Human activities such as mining and smelting of lead (Pb) ores and combustion of alkyllead additives in gasoline have resulted in extensive global Pb pollution. Since the late 1960's studies of polar ice and snow have been undertaken to evaluate the extent of anthropogenic Pb emissions in recent times as well as to investigate changes in anthropogenic Pb emissions in the more distant past. The polar ice sheets have been used to investigate Pb pollution as they offer a long-term record of human activity located far from pollution sources and sample aerosol emissions on a hemispheric scale. Lead isotopes have been previously used to identify sources of Pb in polar snow and ice, while new evaluations of Pb isotopic compositions in aerosols and Pb ore bodies allow more thorough evaluations of anthropogenic Pb emissions. Lead isotopic compositions and Pb and Barium (Ba) concentrations have been measured in snow and ice core samples from Law Dome, East Antarctica, to produce a detailed pollution history between 1530 AD and 1989 AD. Such a record has been produced to evaluate changes in anthropogenic Pb emission levels and sources over the past 500 years, to determine when industrial (anthropogenic) activities first began to influence Antarctica and also to investigate natural Pb fluxes to Antarctica. Additional samples were also collected from Law Dome snow and ice cores to respectively investigate seasonal variations in Pb and Ba deposition, and the influence of the 1815 AD volcanic eruption of Tambora, Indonesia. All samples were measured by thermal ionisation mass spectrometry, for which techniques were developed to reliably analyse Pb isotopic compositions in Antarctic samples containing sub-picogram per gram concentrations of Pb. / Particular attention was given to the quantity of Pb added to the samples during the decontamination and sample storage stages of the sample preparation process. These stages, including the use of a stainless steel chisel for the decontamination, contributed ~5.2 pg to the total sample analysed, amounting to a concentration increase of ~13 fg g-1. In comparison, the mass spectrometer ion source contributed typically 89 +/- 19 fg to the blank, however its influence depended upon the amount of Pb available for analysis and so had the greatest impact when small volumes of samples with a very low concentration were analysed. As a consequence of these careful investigations of the Pb blank contributions to the samples, the corrections made to the Pb isotopic ratios and concentrations measured are smaller than previously reported evaluations of Pb in Antarctica by thermal ionisation mass spectrometry. The data indicate that East Antarctica was relatively pristine until -1884 AD, after which the first influence of anthropogenic Pb in Law Dome is observed. "Natural", pre-industrial, background concentrations of Pb and Ba were - 0.4 pg/g and - 1.3 pg/g, respectively, with Pb isotopic compositions within the range 206Pb/207Pb = 1.20 - 1.25 and 208Pb/207Pb = 2.46 - 2.50 and an average rock and soil dust Pb contribution of 8-12%. A major pollution event was observed at Law Dome between 1884 and 1908 AD, elevating the Pb concentration fourfold and changing 206Pb/207Pb ratios in the ice to ~1.12. Based on Pb isotopic systematics and Pb emissions statistics, this was attributed to Pb mined at Broken Hill and smelted at Broken Hill and Port Pirie, Australia. / Anthropogenic Pb inputs to Law Dome were most significant from ~1900 to 1910 and from ~1960 to 1980. During the 20th century, Ba concentrations were consistently higher than "natural" levels. This was attributed to increased dust production, suggesting the influence of climate change and/or changes in land coverage with vegetation. Law Dome ice dated from 1814 AD to 1819 AD was analysed for Pb isotopes and Pb, Ba and Bismuth (Bi) concentrations to investigate the influence of the 1815 AD volcanic eruption of Tambora, Indonesia. The presence of volcanic debris in the core samples was observed from late-1816 AD to 1818 AD as an increase in sulphate concentrations and electrical conductivity of the ice. Barium concentrations were approximately three times higher than background levels from mid-1816 to mid1818, consistent with increased atmospheric loading of rock and soil dust, while enhanced Pb/Ba and Bi/Ba ratios, associated with deposition of volcanic debris, were observed at mid-1814 and from early-1817 to mid-1818. From the results, it appeared likely that Pb emitted from Tambora was removed from the atmosphere within the 1.6 year period required to transport aerosols to Antarctica. Increased Pb and Bi concentrations observed in Law Dome ice ~1818 AD were attributed to either increased heavy metal emissions from Mount Erebus, or increased fluxes of heavy metals to the Antarctic ice sheet resulting from climate and meteorological modifications following the Tambora eruption. / A non-continuous series of Law Dome snow core samples dating from 1980 to 9185 AD were analysed to investigate seasonal variations in the deposition of Pb and Ba. It was found that Pb and Ba at Law Dome do exhibit seasonal variations in deposition, with higher concentrations of Pb and Ba usually observed during Summer and lower concentrations of Pb and Ba usually observed during the Autumn and Spring seasons. At Law Dome, broad patterns of seasonal Pb and Ba deposition are evident however these appear to be punctuated by short-term deposition events or may even be composed of a continuum of short-term deposition events. This variability suggests that complex meteorological systems are responsible for the transport of Pb and Ba to Law Dome, and probably Antarctica in general.

