Graduate Theses & Dissertations

Pages

Self-Organizing Maps and Galaxy Evolution
Artificial Neural Networks (ANN) have been applied to many areas of research. These techniques use a series of object attributes and can be trained to recognize different classes of objects. The Self-Organizing Map (SOM) is an unsupervised machine learning technique which has been shown to be successful in the mapping of high-dimensional data into a 2D representation referred to as a map. These maps are easier to interpret and aid in the classification of data. In this work, the existing algorithms for the SOM have been extended to generate 3D maps. The higher dimensionality of the map provides for more information to be made available to the interpretation of classifications. The effectiveness of the implementation was verified using three separate standard datasets. Results from these investigations supported the expectation that a 3D SOM would result in a more effective classifier. The 3D SOM algorithm was then applied to an analysis of galaxy morphology classifications. It is postulated that the morphology of a galaxy relates directly to how it will evolve over time. In this work, the Spectral Energy Distribution (SED) will be used as a source for galaxy attributes. The SED data was extracted from the NASA Extragalactic Database (NED). The data was grouped into sample sets of matching frequencies and the 3D SOM application was applied as a morphological classifier. It was shown that the SOMs created were effective as an unsupervised machine learning technique to classify galaxies based solely on their SED. Morphological predictions for a number of galaxies were shown to be in agreement with classifications obtained from new observations in NED. Author Keywords: Galaxy Morphology, Multi-wavelength, parallel, Self-Organizing Maps
Aeolian Impact Ripples in Sand Beds of Varied Texture
A wind tunnel study was conducted to investigate aeolian impact ripples in sand beds of varied texture from coarsely skewed to bimodal. Experimental data is lacking for aeolian megaripples, particularly in considering the influence of wind speed on ripple morphometrics. Additionally, the modelling community requires experimental data for model validation and calibration. Eighteen combinations of wind speed and proportion of coarse mode particles by mass were analysed for both morphometrics and optical indices of spatial segregation. Wind tunnel conditions emulated those found at aeolian megaripple field sites, specifically a unimodal wind regime and particle transport mode segregation. Remote sensing style image classification was applied to investigate the spatial segregation of the two differently coloured size populations. Ripple morphometrics show strong dependency on wind speed. Conversely, morphometric indices are inversely correlated to the proportion of the distribution that was comprised of coarse mode particles. Spatial segregation is highly correlated to wind speed in a positive manner and negatively correlated to the proportion of the distribution that was comprised of coarse mode particles. Results reveal that the degree of spatial segregation within an impact ripple bedform can be higher than previously reported in the literature. Author Keywords: Aeolian, Impact Ripples, Megaripple, Self-organization, Wind Tunnel
TESTING THE ROLE OF BIMODAL CELLS IN NEAR-HAND EFFECTS
We investigated whether hand-proximity effects arise from the recruitment of visual-tactile bimodal cells. In Experiment 1, we executed right-hand open-loop reaching movements to targets, presented either near or far from the resting left-hand, and after applying repetitive transcranial magnetic stimulation (rTMS) to suppress neuronal activity in the PMd and AIP, in Experiment 2. Results from Experiment 1, indicated that near-hand targets improved accuracy and reduced variability. In Experiment 2, suppressing PMd showed similar near-hand effects as above. In contrast, applying rTMS to AIP disrupted the representation of target location, indicating less accuracy and greater error. Near-hand proximity effects possibly arise from the recruitment of visual-tactile bimodal cells within the human AIP. Author Keywords: anterior intraparietal sulcus, peripersonal space, repetitive transcranial magnetic stimulation (rTMS), , rTMS, anterior intraparietal sulcus, AIP., visual-tactile bimodal neurons
Critical Analysis of the Adoption of Maize in Southern Ontario and its Spatial, Demographic, and Ecological Signatures
This thesis centers on analyzing the spatial, temporal, and ecological patterns associated with the introduction of maize horticulture into Southern Ontario - contextualized against social and demographic models of agricultural transition. Two separate analyses are undertaken: a regional analysis of the spread of maize across the Northeast using linear regression of radiocarbon data and a standard Wave of Advance model; and a local analysis of village locational trends in Southern Ontario using a landscape ecological framework, environmental data and known village sites. Through the integration of these two spatial and temporal scales of analysis, this research finds strong support for both migration and local development. A third model of competition and coalescence is presented to describe the patterning in the data. Author Keywords: Demographic Modeling, Environmental Modeling, Geostastical Analysis, Maize, Ontario Archaeology, Spread of Agriculture
