Thesis (Ph. D.) in Forest Resources--University of Maine, 2000. / Includes vita. Includes bibliographical references (leaves 134-143).

No description available.

A method was developed for the isolation of tissue fractions from the middle lamella and secondary wall of black spruce (Picea mariana) tracheids. The concentrations of cellulose and glucomannan were found to be lower in the middle lamella tissue than in the secondary wall tissue, while the reverse was found for galactan and arabian. The content of glucuronoarabinoxylan was essentially the same in both morphological areas. The concentration of phenolic hydroxyl groups in the lignin in the two tissue fractions was determined by turbidimetric spectrophotometry and by pyrolytic gas chromatography. Secondary wall lignin was found to contain more than twice as many phenolic hydroxyl groups per phenylpropane unit as middle lamella lignin. The concentration of methoxyl groups in secondary wall lignin was 1.6 times the concentration in middle lamella lignin, indicating that about 40% of the lignin polymer in the middle lamella consisted of para-hydroxyphenyl residues. Most of the carbonyl groups in the lignin in wood were found in the middle lamella lignin. The carboxyl content of the middle lamella was about three times that of the carboxyl content of the secondary wall. Secondary wall lignin was more reactive towards chlorine, chlorine dioxide and sodium bisulphite. Secondary wall lignin was also more quickly dissolved than middle lamella lignin in kraft, acid-sulphite, and acid-chlorite pulping. The activation energies, towards kraft pulping, of middle lamella and secondary wall "bulk" lignin were found to be essentially the same. However, it was discovered that while most of the secondary wall lignin followed first order reaction kinetics, 70% of middle lamella lignin did not. It was also discovered that the majority of milled wood lignin originated in the secondary wall of the tracheid.

This study develops forestry growth and yield models for two economically important tree species in Manitoba, black spruce [Picea mariana (Mill.) BSP] and jack pine [Pinus banksiana Lamb]. The growth and yield models developed include regression-based individual tree height growth and site index, tree diameter (basal area) growth, tree bole taper, and individual tree mortality models. These regression-based models were developed empirically, using stem analysis, growth and mortality data from 80 permanent sample plots located within the commercially important boreal forests of Manitoba. Model development involved the exploration, comparison and testing of numerous potential regression models and predictor variables. Statistical issues commonly encountered in forest growth and yield modeling, particularly data autocorrelation and variable multicollinearity, were addressed using nonlinear least squares (NLS), generalized nonlinear least squares (GNLS), and nonlinear mixed-effects model regression (NLMM) approaches. Height growth and site index of black spruce and jack pine was modelled using a three-parameter generalized logistic function. NLMM regression was used since the data were spatially autocorrelated. The inclusion of prior measures from individual trees produced more accurate predictions. In the tree diameter (basal area) growth models, tree size variables were significant predictors for black spruce and managed jack pine stands. Site index (a measure of site productivity) was positively correlated, and basal area of trees larger than the target tree (a relative measure of competition) negatively correlated, with diameter increment. Thiessen polygon area, a spatial measure of competition, was a significant predictor for natural jack pine and upland black spruce stands. Tree bole taper was modeled by NLMM approach using a five-parameter equation based on dimensional analysis, with breast height diameter, total height and relative height as predictor variables. The inclusion of a single prior measure from each tree improved model prediction. Black spruce and jack pine mortality was modeled using logistic regression. The black spruce models predicted high survivorship for larger, fast-growing trees in less crowded stands. In the jack pine model, highest survivorship was predicted for larger, less locally crowded trees.

This study develops forestry growth and yield models for two economically important tree species in Manitoba, black spruce [Picea mariana (Mill.) BSP] and jack pine [Pinus banksiana Lamb]. The growth and yield models developed include regression-based individual tree height growth and site index, tree diameter (basal area) growth, tree bole taper, and individual tree mortality models. These regression-based models were developed empirically, using stem analysis, growth and mortality data from 80 permanent sample plots located within the commercially important boreal forests of Manitoba. Model development involved the exploration, comparison and testing of numerous potential regression models and predictor variables. Statistical issues commonly encountered in forest growth and yield modeling, particularly data autocorrelation and variable multicollinearity, were addressed using nonlinear least squares (NLS), generalized nonlinear least squares (GNLS), and nonlinear mixed-effects model regression (NLMM) approaches. Height growth and site index of black spruce and jack pine was modelled using a three-parameter generalized logistic function. NLMM regression was used since the data were spatially autocorrelated. The inclusion of prior measures from individual trees produced more accurate predictions. In the tree diameter (basal area) growth models, tree size variables were significant predictors for black spruce and managed jack pine stands. Site index (a measure of site productivity) was positively correlated, and basal area of trees larger than the target tree (a relative measure of competition) negatively correlated, with diameter increment. Thiessen polygon area, a spatial measure of competition, was a significant predictor for natural jack pine and upland black spruce stands. Tree bole taper was modeled by NLMM approach using a five-parameter equation based on dimensional analysis, with breast height diameter, total height and relative height as predictor variables. The inclusion of a single prior measure from each tree improved model prediction. Black spruce and jack pine mortality was modeled using logistic regression. The black spruce models predicted high survivorship for larger, fast-growing trees in less crowded stands. In the jack pine model, highest survivorship was predicted for larger, less locally crowded trees.