Coleanthus subtilis ist eine durch die FFH-Richtlinie geschützte, weltweit sehr seltene Art, welche periodisch trockenfallende Standorte in Flussauen (Primärhabitate) bzw. durch den Menschen geschaffene Staugewässer (Sekundärhabitate) besiedelt. Die Erarbeitung der aktuellen Verbreitungsbilder der Art in Sachsen und ihrer Besiedlungshistorie waren Ausgangspunkt für die Analysen zur Ökologie dieser Art. Anhand von Analysen zu den Ausbreitungsmechanismen, Standortansprüchen und Reaktionsmöglichkeiten von C. subtilis wurden grundlegende Beiträge zum besseren Verständnis der Strategien dieser Art erbracht. Die Arbeit gibt Hinweise für eine artenschutzgerechte Bewirtschaftung der Staugewässer insbesondere für die Regulierung von Zeitpunkt, Dauer, Intensität und Häufigkeit von Wasserstandsabsenkungen sowie hinsichtlich einzelner Praktiken im Zusammenhang mit einer fischereilichen Nutzung. Weiterhin zeigt die Arbeit auf, wie C. subtilis auf bestimmte, von den durchschnittlichen Wasserstandsschwankungen abweichende, Bedingungen reagieren kann. Damit sind nun wichtige Grundlagen zum nachhaltigen Schutz und zur Förderung von C. subtilis, insbesondere in Sekundärhabitaten, vorhanden. / Coleanthus subtilis is a globally rare species protected by the European Union Habitats Directive. This grass establishes on periodically drying sites in river floodplains (primary habitats) but also in man made ponds and reservoirs (secondary habitats). The identification of the present distribution patterns as well as the reconstruction of the colonisation history of C. subtilis in Saxony were the basis for analyses about the ecology of this species. By analysing dispersal mechanisms, habitat demands and response traits of C. subtilis fundamental contributions to a better understanding of the strategies of this species are given. The thesis provides information about optimum time, duration, intensity and frequency of water level reductions and about the consequences of different management practices particularly with respect to fishery in ponds and reservoirs inhabiting C. subtilis. Further the thesis shows how C. subtilis can cope with deviations from mean water level fluctuations. In summary important fundamental knowledge for the sustainable conservation and promotion of C. subtilis, particularly in secondary habitats, has been established.

No description available.

Shear strength of precast/prestressed hollow-core (PHC) slabs subjected to concentrated or line loads, especially near supports, may be critical and usually is the governing criteria in the design. This study presents the second phase of a research program, undergoing at the University of Manitoba, to calibrate the shear equations in the Canadian code for predicting the shear capacity of PHC slabs. This phase includes both experimental and numerical investigations using a finite element analysis (FEA) software package. The length of bearing, void shape and size, level of prestressing and shear span-to-depth ratio were investigated. The experimental results were compared to the predictions of the Canadian, American and European codes. It was concluded that the Canadian code is unduly conservative. However, the special European code for PHC slabs resulted in better and more consistent predictions. The FEA suggested that the adequate prestressing reinforcement ratio to obtain highest shear capacity ranges between 0.7% and 1.1%.

