This research describes the design, construction, and characterization of a novel annular gap bioreactor. The bioreactor was conceived as a tool for the shear-based control of activated sludge floc particle size in batch and continuous-run experiments. Initial experiments on the feasibility of cultivating aerobic granular sludge in the bioreactor are described. The bioreactor was found to experience turbulent Taylor vortex flow, rather than laminar Couette flow, at all rotational speeds tested. This flow regime is the result of inner cylinder rotation. Despite turbulent flow conditions, the bioreactor was found to behave approximately as a plug-flow device when not aerated, and as a complete-mix reactor when aerated. Floc size control was found to depend on bioreactor rotational speed for two sludges tested, with higher rotational speed leading to smaller particle size. Three experimental attempts at aerobic granular sludge cultivation in the annular gap bioreactor are described in this study. Although none of the three attempts was successful at producing aerobic granules, the experiments allowed critical control issues related to bioreactor operation and influent composition to be identified and addressed. The Bacterial and Eukaryal population dynamics during each run were tracked with denaturing gradient gel electrophoresis and rDNA sequence analysis, using methods developed or streamlined in the course of this research. The foundational work described in this study culminated in the development of a series of protocols and recommendations for the next phase of aerobic granular sludge investigation with this novel annular gap bioreactor.

Prior information regarding the status of black bears (Ursus americanus) in Hunting District 100 in extreme northwestern Montana was confusing and oftentimes conflicting. To gather information on the population and the effects of hunting, especially during the spring hunting season, 13 black bears were radio-collared and monitored during 1986 and 1987. An additional 15 bears were marked and released. Information was gathered on the age and sex structure of the capture sample, survival and mortality rates of marked bears, reproduction, home range sizes, and hunter technique and effort. A density estimate of 1 black bear/15.4 km2 for the study area was determined. The age structure of captured bears was comprised primarily of adult males, which is generally recognized as indicative of a lightly or unexploited population. However, the mortality of marked bears due to hunting was found excessive. A decrease in the number of bears captured from 1986 to 1987, the large proportion of marked bears killed, and the 1987 spring observation data, all indicate that the study area population was well sampled by trapping. A reproductive rate of 0.70 cubs/female/year was determined, with at least 1 female not producing a successful litter until the age of 7. The mean home range size for adult males (5+ years) using the minimum convex polygon method was 98.7 km2 with a range of 35.5 to 192.6 km2 (n ;: 7). The home range size for adult females was 45.8 km2 with a range of 17.5 to 107.4 km2 (n = 4). Driving open roads was the primary hunting technique used during the spring hunting season. Thirty-nine percent of the bears killed during the spring season were shot within 100 mof an open road, while 66% of the bears were killed after first being observed from an open road.

A novel process using high-intensity ultrasonication (HIUS) was developed to isolate fibrils from cellulose fibers. The geometrical characteristics of the fibrils were investigated using polarized light microscopy (PLM), scanning electron microscopy (SEM) and atomic force microscopy (AFM). Results show that small fibrils with diameter ranging from about thirty nanometers to several micrometers were peeled from the fibers. The degree of fibrillation of the fibers was significantly increased. The crystallinities or molecular structures of most of the cellulose materials were changed by HIUS treatment. To evaluate the fibrils degradation by HIUS, a method using AFM was modified and developed to measure the elastic modulus of single cellulose fibrils. The results indicated that it was necessary to consider the penetration of AFM tips to the cellulose fibril surfaces. In the diameter range of 150 to 300 nm, the elastic moduli of Lyocell fibrils did not have significant differences between the HIUS treatment time of 30 min and 60 min. The modulus of Lyocell fibrils with diameters from 150 to 180 nm was evaluated about 98 GPa and it decreased dramatically when the diameter was more than 180 nm. The elastic moduli of cellulose fibrils were not significantly different between isolation methods of HIUS and high-pressure homogenizer for pure cellulose fiber, between different cellulose sources of pulp fibers treated by homogenizer. The elastic modulus of fibrils from regenerated cellulose fibers was higher than that of natural fibers. The treated fiber and separated fibrils were used to reinforce poly(vinyl alcohol), poly(lactic acid), and polypropylene by film casting or compression molding. Both of the tensile modulus and strength of nano-biocomposites reinforced with treated fiber and separated fibrils were higher than those of the untreated fiber reinforced composites. The morphological characteristics of the nanocomposites were investigated with SEM, AFM, and PLM. The dispersion of fibrils was not perfect, and the adhesions between the polymer and fibrils were not good without further modification of the fibrils. The fibrils on the fibers and isolated from the fibers may be the role that the tensile modulus and strength of the treated fiber and separated fibril reinforced composites were higher than those of the untreated fiber.

