For longitudinal data where the response and time-dependent predictors within each individual are measured at distinct time points, traditional longitudinal models such as generalized linear mixed effects models or marginal models cannot be directly applied. Instead, some preprocessing such as smoothing is required to temporally align the response and predictors. In Chapter 2, we propose a binning method, which results in equally spaced bins of time for both the response and predictor(s). Hence, after incorporating binning, traditional models can be applied. The proposed binning approach was applied on a longitudinal hemodialysis study to look for possible contemporaneous and lagged effects between occurrences of a health event (i.e., infection) and levels of a protein marker of inflammation (i.e., C-reactive protein). Both Poisson mixed effects models and zero-inflated Poisson (ZIP) mixed effects models were applied to the subsequent binned data, and some important biological findings about contemporaneous and lagged associations were uncovered. In addition, a simulation study was conducted to investigate various properties of the binning approach. In Chapter 3, asymptotic properties have been derived for the fixed effects association parameter estimates following binning, under different data scenarios. In addition, we propose some leave-one-subject-out cross-validation algorithms for bin size selection. In Chapter 4, in order to identify levels of a predictor that might be indicative of recently occurred event(s), we propose a generalized mixed effects regression tree (GMRTree) based method which estimates the tree by standard tree method such as CART and estimates the random effects by a generalized linear mixed effects model. One of the main steps in this method was to use a linearization technique to change the longitudinal count response into a continuous surrogate response. Simulations have shown that the GMRTree method can effectively detect the underlying tree structure in an applicable longitudinal dataset, and has better predictive performance than either a standard tree approach without random effects or a generalized linear mixed effects model, assuming the underlying model indeed has a tree structure. We have also applied this method to two longitudinal datasets, one from the aforementioned hemodialysis study and the other from an epilepsy study.

Abscisic acid (ABA) was synthesised in dehydrating leaves and roots of willow (<I>Salix dasyclados</I>) and exogenous ABA in the xylem stream was shown to cause decreases in stomatal conductance. A transient decrease in leaf water potential occurred if water was entirely withheld from roots on one side of a willow plant. This was avoided if roots from all sides of the plant were watered and only root tips were allowed to dehydrate. Partial stomatal closure and decreased leaf extension rate then occurred without any initial perturbation in leaf water potential or leaf ABA. The drying event was associated with an increased content of ABA in root tips and xylem sap. The effects were reversible on either rewatering or excision of the affected root tips. It was concluded that partial dehydration of root tips caused partial stomatal closure and decreased leaf extension, and that changes in the ABA content of root tips and the xylem sap were consistent with a possible causal role for root-sourced ABA in the regulation of leaf physiology in response to root water deficit. Stem-girdling experiments indicated that a major pathway of ABA transport, between leaves on different stems in the shoot system, was in the phloem, without an apparent involvement of transport in the xylem. Damage to the shoot apex caused an increase in stomatal conductance. This was associated with a decreased content of ABA in the xylem sap and in fully extended leaves. It is suggested that these changes may have been associated with a possible import of ABA from mature leaves into the growth sites of axillary shoots. Results are discussed within the context of water deficit and the growth and survival of individual stems in a willow plantation.