Shear strength of precast/prestressed hollow-core (PHC) slabs subjected to concentrated or line loads, especially near supports, may be critical and usually is the governing criteria in the design. This study presents the second phase of a research program, undergoing at the University of Manitoba, to calibrate the shear equations in the Canadian code for predicting the shear capacity of PHC slabs. This phase includes both experimental and numerical investigations using a finite element analysis (FEA) software package. The length of bearing, void shape and size, level of prestressing and shear span-to-depth ratio were investigated. The experimental results were compared to the predictions of the Canadian, American and European codes. It was concluded that the Canadian code is unduly conservative. However, the special European code for PHC slabs resulted in better and more consistent predictions. The FEA suggested that the adequate prestressing reinforcement ratio to obtain highest shear capacity ranges between 0.7% and 1.1%.

Αντικείμενο της διατριβής είναι η επίλυση των προβλημάτων της σκέδασης επιπέδων ακουστικών κυμάτων χαμηλών συχνοτήτων από ένα διαπερατό σφαιρικό κέλυφος με έκκεντρο μαλακό, σκληρό ή διαπερατό πυρήνα και από μια μαλακή σφαίρα κάτω από ένα διαπερατό επίπεδο. Η λύση των προβλημάτων σκέδασης στην περιοχή χαμηλών συχνοτήτων επιδέχεται ανάπτυγμα Taylor σε δυνάμεις του κυματικού αριθμού k, όπου οι συντελεστές του αναπτύγματος (προσεγγίσεις χαμηλής συχνότητας) συνιστούν ακολουθία λύσεων στάσιμων προβλημάτων της θεωρίας δυναμικού. Ένα πρόβλημα σκέδασης μπορεί να δεχθεί προσέγγιση χαμηλών συχνοτήτων όταν το μήκος κύματος της κυματικής διαταραχής είναι πολύ μεγαλύτερο από την ακτίνα της ελάχιστης περιγεγραμμένης σφαίρας του σκεδαστή. Το δισφαιρικό σύστημα συντεταγμένων παρέχει κατάλληλο περιβάλλον για την επίλυση προβλημάτων πολλαπλής σκέδασης από δύο σφαίρες Αυτό ισχύει μόνο στη περιοχή των χαμηλών συχνοτήτων δεδομένου ότι η εξίσωση Laplace επιδέχεται διαμορφωμένο χωρισμό στις δισφαιρικές συντεταγμένες, ενώ δεν συμβαίνει το ίδιο στην εξίσωση Helmholtz. Προσαρμόζοντας το δισφαιρικό σύστημα συντεταγμένων στην δεδομένη γεωμετρία του κάθε προβλήματος απλουστεύεται η περιγραφή των χώρων που ορίζονται από το έκκεντρο σφαιρικό κέλυφος και οι σφαιρικές επιφάνειες του προβλήματός μας περιγράφονται από διαφορετικές τιμές της ίδιας συντεταγμένης μεταβλητής, ενώ ο απομακρυσμένος χώρος περιγράφεται από μια γειτονιά της αρχής των συντεταγμένων στο παραμετρικό χώρο των μεταβλητών η, θ. Επιλύοντας τα αντίστοιχα προβλήματα συνοριακών συνθηκών για μηδενική και πρώτης τάξεως προσεγγίσεις, καταλήγουμε σε αντίστοιχες αναγωγικές εξισώσεις ακολουθιών των συντελεστών ή αντίστοιχα συστήματα αναγωγικών εξισώσεων. Δεδομένου ότι οι ακολουθίες των συντελεστών συγκλείνουν ταχύτατα, περιοριζόμαστε στους πρώτους όρους συντελεστών και οι αναδρομικές εξισώσεις ή τα συστήματα αναγωγικών εξισώσεων ανάγονται σε εξισώσεις πινάκων ή γραμμικά συστήματα εξισώσεων με άγνωστους πίνακες στήλες και συντελεστές των αγνώστων τριδιαγώνιοι πίνακες. Με την πρωτότυπη αυτή μέθοδο προσδιορίζονται ακριβώς οι πρώτοι όροι χαμηλών συχνοτήτων των δύο προσεγγίσεων μηδενικής και οι πρώτης τάξεως, και στη συνέχεια οι προσεγγίσεις του πλάτους σκέδασης και των ενεργειακών διατομών σκέδασης. Μειώνοντας την απόσταση d των κέντρων συμπεραίνουμε ότι το πρόβλημα της σκέδασης ομόκεντρου σφαιρικού φλοιού δεν μπορεί να θεωρηθεί ειδική περίπτωση του προαναφερθέντος προβλήματος. / A plane wave is scattered by an acoustically soft, hard or penetrable sphere, covered by a penetrable non-concentric spherical lossless shell which disturbs the propagation of the incident plane wave field. There is exactly one bispherical coordinate system that fits the given two-sphere obstacle. If the wavelength of the incident field is much larger than the radius of the exterior sphere, Low Frequency Theory reduces the scattering problem to a sequence of potential problems which can be solved iteratively Applying the corresponding boundary value problem for each case, a set of two equations results as well as a recurrence equation with three unknown sequence of coefficients for zero-th order, and the first-order approximation is