Investigating the sources and fate of monomethylmercury and dimethylmercury in the Arctic marine boundary layer and waters
Monomethylmercury (MMHg), the most bioavailable form of mercury (Hg) and a potent neurotoxin, is present at elevated concentrations in Arctic marine mammals posing serious health threats to the local populations relying on marine food for their subsistence living. The sources of MMHg in the Arctic Ocean surface water and the role of dimethylmercury (DMHg) as a source of MMHg remain unclear. The objective of this research was to determine the sources and fate of methylated Hg species (MMHg and DMHg) in the marine ecosystem by investigating processes controlling the presence of methylated Hg species in the Arctic Ocean marine boundary layer (MBL) and surface waters. A method based on solid phase adsorption on Bond Elut ENV was developed and successfully used for unprecedented measurement of methylated Hg species in the MBL in Hudson Bay (HB) and the Canadian Arctic Archipelago (CAA). MMHg and DMHg concentrations averaged 2.9 ± 3.6 (mean ± SD) and 3.8 ± 3.1 pg m-3, respectively, and varied significantly among sampling sites. MMHg in the MBL is suspected to be the product of marine DMHg degradation in the atmosphere. MMHg summer (June to September) atmospheric wet deposition rates were estimated to be 188 ± 117.5 ng m-2 and 37 ± 21.7 ng m-2 for HB and CAA, respectively, sustaining MMHg concentrations available for bio-magnification in the pelagic food web. The production and loss of methylated Hg species in surface waters was assessed using enriched stable isotope tracers. MMHg production in surface water was observed from methylation of inorganic Hg (Hg(II)) and, for the first time, from DMHg demethylation with experimentally derived rate constants of 0.92 ± 0.82 x 10-3 d-1 and 0.04 ± 0.02 d-1 respectively. DMHg demethyation rate constant (0.98 ± 0.51 d-1) was higher than that of MMHg (0.35 ± 0.25 d-1). Furthermore, relationships with environmental parameters suggest that methylated Hg species transformations in surface water are mainly biologically driven. We propose that in addition to Hg(II) methylation, the main processes controlling MMHg production in the Arctic Ocean surface waters are DMHg demethylation and deposition of atmospheric MMHg. These results are valuable for a better understanding of the cycle of methylated Hg in the Arctic marine environment. Author Keywords: Arctic Ocean, Atmosphere, Demethylation, Dimethylmercury, Methylation, Monomethylmercury
Demography and habitat selection of Newfoundland caribou
The objective of this thesis is to better understand the demography and habitat selection of Newfoundland caribou. Chapter 1 provides a general introduction of elements of population ecology and behavioural ecology discussed in the thesis. In Chapter 2, I examine the causes of long-term fluctuations among caribou herds. My findings indicate that winter severity and density-dependent degradation of summer range quality offer partial explanations for the observed patterns of population change. In Chapter 3, I investigate the influence of climate, predation and density-dependence on cause-specific neonate survival. I found that when caribou populations are in a period of increase, predation from coyotes and bears is most strongly influenced by the abiotic conditions that precede calving. However, when populations begin to decline, weather conditions during calving also influenced survival. I build on this analysis in Chapter 4 by determining the influence of climate change on the interplay between predation risk and neonate survival. I found that the relative equilibrium between bears and coyotes may not persist in the future as risk from coyotes could increase due to climate change. In Chapter 5, I investigate the relationships in niche overlap between caribou and their predators and how this may influence differential predation risk by affecting encounter rates. For coyotes, seasonal changes in niche overlap mirrored variation in caribou calf risk, but had less association with the rate of encounter with calves. In contrast, changes in niche overlap during the calving season for black bears had little association with these parameters. In Chapter 6, I examine broad-level habitat selection of caribou to study trade-offs between predator avoidance and foraging during the calving season. The results suggest that caribou movements are oriented towards increased access to foraging and the reduction of encounter risk with bears, and to a lesser extent, coyotes. Finally, I synthesize the major findings from this thesis and their relevance to caribou conservation in Chapter 7, to infer that Newfoundland caribou decline is ultimately driven by extrinsic and intrinsic elements related to density-dependence. Reduction in neonate survival emerged from nutritionally-stressed caribou females producing calves with lower survival. Author Keywords: Behavioural ecology, Black bear (Ursus americanus), Coyote (Canis latrans), Population ecology, Predator-prey interactions, Woodland caribou (Rangifer tarandus)
