References of "Doctoral thesis"
     in
Bookmark and Share    
See detailJaponisme et collectionisme : Analyse des collections de Hans de Winiwarter
Bawin, Julie ULg

Doctoral thesis (2004)

Detailed reference viewed: 32 (9 ULg)
Full Text
See detailForecast verification of a 3D model of the Mediterranean Sea. Analysis of model results and observations using wavelets and Empirical Orthogonal Functions.
Alvera Azcarate, Aïda ULg

Doctoral thesis (2004)

The quality assessment of the three-dimensional GHER (GeoHydrodynamics and Environmental Research) model of the Mediterranean Sea is presented in this work. The verification of the model results is done ... [more ▼]

The quality assessment of the three-dimensional GHER (GeoHydrodynamics and Environmental Research) model of the Mediterranean Sea is presented in this work. The verification of the model results is done in a spatio-temporal approach. Traditional error measures (i.e. correlation, mean error, etc) are very useful to assess the quality of a model, but they do not take into account the high complexity of three-dimensional models. The verification process is thus done in three main parts: first, the model is compared to observations and climatology in a qualitative approach, in order to make a preliminar study about the model behaviour. Then, the error assessment is done, using traditional statistic measures. In order to take into account the complexity of the model and observations, the last step in the verification process consists in a spatio-temporal analysis using wavelets and empirical orthogonal functions. This last analysis will allow us to have an insight about the model quality in a more detailed way. This verification process has been applied to the GHER model. This model is implemented in a two-way nesting approach in the Mediterranean Sea, Liguro-Provençal basin and Ligurian Sea, where the highest resolution is reached. Assimilation of sea surface temperature and sea level anomaly is made during a nine-week experiment. Another test is carried out, to assess the quality of sea surface temperature from the SOFT predictor of the Ligurian Sea. The predicted sea surface temperature is assimilated in the model and the quality of the forecast is compared to the first assimilation experiment. The assimilation of the SOFT statistical predictors can be very useful to force models in a real forecast experiment, where no observations are available. [less ▲]

Detailed reference viewed: 27 (4 ULg)
See detailRegions, Political Economics and Economic Performance
Artige, Lionel ULg

Doctoral thesis (2004)

Detailed reference viewed: 29 (0 ULg)
Full Text
See detailSédimentologie de la plate-forme carbonatée frasnienne belge
Da Silva, Anne-Christine ULg

Doctoral thesis (2004)

Detailed reference viewed: 20 (7 ULg)
Full Text
See detailExploring Structure and Reformulations in Different Integer Programming Algorithms
Louveaux, Quentin ULg

Doctoral thesis (2004)

In this thesis we consider four topics all related to using problem reformulations in order to solve integer programs, i.e. optimization problems in which the decision variables must be integer. We first ... [more ▼]

In this thesis we consider four topics all related to using problem reformulations in order to solve integer programs, i.e. optimization problems in which the decision variables must be integer. We first consider the polyhedral approach. We start by addressing the ques- tion of lifting valid inequalities, i.e. finding a valid inequality for a set Y from the knowledge of a valid inequality for a lower-dimensional restriction X of Y . We simplify and clarify the presentation of the procedure. This allows us to derive conditions under which the computation of the lifting is tractable. The second topic is the study of valid inequalities for the single node flow set. The single node flow set is the problem obtained by considering one node of a fixed charge network flow problem. We derive valid inequalities for this set and various generalizations. Our approach is a systematic procedure using only basic tools of integer programming: fixing and complementing variables, mixed- integer rounding and lifting. The method allows us to explain and generate a large range of inequalities describing the convex hull of such sets. The last two topics are based on non-standard approaches for integer pro- gramming. We first show how the group relaxation approach can be used to provide reformulations for the integral basis method. This is based on a study of extended formulations for the group problem. We present four extended for- mulations and show that the projections of three of these formulations provide the convex hull of the original group problem. Initial computational tests of the approach are also reported. Finally we consider a problem that is difficult for the standard branch-and- bound approach even for small instances. A reformulation based on lattice basis reduction is known to be more effective. However the step to compute the re- duced basis is O(n4) and becomes a bottleneck for small to medium instances. By using the structure of the problem, we show that we can decompose the problem and obtain the basis by taking the kronecker product of two smaller bases easier to compute. Furthermore, if the two small bases are reduced, the kronecker product is also reduced up to a reordering of the vectors. Computa- tional results show the gain from such an approach. [less ▲]

Detailed reference viewed: 45 (1 ULg)
See detailLe Paléolithique supérieur de la Moldavie. Essai de synthèse d'une évolution multi-culturelle
Noiret, Pierre ULg

Doctoral thesis (2004)

Les industries lithiques du Paléolithique supérieur en Moldavie sont fondées sur l’exploitation de roches locales et appartiennent principalement à des traditions culturelles connues en Europe ... [more ▼]

Les industries lithiques du Paléolithique supérieur en Moldavie sont fondées sur l’exploitation de roches locales et appartiennent principalement à des traditions culturelles connues en Europe (l’Aurignacien, le Gravettien, puis l’Épigravettien). Cependant, elles coexistent avec d’autres industries marquées par la présence d’outils aménagés par retouche bifaciale, dans des contextes réputés anciens (plus de 30.000 ans) et qui se sont révélées mal datées et souvent peu homogènes. Ces industries appartiennent à trois ou quatre traditions culturelles supplémentaires, dont les définitions restent vagues. Après avoir fouillé les sites de Mitoc–Malu Galben (Roumanie) et de Cosăuţi (République Moldave) en compagnie de leur responsable (V. Chirica et I.A. Borziac) et avec l’aide de nos collègues de l’Université de Liège (le Prof. Marcel Otte, I. López Bayón et V. Ancion) et de l’Institut royal des Sciences naturelles de Belgique (P. Haesaerts et Fr. Damblon), nous avons entrepris l’étude comparative des données chronologiques, stratigraphiques, fauniques et matérielles (outillages lithiques et osseux, témoins esthétiques) de 21 sites pour la période comprise entre 35.000 et 10.000 ans BP. Nous avons constitué un corpus d’une cinquantaine d’ensembles archéologiques. Les vestiges de débitage, les outils domestiques et les armatures liées à la chasse ont été décrits, puis nous avons tenté des approches statistiques par analyse factorielle des correspondances (fondée sur les similarités et les dissimilarités entre les ensembles étudiés). Si l’analyse technologique n’a porté que sur des indications de présence/absence de caractéristiques technologiques, l’analyse typologique a pu être menée sur des données plus complètes (les pourcentages de tous les outils lithiques, dans chaque ensemble étudié). Nous avons ainsi mis en évidence les particularités et l’évolution des traditions culturelles étudiées. Les industries lithiques caractérisées par la présence de pièces bifaciales ne semblent désormais plus pouvoir être décrites en tant que plusieurs entités culturelles distinctes ; au contraire, les analyses ont montré leur relative homogénéité technologique et typologique. Il est probable qu’elles représentent une expression culturelle unique, coexistant avec les autres traditions et localisée dans une zone géographique restreinte ; cela justifie à nos yeux l’emploi de la dénomination de « Culture du Prut ». Cette culture est marquée par la production prédominante d’éclats, employés comme supports à l’outillage, et dans une moindre mesure de lames à partir de nucléus peu préparés et encore assez « plats », alors que dans l’Aurignacien et surtout dans le Gravettien les lames sont les supports principaux, obtenus à partir de nucléus volumétriques, bien préparés et entretenus. Une lecture critique des informations publiées par les fouilleurs a permis une répartition de tous les ensembles étudiés (ou presque) dans un schéma chrono-stratigraphique régional bien assuré et bien daté, élargi en fin de travail à l’ensemble de l’Europe centrale et orientale. Ces ensembles appartiennent à l’Aurignacien typique (33.000-29.000 BP), au Gravettien (29.500-23.000 BP), à la Culture du Prut (27.000-26.000 BP), à l’Aurignacien tardif (vers 25.000 puis 20.000 BP), à l’Épigravettien ancien (20.000-17.000 BP) puis récent (13.500-11.000 BP). L’étude des restes fauniques a permis de mettre en évidence des différences marquées entre traditions culturelles : les Aurignaciens ont chassé le cheval et le bison ; les Gravettiens, le cheval et le renne, puis le bison et peut-être le mammouth. C’est après l’hiatus de 23.000-20.000 BP que l’analyse faunique est la plus significative : les Épigravettiens se sont alors véritablement spécialisés dans la chasse au renne ; dans la plupart des sites, les restes de ces animaux atteignent souvent 70 à 80 % de tous les restes identifiables, devant le cheval et  plus rarement  le bison (ou l’aurochs) et le mammouth. En outre, des petits animaux étaient piégés pour leur fourrure (lièvres, renards) et des poissons et des oiseaux complétaient l’alimentation, ainsi sans doute que des aliments végétaux. À cette spécialisation épigravettienne correspond, selon nous, une entité ethnique homogène, entre 20.000 et 17.000 BP. Des sites à fonction limitée existent alors, à côté de grandes installations occupées peut-être plus longtemps mais surtout plus souvent qu’auparavant. Toutes les installations sont saisonnières. Comme dans les traditions culturelles antérieures, les contacts extra-régionaux semblent limités et les roches locales continuent d’être exploitées. Des comparaisons avec l’Europe centrale et orientale ont permis de lier ce Paléolithique supérieur moldave à quelques phénomènes culturels plus larges : l’exploration par les Aurignaciens d’un territoire s’étendant jusqu’au centre de la Plaine russe et jusqu’aux contreforts septentrionaux de l’Oural (dès 33.000 BP), les origines géographiques multiples du Gravettien (vers 30.000 BP), la présence d’industries lithiques non-aurignaciennes et non-gravettiennes (à pièces bifaciales, sans doute autour de 27.000-26.000 BP), la résurgence de traits techniques d’allure « aurignacoïde » (dès 21.000 BP), et le retour massif de la tradition des outils à dos (à partir de 20.000 BP). Ainsi, les données moldaves s’inscrivent-elles logiquement dans une paléo-histoire large, à l’échelle de l’ensemble du continent. [less ▲]