obtained, by solving two sets of two equations and a recurrence equation with three unknown sequence coefficients each for the soft core or the calculation of the zero–th order coefficients of the hard or penetrable core, leads to a solution of a linear system of two equations with two unknown columns and tri-diagonal square matrices are coefficients of the unknown columns, while the first-order approximation is obtained, by solving two linear systems of two equations with four unknown columns and eight tri-diagonal matrices as coefficients of the unknown columns. Applying the cut-off method for soft, hard and penetrable sphere, the low-frequency coefficients of the zero-th and first-order for the near field as well as the first and second-order coefficients are obtained for the normalized scattering amplitude and cross section. Decreasing the distance d of the centres we conclude that the problem of scattering concentric cell cannot be considered special case of mentioned before problem. A plane wave is scattered by an acoustical soft acoustic sphere embedded into an acoustically lossless half space, which disturbs the propagation of the incident wave field. In the first step, the problem of sound diffraction by only a penetrable plane is solved, were the amplitudes of reflective and diffractive acoustical waves are calculated. In the second step the diffractive as an incident wave is scattered by the embedded acoustical soft sphere. The low frequency zero-th and first order coefficients of the near field are calculated for the soft scatterer and finally the scattering amplitude and cross-section are determined.

Les systèmes turbiditiques forment de grands édifices sédimentaires sous-marins situés au large des fleuves en pied de pente continentale et constituent les dépôts terrigènes les plus distaux d’un système fluviatile. Leur structure interne, définie comme un empilement de systèmes chenal-levées, montre des changements architecturaux au cours du temps dont les forçages sont encore mal connus, et le rôle respectif des facteurs de contrôle interne (lié au fonctionnement propre du système) ou externe (climat, variations du niveau marin, tectonique) reste sujet à débats. Afin de mieux comprendre le rôle de ces facteurs de contrôle, une étude détaillée de l’architecture du système Congo a été réalisée. Cette étude a été menée dans le cadre du projet de recherche Reprezaï (Ifremer/IUEM, depuis 2006), faisant suite aux projets Guiness et ZaïAngo (Ifremer/Total, 1992-2003), projets au cours desquels de nombreuses campagnes en mer ont permis l'acquisition d’une importante base de données géophysiques et géologiques. Une analyse quantitative de différents paramètres architecturaux des chenaux (longueur totale, longueur construite après avulsion, entre autres) de l’Edifice Axial du Congo (derniers 200 ka) révèle des cycles sédimentaires de progradation et rétrogradation des dépôt-centres, les plus grandes rétrogradations correspondant à des avulsions situées très en amont dans l’édifice turbiditique. Les dépôts-centres identifiés à la terminaison des chenaux correspondent à des complexes de lobes (selon la nomenclature de Prélat et al., 2009) relativement allongés dont les volumes (de 3 à 196 km3) varient considérablement dans le temps et l'espace sans pour autant montrer de cycles de variation comme pour les autres paramètres. Le volume cumulé de ces complexes de lobes reconnus représente jusqu'à 31% du volume de l’Edifice Axial. L’étude multiproxies (datations 14C, δ18O, mesures XRF, Carbone Organique Total, …) de carottes prélevées sur les chenaux les plus progradants et les plus rétrogradants, a fourni un cadre chronologique permettant de caler temporellement ces cycles sédimentaires. L’évolution temporelle des paramètres architecturaux a été comparée avec les variations des signaux paléoenvironnementaux et paléoclimatiques (pollens, COT, mesures XRF et MSCL, argiles…) enregistrés par une carotte de référence située en domaine hémipélagique et contemporaine de la construction de l’Edifice Axial. Les résultats mettent en évidence un lien étroit entre l’évolution architecturale de l’édifice turbiditique et les décharges fluviatiles du Congo. Ces dernières, mais