The growth and productivity of British Columbia’s interior forests is largely limited by soil nutrient availability. Fertilization has been shown to be an effective silvicultural tool for increasing the development of immature stands throughout the region. This has lead to increased interest in long-term, repeated fertilization as a means of addressing timber-supply shortfalls as a result of the current mountain pine beetle (Dendroctonus ponderosae) outbreak. However, there is little information related to the impacts of repeated fertilization on the cycling of nutrients in many of these stands. This study makes use of a long-term (13-15 year) fertilization experiment in two lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm) and two interior spruce (Picea glauca [Moench] Voss and Picea engelmannii Parry) forests in the central interior of British Columbia subject to two levels (periodic and annual) of nitrogen(N)-based fertilization. The primary goal of the project was to examine the effects of different fertilizer regimes on aspects of soil chemistry. Specifically, this project was concerned with the impacts of repeated fertilization on: 1) soil carbon (C) and N cycling, and 2) soil base cation (e.g., Ca, Mg & K) availability. Soil and foliar nutrient regimes were quantified throughout the 2008 and 2009 growing seasons using ion-exchange membrane (IEM) plant root simulator (PRS) probes and traditional soil and foliar analyses. Fertilization increased N cycling at all sites, with generally elevated soil and foliar N and significant soil-foliar N relationships in several cases. Nitrate (NO3 -) increased in the fertilized plots in several cases; however, there was minimal evidence of NO3 - leaching. Greater than 90% of fertilizer-N inputs were retained onsite, suggesting these forests are not N-saturated. Soil, tree and total ecosystem C generally increased in response to fertilization, with the spruce sites exhibiting greater C accrual per unit of fertilizer N than the pine sites. Further, significant linear relationships between soil C and N were evident at all sites. At sites with poorly buffered soils (pH < 4), fertilizer treatments generally led to increased soil acidification and decreases in soil and foliar Ca. Decreases in soil Ca may have been due to significant increases in sulfate leaching; whereas foliar Ca decreases appear to be related to compromised uptake systems, potentially from increased soil aluminum. Buffering capacities, rather than forest type, appear to be the best predictor of soil and foliar Ca responses to fertilization. Despite significant changes in soil chemistry at all four sites, it does not appear that current fertilization rates are detrimentally affecting tree growth. / Graduate

Significant species differences in foliar litter quality have been well-documented but differences within species, as a function of variation in genotype or environment, have not been directly addressed and, thus, remain poorly understood. We studied the significance of genetics, the environment, and their interaction in regulating forest nutrient dynamics and productivity using trees from four full-sib black spruce (Picea mariana (Mill.) B.S.P.) families of a complete 7 x 7 diallel cross experiment at the Petawawa National Forestry Institute (Chalk River, Ontario). Foliar litter composites from each family-plot on two sites of contrasting water availability were collected in October, 1993, and sub-samples were analyzed for initial N, P, K, Ca, Mg, cellulose and lignin concentrations. Inherent decomposition rates and nutrient dynamics of the litter composites were followed for one year. Samples of forest floor and mineral soil collected from within each plot were used to determine organic matter accumulation, pH, total nutrient contents in the forest floor and extractable nutrient concentrations in the mineral soil. Height, dbh, and total tree cubic volume, were used to characterize tree productivity. (Abstract shortened by UMI.)

Thesis (M. Sc.)--Carleton University, 2003. / Includes bibliographical references (p. 67-75). Also available in electronic format on the Internet.

Mémoire (M.Ress.Renouv.) -- Université du Québec à Chicoutimi, 1992. / Document électronique également accessible en format PDF. CaQCU

Thèse (N. Ress. Renouv.) -- Université du Québec à Chicoutimi, 2004. / Bibliogr.: f. [46]-54. Document électronique également accessible en format PDF. CaQCU