Geology of the Gander Group, Gander River Ultramafic Belt and Davidsville Group in the Jonathan's Pond - Weir's Pond Area, northeast Newfoundland /

The study area is situated in NE Newfoundland between Gander Lake and the north coast and on the boundary between the Gander and Botwood tectonostratigraphic zones (Williams et al., 1974). The area is underlain by three NE trending units; the Gander Group, the Gander River Ultramafic Belt (the GRUB) and the Davidsville Group. The easternmost Gander Group consists of a thick, psammitic unit composed predominantly of psammitic schist and a thinner, mixed unit of semipelitic and pelitic schist with minor psammite. The mixed unit may stratigraphically overlie the psammitic unit or be a lateral facies equivalent of the latter. No fossils have been recovered from the Gander Group. The GRUB is a terrain of mafic and ultramafic plutonic rocks with minor pillow lava and plagiogranite. It is interpreted to be a dismembered ophiolite in thrust contact with the Gander Group. The westernmost Davidsville Group consists of a basal conglomerate, believed deposited unconformably upon the GRUB from which it was derived, and an upper unit of greywacke and slate, mostly of turbidite origin, with minor limestone and calcareous sandstone. The limestone, which lies near the base of the unit, contains Upper Llanvirn to Lower Llandeilo fossils. The Gander and Davidsville Groups display distinctly different sedimentological , structural and metamorphic histories. The Gander Group consists of quartz-rich, relatively mature sediment. It has suffered three pre-Llanvirn deformations, of which the main deformation, Dp produced a major, NE-N-facing recumbent anticline in the southern part of the study area. Middle greenschist conditions existed from D^ to D- with growth of metamorphic minerals during each dynamic and static phase. In contrast, the mineralogically immature Davidsville Group sediment contains abundant mafic and ultramafic detritus which is absent from the Gander Group. The Davidsville Group displays the effects of a single penetrative deformation with localized D_ and D_ features, all of which can be shown to postdate D_ in the Gander Group. Rotation of the flat Gander S- into a subvertical orientation near the contact with the GRUB and the Davidsville Group is believed to be a Davidsville D^ feature. Regional metamorphism in the Davidsville Group is lower greenschist with a single growth phase, MS . These sedimentological, structural and metamorphic differences between the Gander and Davidsville Groups persist even where the GRUB is absent and the two units are in contact, indicating that the tectonic histories of the Gander and Davidsville Groups are distinctly different. Structural features in the GRUB, locally the result of multiple deformations, may be the result of Gander and/or Davidsville deformations. Metamorphism is in the greenschist facies. Geochemical analyses of the pillow lava suggest that these rocks were formed in a back-arc basin. Mafic intrusives in the Gander Group appear to be the result of magraatism separate from that producing the pillow lava. The Gander Group is interpreted to be a continental rise prism deposited on the eastern margin of the Late Precambrian-Lower Paleozoic lapetus Ocean. The GRUB, oceanic crust possibly formed in a marginal basin to the west, is believed to have been thrust eastward over the Gander Group, deforming the latter, during the pre-Llanvirnian, possibly Precambrian, Ganderian Orogeny. The Middle Ordovician and younger Davidsville Group was derived from, and deposited unconformably on, this deformed terrain. Deformation of the Davidsville Group occurred during the Middle Devonian Acadian Orogeny.

The effects of static and diurnally-cycling temperature acclimations on thermal tolerances of rainbow trout, Salmo gairdneri