Detailed reference viewed: 39 (2 ULg)
See detailProbing the interactions of early-type stars with their surroundings through X-ray and optical observations
Nazé, Yaël ULg

Doctoral thesis (2004)

This work is devoted to the study of the interactions of massive stars with their environment, either through their intense radiation fields or through their powerful stellar winds. First, we have studied ... [more ▼]

This work is devoted to the study of the interactions of massive stars with their environment, either through their intense radiation fields or through their powerful stellar winds. First, we have studied the close neighbourhood of the peculiar Of?p star HD108, i.e. its stellar wind and associated emissions. We have notably uncovered the important variations of the emission lines seen in the visible domain. At the same time, in contrast, the X-ray emission remained stable. To better understand the X-rays emitted by the star, we have developed a program for modelling the ionisation of the stellar wind, that enabled us to determine the absorption of this ionised wind in the X-ray domain. Our X-ray observations have then permitted us to discard several models proposed to explain the nature of the system, and we have discussed the remaining possibilities in the light of the behaviour of the two other Of?p stars, the possible ``twins" of HD108. The wind modelling program was also used to study the properties, in the X-ray range, of the Wolf-Rayet star WR40: combined to the observations taken by XMM-Newton, we have shown that the non-detection of X-rays from this star can be explained by the very large opacity of its wind. The study of the ionisation of the surrounding gas by Wolf-Rayet stars was extended through the analysis of HeII nebulae. For the first time, high quality images of the highly excited regions surrounding Wolf-Rayet stars were taken and analysed, enabling to derive the extreme UV flux of these stars, and thus their temperature, from ground-based observations. In addition to the ionisation by massive stars, the impact of their stellar winds was also considered in this work, especially the wind-blown structures called ``bubbles''. These bubbles arise from the action of the winds of massive stars, in isolation or in clusters, onto the interstellar medium. We first focused on the bubbles blown by isolated stars, and we started by the study of the high energy properties of the bubble blown by WR40. Thanks to XMM-Newton data, we showed that even the most recent theoretical models cannot account for the lack of X-ray emission of this bubble. To better understand the discrepancies between the theory and the observations of wind-blown bubbles, we then turned to the most simple structures, the ones blown by single main sequence massive stars. Several bubbles have been discovered in N11B, N180B, and N44, and their properties agree better with theoretical expectations than in the case of WR bubbles and superbubbles. Before our study, only a few interstellar bubbles were known: we showed that a simple morphological search was insufficient, and that kinematic studies were essential. In fact, the low-velocity expansion of these bubbles implies a weak compression of the gas, that can inhibit the morphological detection. Finally, we investigated the properties of the interaction of massive star clusters with their surroundings in the high-energy range. The clusters of N11 were observed in X-rays and UV thanks to the XMM-Newton satellite: their differences or similarities appear clearly at these energies, and we note that the diffuse X-ray emission always exceeded the expected level. Moreover, a unique dataset composed of Chandra and XMM-Newton observations enabled us to study the largest star formation region of the Small Magellanic Cloud, NGC346. Notably, the X-ray emissions from the NGC346 cluster itself and from the peculiar WR binary system HD5980 were discovered and analysed in details. Our work underlines the complementarity between the visible, UV and X-ray domains, and the necessity to have observatories on the ground as well as in space. It also illustrates the important contribution of multiwavelength datasets to the study of massive stars. These observations indeed play a crucial role in the determination of the properties of early-type stars but they also provide critical tests for theoretical models (present or future). [less ▲]

Detailed reference viewed: 12 (7 ULg)
Full Text
See detailUn modèle pédagogique global pour une approche du paysage fondée sur les nouvelles technologies de l'information et de la communication (NTIC)
Partoune, Christine ULg

Doctoral thesis (2004)

How to develop global and complex thinking ? By conceiving and making an hyperlandscape with appropriate pedagogical and technical constraints. But is it enough ? No, of course ! I requires an inventive ... [more ▼]

How to develop global and complex thinking ? By conceiving and making an hyperlandscape with appropriate pedagogical and technical constraints. But is it enough ? No, of course ! I requires an inventive approach for giving a sense to that task. Here is described the way followed during a research-action in various contexts, from primary school to high school. A global pedagogical model is proposed as a tool for environmental education by and for the landscape. [less ▲]

Detailed reference viewed: 62 (6 ULg)
See detailEn/Gendering Representations of Childbirth in Fifteenth-Century Franco-Flemish Devotional Manuscripts
L'Estrange, Elizabeth ULg

Doctoral thesis (2004)

Representations of childbirth in fifteenth-century devotional and historical manuscripts are invariably depicted as post-partum confinement scenes in a domestic interior. These images appear to show a ... [more ▼]

Representations of childbirth in fifteenth-century devotional and historical manuscripts are invariably depicted as post-partum confinement scenes in a domestic interior. These images appear to show a ‘gendered space’ in which women care for each other and men are marginalized. Neglected by medieval art historians, such pictures have been uncritically used by historians of obstetric and social history to prove that childbirth was the one time when medieval women exercised power and control in an otherwise patriarchal society. However, as with all historical evidence, these images do not offer us unmediated access to the past. This thesis brings these domestic, post-partum pictures of childbirth to the centre of an art historical enquiry by undertaking a survey of this iconography in some fifty fifteenth-century manuscripts and incunables. Since the occurrence of this generic iconography cannot be consistently associated with female spectators, it has been necessary to reassess in what way they might be en/gendered: how they were received by their original viewers (male and female) and how we can bring them into meaning as sources for reconstructing the lives of medieval women. To avoid equating these images with reality and reducing the female sex at large to the maternal function, I develop a methodology to show how the social viewing positions occupied by certain spectators would have rendered them sensitive to images of maternity and childbirth. Specifically, I argue that the images of childbirth in a group of fifteenth-century Books of Hours made for male and female members of the houses of Anjou and Brittany would have been seen with a ‘situational eye’ that was informed by the requirements of patriarchal, aristocratic families, and by the dangers surrounding childbirth. My thesis demonstrates that this situational eye can be extended beyond the field of art history to show how other sources from fifteenth-century childbearing such as charms, lying-in, and churching, can be brought into meaning for the women whose social position required them to conceive and give birth to male heirs. [less ▲]

Detailed reference viewed: 31 (1 ULg)
Full Text
See detailExporation cognitive et neuropsychologique du fonctionnement de la mémoire épisodique: Les processus de reconnaissance et la mémoire du contexte temporel
Bastin, Christine ULg

Doctoral thesis (2004)

The main objective of our dissertation was to contribute to a better understanding of the cognitive processes involved in episodic memory and their cerebral substrates. More specifically, our work focused ... [more ▼]

The main objective of our dissertation was to contribute to a better understanding of the cognitive processes involved in episodic memory and their cerebral substrates. More specifically, our work focused on two questions. First, we were interested in the contribution of recollection and familiarity to recognition memory. On the one hand, we examined the performance on recognition tasks when recollection cannot be used anymore (studies 1, 2 and 3) and, on the other hand, the influence of frontal lobe lesions on the processes involved in a recognition memory task (study 4). Second, we explored the processes recruited in an important aspect of episodic memory, namely, memory for the temporal context of events, by studying the nature of the difficulties encountered by older adults on a task assessing memory for temporal information (studies 5 and 6). [less ▲]

Detailed reference viewed: 88 (16 ULg)
Full Text
See detailEpidémiosurveillance des événements rares chez les bovins en Belgique
Saegerman, Claude ULg

Doctoral thesis (2004)

Detailed reference viewed: 28 (4 ULg)
Full Text
See detailContributions aux algorithmes d'intégration temporelle conservant l'énergie en dynamique non-linéaire des structures
Noels, Ludovic ULg

Doctoral thesis (2004)

Les points fondamentaux de mes recherches consistent en une étude mathématique des algorithmes d'intégration temporelle d'une discrétisation de type éléments-finis. Les algorithmes traditionnels d ... [more ▼]

Les points fondamentaux de mes recherches consistent en une étude mathématique des algorithmes d'intégration temporelle d'une discrétisation de type éléments-finis. Les algorithmes traditionnels d'intégration temporelle (Newmark, famille alpha-généralisée, ...) peuvent introduire numériquement de l'énergie dans les systèmes non-linéaires. De nouveaux algorithmes sont dès lors apparus pour permettre d'éviter cette perte de stabilité. Lors de mes études, ces algorithmes ont été étendus, de manière originale, au traitement de la plasticité à l'aide d'un modèle de type hypo-élastique. Ils ont également été étendus au traitement généralisé, dans le cadre tridimensionnel, de l'interaction de contact. Ensuite la mise en oeuvre d'une méthode permettant de combiner, au cours du temps un algorithme implicite stable, avec un algorithme explicite traditionnel a été envisagée. La stabilité du passage d'une méthode explicite vers une méthode implicite a été démontrée par nos soins. Une définition originale de critères automatiques de choix de basculement entre les méthodes a été proposée. Des exemples numériques complexes ont mis en évidences les bonnes performances des algorithmes développés, tant du point de vue de la précision des résultats - confrontés aux résultats expérimentaux si ces derniers existent -, que du point de vue de la réduction des coûts de calcul lorsque la méthode combinée est utilisée. [less ▲]

Detailed reference viewed: 240 (32 ULg)
Full Text
See detailAssimilation of sea surface temperature and sea surface height in a two-way nested primitive equation model of the Ligurian Sea
Barth, Alexander ULg

Doctoral thesis (2004)

A coarse grid primitive equation model of 1/4° resolution is implemented covering the whole Mediterranean Sea. Within this grid a 1/20° resolution model of the Liguro-Provençal basin and the northern part ... [more ▼]

