Thèse (M.Sc.Geogr.)--Université Laval, 2006. / Titre de l'écran-titre (visionné le 18 sept. 2007). Bibliogr.

Thesis (Ph. M.)--University of Wisconsin--Madison, 1945. / Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 85-88).

No description available.

<p>This thesis describes a grey-box model for the dimensionalstability properties (i.e. curl and twist) of the carton boardproduced at AssiDomän Frövi paper mill in Sweden.AssiDomän Frövi AB is one of Sweden major cartonboard manufacturer, and produces some 350000 ton of board peryear.</p><p>Curl is defined as the departure from a at form, and it mayseriously affect the processing of the paper. For this reason,customers impose quite restrictive limits on the allowedcurvatures of the board. So, it is becoming more and moreimportant to be able to produce a carton board with a curlwithin certain limits. Due to the economic significance of thecurl problem, much research has been performed to find sheetdesign and processing strategies to eliminate or reducecurl.</p><p>The approach we used to tackle this problem is based ongrey-box modelling. The reasons for such an approach is thatthe physical process is very complex and nonlinear. The inuenceof some inputs is not entirely understood, and besides itdepends on a number of unknown parameters andunmodelled/unmesurable disturbances.</p><p>One of the main part of the model is based on classicallaminate theory which is used to model the dimensionalstability of multi-ply board. The main assumption is that eachlayer is considered as an homogeneous elastic medium.</p><p>The model is then complemented with a sub-model forunmodelled/umeasurable disturbances which are described asstates of a dynamical system, and estimated by means of anextended Kalman filter.</p><p>The simulated curvatures show a general agreement with themeasurements. However, the prediction errors are too large forthe model to be used in an effective way, and a bigger efforthas to be carried out in order to improve the physicalsub-models.</p><p>A chapter of this thesis discusses the modelling of thewet-end part of the paper machine with Dymola, a modelling toolfor simulation of large systems based on Modelica language.</p>

No description available.

No description available.

No description available.

Hybridisation is increasingly acknowledged as a conservation problem. The widespread hybridisation between grey duck (Anas superciliosa) and mallard (A. platyrhynchos) in New Zealand is a good example of a native species hybridising with a foreign one, and forms the main focus of this thesis. Mallards were introduced into New Zealand from Europe, and hybrids were soon observed. I surveyed the extent of the hybridisation on the West Coast of the South Island and found that, based on phenotype, at least half of population is now hybrids. Mallards and mallard-like hybrids dominate in the eastern South Island, while grey ducks occur only in some areas of the West Coast. Comparison with historical data suggests that the decline of the grey duck and the spread of hybrids has not stabilised and is ongoing. Contrary to expectations most grey ducks were found on agricultural land and most mallards on natural lakes or rivers, so grey ducks probably do not have an advantage over mallards on the less developed West Coast. An alternative theory is proposed here that explains the spatial distribution of hybridisation as a reflection of a temporal pattern. As mallards were first released in the east, the delay taken to cross the Southern Alps could also explain the pattern observed. This hypothesis suggests that the grey duck will persist in the southern West Coast. An analysis of the phenotypes of partners in pairs suggests that mating is positively assortative within each species and within hybrids. In fact, not a single pair of pure grey duck mated with pure mallard was observed in almost a thousand pairs, raising the question of how hybridisation started. There was a tendency for males to be more mallard-like in phenotype than their partners, suggesting there might be a selective advantage to the mallard male phenotype. This may be one factor explaining the dominance of mallards in the hybrid swarm. To analyse hybridisation at the genetic level, I analysed samples from grey ducks, mallards and domestic ducks with 11 microsatellite loci. This genotyping profile was then compared to ducks captured and shot in New Zealand. Genetic analysis confirms that the ducks in New Zealand were almost exclusively of hybrid origin. Phenotypic hybrid scores correlated with the established genotypic scores, but the correlation was imperfect, suggesting inaccuracies in either or both measures. As the spread of hybrids might be due to the differences in their fitness relative to either parent species, I compared the relative fitness of hybrid ducks using a range of health-related measures such as ecto- and endoparasite loads, immunocompetence, body condition, and heterophil to lymphocyte ratios. Overall, I found no conclusive evidence for any differences between grey duck-like and mallard-like individuals. However, as my sample consisted nearly entirely of hybrids, it is possible that fitness may differ from the parental species. To understand the outcome of hybridisation between two species, I next constructed a mathematical model to simulate hybridisation, and which allowed the specification of parameters describing mating patterns, differential survival, and differential reproductive output. The model successfully predicted the outcomes of two known hybridisation cases. In a sensitivity analysis for mallard and grey duck, the model predicted that this species pair is likely to hybridise under any set of conditions likely to be encountered across their shared range. Finally, in a study within the more general context of hybridisation, the influence of inbreeding on hybridisation rates was investigated using inbred and outbred lines of Drosophila species. I found evidence for increased hybridisation in inbred lines, and although further studies are needed to confirm the generality of this pattern, my results have implications for the management of hybridisation, and for the use of hybridisation as an adaptive strategy. In conclusion, my work suggests it is very likely that the grey duck will become extinct as a separate species in New Zealand in the near future, and that it is likely to be threatened in other areas of its range were it co-occurs with the mallard. The options for management of this situation are limited, as large areas without mallards are lacking. Captive breeding, or the management of grey duck populations on isolated islands appear the only feasible options. It seems unlikely that hybridisation can be reversed on the mainland, and a homogenous hybrid population is likely to eventually occupy the entire range of the grey duck across New Zealand.

No description available.

This thesis describes a grey-box model for the dimensionalstability properties (i.e. curl and twist) of the carton boardproduced at AssiDomän Frövi paper mill in Sweden.AssiDomän Frövi AB is one of Sweden major cartonboard manufacturer, and produces some 350000 ton of board peryear. Curl is defined as the departure from a at form, and it mayseriously affect the processing of the paper. For this reason,customers impose quite restrictive limits on the allowedcurvatures of the board. So, it is becoming more and moreimportant to be able to produce a carton board with a curlwithin certain limits. Due to the economic significance of thecurl problem, much research has been performed to find sheetdesign and processing strategies to eliminate or reducecurl. The approach we used to tackle this problem is based ongrey-box modelling. The reasons for such an approach is thatthe physical process is very complex and nonlinear. The inuenceof some inputs is not entirely understood, and besides itdepends on a number of unknown parameters andunmodelled/unmesurable disturbances. One of the main part of the model is based on classicallaminate theory which is used to model the dimensionalstability of multi-ply board. The main assumption is that eachlayer is considered as an homogeneous elastic medium. The model is then complemented with a sub-model forunmodelled/umeasurable disturbances which are described asstates of a dynamical system, and estimated by means of anextended Kalman filter. The simulated curvatures show a general agreement with themeasurements. However, the prediction errors are too large forthe model to be used in an effective way, and a bigger efforthas to be carried out in order to improve the physicalsub-models. A chapter of this thesis discusses the modelling of thewet-end part of the paper machine with Dymola, a modelling toolfor simulation of large systems based on Modelica language. / NR 20140805