GIS-based Spatial Analysis of Visibility and Movement Using the Ancient Maya Center of Minanha, Belize
It has long been hypothesized the location of the ancient Maya center of Minanha was a strategic one based on its ability to control the flow of communication and key resources between major geopolitical zones. Situated in the Vaca Plateau, at the nexus of the Belize River Valley, the Petén District of Guatemala, and the Maya Mountains, Minanha became a Late Classic polity capital that was tapped into a regional economy as well as long-distance trade networks. In this thesis I present a GIS-based spatial analysis that includes viewshed and cost surface analysis (CSA) to model visibility and movement within the north Vaca Plateau and neighboring regions to address specific questions concerning Minanha's strategic value. The results indicate that Minanha occupied a visually prominent location in proximity to major corridors of movement that suggest it was strategically, and in fact ideally located, as a polity capital with the ability to monitor the movement of people and resources. Author Keywords: Belize, GIS, Least Cost Path, Maya, Minanha, Viewshed
Factors Controlling Peat Chemistry and Vegetation Composition in Sudbury Peatlands after 30 Years of Emission Reductions
Peatlands are prevalent in the Sudbury, Ontario region. Compared with the well documented devastation to the terrestrial and aquatic ecosystems in this region, relatively little work has been conducted on the peatlands. The objective of this research was to assess factors controlling peat and plant chemistry, and vegetation composition in 18 peatlands in Sudbury after over 30 years of emission reductions. Peatland chemistry and the degree of humification varies considerably, but sites closer to the main smelter had more humified peat and the surface horizons were enriched in copper (Cu) and nickel (Ni). Copper and Ni concentrations in peat were significantly correlated with Cu and Ni in the plant tissue of leatherleaf, although the increased foliar metal content did not obviously impact secondary chemistry stress indicators. The pH and mineral content of peat were the strongest determining factors for species richness, diversity and community composition. The bryophyte communities appear to be acid and metal tolerant, although Sphagnum mosses are showing limited recovery. Author Keywords: anthropogenic emissions, bryophytes, community comspoition, heavy metals, peatlands, wetland vegetation
Variation in the δ15N and δ13C composition of POM in the Lake Simcoe watershed
The purpose of this study was to quantify the variation of baseline carbon and nitrogen stable isotope signatures in the Lake Simcoe watershed and relate that variation to various physicochemical parameters. Particulate organic matter samples from 2009 and 2011 were used as representatives of baseline isotopic values. Temporal data from two offshore lake stations revealed that δ15N of POM was lowest mid-summer and highest after the fall turnover. POM δ13C was variable throughout the summer before declining after fall turnover. Spatial data from the lake and the tributaries revealed that POM stable isotope signatures were highly variable. Various physicochemical parameters indicative of phytoplankton biomass were significantly positively correlated with POM δ15N and significantly negatively correlated with POM δ13C. The correlations were mostly significant in the tributaries, not the lake. Moreover, many of the correlations involving δ15N of POM were driven by extreme values in Cook's Bay and its tributaries. In general, it's likely that different processes or combination of processes were affecting the δ15N and δ13C POM in the Lake Simcoe watershed as physicochemical parameters alone could not explain the variability. Measuring the δ15N of ammonium and nitrate, as well as the δ13C of DIC would help discern the dominant nitrogen and inorganic carbon cycling processes occurring in the Lake Simcoe watershed. Author Keywords: δ13C, δ15N, isotopic baseline, particulate organic matter, spatial variation, stable isotopes
effects of environmental variables and dissolved organic matter characteristics on the diffusion coefficient of dissolved organic matter using diffusive gradients in thin films