A coarse grid primitive equation model of 1/4° resolution is implemented covering the whole Mediterranean Sea. Within this grid a 1/20° resolution model of the Liguro-Provençal basin and the northern part of the Tyrrhenian Sea is embedded. A third fine resolution model of 1/60° is nested in the latter one and simulates the dynamics of the Ligurian Sea. Comparisons between one-way and two-way nesting in representing the Northern Current (NC) are made. This system of nested models is coupled with a simplified Kalman-filter based assimilation method. The state vector for the data assimilation is composed by the temperature, salinity and elevation of the three models. The forecast error is estimated by an ensemble run by perturbing initial conditions and atmospheric forcings. The leading empirical orthogonal functions (EOF) of this ensemble are taken as the error covariance of the model forecast. This error covariance is assumed to be constant in time. Sea surface temperature (SST) and sea surface height (SSH) are assimilated in this system. Data assimilation is also used to couple the hydrodynamic model with a statistical predictor of SST in the Ligurian Sea. The forecast improvement of this hybrid modelling system is shown and applications to operational models are highlighted. [less ▲]

Detailed reference viewed: 28 (5 ULg)
Full Text
See detailTrajectoires professionnelles féminines : flexibilités et enjeux de genre
Gavray, Claire ULg

Doctoral thesis (2003)

Professional positions and transitions, and the capacities of autonomy which result from it, vary very significantly according to three articulated variables: the gender group, the level of certification ... [more ▼]

Professional positions and transitions, and the capacities of autonomy which result from it, vary very significantly according to three articulated variables: the gender group, the level of certification and the life course development. The gendered social hierarchisation present in every field of social insertion -although different according to generation- occupies a significant position in the construction of this course. It's important to understand objective and subjective factors (through values, choices, rôles, plans, opportunities and constraints) influencing professional paths within each gender group . These comparisons are essential. Neglecting the gender perspective and theory forbids the comprehension of the work and labor offer but also of the distribution of places on labour market. Decompartmentalization of `employment, formation, family, social capital' is also more than ever necessary to analyse individual trajectories and to imagine fieldeffective and nondiscriminating policies. For exemple, time -and its value- seems a crucial issue for the families, the citizens, the democratic society in change. [less ▲]

Detailed reference viewed: 190 (21 ULg)
Full Text
See detailPreventive assessment and enhancement of power system voltage stability: an integrated approach of thermal and voltage security
Capitanescu, Florin ULg

Doctoral thesis (2003)

This thesis is devoted to the preventive assessment and enhancement of voltage stability and security in electric power systems. However, in the course of deriving all the proposed methods we have paid ... [more ▼]

This thesis is devoted to the preventive assessment and enhancement of voltage stability and security in electric power systems. However, in the course of deriving all the proposed methods we have paid attention to keeping them compatible with the (more traditional) handling of thermal overloads, thereby providing a unified treatment of voltage and thermal security. The approaches presented in this work apply to both deregulated environments and classical, vertically integrated ones. The heart of most methods developed in this thesis is : (i) the derivation of sensitivities indicating the relative efficiency of the various bus injections to restore voltage stability or increase an insufficient voltage security margin, and (ii) the use of these sensitivities in linearized security constraints that can be incorporated to various optimization problems. Using this formulation, we deal with three different problems of interest in preventive security analysis: 1) Congestion management. We propose two optimization-based approaches to manage congestions due to voltage instability and/or thermal overload. The control variables are either power injections (generation rescheduling and load curtailment) or power transactions; 2) Computation of Available Transfer Capabilities (ATCs). We determine the simultaneous ATCs of multiple transactions by means of a single optimization-based computation; 3) evaluation of security margins interval. To face the uncertainty affecting power transfers, we present an optimization-based computation of the minimal and maximal margins under the assumption that individual injections vary within specified bounds. Besides this main theme, the thesis offers additional reflections on the: 1) Filtering of contingencies. We propose a simple and reliable technique to filter out harmless contingencies when computing voltage security margins of a large set of contingencies; 2) Evaluation of reactive reserves with respect to a contingency, an important topic for voltage security reasons as well as within the context of a deregulated market where providing reactive reserves is an ancillary service which should be properly paid. Most of the methods proposed in this thesis were successfully tested on realistic power system models. From a practical viewpoint all the above computations have been coupled to the fast time-domain quasi steady-state simulation used in the ASTRE software developed at the University of Liège. [less ▲]

Detailed reference viewed: 55 (5 ULg)
Full Text
See detailEtude expérimentale des sons auto-entretenus produits par un jet issu d'un conduit et heurtant une plaque fendue
Billon, Alexis ULg

Doctoral thesis (2003)

The production of self-sustained tones (up to 125 dB) by a low Mach number free plane jet impinging on a slotted plate, known as slot-tone, is experimentally studied. For all geometries of the slot’s edge ... [more ▼]

The production of self-sustained tones (up to 125 dB) by a low Mach number free plane jet impinging on a slotted plate, known as slot-tone, is experimentally studied. For all geometries of the slot’s edge tested, the tones are generated from 6 m/s and when a resonant mode of the duct from which the jet flows out is excited. The Strouhal number associated with the tones fundamental frequency describes stages. These one are linked to the number of vortices present, with vortex pairing possibilities. On the other hand, the frequency domain of the fundamental is defined by the two instability modes of the jet. If the edge of the slot is beveled and if this one is placed backwards to the flow, the tones occurs at a lower speed (3.5 m/s) as soon as the jet becomes turbulent, and the excitation of an acoustic resonator is then not necessary. [less ▲]

Detailed reference viewed: 30 (4 ULg)
See detailA cognitive approach to hallucination-proneness in non-clinical subjects
Laroi, Frank ULg

Doctoral thesis (2003)

Detailed reference viewed: 4 (0 ULg)
Full Text
See detailOrigine des troubles morphosyntaxiques chez des enfants dysphasiques
Maillart, Christelle ULg

Doctoral thesis (2003)

Detailed reference viewed: 145 (11 ULg)
Full Text
See detailModélisation de la répartition dynamique des produits phytopharmaceutiques sous une rampe de pulvérisation
Lebeau, Frédéric ULg

Doctoral thesis (2003)

A mathematical model aimed at estimating the distribution of spray deposits of phytopharmaceutical <br />products under a spray boom is proposed. It focuses on the need to take account of the dynamic ... [more ▼]

A mathematical model aimed at estimating the distribution of spray deposits of phytopharmaceutical <br />products under a spray boom is proposed. It focuses on the need to take account of the dynamic effects of <br />the forward movement of the boom. These are related both to the horizontal and vertical boom <br />movement and to the influence of aerodynamical factors on the nozzle spray distribution. <br />The model allows one to compute the spray deposits distribution by multiplying the nozzle spray <br />distribution by the time needed to move from one position to the next. Mathematically, this is expressed <br />by a convolution of the trajectory with the nozzle spray distribution. <br />The model is validated through a dynamic test bench aimed at reproducing the boom movements <br />observed in the field. The chosen method to measure spray deposits distribution is a chemical dosage of <br />the sprayed KCL solution collected in Petri dishes. This method is accurate if the spacing between <br />collectors is small enough. With constant speed and height, the longitudinal variability approaches 5 % <br />and the transversal spray deposits distribution is a function of the nozzle characteristics (as measured on <br />the static spray patternator) and of the nozzle speed. <br />A Pulse Width Modulation (PWM) nozzle boss fitted on the test bench is used to generate a dynamic <br />spray deposits distribution from which the dynamic bidimensional nozzle spray pattern is reconstructed. <br />This dynamic nozzle spray pattern introduced in the model allows a far better estimation of the spray <br />deposits distribution to be made than the one obtained using the static nozzle spray pattern computed <br />using filtered back-projection. <br />The use of the model for estimating the spray deposits distribution under a crop sprayer allows one <br />to identify the influential parameters and to calculate the dosages according to specific local <br />circumstances. Statistical parameters aimed at characterising the treatment quality are proposed. [less ▲]

Detailed reference viewed: 20 (11 ULg)
Full Text
See detailStudie zur sonographischen Anatomie des Hufgelenks und der Hufrolle beim Pferd - ein Beitrag zur Diagnose der Podotrochlose
Schneider, Nicole ULg

Doctoral thesis (2003)

Ultrasound is becoming more common in the examination of the coffin joint and podotrochlear apparatus in horses. Purpose of this study is an approach to ultrasound anatomy of the coffin joint and ... [more ▼]

Ultrasound is becoming more common in the examination of the coffin joint and podotrochlear apparatus in horses. Purpose of this study is an approach to ultrasound anatomy of the coffin joint and detectable associated structures. Ultrasound is compared to photographs of dissected limbs to prove sensibility and reliability. Anatomy, artefacts, methods and other difficulties concerning realisation during an ultrasound imaging procedure of the distal interphalangeal joint and podotrochlear apparatus are discussed. This study is based on ten physiologic and seven pathologic reference sonograms, selected from patients at the Ecole Nationale Vétérinaire d’Alfort and the faculty of veterinary medicine of the university of Liège. Transcuneal views of the podotrochlear apparatus of an 18 months old foal have been established post mortem at the institute of veterinary diagnostic imaging at Liège. Modus of registration is real-time-motion, linear and sectorscanners working with 7,5 MHz are used. Echoanatomic documentation shows most of the soft tissue structures in the equine pastern in a reliable and sufficient way, some examples of pathologic alterations detected by ultrasound demonstrate that it is a suitable method to enrich informations achieved by common imaging procedures at the distal region of the equine limb. [less ▲]

Detailed reference viewed: 51 (3 ULg)
Full Text
See detailNear Optimal Closed-Loop Control. Application to Electric Power Systems
Ernst, Damien ULg

Doctoral thesis (2003)