A complex simulation model, MAESTRO, has been developed and validated against field measurements in plantation in both Scotland and Australia. It has been shown that MAESTRO can reasonably predict the daily course of PAR (photosynetically active radiation) transmittance at points below the canopies of radiata pine and Sitka spruce plantations. 1. Four structural properties of the Sitka spruce tree crown have been identified and evaluation in relation to PAR absorption, photosynthesis and transpiration. They are : the total amount leaves within the tree crown, leaf inclination angle distribution and crown shape. The total area of leaves and their spatial distribution within the tree crown are ranked as the two most significant properties. 2. Among the four different structural properties studied, crown shape is least important for PAR absorption, photosynthesis and transpiration. For the Sitka spruce stand studied, transpiration was most sensitive to the total area of leaves within the tree crown, whereas photosynthesis can be significantly affected by the leaf angle distribution. The influence of crown structure also depends on the beam fraction, radiance of diffuse radiation, incident angle of beam radiation. 3. Application of MAESTRO has shown the importance of the spatial distribution of leaf area density distribution and the age structure of the leaf population within a dense Sitka spruce canopy for the radiation regime and for photosynthesis. It has been found that the non-uniform canopy structure. The actual leaf age structure within a dense Sitka spruce canopy has been compared with the leaf age structure that is optimal for maximising the PAR use efficiency of the canopy and it has been found that PAR use efficiencies of both canopies are very close. 4. The non-uniform structure of a tree crown has been described using a two-dimensional leaf area density distribution function within the tree crown. It was found that the spatial distribution of leaf area density and the age structure of the leaf population within the crowns of radiata pine trees were significantly affected by water stress and nutrient deficiency during the growing season. 5. The radiance distribution of diffuse radiation from the sky and the beam fraction of the incident radiation also affect PAR absorption, photosynthesis and transpiration by altering the uniformity of the radiation regime within the tree crown. The light use efficiency of the canopy decreases with the increase in beam fraction. 6. The stomatal and photosynthetic properties of the shoots of Sitka spruce were measure in an open path gas exchange system to parameterize a biochemical photosynthesis model. It was found that co-limitation of photosynthesis plays an important role in the PAR responses of both an isolated shoot and of a canopy. It was shown that co-limitation of canopy photosynthesis enables the canopy to use a high flux density of incident PAR efficiently. 7. The dry matter production of the above-ground biomass of a Sitka spruce stand was studied in relation to PAR absorption to test the Monteith hypothesis. It was found the relationship between the above-ground dry matter production and PAR absorption is approximately linear. The slope of this linear relationship is affected by nitrogen fertilization, but not by thinning for the Sitka spruce stands studied. Crown structure plays an important role in the processes or radiation transfer, photosynthesis and transpiration within the tree crown. Detailed descriptions of the important structural properties of the tree crown are necessary for adequate understanding of canopy processes, such as photosynthesis. Co-limitation of crown photosynthesis, determined by the structure of all the tree crowns in the stand, results in an efficient use of the incident solar radiation by the trees in the stand. Photosynthesis of the forest canopy is roughly proportional to the incident PAR. PAR absorption is the primary determinant of dry matter production of the above-ground biomass for the Sitka spruce stand studied.

The dendrochronological potential of wood from two crannogs was evaluated. Consisting predominantly of non-oak species with fewer than 50 year-rings, the assemblages required the development of new techniques and approaches. The problem of validating cross-matches between short ring sequences was resolved by a new methodology based on internal consistency. A group of internally consistent sequences contains no conflicting chronological relationships in the full set of its pairwise correlations. The SORT.STRING program, designed to identify sets of internally consistent groups in large data-sets, was tested on two 'known' data-sets. It successfully identified the correct chronological relationships in these data-sets. Applied to the assemblages from the crannogs, it identified small, mutually exclusive groups of sequences which could not be satisfactorily merged to form a site chronology. Reasons for this are explored and indications for future work identified. The dendrochronological potential of alder (Alnus glutinosa)was examined and a suitable methodology developed. Cross-correlation and chronology formation are possible but extreme ring-width values, i.e signature years, and compression of the outermost rings re identified as problematic features. Evidence for woodland management practices in the crannog assemblages was evaluated. A model for woodland management was formulated, on the basis of two samples of modern coppiced material. This distinguishes between adventitious and formal coppice. Application of the model to hurdles found on the Irish crannog indicates that they are the products of adventituous coppice.

No description available.

No description available.

No description available.