The efficacy of the diffusive gradients in thin films (DGT) passive samplers to provide accurate measurements of free metal ions and those complexed with dissolved organic matter (DOM) was investigated. DOM controls the diffusive properties of DOM-complexed metal species in natural systems. Knowing the diffusion coeiffiecent (D) for DOM of different molecular weights (MW) and the major environmental variables influencing D is critical in developing the use of DGT passive samplers and understanding labile species. D and MW were determined for natural and standard DOM. No noticeable changes in DOM MW were observed during the diffusion process, suggesting that DOM remains intact following diffusion across the diffusive gel. Data analysis revealed that MW had the greatest influence on D, with a negative relationship between D and MW, except in tidal areas where ionic strength influence on D was significant. This study provides further characterization of the variables influencing D using the DGT technique. Author Keywords: Diffusion coefficient, Diffusive gradients in thin films, Dissolved organic matter, Flow field-flow fractionation, Principal Component Analysis, UV-Vis Spectroscopy
Phosphorus forms and response to changes in pH in acid-sensitive soils on the Precambrian Shield
Catchment soil acidification has been suggested as a possible mechanism for reducing phosphorus (P) loading to surface waters in North America and northern Europe, but much of the research that has been conducted regarding P immobilization in pH manipulated soils has been performed at high P concentrations (> 130 μM). This study investigated how soil acidity was related to P fractionation and P sorption at environmentally relevant P concentrations to evaluate the potential influence of long term changes in soil pH on P release to surface waters. Total phosphorus (TP) concentrations declined between 1980 and 2000 in many lakes and streams in central Ontario; over the same time period forest soils in this region became more acidic. Soils were collected from 18 soil pits at three forested catchments with similar bedrock geology but varying TP export loads. The soil pH at the 18 study soil pits spanned the historic soil pH range, allowing for `space for time' comparison of soil P factions. Soils were analysed by horizon for P fractions via Hedley P fractionation. Batch P sorption experiments were performed on selected B-horizon soils at varied solution pH. Soil P fractions varied by horizon but were comparable among the three catchments, with only apatite (PHCl) differing significantly across catchments. Contrary to expectation, both soluble and labile P showed negative relationships with pH in some horizons. Mineral soils were able to sorb almost all (> 90 %) of the P in solution at environmentally relevant P concentrations (4.5 - 45.2 μM). Phosphorus sorption at environmentally relevant P concentrations was unrelated to solution pH but at high P concentration there was a positive relationship between P sorption and solution pH, suggesting a P concentration dependant P sorption mechanism. Phosphorus budgets indicate that P is accumulating within catchments, suggesting that P is being immobilized in the terrestrial environment. An alternative hypothesis, which attempts to explain both the decline in stream TP export and terrestrial P accumulation, is discussed. The results from this study suggest that acidification induced P sorption in upland soils are not a contributing factor to decreases in stream TP concentration in the study catchments. Author Keywords: central Ontario, Hedley fractionation, phosphorus, podzols, soil acidification, sorption
Talking it out
The literature to date that investigates the development of social perspective taking in children primarily focuses on preschool aged children. These studies provide evidence that implicates language as being crucial for social perspective taking in young children but less is known about the importance of language to social perspective taking during middle childhood and early adolescence. The current study uses Selman's theory of socio-cognitive development to investigate the maturation of social perspective taking and the importance of language to social problem solving in 8 year olds (n = 111) and 12 year olds (n = 112). Analysis of variance and scalogram analysis shows a developmental progression of social perspective taking across the social problem solving process. Children may be able to demonstrate reciprocal perspective taking when generating strategies before they are able to demonstrate reciprocal perspective taking for other steps of social problem solving. Flexibility in interpersonal orientation is shown to be a predictor of social problem solving ability. Correlations and multiple regression analysis demonstrate that language is important to overall social problem solving but that the role of semantic and syntactic language may differ at age 8 compared to age 12. Author Keywords: interpersonal orientation, language, Selman, social cognition, social perspective taking, social problem solving

Pages

Search Our Digital Collections

Query

Enabled Filters

  • (-) ≠ Biology

Filter Results

Date

1974 - 2024
(decades)
Specify date range: Show
Format: 2024/02/25