aussi l’extension du couvert végétal sur le bassin versant, contrôlent le rapport sable/argile et donc, en partie, la capacité de transport des courants turbiditiques. Ces deux facteurs dépendent des précipitations liées aux variations d’intensité de la mousson en relation avec les cycles de précession (19-23 ka, Milankovitch). Un scénario de mise en place des systèmes chenal-levées et lobes de l’Edifice Axial en fonction des variations climatiques arides/humides sur le bassin versant a ainsi pu être proposé grâce à de bonnes contraintes stratigraphiques pour les derniers 40 ka. En période aride, les courants turbiditiques, peu fréquents et avec un fort rapport sable/argile qui diminue leur capacité de transport, sont propices à l'aggradation des chenaux, engendrant des conditions favorables pour la création ultérieure d’avulsions en amont. En période de transition aride/humide, lorsque le couvert végétal est encore peu développé, la décharge fluviatile accompagnée d’une charge solide importante augmente le volume des courants turbiditiques, favorisant les avulsions en amont du système. Enfin, les périodes humides, caractérisées par un débit liquide fort associé à une charge solide essentiellement argileuse qui décroît au fur et à mesure que le couvert végétal s’étend, génèrent des courants de turbidité dont la capacité de transport diminue au cours du temps […] / Turbidite systems are huge submarine sedimentary fans located off rivers, at the foot of the continental slope. They constitute the most distal terrigeneous deposits of a fluvial system. Their internal structure, defined as a stacking of channel-levee systems, show architectural changes through time. Forcing factors of these architectural changesare still poorly understood, and the respective role of internal (related to the own functioning of the system) or external forcing factors (climate, seal-level variations, tectonics) remains debated. To better understand the role of these controlling factors, a detailed study of the Congo system architecture was carried out. This study was conducted as part of the Reprezaï research project (Ifremer/IUEM, since 2006), following the Guiness and ZaïAngo projects (Ifremer/Total, 1992-2003), during which many oceanographic surveys allowed acquiring an important geophysical and geological data base. A quantitative analysis of different architectural parameters from the channels (e.g. total length, length built after avulsion,…) of the Congo Axial Fan (last 200 ka) reveals progradational-retrogradational sedimentary cycles of the depocenters, the highest retrogradations corresponding to avulsions located very upfan. Depocenters identified at the termination of the channels correspond to relatively elongated lobe complexes (according to the nomenclature of Prélat et al., 2009) with greatly variable volumes (from 3 to 196 km3) both in time and space, without any cyclicity like that identified by other parameters. The cumulative volume of these lobe complexes represents up to 31% of the Axial Fan volume. The multiproxies study (14C dating, δ18O, XRF measurements, Total Organic Carbon…) of cores sampled on the most prograding and retrograding channels provided a chronological framework to these sedimentary cycles. The temporal evolution of the architectural parameters was compared with changes in paleoenvironmental and paleoclimatic signals (pollens, TOC, XRF and MSCL measurements, clay mineralogy,…) recorded in the sediments of a reference core deposited simultaneously to the Axial Fan, but located outside the turbidite flows. Results highlight a strong link between the architectural evolution of the turbidite system and the Congo River discharge. These, as well as the expansion of the vegetation cover in the catchment area, control the sand/clay ratio and thus, at least partly, the transport capacity of turbidity currents that build the channel-levee systems. Both factors depend on rainfall related to the monsoon intensity variations in relation with precession (19-23 ka Milankovitch). A scenario for the deposition of channel-levee systems and lobes of the Axial Fan in link with arid/humid climate variations in the catchment area has been proposed thanks