Detailed reference viewed: 42 (16 ULg)
Full Text
See detailAn observational study of line-profile variable B stars
Briquet, Maryline ULg

Doctoral thesis (2003)

Detailed reference viewed: 16 (1 ULg)
Full Text
See detailClonage, caractérisation et étude de la régulation transcriptionnelle du gène Aox1 encodant l'oxydase alternative chez Chlamydomonas reinhardtii
Baurain, Denis ULg

Doctoral thesis (2003)

Au sein de la membrane interne des mitochondries, quatre complexes multiprotéiques sont impliqués dans le transfert des électrons depuis les équivalents réducteurs jusqu'à l'oxygène moléculaire. L'énergie ... [more ▼]

Au sein de la membrane interne des mitochondries, quatre complexes multiprotéiques sont impliqués dans le transfert des électrons depuis les équivalents réducteurs jusqu'à l'oxygène moléculaire. L'énergie associée à ce transport au travers des complexes I, III et IV est couplée à la synthèse d'ATP par l'intermédiaire d'un gradient de protons. Chez les plantes supérieures, de nombreux champignons et quelques protistes, une seconde voie de transfert diverge de la voie principale au niveau du pool d'ubiquinone, celui-ci étant alors oxydé directement par l'oxygène moléculaire. Lorsque les électrons empruntent cette voie alternative, deux sites d'éjection de protons sont court-circuités et l'énergie produite est dissipée sous forme de chaleur. Cette réaction est catalysée par une enzyme unique, l'oxydase alternative (AOX), souvent encodée par une petite famille multigénique chez les plantes supérieures. Une activité accrue de la voie alternative est observée suite à divers stimuli développementaux et environnementaux, en particulier en conditions de stress. Cette augmentation d'activité résulte d'une activation transcriptionnelle du gène Aox et/ou de modifications post-traductionnelles de la protéine mature. L'AOX de l'algue verte unicellulaire Chlamydomonas reinhardtii est encodée par deux gènes différents, Aox1 et Aox2, le premier étant beaucoup plus transcrit que le second. Les cDNAs Aox1 et Aox2, de même que la séquence génomique Aox2, ont été isolés et caractérisés dans notre laboratoire. Dans un premier temps, nous avons entrepris le clonage et la caractérisation de la séquence génomique Aox1, ce qui nous a permis de comparer sa structure avec celle de son homologue Aox2. Ensuite, afin d'étudier sa régulation transcriptionnelle, nous avons fusionné un segment de 1,4 kb contenant la région promotrice Aox1 à la région codante du gène (Ars) de l'arylsulfatase et mesuré les activités ARS dans des transformants porteurs de la construction chimérique. Nous avons ainsi montré que le promoteur Aox1 est insensible à la plupart des inducteurs classiques de l'AOX, parmi lesquels des agents de stress, des inhibiteurs respiratoires et des métabolites. En revanche, l'expression du gène Aox1 répond à la nature de la source d'azote, sa transcription étant réprimée par l'ammonium et stimulée par le nitrate. De plus, en milieu contenant du nitrate, l'inactivation de la nitrate réductase (première enzyme de la voie d'assimilation du nitrate) conduit à une expression du gène Aox1 encore plus importante. Nous avons en outre observé que cette stimulation par le nitrate se répercute aux niveaux protéique et respiratoire. Une étude de délétion de la région promotrice Aox1 indique qu'un segment court (de −253 à +59 par rapport à l'origine de transcription) est suffisant pour assurer la transcription et la régulation du gène, mais que son expression maximale requiert également des éléments distaux. Aucun motif nucléotidique susceptible d'intervenir dans l'expression du gène Aox1 n'a été identifié à l'issue d'une analyse bioinformatique du promoteur. L'effet de la nature de la source d'azote sur l'expression de l'AOX est interprété sous l'angle d'une optimisation de la synthèse d'ATP mitochondrial sans modification de l'activité respiratoire, en relation avec un possible accroissement de la production d'ATP photosynthétique lorsque le nitrate est utilisé comme source d'azote. [less ▲]

Detailed reference viewed: 44 (7 ULg)
See detailSome wavelet applications to signal and image processing.
Charles, Catherine ULg

Doctoral thesis (2003)

Detailed reference viewed: 15 (4 ULg)
Full Text
See detailL'alliance délicate de la gestion forestière et de la biodiversité dans les forêts du centre du Gabon
Doucet, Jean-Louis ULg

Doctoral thesis (2003)

Afin de garantir une exploitation durable de leurs forêts, la plupart des pays d’Afrique centrale ont récemment adapté leur législation forestière. Contraints de suivre cette évolution, les exploitants ... [more ▼]

Afin de garantir une exploitation durable de leurs forêts, la plupart des pays d’Afrique centrale ont récemment adapté leur législation forestière. Contraints de suivre cette évolution, les exploitants forestiers se sont impliqués dans la conception de plans d’aménagement durable. Malheureusement cette « durabilité » se réduit généralement à une production soutenue sur une seule rotation, soit une période d’une trentaine d’années. L’approche proposée ici vise à assurer une exploitation réellement durable par la prise en compte de deux facteurs fondamentaux : la diversité végétale et la dynamique des populations des essences commerciales. [less ▲]

Detailed reference viewed: 55 (14 ULg)
Full Text
See detailEléments traces métalliques dans les sols. Contribution à l'étude de leur déterminisme spatial en Région limoneuse belge.
Colinet, Gilles ULg

Doctoral thesis (2003)

The natural background levels of metallic trace elements (MTEs) in cultiva-ted soils should be known in order (i) to understand the processes in which MTEs are involved and (ii) settle a reference system ... [more ▼]

The natural background levels of metallic trace elements (MTEs) in cultiva-ted soils should be known in order (i) to understand the processes in which MTEs are involved and (ii) settle a reference system dedicated to the the ap-praisal of agricultural practices, to the detection of contaminations, or to the time-course follow-up of regional values. A methodological framework has thus been set up to evaluate the importance of natural driving factors, the pe-dogeochemical background, of the spatial distribution of MTEs in soils and also to compare the current levels of MTEs content in agricultural soils from the Loamy Region with natural values. As main results from the study of parent materials, (i) lithological units are now characterized with element content signatures, (ii) the prominent part of clay, total iron, and total aluminium contents as driving factors of the content in MTEs has been measured, and (iii) the geographical homogeneity as well as the absence of spatial continuity of the cover loessic parent material have been showed up. The relations between MTEs and driving variables have been confirmed through the study of the influences of pedogenesis factors on the vertical and lateral differenciations of MTEs content in soils. It was also demonstrated that taking into account the natural volumes of soils when sam-pling facilitates the interpretation of the results and allows finer estimations of the MTEs content, even in homogeneous context. The reference values about natural and usual content in MTEs for various soil horizons that are proposed reveal a relatively weak level of man-made contamination in the Loamy Re-gion. It has also been shown that a regional reference system dedicated to the detec-tion of local contaminations gains in relevance as it integers ancillary datas as well as the local specificities of small natural regions. The spatial variability of MTEs content has been evaluated through various cartographic approaches which led to the conclusion that the Soil Map of Belgium is an extremely worthy document for geochemical mapping. [less ▲]

Detailed reference viewed: 92 (11 ULg)
Full Text
See detailConception d’un système d’aide à la décision pour le choix d'un scénario sylvicole : Application aux peuplements de mélèze en Région wallonne
Pauwels, Dominique

Doctoral thesis (2003)

A silvicultural decision support system (SDSS) has been developed for pure and even-aged larch stands (Larix sp.) in Walloon Region, where Larix seems to be a species with good prospects in a ... [more ▼]

A silvicultural decision support system (SDSS) has been developed for pure and even-aged larch stands (Larix sp.) in Walloon Region, where Larix seems to be a species with good prospects in a multifunctional management context. It main purposes are: (i) to predict the influence of silvicultural treatments on stand evolution and (ii) to help forest managers compare different scenarios according to preset goals. The SDSS comprises three modules designed for (i) larch growth prediction based on initial tree density, thinning regime and site index (scenarios building), (ii) assessment of a set of indicators describing scenarios and (iii) the comparison of scenarios according to appropriate calculated indicators (Multi-Criteria Decision Making or MCDM approach). Financial, technico-economic and ecological indicators are calculated in order to characterise wood production both qualitatively and quantitatively. One of them also tries to express a risk assessment (windthrow hazard). The MCDM method is based on partial aggregation (Electre III) and is used to compare scenarios according to the indicators. This SDSS is integrated in a user-friendly designed software “MGC_Larch” (Make Good Choice for Larch) that helps managers formulate, evaluate and compare different larch silvicultural scenarios according to their priority goals. [less ▲]

Detailed reference viewed: 13 (4 ULg)
Full Text
See detailOn the Model Validation in Nonlinear structural Dynamics,
Kerschen, Gaëtan ULg

Doctoral thesis (2002)

Detailed reference viewed: 53 (12 ULg)
Full Text
See detailContamination and detoxication processes in marine mammals from European coasts
Das, Krishna ULg

Doctoral thesis (2002)

The aim of the present work was to determine the relationships between trace metal levels and different factors such as species, geographic origin, diet, age and general body condition of marine mammals ... [more ▼]

The aim of the present work was to determine the relationships between trace metal levels and different factors such as species, geographic origin, diet, age and general body condition of marine mammals from European seas [less ▲]

Detailed reference viewed: 47 (18 ULg)
Full Text
See detailLes conflits de procédure - étude de droit international privé comparé
Wautelet, Patrick ULg

Doctoral thesis (2002)

The dissertation focuses on multiple (or parallel) proceedings in private international litigation. In international civil disputes it is not unusual that jurisdiction for a particular dispute exists in ... [more ▼]