The forest products industry is undergoing unprecedented change from international competition, increasing fiber costs, rising energy prices and falling product prices. Competitive businesses have the key ability to adapt quickly to change through improved knowledge. Among adaptations to change are better product development, improved process efficiency and superior product quality. This dissertation is directly related to improving the knowledge of forest products manufacturers by investigating data mining (DM) methods that improve the ability to quantify causality of sources of variation. A contemporary DM method related to decision theory is decision trees (DTs). DTs are designed for heterogeneous data and are highly resistant to irrelevant regressors. The tree structures of DTs are also easy to interpret. The research hypothesis of this dissertation is that there is no significant difference in the explanatory or predictive capabilities of multiple linear regression (MLR) models, parametric regression trees (RTs) and non-parametric quantile RTs. To test this hypothesis 1,335 statistical models are developed. Box Cox transforms of Y are considered. Models are developed for the internal bond (IB) of medium density fiberboard (MDF) and the IB (and Parallel EI) of oriented strand board (OSB) from automatically fused data of destructive test data and real-time production line sensor data. Models with good predictability of the validation data set are possible for MDF IB when using traditional MLR methods with short record lengths without Box Cox transforms. Significant regressors (α < 0.01) for MDF MLR models are related to overall pressing time and press pre-position time settings. Parametric and non-parametric RT models without Box Cox transforms outperform the predictability of MLR models. For MDF IB, process variables related to overall pressing time, press position times and core fiber moisture are significant (α < 0.01). RT models with Box Cox transforms of OSB IB improve predictability for record lengths less than 100. Significant regressors (α < 0.01) of OSB IB are related to pressing times and core layer moisture. Significant regressors (α < 0.01) of OSB Parallel EI are related to forming speed and pressing times. There is evidence from the extensive investigation of 1,335 models to support the alternative research hypothesis.

Hardwood plantations are becoming increasingly important in the United States. To date, many foresters have relied on a conifer plantation model as the basis of establishing and managing hardwood plantations. The monospecific approach suggested by the conifer plantation model does not appear to provide for the development of quality hardwood logs similar to those found in natural hardwood stands. Thus, there is interest in creating mixed species plantations to simulate natural hardwood stand development. A simulation system, CherrybarkSQ, was developed to provide a platform for investigating the impacts of mixed species management of hardwood plantations on the stem quality of cherrybark oak (Quercus pagoda). Stem wood quality was defined by the proportion of clear wood (without knots) grown over a knotty core expressed by a Clear Wood Index (CWI) value. The construction of CherrybarkSQ consisted of developing models for predicting first-order branch characteristics and the occlusion of first-order branches for a distance-dependent individual tree model. CherrybarkSQ tracks the production of the knotty-core and clear wood over time through relationships developed between crown length and branch diameter and branch diameter and the overwood needed to occlude the branches. CherrybarkSQ was used to simulate the development of four hardwood plantation designs, three cherrybark oak monoculture designs (Pure, PureThin and Pure25) and a cherrybark oak and sweetgum mixed design (Mix) over a 50-year period. Among designs with similar initial stand density (Pure, PureThin and Mix), the Mix design produced the lowest CWI, an indication of greater clear wood production. The Mix plantation design consistently out-performed the Pure and PureThin designs. The low-density Pure25 design had the lowest CWI and the largest average diameter. The results of the CherrybarkSQ simulations indicate that when initial stand density is similar, a mixed species approach to cherrybark oak plantation management produces greater amounts of clear wood. A low-density monospecific approach produces a similar proportion of overwood, but the distribution of that overwood is on a shorter branch-free bole compared to the longer branch-free bole of trees at the greater density. The impact of hardwood plantation management decisions on clear wood or stem quality production, and therefore value, offers a great opportunity for evaluating hardwood plantations. Models and evaluations like those provided by CherrybarkSQ can assist with the interpretation, management and future development and value of mixed species plantations.

