discussed and a process is developed for using indigenous vegetation to revegetate severely disturbed areas. This use of indigenous vegetation represents a significant and innovative move away from the traditional use of commercially available grasses in South Africa. Fynbos is adapted to periodic droughts and nutrient poor soils and as such has practical advantages over artificial plant covers in terms of its low maintenance requirements for long term establishment. Theoretical perspectives of disturbance and succession are contrasted with the requirements for revegetation as observed in the field.

A long-standing question in ecology is understanding how the environment structures species occupancy in space and time. Specifically, identifying associations between environmental drivers - climate and land cover - and species occupancy is crucial to predicting species distributional dynamics in the future. Over the last century, research on the abiotic drivers of species occupancy has largely focused on temperate regions. Tropical mountain ecosystems harbor extraordinary levels of diversity and face some of the highest anthropogenic pressures of climate and land cover change. Yet, such regions have remained historically understudied. Bird species, due to their sheer diversity and occurrence across climatic zones and land cover types, are an ideal model for understanding how climate and land cover structure occupancy in space and time. The goal of this dissertation is to disentangle spatial and temporal associations between environmental drivers - climate and land cover - and bird species occupancy along a tropical montane gradient. I use an integrative approach that relies on citizen science, historical ecology, and bioacoustics to study bird communities in the Western Ghats biodiversity hotspot of southern India. In Chapter 1, we used observations from the world’s largest citizen science database, eBird, to ask how contemporary climate and land cover are associated with bird species occupancy across the Nilgiri and the Anamalai hills of the Western Ghats. We show that the occupancy of several forest specialist birds was negatively associated with temperature seasonality, highlighting narrow thermal niches for such species. We also show that a small number of generalist bird species are positively associated with human-modified land cover types. In Chapter 2, we combined colonial-era and modern datasets on bird species observations and land cover to ask how a century of landscape change across the Nilgiri hills has impacted bird communities. Between 1848 and 2017, 75% of grassland habitat across the Nilgiri hills was lost toward timber plantation and cash crop expansion. Such drastic declines in grassland habitat have resulted in declines in species persistence and relative abundance of grassland specialist birds over the last century. As a result, the functional trait space has undergone biotic homogenization. In Chapter 3, we ask if the reversal of landscape changes significantly affects bird communities. Using passive acoustic monitoring, we examined the impacts of ecological restoration on bird communities and other vocalizing fauna along a gradient of forest regeneration (consisting of actively restored, naturally regenerating, and mature benchmark sites) in the Anamalai hills. Encouragingly, we show that the bird community composition of actively restored sites is transitioning toward mature benchmark sites. However, when we moved beyond birds, we found that vocalizations at higher frequencies (> 12 kHz) were largely missing from actively restored and naturally regenerating sites, while the same frequency space was occupied in mature benchmark sites. Taken together, we find that climate and land cover are key determinants of bird species occupancy in the Western Ghats, and in a globally changing world, conservation interventions such as ecological restoration along with the preservation of naturally occurring land cover types are key to sustaining montane avifauna in the long run.

In this study I investigated patterns of invertebrate diversity in Limpopo Province indigenous forests, in order to highlight forests and taxa of special conservation significance. Invertebrates from seven target taxa were sampled in 11 patches of indigenous forest in Limpopo Province from February 2001 to January 2002, including six forests in the Soutpansberg and five forests in the northern Drakensberg. Selected forests comprise three distinct vegetation subtypes and the target taxa selected were millipedes, centipedes, earthworms, terrestrial molluscs, spiders, scorpions and amphipods. Invertebrates were sampled by active searching of quadrats and line transects and pitfall traps. A total of 11 969 indigenous target group individuals were sampled, comprising 14 orders, 50 families, 86 genera and 142 species (including at least nine new). There was a significant difference in the total invertebrate species richness and diversity of forest patches but results varied considerably when different target group figures were analyzed. With the exception of spiders, the factors influencing total and individual target group richness in forests could not be determined. Introduced invertebrates comprised a large proportion of the species and individuals sampled, but were not shown to affect indigenous fauna. Invertebrate species assemblages were most similar between forests sharing the same vegetation subtype and between forests in the same mountain region. However, each forest patch had unique species and some even had unique families. Limpopo Province forests support high numbers of endemic invertebrates. A total of 47 endemic invertebrate species were sampled, including six site endemics, eight local endemics, nine regional endemics and 24 national endemics. The numbers and scales of endemism varied by target group. Invertebrate species’ distributions in Limpopo Province forests generally support the biogeographic theories of Pleistocene forest refugia and the Limpopo River valley as a radiation barrier, although some important contradictions were found. Local endemism in Limpopo Province forests is likely the product of historical processes. Although some significant relationships were found between surrogate and true measures, single taxon biodiversity indicators, the higher taxon method, morphospecies and land classes could not accurately predict patterns of target invertebrate species richness in Limpopo Province forests. Results show that formal species identification should be used if accurate richness estimates are desired; the use of surrogates is not supported by this study. Conservation of Limpopo Province forests is vital for the preservation of valuable invertebrate communities. No forest sampled in this study can be considered unimportant. Effective forest conservation and management is dependent upon the protection of forests of varying patch size, careful evaluation and control of utilization and the establishment and maintenance of corridors linking isolated forest patches. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2004.

Not available / text

Non-forested ecosystems dominate high elevation sites in coastal British Columbia, yet there has never been a comprehensive classification or mapping of all high-elevation community types. The objective of this study is to collate and expand upon previous classifications, and thereby to increase our understanding of the habitats and composition of these plant communities.