Fire suppression has resulted in dramatic changes to species composition and structural diversity in the Sierra Nevada mixed-conifer forests of California. Advancing the ecological understanding and management of these forests requires a better understanding of changes that occurred during the fire suppression era, but empirical historical datasets are rare and methodologies for developing new historical reference information are subject to many limitations. I sought to develop historical reference information for the Yosemite Forest Dynamics Plot (YFDP), a research plot located in an old-growth mixed-conifer stand in Yosemite National Park. I performed a dendrochronological fire history analysis to characterize the historical fire regime of the YFDP (pre-1900 fire return interval: 29.5 years). I then developed two different forest reconstruction models to estimate pre-suppression forest conditions. Two alternative tree growth models, one regionally-parameterized and one locally-parameterized, and a decay model based on published estimates of tree decay rates were evaluated. Limited tree decay data available in the literature is a source of uncertainty in forest reconstructions, both for this study and other studies in the Sierra Nevada. Model analysis demonstrated that the regionally-parameterized growth model resulted in unreasonably fast tree growth rates. The site-specific growth model produced results similar to empirical historical datasets (84.5 trees ha-1 and 25.7 m2 ha-1 in 1900) I utilized these results to investigate patterns of tree establishment during the fire suppression era. I found evidence for spatial attraction between early ingrowth (trees that established between 1930 and 1970) sugar pine and legacy trees (trees established before 1930) and spatial repulsion between late ingrowth (trees that established after 1970) sugar pine and legacy trees. This indicates that fire suppression is driving changes to intertree relationships, causing current tree spatial patterns to be outside of their historical range and variability. These results highlight the need for a substantial increase in research efforts regarding tree decay data for Sierra Nevada mixed-conifer forests. When developing restoration targets, managers should avoid a one size fits all approach and consider site-specific factors, such as parent material and the historical fire regime, which have influenced changes to forest conditions during the fire suppression era.

In Canada’s eastern boreal forest region, partial-harvest silviculture has garnered increasing support for maintaining wildlife species and habitat structure associated with late-successional forests. If late-successional species can find suitable habitat in managed stands that retain a certain number, type, and pattern of live trees, then partial harvesting might represent a viable tool for maintaining species associated with old and complex forests. I used several indirect forms of inference to evaluate whether late-successional vertebrate species can be maintained within partially harvested stands in the eastern boreal forest. A meta-analysis of studies across North America showed that no bird species decreased in abundance by half where light harvesting retained at least 70% of live trees. However, adverse effects occurred at lower levels of retention, with some bird species unlikely to use harvested stands with less than 50% retention until appropriate habitat structure returned. A spatially explicit stand dynamics model showed that while partial harvesting can promote development of understory saplings, downed wood, and heterogeneity, it can also induce long-term decreases in the abundances of large trees and snags. Consequently, species dependent on the latter, such as brown creepers (Certhia americana) and northern flying squirrels (Glaucomys sabrinus), were projected to be more susceptible to partial harvesting than those associated with other types of structure. At a more detailed scale, a neighbourhood model developed from live-trapping data revealed that southern red-backed voles (Myodes gapperi) exhibited local associations with several late-successional features within boreal mixedwood stands. Their associations with some features depended on stand-level habitat conditions, which suggested that vole habitat in managed stands could be improved by retaining live trees and downed wood. A spatially explicit model of optimal home range establishment that incorporated these relationships fit vole abundance data marginally better than an aspatial habitat model. When the home range model was applied to simulated partially harvested stands, it predicted that spatial heterogeneity could have a positive effect on vole abundance, but only at harvest intensities of 70-90% with suppressed shrub cover. With careful attention to issues such as these, partial-harvest silviculture could be useful in maintaining vertebrate biodiversity within eastern boreal forests.

The question was asked: is it possible to realistically and spatially simulate the water balance of forests at large scales for long periods in a way which responds to land cover change. The answer to this question was pursued by attempting to construct a new model and applying it to a 163 km2 catchment area near Melbourne, Australia.

Thesis (Ph. D.)--Oregon State University, 1988. / Typescript (photocopy). Includes bibliographical references (leaves 127-138). Also available on the World Wide Web.

Thesis (M. S. - Renewable Natural Resources)--University of Arizona, 1990. / Includes bibliographical references (leaves 52-54).

Thesis (M.S.)--Oregon State College, 1952. / Typescript. Includes bibliographical references (leaf 44). Also available on the World Wide Web.

Thesis (doctoral)--Wageningen Universiteit, 2006. / Title from opening screen of PDF (viewed Jan. 8, 2008). Vita. Includes bibliographical references (p. 183-196). Also available in print.

Thesis (Ph.D.) in Ecology and Environmental Science--University of Maine, 2007. / Includes vita. Includes bibliographical references (leaves 103-119).

Thesis (M.S.)--Oregon State University, 1988. / Typescript (photocopy). Includes bibliographical references (leaves 49-51). Also available on the World Wide Web.

Thesis (M.S.)--Oregon State University, 1988. / Typescript (photocopy). Includes bibliographical references (leaves 68-69). Also available on the World Wide Web.