The dissertation focuses on multiple (or parallel) proceedings in private international litigation. In international civil disputes it is not unusual that jurisdiction for a particular dispute exists in several countries. In that case, each party may be tempted to start litigation in the forum most favorable to it, for example to take advantage of the forum’s more generous recovery rules or the extended possibilities of discovery. Similarly, one party can also choose to initiate proceedings in different countries for the same dispute. The question arises how the legal systems involved will and should deal with such ‘multi-fora’ disputes. The first question addressed is whether and to what extent the international legal community should strive to prevent and/or eliminate such duplicate proceedings. It is undubitable that multiple proceedings could lead to undesirable consequences, such as conflicting judgments. Concurrent jurisdiction also brings about a waste of (scarce) judicial resources, and unnecessary expenses for the litigants. To prevent altogether parallel proceedings from happening however, seems at best utopic or even dangerous. Even if the international community were to achieve a global agreement on a comprehensive set of jurisdiction rules, one cannot imagine that there rules would designate one and only one court for each particular dispute. The desire to concentrate jurisdiction for all related disputes in the hands of one court is indeed but one of the numerous factors influencing the allocation of jurisdiction in international matters. The dissertation therefore undertakes to study the rules and instruments that can be used to resolve parallel litigation once multiple proceedings have been issued. The study learns that courts can adopt a variety of attitudes when faced with parallel litigation. A legal system can decide to ignore altogether the fact that a foreign court is seised of parallel proceedings, relying on the res judicata effect of the first decision to be rendered. It can also decide to abstain from taking up the case in deference to the foreign proceedings. Finally it can try to put an end to the parallel proceedings, by enjoining a party from further proceeding before the foreign court. After an in-depth analysis of these different policies and of the various rules and instruments in which they are embodied, my conclusion is that the preference should be given to mechanisms whereby one court voluntarily declines jurisdiction in favor of the other. The preference I express for a mechanism of ‘international abstention’ is however qualified. Whether or not a court declines jurisdiction in favor of a foreign court, will depend inter alia on the confidence it has that justice will be done in the foreign forum. Even with a minimum level of trust and confidence between jurisdictions, the abstention rule can follow different models, from the continental lis alibi pendens rule, which offers almost automatic priority to the court first seised, to the common law doctrine of ‘forum non conveniens’, where priority goes to the more appropriate forum. Between these two extremes lies a world of nuances. The dissertation therefore attempts to formulate an abstention rule which will reconcile these seemingly opposed visions and explain in which circumstances and under which conditions a court should or can decline jurisdiction in deference to litigation pending before a foreign court. [less ▲]

Detailed reference viewed: 455 (11 ULg)
See detailLa rationalisation dans l'industrie belge de l'Entre-deux-guerres
Geerkens Provin, Eric ULg

Doctoral thesis (2002)

L'industrie belge a-t-elle fait, dans la première moitié du XXème siècle, l'expérience de ce qui, dans des pays voisins, a été étudié sous le nom de rationalisation, ou en a-t-elle été empêchée par la ... [more ▼]

L'industrie belge a-t-elle fait, dans la première moitié du XXème siècle, l'expérience de ce qui, dans des pays voisins, a été étudié sous le nom de rationalisation, ou en a-t-elle été empêchée par la prépondérance en son sein d'activités anciennes, nées de la Révolution industrielle? Pour répondre à cette question, ce livre fait le choix de donner un très large aperçu de mesures d'organisation, saisies autant que possible à l'échelle de l'atelier; il prend appui sur le dépouillement de nombreux fonds d'archives d'entreprises et d'une abondante littérature technique, patronale et syndicale. Il s'attache d'abord aux pratiques d'organisation présentées en référence non exclusive à la taylorisation, à la recherche de la fluidité du processus de production et aux techniques de contrôle centralisé des exploitations; il réserve ensuite une large attention à la gestion du facteur humain, depuis la formation et la sélection professionnelles jusqu'à la politique sociale patronale, en passant par la question du commandement. Enfin, il tente de mesurer l'impact du changement organisationnel sur les conditions d'activité des travailleurs, de saisir leurs réactions et celles du mouvement ouvrier organisé. S'en dégage un bilan nuancé de la rationalisation dans l'industrie belge à la veille de la Seconde Guerre mondiale, prenant en compte le poids de la conjoncture, les contraintes structurelles, internes et externes aux entreprises, et le jeu — conflictuel ou coopératif — des acteurs. [less ▲]

Detailed reference viewed: 27 (6 ULg)
Full Text
See detailApport de la réaction de polymérisation en chaine dans le diagnostic de l'aspergillose invasive
Hayette, Marie-Pierre ULg

Doctoral thesis (2002)

Les aspergilloses invasives surviennent chez les patients immunodéprimés pour la plupart neutropéniques. leur diagnostic est souvent trop tardif et la mise sous traitement aboutit à l'échec dans plus de ... [more ▼]

Les aspergilloses invasives surviennent chez les patients immunodéprimés pour la plupart neutropéniques. leur diagnostic est souvent trop tardif et la mise sous traitement aboutit à l'échec dans plus de 50% des cas. Aussi, il est nécessaire de développer de nouvelles techniques de diagnostic plus sensibles. Dans cette étude deux volets ont été développés: tout d'abord la mise au point d'une technique de polymérisation en chaîne pour la détection d'ADN d'Aspergillus sp. et son utilisation dans les lavages bronchoalvéolaires. Cette étude a montré que la technique développée était sensible et spécifique mais ne permettait pas de distinguer entre infection et colonisation. La seconde étude a consisté en la mise au point d'un modèle expérimental d'aspergillose invasive sur le lapin, dans le but d'évaluer la détection d'ADN fongique par PCR sur sang périphérique, parallèlement à la détection d'antigène galactomannane sérique par ELISA. Les résultats de cette étude ont montré qu'il était possible de détecter de l'ADN fongique dans le sang mais de façon inconstante et souvent postérieurement à la détection d'antigène galactomannane par ELISA. Cela permet de conclure que la place de la PCR Aspergillus sur sang périphérique reste encore à définir dans le diagnositic des aspergilloses invasives. Cependant elle peut être utilisée en complément d'autres techniques diagnostiques. [less ▲]

Detailed reference viewed: 63 (4 ULg)
Full Text
See detailSynthesis and potential of new functional aliphatic polyesters
Xudong, Lou

Doctoral thesis (2002)

Detailed reference viewed: 11 (5 ULg)
See detailEtude de l'effet d'un stimulus gravitationnel induit artificiellement sur la formation du bois de tension et du bois opposé dans de jeunes pousses de peuplier (Populus euramericana cv "Ghoy"
Jourez, Benoît ULg

Doctoral thesis (2002)

The tension wood of arborescent dicotyledons can be distinguished by its anatomical, chemical, physical and mechanical characteristics which give to it a specific behavior. In its development interfere ... [more ▼]

The tension wood of arborescent dicotyledons can be distinguished by its anatomical, chemical, physical and mechanical characteristics which give to it a specific behavior. In its development interfere complex mechanisms able to perceive endogenous and exogenous stimuli acting on the cambial zone during the new cell differentiation and maturation. Its particular distribution in the shoot induces a dissymetrical distribution of growth stresses capable to modify their orientation. The supporting thesis of this study maintains the idea of a quantitative responses of the young shoots of poplar which develop under controlled conditions, as a reaction to a gravitational stimulus artificially induced. To demonstrate this, the experimental approach consisted on one hand, to determine the reaction induced by the stimulus in the wood structure. On the other hand, to attempt to establish the existence of a modulation of the shoots reaction in regard of the duration and of the intensity of the stimulus application, while applying increasing levels of these parameters. The anatomical observations were based on microscopic cross sections taken from the base of the shoots. The analysis of these sections were realized using an image analyzing technology. The anatomical modifications induced by the stimulus concern all the elements of the wood structure. They generate repercussions on the physical properties of the wood of micro samples composed exclusively of gelatinous fibers. As the stimulus is applied during increasing duration it appears that the formation of the tissues in the opposite wood and in the tension wood requires a relatively short time of presentation. It shows also the rapidity of reaction of the shoots estimated by the latent period. Short duration events are capable to produce repercussions on the wood structure. The rate of cellular division, more important in the tension wood tissues, as measured on the upper face of a bent axis leads to a pith eccentricity in the shoot. The formation of tension wood, a modification of the secondary growth and the movement of the axis are the major modifications quantitatively linked to the intensity of the stimulus. Moreover, the evolution of the measured parameters demonstrates the existence of a natural equilibrium position of the shoots, different from the vertical position, and to which they attempt to return by an straightening or a sagging motion induced by the tension wood tissue. All these results indicated that tension wood should not be considered as a defect but rather as a efficient mechanism for the shoot allowing adaptation and harmonious development in its own environment. [less ▲]

Detailed reference viewed: 16 (3 ULg)
Full Text
See detailRésonances et dissipations
Jancart, Sylvie ULg

Doctoral thesis (2002)

The discovery of protosolar disks and exoplanets around several stars has revitalized the research combining mean motion resonances and dissipative forces in the elliptic spatial restricted three-body ... [more ▼]

The discovery of protosolar disks and exoplanets around several stars has revitalized the research combining mean motion resonances and dissipative forces in the elliptic spatial restricted three-body problem. This thesis presents that combination in three different ways. A semi-numerical model simulates the behaviour of test particles under the effects of well- known dissipative forces or generic forces. This model is valid for any orbital eccentricities or inclinations, even at high values, and captures around sym- metric and asymmetric equilibria are reproduced. The most common result is the capture in Kozai resonance. Analytical models have been developed to understand the influence of each particular term in the development of the hamiltonian function. An other faster numerical model allows the possibility of following the evolution of thousands of test particles. This mapping has been applied in the study of the stability of the F-ring of Saturn. [less ▲]

Detailed reference viewed: 10 (3 ULg)
Full Text
See detailQualité spectrales des tests standardisés universitaires - Mise au point d'indices édumétriques d'analyse de la qualité spectrale des évaluations des acquis des étudiants universitaires et application aux épreuves MOHICAN check up '99
Gilles, Jean-Luc ULg

