Thesis (M.S.)--Oregon State University, 2010. / Printout. Includes bibliographical references (leaves 87-92). Also available on the World Wide Web.

Thesis (M.S.)--Oregon State University, 2009. / Printout. Includes bibliographical references (leaves 58-61). Also available on the World Wide Web.

Disturbance is fundamental to forest ecosystem function and overall health, but climate change is likely to increase both disturbance frequency and intensity in the future. Forests subject to increasingly frequent and intense disturbances are more likely to experience compounding disturbance effects. Compounding disturbances may exert unpredicted, non-additive stresses on ecosystems, leading to novel conditions that may exceed the capacity for local species to survive and regenerate. I further hypothesize that compounding disturbances could create conditions misaligned with species’ adaptations by altering physical and chemical growing conditions in forest soils, affecting forest composition, structure, and, subsequently, function for many years following disturbance. A better understanding of these remnant effects will be essential to managing and conserving coast redwood forests, which are projected to see increased frequency of fire under future climate scenarios. My objectives in this study were to quantify the effects of time-since-fire and single vs. compounding disturbances on coast redwood forest structure, composition, and regeneration dynamics and to evaluate the effects of abiotic soil qualities on post-fire regeneration. I mapped and sampled coast redwood forests burned in 1985, both 1985 and 1999, 2008, and 2013; modeled regeneration as a function of burn history, understory light, and post-fire nutrient levels; and tested redwood seed regeneration in post-fire soils in a greenhouse experiment. Forest structure, composition, and regeneration following compounding disturbance were most similar to the homogenous, redwood-dominated forest of the recent 2013 burn. There were no unique effects of compounding disturbance on soil nutrient levels, although variations in nutrient levels generally followed patterns seen in previous studies. Soil nitrate was positively associated with coast redwood regeneration levels, showing that soil nutrients may be influential in regeneration processes following disturbance. Time since burn and single burn histories were negatively associated with regeneration levels in the field, and there were no differences in seed germination in the greenhouse between soils from different fire histories. Increases in coast redwood forest dominance accompanied declines in bay laurel and tanoak presence, indicating a shift in post-fire forest structure and composition resulting from compounding disturbance. These results illustrate a complex relationship between regeneration dynamics, post-fire soil quality, and disturbance histories. Forest homogenization from compounding disturbances may have negative implications for ecosystem services and overall function if compounding disturbances are more frequent as predicted under future climate conditions.

Phytophthora ramorum, an oomycete plant pathogen, is the causal agent of sudden oak death, a serious disease of Fagaceous trees in California and Oregon over the last decade. Tanoak (Notholithocarpus densiflorus) is one of the most susceptible host species, but the cause of host mortality is poorly understood. Previous research has implicated disruption in stem water transport, phloem girdling, and activity of a class of secreted proteins known as elicitins as possible mechanisms of pathogenesis. In this study I investigated certain physiological impacts of P. ramorum infection on tanoak saplings and tanoak seedlings. In growth chamber experiments, stems of plants were inoculated with isolates that differed in the amount of elicitin secreted in vitro. Stem-wounded, non-inoculated plants served as controls. Parameters measured included net photosynthetic rate, stomatal conductance, whole plant water usage, stem specific hydraulic conductivity, tylosis production, starch partitioning, and mortality. Inoculated saplings exhibited a reduction in whole plant water usage, followed by a reduction in stem specific hydraulic conductivity implicating an interruption in stem water transport as the primary symptom. A reduction in net photosynthetic rate and stomatal conductance occurred one week later. Experiments conducted on inoculated tanoak seedlings supported the hypothesis that a reduction in stem water transport is the primary disease symptom. Stem specific hydraulic conductivity was the only parameter that appeared to be significantly impacted when treatments were compared during each measurement period. There was, however, a significant difference between treatments over the course of the entire experiment. Due to differences in isolate growth rates and similar levels of elicitin secretion, symptom expression could not be tied to elicitin production. To determine where elicitins are produced in planta, an immunolabeling technique was tested utilizing an elicitin-specific fluorescent antibody. The elicitin protein was most apparent in paratracheal parenchyma cells, although nonspecific staining in control samples confounded interpretation. / Graduation date: 2012

The phytopathogen Phytophthora ramorum (Werres, DeCock & Man in't Veld), causal agent of Sudden Oak Death (SOD) of oaks (Quercus spp.) and tanoaks (Notholithocarpus densiflorus syn. Lithocarpus densiflorus), is established in coastal forests of the western United States. Since the discovery of SOD in the Douglas-fir / tanoak forests of southwest Oregon in 2001, a multiagency effort has ensued with the goal of fully eliminating P. ramorum from this originally small and isolated area. In this study we investigated the epidemiology of SOD in Oregon, particularly as it affects the success of the eradication program. Two approaches were taken to discern the mechanism of long distance dispersal: first, a landscape analysis of the spatial relationship between SOD sites and roads or streams, features associated with movement of infested soils, and, second, a local analysis to discern if understory infection is originating from soil or stream-borne inoculum. Using a restricted randomization test we concluded that SOD sites were no closer to roads than expected by chance, which is inconsistent with soil dispersal by people. While we found evidence that SOD sites were preferentially closer to waterways, inoculum had not moved away from streams into adjacent understory foliage. The local distribution of understory infection around SOD positive trees indicated that primary inoculum is infecting overstory canopies first, suggesting that P. ramorum is dispersing in air currents. Regression modeling indicated that weather conditions two years before detection could explain variation in the maximum distance inoculum moved each year of the epidemic between 2001 and 2010. This two year delay between infection and detection has allowed ample time for infested sites to contribute to further spread. Model results were consistent with observations made the summer of 2011, when trees likely infected by secondary inoculum at non-eradicated sites developed symptoms but were still undetectable by aerial surveys. Due to the prevalence of infection on tanoak, opportunities for sporulation and infection occur more often in Oregon than in California. These data can explain the failure to eliminate P. ramorum. Nevertheless, we did find evidence that the eradication program has significantly reduced the potential size of the SOD epidemic in Oregon. / Graduation date: 2012