The Cumberland Plateau in Tennessee currently is experiencing widespread forest parcelization and changes in species composition as a result of changes in land use and ownership. These changes can be attributed partially to industrial forest land divestiture, the lingering effects of the 1998 – 2002 Southern Pine Beetle (SPB) epidemic, and expanding development due to increasing population growth. The region has recently become a focus of debate concerning land use change, land management practices, and the effects on biodiversity. A stratified random sample of 1600 Nonindustrial Private Forest (NIPF) landowners owning 40 or more acres of forestland were surveyed obtaining a 39 percent response rate. The survey was undertaken to gain new insight on the socio-demographics of Plateau NIPF landowners and to understand their forest management objectives and intentions for future timber harvesting activities. The findings reveal almost 50 percent of respondents were retired or employed as professionals and lived on or within 60 miles of their forest land. Ninety-one percent of all respondents had either purchased or inherited their forest land, and the majority indicated they intended to pass their forest land on to their heirs. The top three nonconsumptive ownership objectives were to enjoy scenery (m = 3.98), for peacefulness (m = 3.94), and to preserve nature (m = 3.83). Timber management was ranked as only moderately important (m = 2.60). Forty-five percent of all respondents indicated that they had previously sold or harvested timber from their forest land, but only 30 percent indicated they intended to sell timber in the future. Logit regression (n = 438) and factor analysis (n = 344) were used to model the respondents’ willingness to sell timber in the future. NIPF landowners who indicated they would most likely consider a future timber sale had sold timber in the past, had a higher interest in timber production, had received forest management advice in the past, and had a higher interest in maintaining the health of their forest. Factor analysis revealed landowners most likely to consider selling timber in the future fit into three principle component groupings: 1) Improvers; 2) Investors; 3) or Legacy Owners.

As the traditional forestry Extension clientele in Tennessee changes, Extension professionals need to adapt and deliver programs in such a way as to reach the maximum audience while ensuring the educational benefit of the program and remaining within budget. The development of the Internet and associated web-based programs presents a new tool by which Extension professionals can develop and deliver educational programs to clientele. This study compared two Extension program delivery methods, web-based and traditional field-based. A series of field workshops were held and participants were surveyed. From these field workshops, a web-based program was developed and was marketed to a similar audience; these participants were also surveyed. Demographic information, learning preferences, and other measures of program preference were collected. Ninety-seven percent of participants in the field-based program and 68% of the web-based sample preferred a field program over a web-based program. If only a web-based program was made available to participants, at least 50% of each sample would be interested. Educational level was the only demographic characteristic that was significantly different between those who chose a web-based program and those who chose a field-based program. Participants with higher educational levels were more likely to choose a web-based program. Educational level was positively correlated with income level, computer ownership, and Internet access. Educational level was negatively correlated with age. Knowledge gain was significantly higher in the field-based sample. Web-based participants showed average knowledge gain of 12.2 to 28.7% whereas the knowledge gain for people attending the traditional field day program averaged 16.5 to 46.1%. Level of computer ownership was high in both samples as was the use of the computer on a regular basis. Internet access was high in both groups. The web-based sample had significantly higher levels of high-speed access. Those with high-speed connections were more likely to choose a web-based program, although both groups used the computer/Internet regularly for gathering information. Current forestry Extension clientele preferred a field-based program delivery method. Some clientele, though, preferred web-based programming and these types of programs should be offered when suitable. Using a variety of delivery methods, Extension can adapt programs to a wide range of learning styles to reach new audiences, including younger generations who are more comfortable with Internet learning.