Doctoral thesis (2002)

For several decades almost all European university institutions have been faced with a strong increase in their number of students whereas the allocated budgets have not increased in proportion (Gibbs ... [more ▼]

For several decades almost all European university institutions have been faced with a strong increase in their number of students whereas the allocated budgets have not increased in proportion (Gibbs & Jenkins, 1992). The universities of the French Community of Belgium do not escape this heavy tendency: compared with the 1972 situation, the number of students has increased to 150% and, in constant francs, subsidies have remained unchanged (Debry & al., 1998). This situation leads in the first cycles of studies, where the students are most numerous, to a massive use of standardized examinations with multiple choice questions (MCQ) which makes the assessment of great groups of students possible within reasonable time delays. The confidence degrees technique associated to the MCQ makes it possible to bypass the "binary" character of students assessment performance (the selected proposal is either correct, or incorrect) provided that a series of methodological rules are followed called "admissible probability measurement procedures" by Shufford & al. (1966). Among those rules: ask the students to express their confidence in a numerical (probabilistic) scale. The one adopted offers 6 degrees of subtlety: 0%, 20%, 40%, 60%, 80% and 100% of confidence. By inviting the student to accompany his choice by the percentage of chances to be correct he attributes to his answer, we allow more subtleties in the analysis of his performances. At one extreme, the choice of a wrong answer accompanied by the maximal percentage of certainty (100%) presents the worst situation, in which the student provides an erroneous answer by estimating that he has a maximal chances to be correct. On the other hand, the student who answers correctly with a maximal certainty shows an assured performance. Between these two extremes, other zones can be identified in the "spectral analysis" of the performances. Jans & Leclercq (1999) propose a specific terminology. They distinguish "ignorance" (correct answer and weak certainty), from "partial knowledge" (correct answer and average certainty), and "perfect knowledge" (correct answer and high certainty). Such spectral gradations were also considered by these authors in the case of an incorrect answer ("mistaken knowledge" and " fallacious knowledge" or "dangerous knowledge"). Usually, the confidence percentages which accompany the MCQ answers are used to deliver more subtler feedback on each student's spectral performances. The innovative aspect of our approach lies in the fact that we have exploited the confidence percentages provided by the students to provide spectral information on the quality of the questions (as opposed to information on the quality of students performances). Our research thus led to the development of a series of original indices for the analysis of assessments' spectral quality. These spectral indices are intended to be used when the assessor must highlight problematic MCQ and, within those, the proposals which contain anomalies. Our starting intuition for the construction of these new indices is as follows: logically the students who answer a question correctly should provide percentages of certainty higher than the students who answer incorrectly. Thus, for a multiple choice question which functions normally from the point of view of the certainty percentages use, we should observe among the students who choose the correct answer a tendency to answer with higher percentages of certainty and, in parallel, among the students who choose a wrong proposal a tendency to answer with lower percentages of certainty. We will then say that there is "spectral coherence". In the case where this situation does not arise, for example when students choose higher certainty percentages for one of the incorrect answers rather than for the correct answer, we are confronted with a problem of inconsistency in the use of the percentages of certainty, we will then speak of "spectral inconsistency". To measure spectral coherence we created two new types of indices starting from the classical point bi serial correlation coefficient (classical rpbis) calculation principle. Let us recall that in the case of the rpbis, the choices or the rejections (1 or 0) of each proposal of a MCQ are correlated with the numbers of correct answers obtained to the whole test. The classical rpbis makes it possible to evaluate up to what extent each proposed alternatives solutions of each MCQ discriminates the students according to the criterion of the number of correct answers. Logically, one expects that the students who collect a high number of correct answers tend to choose the correct answer for a given question and that the students which collect a lower number have a tendency to choose an incorrect proposal. The two new types of spectral coherence indices of measurement are: (1) the Spectral Contrasted rpbis (rpbis SC) and (2) the Spectral Contrasted rpbis calculated after Turbo analysis (rpbis SCT). During a former research we had already used information related to the confidence degrees to calculate a new type of rpbis, the Spectral rpbis or rpbis S (Gilles, 1998). The rpbis S was developed in order to analyze the tendency to use higher certainty in case of correct answers than in case of incorrect answers. Within the framework of this thesis we propose a first improvement of the rpbis S by implementing a "contrasted treatment" for the MCQ incorrect proposals. We use the name "rpbis SC" to indicate the rpbis S is constructed with a "Contrasted treatment" which consists in the computation of the rpbis SC of an incorrect answer by using the data of the students who choose this wrong response in contrast with the data restricted to the students who choose the correct answer excluding the data from the students who choose an other incorrect answer. The advantage lies in the elimination of the data of the students who have chosen the other incorrect answers. This avoids introducing into the measurement of spectral coherence, the "background noise" generated by the data of the other incorrect answers. The principle of "turbo analysis", consists in operating a selection in the data used for rpbis SC calculation on the basis of the level of realism reached by the students. We can thus increase the reliability of information related to the spectral indices by restricting the data to the students who make less errors in their self-assessments (in their use of certainty percentages). We measure the level of errors of self-assessments made by the students by using the index of realism which varies from 0 to 100 (Leclercq & al, 2000). The name rpbis SCT indicates rpbis SC calculated with a Turbo analysis. The word "turbo" refers to the rise to power of the instrument in terms of the quality of information provided as one restricts oneself progressively to data from sets of students who make less and less errors in their self-assessments. We mention in the index's name the threshold of realism used to select the data. For instance rpbis SCT80 was calculated starting from the data of the students whose realism is equal to or higher than 80 (those who make between 0% and 20% of errors in their self-assessments). The new indices rpbis SC and rpbis SCT, which are in the heart of this research, are designed for the detection of problems located at the "alternatives level" inside the MCQ. We also adapted other spectral indices initially planned for the analysis of students performances so that these indices deliver information on the MCQ performances, at a "question level". On one hand, the index of Realization of the predictions by question (Rq) or the quantity of errors of self-assessment contained in the results for a question and, on the other hand, the index of Centration per question (Cq) or the level of over or under confidence in the results for a question. Lastly, we also measured the level of spectral coherence of a question (NCSq) by taking into account the rpbis SC of the various proposals of a MCQ. The "turbo analysis" principle was also applied to the calculation of these spectral indices. We tried to test these new spectral indices designed for detection of suspect alternatives within suspect MCQ using the data of several thousands of answers and certainty percentages collected during the ten standardized tests of the MOHICAN project (Leclercq & al., 2001). They consisted in ten tests of knowledge of the principal subjects at the level of the end of secondary education which were submitted to groups of students entering first year studies in eight of the nine university institutions of the French Community of Belgium (the number of questioned students varied between 1.392 and 3.846 according to tests). These standardized tests consisted of MCQ for which students were invited to choose one answer and systematically accompany it with a percentage of certainty. The students had, not only to state which was their correct proposal, but also which was the percentage of chances to be correct that they granted to each one of their answers. No academic scores were delivered for these (Check up) MOHICAN tests (each student received a diagnostic feedback and the assessors a global feedback). Anonymity was guaranteed. The choice of the percentages of certainty was thus influenced neither by a scale of tariff of points nor by granting a final score which could have affected the later academic course of the student. The total number of MCQ for the ten MOHICAN tests was 173. For two MCQ among them, the 3rd and the 20th question of the general knowledge test in History and Socio-Economy, the values obtained with the spectral rpbis indicate situations of marked spectral inconsistency, the students tending to give lower percentages of certainty for the answers considered as correct and higher percentages of certainty for the incorrect answers. Studying the proposals of the two problematic MCQ by using the classical rpbis indices, we notice that the two MCQ do not function correctly from the point of view of classical discrimination. When we ask for the opinion of the experts of the contents, they confirm that the two MCQ display problematic results: for one of the questions a wrong proposal could also be regarded as being a correct answer and for the other, there is an error in the encoding of the correct answer. For these two questions there is thus convergence of three different perspectives: (1) that of the classical rpbis, (2) that of the experts and, (3) that of the spectral coherence measured using the spectral rpbis. In the case of the MOHICAN tests, the spectral analysis thus allows to highlight two questions that a more qualitative analysis (posterior opinions of the experts) as well as a classical analysis of discrimination (classical rpbis) also indicate as questions with problems. Does the spectral analysis make it possible to perform better than the classical analysis of discrimination (classical rpbis) for locating the problematic questions and the proposals which contain anomalies? This issue has been tested. We analyzed the 173 MCQ of the 10 MOHICAN tests by using the spectral rpbis (rpbis SC, rpbis SCT80 and rpbis SCT90) as well as the classical rpbis indices. We also reviewed the comments formulated by the experts of the contents in connection with each question. From these analyses it comes out that in addition to the two MCQ already discovered previously, 14 other questions are highlighted. The classical rpbis indicates anomalies in each one of these 14 MCQ. Six MCQ present abnormal values at rpbis SC. Only one MCQ obtains an abnormal rpbis SCT80. No MCQ obtains an abnormal rpbis SCT90. Among these 14 questions, only 3 are pointed as problematic by the experts. With regard to the three questions singled out by the experts, they lead to a set of proposals that only the rpbis SC designates, whereas for the rpbis SCT80 and the rpbis SCT90, they are not highlighted. As far as the classical rpbis is concerned one only of the two problematic proposals for only one of the three questions appears. Therefore, from the point of view of "detection", rpbis SC were more effective to highlight the problems raised by the experts than were the other indices. This analysis shows that the various types of rpbis induce also "false alarms", abnormal values collected by proposals whereas the experts of the contents do not detect particular problems. From this point of view, the rpbis SC, with 7 false alarms, is less effective than the rpbis SCT80 which causes only one of them and less than the rpbis SCT90 which starts none, but rpbis SCT80 and rpbis SCT90 fail by "undetecting" the three questions pointed by the experts. On the other hand the rpbis SC starts less false alarms than the classical rpbis which has 10 false alarms. These qualities of lower "undetection" and less "false alarms" are crucial when the question arises of highlighting a problematic MCQ. When we correct the anomalies contained in certain answers within the MCQ, we can, from now on, not only evaluate the spectral impact of these corrections on the answers ("alternatives level"), but also on the whole question by comparing the values obtained with the indices of NCSq, Rq and Cq before and after changes are operated (at the "question level"). We did this for the two most problematic questions of the general knowledge test in History and Socio Economy and quantified the gains in spectral coherence. The improvement of the spectral coherence of the test was also measured by calculating the average values of the spectral indices at the "question level". These average indices thus made it possible to evaluate the spectral impact on a third "test level" of the assessment. In parallel, we also observed an improvement of the fidelity of the test using the classical Cronbach's alpha coefficient. Using the spectral indices developed within the framework of our thesis and usable with three levels of spectral analysis: "ALTERNATIVES", "MCQ" and "TEST", we open a new way for the analysis of the quality of standardized tests and their regulation. We are now able to evaluate the spectral quality of higher education standardized tests using confidence degrees, to highlight possible anomalies in the questions; and, after corrections, to evaluate the spectral impact of the improvements. This is the main contribution of our thesis to the improvement of procedures that control the quality of standardized tests and, by extension, to the improvement of the reliability of the students scores, which, in fine, constitutes the stake of our edumetric concerns. [less ▲]