to good stratigraphic constraints for the last 40 ka. During arid periods, turbidity currents are infrequent and present a high sand/clay ratio which decreases their transport capacity. These turbidity currents are suitable to channel aggradation, generating favorable conditions for the subsequent creation of upfan avulsion. During periods of transition of arid to humid conditions, when the canopy is still underdeveloped, fluvial discharge and significant sediment load increases result in an increase of the turbidity currents volume, which favors upfan avulsions. Finally, during humid periods, Congo discharge is characterized by a strong liquid flow and a mainly clayey solid discharge which decreases gradually as the canopy extends. These conditions generate turbidity currents with decreasing transport capacity through time. However the transport capacity of these currents remains sufficient to reach distal portions of the systems and favor channels progradation. This climatic factor seems therefore to regularly disrupt the more perennial internal control which is also highlighted by the Congo Axial Fan architecture.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-07-27T18:13:50Z No. of bitstreams: 1 LUIZ CLÁUDIO ALENCAR DA SILVA - DISSERTAÇÃO PPGEE 2007..pdf: 3068941 bytes, checksum: c6fb80d45b04ea96430f56aef0a5a1f5 (MD5) / Made available in DSpace on 2018-07-27T18:13:50Z (GMT). No. of bitstreams: 1 LUIZ CLÁUDIO ALENCAR DA SILVA - DISSERTAÇÃO PPGEE 2007..pdf: 3068941 bytes, checksum: c6fb80d45b04ea96430f56aef0a5a1f5 (MD5) Previous issue date: 2007-06-11 / Um sistema telerobótico baseado na Internet foi desenvolvido neste trabalho de pesquisa. O sistema consiste de uma interface gráfica de usuário com representação virtual do robô e do ambiente de trabalho, e um painel para programação o -line de tarefas do robô. As tarefas podem ser enviadas via Internet ao ambiente remoto,onde são executados, e o resultado, pode ser visualizado de forma on-line em um ambiente virtual no lado de cliente. O ambiente virtual recebe atualizações via Internet de posições do robô e dos objetos da área de trabalho, permitindo eliminar a necessidade de uma transmissão de imagens de vídeo. Ou so do ambiente virtual neste sistema permite que:(1) atrasos de tempo inerentes à Internet sejam suprimidos; (2) o trabalho do operador seja simplificado e acelerado, se comparado com sistemas que usam imagens de vídeo. Entretanto, os ambientes virtual e físico devem se manter sincronizados de modo que o ambiente virtual possa mapear de forma mais realista o ambiente físico, principalmente se este for dinâmico. Para tratar com essa questão, neste trabalho é apresentada uma solução baseada no uso de redes de Petri para fazer a integração dos ambientes virtual e físico através de técnicas de processamento de imagens, de modo que os mesmos se mantenham sincronizados garantindo atualizações do ambiente virtual para operações subseqüentes. / An Internet-Based-Telerobotic system has been developed in the research here reported. The system consists of a graphic user interface with virtual representation of the robot and of the worksite, and a panel for off-line planning robot tasks. the tasks can be sent via internet to the remote environement, where they are executed, and the result, can be visualized on-line in a virtual environement, at the client side. the virtual environement receives updapes of robot and objects positions at the robot worksite via Internet, thus eliminating the need for a full image transmission. The use of virtual environement in this systems allows: (1) time delays inherent to Internet to be suppressed; (2) work of the operator to be simplified and accelerated, compared to sistems that useTV images. However, the virtual and physical environments have to remain synchronized so that the virtual environment can fit the physical environment in a more realistic way, mainly if it is a dynamic one.To deal with this issue, this work proposes a solution based on Petri nets to do the integration of virtual and physical environments through the image processing techniques, so that they remain synchronized and assuring updates of virtual environment for subsequent operations.