Partners in Flight (PIF) recommends using silviculture to improve breeding habitat conditions for migrant landbirds. Alternative thinning treatments may benefit priority landbird species by increasing structural complexity in second-growth forests. However, the effects of thinning on landbird populations in oak (Quercus spp.)-hickory (Carya spp.) forests have not been experimentally demonstrated. I used a randomized and replicated large-scale manipulative experiment to evaluate the effects of thinning (i.e., crown-release and gap creation) on forest habitat characteristics and avian populations at the Tennessee National Wildlife Refuge. I collected data during 2001 (pre-treatment) and from 2002 to 2005 (1 to 4 years post-treatment) in 20-ha thinned (n = 8) and control (n = 4) plots. Using mixed model ANOVA with covariates, I compared habitat attributes, tree regeneration, avian population densities, daily nest survival rates, realized brood sizes (# fledged per successful nest), rates of brown-headed cowbird (Molothrus ater) brood parasitism, and nest-site characteristics between treatments. In addition, I used Program MARK to evaluate the influence of habitat factors at multiple spatial scales on predation rates of Acadian flycatcher (Empidonax virescens) and wood thrush (Hylocichla mustelina) nests. Forest habitat attributes, avian population densities, and nest survival rates did not differ between control and thinned plots prior to treatment, indicating my experimental design (including the random allocation of treatments to plots, blocking, and the interspersion of plots across the study area) was sufficient for detecting treatment effects. Thinning resulted in a 29% difference in basal area between treatments (thinned = 20.3 m2 ha-1; control = 28.5 m2 ha-1). Compared to controls, thinned plots had significantly less overstory cover and midstory cover and significantly more downed wood and herbaceous and woody vegetation in the lower forest strata. Specifically, I detected greater densities of oak (Quercus spp.), yellow-poplar (Liriodendron tulipifera), and sourwood (Oxydenrum arboretum) saplings, and greater cover in poison ivy (Toxicodendron radicans) and blackberry (Rubus spp.) in thinned than control plots. I used spot-mapping to estimate the densities of PIF priority species. Thinning had positive effects on the densities of seven species (eastern towhee [Pipilo erythropthalmus], eastern-wood pewee [Contopus virens], indigo bunting [Passerina cyanea], Kentucky warbler [Oporornis formosus], white-eyed vireo [Vireo griseus], yellow-breasted chat [Icteria virens], and yellow-throated vireo [Vireo flavifrons]), inconclusive or negligible effects on the densities of two species (Louisiana waterthrush [Seiurus motacilla] and worm-eating warbler [Helmitheros vermivorus]), and negative effects on the densities of two species (Acadian flycatcher and wood thrush). I monitored 1,149 nests of 28 species. Predation accounted for 80% of all nest failures. Mayfield-adjusted nest daily survival rates of all species combined did not significantly differ between treatments. For all species combined, rates of cowbird parasitism varied annually but did not significantly differ between thinned (20.8%, SE = 2.3) and control (18.5%, SE = 3.7) plots. I assigned bird species to functional groups for further analyses. PIF priority mature-forest species exhibited nest daily survival rates (0.972 vs. 0.969), realized brood sizes (2.8 vs. 2.6), and parasitism rates (16.9 vs. 10.4%) that were comparable between thinned and control plots. Based on 162 nests in thinned plots, PIF shrubland species had nest daily survival rates of 0.958, realized brood sizes of 2.9, and parasitism rates of 13.6%; this functional group nested too rarely in control plots for analysis. Treatment effects were significant for the overstory and midstory nesting functional groups. Overstory nesters exhibited nest daily survival rates that were greater in thinned (0.982) than control (0.963) plots. Midstory nesters experienced greater parasitism rates in thinned (30.0%) than control (17.9%) plots. I evaluated nest-site selection and factors affecting nest predation rates using 132 Acadian flycatcher and 112 wood thrush nests. In thinned plots, both species selected nest sites with greater overstory and midstory cover than found at random. I found little evidence that nest predation rates were influenced by the amount of agriculture in the local (314 ha) landscape or by distance to anthropogenic edge, perhaps because the landscape was predominantly forested (agriculture < 4%) and most nests were >350 m from an edge. In thinned plots, predation rates on wood thrush nests decreased with increasing overstory cover and increasing basal area in large trees; predation rates increased with increasing basal area in small-diameter trees. None of the habitat predictors I measured had a strong relationship to Acadian flycatcher nest predation rates in thinned or control plots. Model-averaged nest survival estimates for wood thrushes were 27.8% and 26.8% in thinned and control plots, respectively. Acadian flycatcher model-averaged nest survival estimates were 53.5% in thinned and 56.4% in control plots. In summary, my results indicate that thinning had strong effects on forest habitat attributes and the demographics of some priority bird species. In the short term (1 to 4 years post-treatment), thinning appears to provide suitable breeding habitat for priority bird species that prefer dense understory vegetation or partially-opened overstories for nesting. Conversely, thinning had neutral or negative effects on some species and functional groups that nest in midstory vegetation, indicating there may be an ecological cost, in the short-term, associated with implementing this treatment. This treatment likely will have differential costs and benefits for avian populations as forest habitat conditions continue responding via successional dynamics and vegetative growth to the initial thinning operation.