Detailed reference viewed: 215 (18 ULg)
Full Text
See detailValidation et exploitation de nouvelles techniques d’imagerie numérique pour la caractérisation des profils comportementaux chez les poissons. Etude de l’influence de facteurs abiotiques et biotiques chez Symphodus ocellatus (Forsskål, 1775) et Chromis chromis Linné, 1758, Labridé et Pomacentridé méditerranéens
Ylieff, Marc ULg

Doctoral thesis (2002)

In many fields dedicated to behavioural research, digital imaging techniques (D.I.T.) now allow the conceptualisation of tools and systems from which the descriptive precision, the capabilities to detect ... [more ▼]

In many fields dedicated to behavioural research, digital imaging techniques (D.I.T.) now allow the conceptualisation of tools and systems from which the descriptive precision, the capabilities to detect, to analyse and to quantify behavioural parameters open original and very promising prospects. By the computing power offered by these new techniques, behavioural scientists have the possibility to quantify a large number of variables, which generally cannot be measured by usual methods of direct observation. The research works presented in this thesis, using new ethometric techniques based on D.I.T., concerned several questions relating to the study of effects of environmental factors on behavioural patterns in fish. By this way, we adjusted, experienced and operated complementary D.I.T. through field and lab researches. After a extensive general introduction devoted to D.I.T. developments and to its contributions in the study of animal behaviour, we present our works in two major parts. These two parts concern: 1.- Field studies conducted with the use of Digital Elevation Modelling (DEM) techniques and using classical methods of SCUBA diving direct observation of fish behaviour; 2.- Lab works using classical methods of direct observation and the techniques of computerised vision systems via the ETHOVISION® Color-Pro computerised video tracking system. The first part exposes research devoted to: Chapter 1 – the Use of DEM of natural aquatic habitats to characterised space and time utilisation modes during reproduction in the Mediterranean wrasse Symphodus ocellatus (Forsskål, 1775). The second part exposes research devoted to: Chapter 2 – the Study of effects of water temperatures and photoperiod (abiotic variables) on the display of reproductive behavioural patterns in the ocellated wrasse (Symphodus ocellatus) acclimated in large tank (5 M³ mesocosm) replicating a Mediterranean biotope; Chapter 3 – the Adjustment and the validation of the ETHOVISION® Color-Pro computerised video tracking system to the characterisation of behavioural patterns in aquarium swimming freely fishes. Study of effects of water temperature (abiotic variable) and congeners density (biotic variable) on spontaneous swimming behaviour in the Mediterranean damselfish Chromis chromis Linné, 1758. A general discussion mainly centred on new potential applications of digital imaging techniques to the fields of behavioural and environmental sciences, concludes this PhD dissertation. [less ▲]

Detailed reference viewed: 27 (5 ULg)
Full Text
See detailContributions to decision tree induction: bias/variance tradeoff and time series classification
Geurts, Pierre ULg

Doctoral thesis (2002)

Because of the rapid progress of computer and information technology, large amounts of data are nowadays available in a lot of domains. Automatic learning aims at developing algorithms able to produce ... [more ▼]

Because of the rapid progress of computer and information technology, large amounts of data are nowadays available in a lot of domains. Automatic learning aims at developing algorithms able to produce synthetic high-level information, or models, from this data. Learning algorithms are generally evaluated according to three different criteria: interpretability (how well the model helps to understand the data), predictive accuracy (how well the model can predict unseen situations), and computational efficiency (how fast is the algorithm and how it scales to large databases). This thesis explores two issues in automatic learning: the improvement of the well-known decision tree induction method and the problem of learning classification models for time series data. Decision tree induction method is an automatic learning algorithm which focuses on the modeling of input/output relationships. While this algorithm is among the fastest and most interpretable methods, its accuracy is not always competitive with respect to other algorithms. It is commonly admitted that this suboptimality is due to the excessive variance of this method. We first carry out an empirical study which shows quantitatively how important this variance is, i.e. how strongly decision trees depend on the random nature of the database used to infer them. These experiments confirm that this variance is detrimental not only from the point of view of accuracy but also from the point of view of interpretability. With the goal of improving both interpretability and accuracy, we consider three variance reduction techniques for decision trees. First, in the goal of improving mainly interpretability, we propose several methods which try to stabilize the parameters chosen during tree induction. While these methods succeed in reducing the variability of the parameters, they produce only a slight improvement of the accuracy. Then, we consider perturb and combine algorithms (e.g. bagging, boosting) which consist in combining the predictions of several models obtained by randomizing in some way the learning process. Inspired by the high variance of the parameters defining a decision tree, we propose an extremely randomized decision tree induction algorithm, called extra-tree, which chooses all parameters at random during induction. The aggregation of several of these extra-trees gives an important reduction of variance and this algorithm compares favorably in terms of accuracy and computational efficiency with both bagging and boosting. Because of the randomization of the parameters, the resulting method is also competitive with classical decision tree induction in terms of computational efficiency. In addition to these two approaches, we propose a ``dual'' perturb and combine algorithm which delays the perturbation at the prediction stage and hence requires only one model. In combination with decision tree, this method actually bridges the gap between single decision trees and perturb and combine algorithms. Of the first, it saves the interpretability (by using only one model), and with perturb and combine algorithm, it shares some of the accuracy (by reducing the variance). The second topic of the thesis is the problem of time series classification. The most direct way to solve this problem is to apply existing learning algorithms on low-level variables which correspond to the values of a time series at several time points. Experiments with the tree-based algorithms studied in the first part of the thesis shows that this approach is limited. A variance reduction techniques is then proposed specifically for this kind of data which consists in aggregating the prediction given by a classification model for subsequences of time series. Since this method does not provide interpretable models, we propose a second method which extends decision tree tests by allowing them to detect local shift invariant properties, or patterns, in time series. The study proposed in this part of the thesis is only a first step in the domain but our conclusions give some future work directions for handling complex type of data with automatic learning methods. [less ▲]

Detailed reference viewed: 231 (22 ULg)
Full Text
See detailPurificacao, caracterizacao e dosagem radioimunologica de glicoproteinas associadas a gestacao em zebuinos
Melo de Sousa, Noelita ULg

Doctoral thesis (2002)

The pregnancy-associated glycoproteins (PAGs) constitute a large family of aspartic proteinases expressed in the outer epithelial cell layer of the Artyodactyla placenta. In the first part of the present ... [more ▼]

The pregnancy-associated glycoproteins (PAGs) constitute a large family of aspartic proteinases expressed in the outer epithelial cell layer of the Artyodactyla placenta. In the first part of the present work, two biochemical approaches were used to characterize PAGs isolated from zebu (Bos indicus) fetal cotyledons. The first procedure, used to isolate PAG from zebu placenta removed late in pregnancy, included extraction of proteins at neutral pH, acidic and ammonium sulfate precipitations, anion and cation exchange chromatographies. The second procedure, used to investigate PAG in placentas removed at early and mid gestational periods, included protein extractions at acid or alkaline pH followed by pepstatin-A agarose affinity chromatography. A bovine PAG radioimmunoassay was used to monitor the immunoreactivity throughout the isolation procedures. The most immunoreactive fractions issued from cation exchange and affinity chromatographies were analyzed by SDS-PAGE and Western Blot, before transfer to polyvinylidene difluoride (PVDF) membrane for NH2-microsequence determination. By use of SDS-PAGE and Western Blot, different isoforms of PAG with apparent molecular masses from 51 to 69 kDa and isoeletric points varying from 3.1 to 6.7 were identified in placentas from different gestational ages. After CM ceramic chromatography of all except 0.32 M NaCl DEAE fraction, the most immunoreactive proteins revealed N-terminal amino acid sequences (10 to 25 aa long) which were 100% identical to bovine PAG-1. The same sequence (14 aa long) was found after pepstatin-agarose affinity chromatography of proteins extracted from placentas removed earlier in pregnancy. These results converged towards the expression of one major N-terminal PAG amino acid sequence in zebu placentas at different gestational ages. In the second part of this study, two specific RIA systems were developed then used to measure plasmatic PAG concentrations during gestation and postpartum period in Azawak zebu cows. Twelve females palpated per rectum and diagnosed as pregnant were bled at 5-10 days interval approximately from Week 8 of gestation till Week 10 postpartum (pp). One zebu cow initially diagnosed as pregnant showed PAG concentrations lower than the assay sensitivity (<0.20 ng/ml) and did not calve. Another cow showed abnormally high PAG concentrations during gestation, being excluded from the general PAG profile. The 10 other zebu cows gave a very homogeneous PAG profile. In these animals, concentrations increased progressively from Week 8 to Week 35 of gestation (from 6.04.2 to 196.034.8 ng/ml), remaining relatively constant until Week 39 (210.874.8 ng/ml), when they increased sharply to reach their highest level (1,095.6607.2 ng/ml) at parturition. After delivery, PAG concentrations declined significantly (P<0.05) till the Week 2 postpartum (348.4  85.6 ng/ml) and slowly till the Week 10 postpartum. These results revealed that PAG concentrations in zebu cattle were lower than those previously described in taurine breeds between week 35 and 39 of gestation. [less ▲]