The impact of a canker disease of Corymbia calophylla (marri) in the southwest of Western Australia (WA) has increased substantially since it was first observed causing decline and death of this species in the 1970s. By the early 1990s there were expressions of concern and calls to determine the cause and management options. Despite this, there has been very little research into the incidence, severity and possible causes of the disease. There are, however, historical reports dating back to the 1920s of a canker disease of amenity planted C. ficifolia caused by Sporotrichum destructor, though the diagnosis and Latin description were never published. It has been suggested that there may be links between this species and the genus Quambalaria, a group containing leaf and shoot pathogens of species of Eucalyptus and Corymbia. This study examined the incidence and symptomology of the disease, the range of fungal species associated with healthy and diseased C. calophylla, and the pathogenicity of isolates obtained from these surveys. Also investigated was the identity of the pathogen, S. destructor, historically attributed to canker disease of C. ficifolia, to determine whether this pathogen is responsible for the current epidemic of C. calophylla canker and if it is synonymous with Quambalaria. Cankers were present on trees across the range of surveyed sites, with lesions occurring on trunks, branches or twigs of 25.7 % of the C. calophylla surveyed. Canker incidence was significantly greater on trees present at remnant sites, such as roadsides and in paddocks, than forest trees being 38 % and 13.3 %, respectively. Tree height, trunk diameter at breast height and crown position, size and health ratings were significant predictors of canker presence, with cankers more common on larger, older trees and trees with poor crown condition. Bark cracks exuding kino were present on 48 % of the surveyed trees, and when dissected, lesions were observed on 40 %, suggesting that these cracks could be the initial stages of canker disease. This was confirmed by the observation of a number of cracks that developed into perennial cankers during the three year study. Monitoring of canker development and the examination of transverse sections showed the circumvention of host defenses by the pathogen and the subsequential walling off response of the host (which typifies perennial cankers) was not necessarily an annual event, with no change observed in some cankers over the three year period, while others progressed rapidly in that time, occasionally to the point of girdling and killing the host. Initial surveys isolated 44 fungal species from healthy and diseased C. calophylla, with opportunistic pathogens including Endothiella eucalypti and Cytospora eucalypticola common. Subsequent surveys foccussed more on a potential pathogen in the genus Quambalaria, which was rarely isolated from active lesions, presumably because of its slow growth rate, but which sporulated consistently on the surface of older sections of the cankers. DNA sequences confirmed that Q. cyanescens and Q. pitereka are present in southwest WA, with the latter associated with leaf and shoot disease. A third and new species of Quambalaria was isolated from cankers. Comparisons of disease symptoms and conidiogenesis indicate this species is synonymous with S. destructor. The species is formally described here as Q. coyrecup T. Paap sp. nov. A pathogenicity trial was unsuccessful in causing disease symptoms in trees inoculated with core plugs taken from canker lesion margins of diseased trees, though the time frame and environmental factors may not have been adequate for disease development. The core plug inoculation method may also have failed because opportunistic pathogens which were frequently isolated from lesions out-competed Q. coyrecup (paralleling the results achieved by culturing from lesions). Quambalaria coyrecup caused symptoms matching those observed in natural infections when suitable hosts were inoculated, confirming it is the fungus responsible for the current canker disease of C. calophylla and C. ficifolia. Endothiella eucalypti also caused significant lesions, though these were not typical of natural infections, which together with its frequent isolation from both healthy and diseased trees suggests it is an opportunistic pathogen, potentially contributing to disease development in trees already infected with Q. coyrecup. Isolates of Q. pitereka from WA and eastern Australia both caused typical shoot blight symptoms in the WA hosts C. calophylla and C. ficifolia, and the eastern Australian host C. maculata, though a larger path trial is required to examine the possibility of host specificity. Quambalaria cyanescens was non-pathogenic in all inoculation trials. The current cause of cankers in C. calophylla is now known to be the same as the fungus historically implicated in the canker disease of C. ficifolia, when at the time it was described as an endophyte doing little or no damage in C. calophylla. Thus, it is of immediate importance to determine the factors driving this decline, and develop control and management options.

Thesis (M.S.)--University of Arizona, 1993. / Includes abstract. Includes bibliographical references (p. 134-138).

Thesis (Ph. D.)--Oregon State College, 1955. / Typescript. Includes bibliographical references (leaves 150-158). Also available online.