La perspective de technologies nanométriques permet d'envisager l'avènement de processeurs constitués de centaines de coeurs de calcul. Néanmoins, l'utilisation de ces processeurs nécessitera de pallier aux problèmes de fiabilité et de variabilité inhérents à ces procédés de fabrication agressifs. Dans cette thèse, nous présentons un ensemble cohérent de techniques pour l'utilisation de processeurs multi-coeurs massivement parallèles, soumis à de forts taux de variabilité et de défaillance. Tout d' abord, la fiabilité du réseau d'interconnexion est abordée, avec la présentation de plusieurs algorithmes de routage tolérants aux fautes, sans interblocages et sans table de routage pour une meilleure scalabilité. Les différentes variantes de ces algorithmes permettent d'ajuster la complexité du réseau sur puce, en fonction des besoins en fiabilité des applications. A titre d'exemple, le plus performant des algorithmes de routage peut acheminer les paquets tant qu'il existe un chemin sans défaillance, et ce jusqu'à 40% de ressources défectueuses. Plusieurs évolutions ont également été étudiées afin d'améliorer les performances du réseau en présence d'un nombre important de fautes. Ensuite, nous proposons une technique auto-adaptative de gestion des applications parallèles, basée sur un routage tolérant aux fautes. L'affectation dynamique des tâches se base sur la recherche adaptative des noeuds de calcul, afin de diminuer la consommation énergétique de l'application en présence de variabilité. Enfin, nous présentons un modèle de simulation de haut-niveau appelé VOCIS (Versatile On-Chip Interconnect Simulator), développé pendant cette thèse. Il permet l'étude approfondie des réseaux d'interconnexion et des routages tolérants aux fautes dans des conditions complexes, afin de répondre aux contraintes propres à ce travail. Nous décrivons son architecture et ses capacités de visualisation. Finalement, nous analysons et illustrons plusieurs résultats expérimentaux originaux obtenus avec ce modèle. / The perspective of nanometric technologies foreshadows the advent of processors consisting of hundreds of computation cores. However, the exploitation of these processors will require to cope with reliability and variability issues inherent to these aggressive manufacturing processes. In this thesis, we present a coherent set of techniques for the utilization of many-cores processors subject to high defect and variability rates. First, the interconnection network reliability is addressed, with the presentation of several deadlock-free fault-tolerant routing algorithms, without routing tables for improving their scalability. The different variants of these algorithms allow for the tune-up of NoC complexity, depending on applications' reliability requirements. For example, the most performant routing algorithm is able to transmit packets as long as a fault-free path exists, with defect rates as high as 40%. Evolutions have also been studied, in order to improve the interconnect performances in the presence of a large number of faults. Second, we propose a self-adaptive technique for the management of parallel applications, based on a fault-tolerant interconnect. The dynamic tasks mapping is based on the adaptive search of computing nodes, in order to reduce the application's energy consumption in the presnece of variability. Third, we present a high-level simulation model named VOCIS (Versatile On-Chip Interconnect Simulator), developed during this thesis. The model allows in-depth study of interconnection networks and fault-tolerant routings under complex settings, in order to meet the specific constraints of this work. The architecture and visualization features are described. Finally, we analyse and illustrate original experimental results obtained with this model.