Detailed reference viewed: 46 (1 ULg)
Full Text
See detailA study of noncovalent interactions by electrospray mass spectrometry
Gabelica, Valérie ULg

Doctoral thesis (2002)

This work describes a study of noncovalent interactions by electrospray mass spectrometry (ES-MS). In particular, we studied DNA duplexes, DNA complexes with drugs and cyclodextrin complexes with ... [more ▼]

This work describes a study of noncovalent interactions by electrospray mass spectrometry (ES-MS). In particular, we studied DNA duplexes, DNA complexes with drugs and cyclodextrin complexes with aliphatic acids. ES-MS experiments were performed on a hybrid quadrupole-TOF instrument and on an ion trap instrument. Basically two kinds of information can be obtained on the complexes. First, the full scan mass spectra give information on the composition of the solution that is injected, and therefore on the solution-phase stability of the complexes. We compare the ESMS data with the solution-phase data, and discuss the specificity of the observed complexes. The problem of nonspecific aggregation was encountered for the hydrophobic cyclodextrin complexes. We developed a new method for determining simultaneously the equilibrium association constant of [1:1] complexes and the ratio between the electrospray response factors of the complex and the free substrate. A change in the substrate conformation upon ligand binding can be detected by measuring the response factors of the complex and the free host. Second, the collision-induced dissociation of the complexes in the gas phase and the measurement of the amounts of fragments resulting from this dissociation give information on the gas phase kinetic stability of the complexes. It is shown that different collision regimes allow the system to fragment via different pathways, and that fast activation conditions favor the noncovalent dissociation of the complex because this process is entropy favored. The gas-phase kinetic stability measurements on series of homologous complexes suggest that intermolecular interactions like hydrogen bonding, electrostatic interactions, and stacking are conserved in the gas phase. Finally, on the basis of the results obtained on DNA duplexes, we propose a general multistep mechanism for the dissociation of noncovalent complexes. [less ▲]

Detailed reference viewed: 74 (4 ULg)
See detailDevelopment of a novel class of biodegradable (co)polyesters
Latere, Jean-Pierre

Doctoral thesis (2002)

Detailed reference viewed: 15 (5 ULg)
See detailRéactivité du tétroxyde d'osmium et des osmates vis-à-vis des séléniures et des sélénoxydes: application à la synthèse d'alcools allyliques et à la dihydroxylation énantiosélective d'oléfines
Colaux, Catherine ULg

Doctoral thesis (2002)

Ce travail traite de l’équilibre réactionnel établi entre le tétroxyde d’osmium et les séléniures d’une part et les osmates (VI) et les sélénoxydes d’autre part. La thèse se divise en trois grandes ... [more ▼]

Ce travail traite de l’équilibre réactionnel établi entre le tétroxyde d’osmium et les séléniures d’une part et les osmates (VI) et les sélénoxydes d’autre part. La thèse se divise en trois grandes parties : 1. L’oxydation des allylséléniures par l’AD-mix 2. L’utilisation des sélénoxydes comme oxydants des sels osmiques (VI) 3. L’établissement de l’existence de l’équilibre réactionnel précité La réaction d’oxydation des allylséléniures par l’AD-mix nous a permis de décrire une réaction dont la chemosélectivité dépend largement de la nature du groupement sélénié mais également de la nature et du degré de substitution de la double liaison. Nous avons au cours de ce travail comparé la réactivité d’une double liaison carbone-carbone comportant une espèce séléniée en position allylique avec celle de son homologue alkylé. Enfin, nous avons dans deux cas déterminé la stéréochimie absolue et relative des diols séléniés générés par oxydation des doubles liaisons carbone-carbone. L’utilisation des sélénoxydes comme co-oxydants des sels osmiques (VI) a conduit à un nouveau système de dihydroxylation catalytique énantiosélective d’oléfines générant dans tous les cas les diols correspondants avec de bons rendements chimiques et d’excellents excès énantiomériques. Une étude de l’influence exercée par la nature du sélénoxyde sur la réaction nous a permis d’une part de sélectionner le sélénoxyde le plus adapté à notre système mais également de déterminer l’étape limitante de vitesse. Enfin, nous avons posé les bases d’un système de dihydroxylation catalytique énantiosélective impliquant la génération in situ des sélénoxydes par oxydation du séléniure présent en quantité catalytique avec l’oxygène singulet généré photochimiquement. Enfin, nous avons démontré l’existence de l’équilibre reliant les quatre espèces (osmium (VIII), séléniure et osmate (VI), sélénoxyde). Cet équilibre a été mis à profit pour générer sélectivement des alcool allyliques. Dissertation doctorale en Sciences Chimiques Le 28 Janvier 2002 Laboratoire de Chimie Organique de Synthèse Promoteur : Prof. A. Krief [less ▲]

Detailed reference viewed: 11 (2 ULg)
Full Text
See detail2D1/2 Thermal-Mechanical Model of Continuous Casting of Steel Using Finite Element Method
Pascon, Frédéric ULg

Doctoral thesis (2002)

The thesis deals with a numerical model based on the finite element method applied to the continuous casting of steel. The model is developed using the LAGAMINE finite element code and two industrial ... [more ▼]

The thesis deals with a numerical model based on the finite element method applied to the continuous casting of steel. The model is developed using the LAGAMINE finite element code and two industrial applications of steel casting are presented to illustrate the capabilities of the model. The first chapter introduces the continuous casting process, defining the essential terminology and the industrial challenges. Then, the objectives of the model are detailed: the model is applied to the upper part of the caster (in the mould region) as well as to the bending and straightening zones. Chapters two, three and four are dedicated to the theoretical developments of thermal aspects (phase transformations, thermal contraction, heat flow and thermal boundary conditions), mechanical aspects (generalized plain strain state, material constitutive law, ferrostatic pressure and mechanical contact) and the staggered analysis used for thermal / mechanical coupling. In the two next chapters, some numerical results are presented. The model has been first applied to a 125mm square billet in the mould region. The industrial goal is to evaluate the effect of the mould taper on the cooling of the billet (temperature evolution, solid shell growing, stress and strain development). The second application focuses on the risk of transverse cracking during straightening of steel slabs and the effect of some local defects (reduction of secondary cooling, locking or misalignment of rolls). The thesis ends with some concluding remarks and perspectives for the model. [less ▲]

Detailed reference viewed: 69 (16 ULg)
See detailEstimation of additive and dominance genetic variances with Method R
Druet, Tom ULg

Doctoral thesis (2002)

Detailed reference viewed: 12 (4 ULg)
Full Text
See detailDistribution et comportement biogéochimique des métaux en trace dans les sédiments estuariens de l'Escaut.
Bouezmarni, Mohamed ULg

Doctoral thesis (2002)

L’objectif de cette étude est la détermination de la distribution et des propriétés des métaux traces dans les sédiments de l’estuaire de l’Escaut, en vue d’améliorer la compréhension de leur comportement ... [more ▼]

L’objectif de cette étude est la détermination de la distribution et des propriétés des métaux traces dans les sédiments de l’estuaire de l’Escaut, en vue d’améliorer la compréhension de leur comportement biogéochimique. Dans ce but, les distributions des éléments majeurs (Si, Al, Fe, Ca, Mg, Na, K, P, matières organiques et pertes au feu) et mineurs (Cd, Pb, Cu, Zn, Cr, Ni, Co, Mn et Li) ont été examinées dans 59 échantillons de sédiments de surface. L’analyse factorielle des résultats de la composition des dépôts montre que le processus de piégeage dans la zone de turbidité maximale constitue le facteur le plus important qui explique les teneurs des métaux traces dans les sédiments. De plus, la distribution de ces éléments a été étudiée dans différentes fractions granulométriques démontrant l’accumulation des métaux toxiques dans les fractions les plus fines. Elle permis de discuter de manière critique les méthodes de normalisation les plus utilisées dans la littérature et d’adopter la normalisation par le Li comme la méthode la plus adéquate dans le cas de l’Escaut, l’Al reste un bon élément de normalisation sauf pour les sables. La normalisation des concentrations métalliques des sédiments par le Li, montre une accumulation importantes du Cd, du Zn, du Cu, du Pb et du Mn limitée à la zone amont et un comportement conservatif pour le Cr, le Ni et le Co. Une étude de spéciation géochimique des solides a été réalisée par une nouvelle approche basée sur la titration acide des sédiments mis en suspension. Cette méthode permet, contrairement à la méthode de digestion sélective, de préciser la spéciation, la mobilité et la réactivité des métaux traces dans les sédiments. Les fractions les plus réactives des métaux étudiés sont notées dans le cas du Cd (80%), du Pb (70%), du Cu (70%), du Zn (80%) et du Mn (80%), notamment dans les vases anaérobiques de la zone amont. Le piégeage de ces métaux d’origine anthropogénique s’explique par la précipitation du Cd et du Zn sous forme de sulfures, la précipitation du Mn avec les carbonates et le recyclage du Pb avec les oxyhydroxydes de Fer à l’interface eau/sédiment. [less ▲]

Detailed reference viewed: 39 (8 ULg)