:ofiedian lethal temperatures ( LT50' s ) were determined for rainbow trout, Salmo gairdnerii, acclimated for a minimum of 21 days at 5 c onstant temperatures between 4 and 20 0 C. and 2 diel temperature fluctuations ( sinewave curves of amplitudes ± 4 and ± 7 0 C. about a mean temperature of 12 0 C. ) . Twenty-four-, 48-, and 96-hour LT50 estimates were c alculated f ollowing standard flow-through aquatic bioassay techniques and probi t transformation of mortality data. The phenomenon of delayed thermal mortality was also investigated. Shifts in upper incipient lethal temperature occurred as a result of previous thermal conditioning. It was shown that increases in constant acclimation temperature result in proportional l inear increases in thermal tolerances. The increase i n estimated 96-hour LT50's was approximately 0.13 0 c. X 1 0 C:1 between 8 and 20 0 C. The effect of acclimation to both cyclic temperature regimes was an increase in LT50 to values between the mean and maximum constant equivalent daily temperatures of the cycles. Twenty-four-, 48-, and 96-hour LT50 estimates of both cycles corresponded approximately to the LT50 values of the 16 0 C. c onstant temperature equivalent . This increase i n thermal tolerance was further demonstrated by the delayed thermal mortality experiments . Cycle amplitudes appeared to i nfluence thermal resistance through alterations in initi al mortality since mortality patterns characteristic of base temperature acclimations re-appeared after approximately 68 hours exposure to test temperatures for the 12 + 4 0 C. group, whereas mortality patterns stabilized and remained constant for a period greater than 192 hours with the larger therma l cycle ( 12 + 7 0 C. ). NO s ignificant corre lations between s pecimen weight and time-to-death was apparent. Data are discussed in relation to the establishment of thermal criteria for important commercial and sport fishes , such as the salmonids , as is the question whether previously reported values on lethal temperature s may have been under estimated.

Evolutionary synthesis of stochastic gene network models using feature-based search spaces

A feature-based fitness function is applied in a genetic programming system to synthesize stochastic gene regulatory network models whose behaviour is defined by a time course of protein expression levels. Typically, when targeting time series data, the fitness function is based on a sum-of-errors involving the values of the fluctuating signal. While this approach is successful in many instances, its performance can deteriorate in the presence of noise. This thesis explores a fitness measure determined from a set of statistical features characterizing the time series' sequence of values, rather than the actual values themselves. Through a series of experiments involving symbolic regression with added noise and gene regulatory network models based on the stochastic 'if-calculus, it is shown to successfully target oscillating and non-oscillating signals. This practical and versatile fitness function offers an alternate approach, worthy of consideration for use in algorithms that evaluate noisy or stochastic behaviour.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and deterministic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel metaheuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS metaheuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

Feature Selection and Classification Using Age Layered Population Structure Genetic Programming

The curse of dimensionality is a major problem in the fields of machine learning, data mining and knowledge discovery. Exhaustive search for the most optimal subset of relevant features from a high dimensional dataset is NP hard. Sub–optimal population based stochastic algorithms such as GP and GA are good choices for searching through large search spaces, and are usually more feasible than exhaustive and determinis- tic search algorithms. On the other hand, population based stochastic algorithms often suffer from premature convergence on mediocre sub–optimal solutions. The Age Layered Population Structure (ALPS) is a novel meta–heuristic for overcoming the problem of premature convergence in evolutionary algorithms, and for improving search in the fitness landscape. The ALPS paradigm uses an age–measure to control breeding and competition between individuals in the population. This thesis uses a modification of the ALPS GP strategy called Feature Selection ALPS (FSALPS) for feature subset selection and classification of varied supervised learning tasks. FSALPS uses a novel frequency count system to rank features in the GP population based on evolved feature frequencies. The ranked features are translated into probabilities, which are used to control evolutionary processes such as terminal–symbol selection for the construction of GP trees/sub-trees. The FSALPS meta–heuristic continuously refines the feature subset selection process whiles simultaneously evolving efficient classifiers through a non–converging evolutionary process that favors selection of features with high discrimination of class labels. We investigated and compared the performance of canonical GP, ALPS and FSALPS on high–dimensional benchmark classification datasets, including a hyperspectral image. Using Tukey’s HSD ANOVA test at a 95% confidence interval, ALPS and FSALPS dominated canonical GP in evolving smaller but efficient trees with less bloat expressions. FSALPS significantly outperformed canonical GP and ALPS and some reported feature selection strategies in related literature on dimensionality reduction.

New Feature at Niagara – Clark Hill Islands

New Feature at Niagara – Clark Hill Islands (5 islands situated in the rapids of the Niagara River). These islands are currently known as Dufferin Islands, 22 ½ cm. x 15 ½ cm, n.d.