Four experiments were performed in a greenhouse or growth chamber to study the quantity and quality of flavonolignans in response to environmental stress in milk thistle (Silybum marianum L., Gaertn.). A preliminary experiment was conducted in spring 2004 to study effects of leaf harvests on growth, development, and flavonolignan content in milk thistle seeds. Taxifolin, a component of silymarin and precursor to flavonolignans, was significantly reduced due to leaf harvest treatments. The weedy nature (sporadic germination) of Silybum made germinating enough plants for experimentation problematic. Initial germination studies to determine imbibition and pre-chilling requirements were inconclusive due to heavy fungal infections. A sterilization procedure was chosen to treat seeds before experimental use and flavonolignan analysis. Germination trials were also performed with seven other seed sources. Seeds harvested in Croatia showed the best germination and were used for subsequent experiments. Flavonolignans in Silybum seeds are reported to vary depending on environmental conditions and genetic diversity within a population. In another experiment, total silymarin concentrations determined from eight seed sources from around the world ranged from 29.6 to 56.9 mg/gram of seed meal. Individual flavonolignans varied significantly in and among seed sources. In the first stress experiment was, plants were grown in pine bark media in polyethylene bags. Plant densities from 1-24 plants/bag were established. Immature and total seed counts and yields decreased with increasing density. Number of blooms per plant, bloom diameter, and mature seed count and yield were negatively correlated to density. There was no significant effect of plant density on flavonolignan content. In the second stress experiment, milk thistle plants were grown in perlite in polyethylene bags. Water treatments (200, 650, 1100, 1550, and 2000 mL/day including fertigation) were created using pressure-compensated emitters. The lowest watering rate significantly reduced stem height and bloom diameter. The highest water treatment showed the highest count of immature seeds. In primary blooms, the lowest water rate yielded the highest taxifolin concentration (0.89 mg/g). Flavonolignan content was not significantly affected in secondary blooms. Silybum growth and development was affected by environmental stress. However, no significant effect on silymarin concentration or composition was established.

Episodic stream acidification occurs as storm events temporarily reduce acid neutralizing capacity (ANC) and pH. Stream acidification is suspected to have damaging effects on the health of aquatic ecosystems and biota. The objectives of this research are to 1) characterize stream baseflow and stormflow chemistries in three watersheds in the Great Smoky Mountains National Park (GRSM), 2) understand potential mechanisms responsible for episodic acidification, and 3) understand the relationship between storm event magnitude, antecedent soil moisture condition, and the stream’s pH response. Three remote, forested, high-elevation streams (Middle Prong, Ramsey Prong, and Eagle Rocks Prong) were selected in the Middle Prong of the Little Pigeon River Watershed. Multi-parameter data sondes were installed at each site to record continuous stream data. Autosamplers were set up in connection with the sondes to collect samples during storm events. Stormflow, baseflow, and precipitation samples were analyzed for pH, ANC, and a broad spectrum of cations and anions that contribute to the ion balance. During stormflow, ANC and pH depressions were observed for all storms at each study site. Sulfate, nitrate, and organic acid concentrations increased during each storm. Base cation concentrations generally increased during stormflow at Middle and Ramsey Prongs, but diluted occasionally on Eagle Rocks Prong. The relative changes in ion concentrations were used to determine which ions (acids) were most responsible for ANC depression. ANC contribution analysis indicates acid deposition may be the primary cause of episodic acidification, but it appears organic acids and cation dilution may also contribute. Pyritic geology is also suspected to contribute to baseflow and stormflow acidity in the Eagle Rocks Prong. Data exploration indicates large storms preceded by long, dry periods cause the largest pH depressions. It appears stream acidification may be driven by acid deposition, but additional inputs from varying vegetation and geology create unique and complex response to the observed stream acidification.