INTRODUÇÃO: A orbitopatia de Graves é uma doença autoimune cujas manifestações clínicas ocorrem por infiltração tecidual inflamatória. Apresenta estágio ativo inicial e estágio sequelar posteriormente, podendo se manifestar nas formas clínicas miogênica ou lipogênica. A inflamação da órbita pode gerar importante congestão, agravando manifestações clínicas e comprometendo o sucesso da terapia específica em alguns pacientes. O objetivo deste trabalho é a avaliação da congestão orbitária feita pelo estudo das características do fluxo sanguíneo na veia oftálmica superior, por meio do Doppler colorido de órbitas, nos grupos de pacientes com orbitopatia de graves em fase ativa, em fase sequelar miogênica e em fase sequelar lipogênica, além do grupo controle. MÉTODOS: Estudo transversal prospectivo realizado entre maio de 2006 e abril de 2008 no Serviço de Órbita do HCFMUSP. Noventa órbitas de 46 pacientes com orbitopatia de Graves foram incluídas, assim como 38 órbitas de 20 pacientes do grupo controle. Foi feito exame oftalmológico completo e avaliação pelo DCO em todos os pacientes. Pacientes com orbitopatia de Graves foram divididos em grupos com doença em atividade e doença em estágio sequelar nas apresentações miogênicas ou lipogênicas. O sentido, as velocidades máxima e mínima do fluxo sanguíneo na veia oftálmica superior, e o intervalo entre as mesmas, foram aferidos. Os achados foram comparados entre os diferentes grupos e analisados também em relação à restrição muscular, exoftalmometria, escore de atividade clínica, idade, sexo, lado estudado, comorbidades e tabagismo. RESULTADOS: Foi encontrada diferença no sentido do fluxo sanguíneo da veia oftálmica superior (p<0,001), na velocidade máxima (p=0,001), na velocidade mínima (p=0,002) e no intervalo entre a velocidade máxima e a mínima do mesmo (p<0,001) entre os grupos de estudo. A velocidade máxima do fluxo sanguíneo apresentou correlação com a restrição muscular (p=0,003), o escore de atividade clínica (p<0,001) e a exoftalmometria (p=0,015), enquanto a velocidade mínima correlacionou-se com restrição muscular (p<0,001) e escore de atividade clínica (p=0,006). CONCLUSÕES: Há diferença mensurável na velocidade e no sentido do fluxo sanguíneo da veia oftálmica superior entre pacientes dos diferentes grupos de estudo / INTRODUCTION: Graves orbitopathy is an autoimmune disease whose clinical manifestations are secondary to inflammatory tissue infiltration. It presents an initial active stage and a later latent stage, and can manifest myogenic or lipogenic clinical forms. Inflammation of the orbit can generate significant congestion, worsening clinical manifestations and limiting the success of specific therapy in some patients. The objective of this study is to evaluate orbital congestion studying the characteristics of blood flow in the superior ophthalmic vein with the color Doppler in groups of patients with Graves Orbitopathy in active stage, lipogenic latent stage and myogenic latent stage, and patients from the control group. METHODS: Transversal study made between May 2006 and April 2008 in the Orbit Department of University of São Paulo Medical School. Ninety orbits from 46 patients with Graves orbitopathy and 38 orbits from 20 patients of the control group were included. All patients underwent complete ophthalmological exams including color Doppler of the orbits. Patients with Graves orbitopathy were divided in active stage, latent lipogenic stage and latent myogenic stage groups. The direction of venous blood flow in the superior ophthalmic vein and its maximum and minimum velocities - including the interval between them - were measured. The results were compared among the groups and analyzed also in relation to muscular restriction, exophthalmometry, clinical activity score, age, sex, studied orbit side, comorbidities and use of tobacco. RESULTS: There was difference in blood flow direction of the superior ophthalmic vein (p<0,001), blood flow maximum speed (p=0,001), blood flow minimum speed (p=0,002) and blood flow speed interval (p<0,001) among the groups of this study. Blood flow maximum speed presented correlation with muscle restriction (p=0,003), clinical activity score (p<0,001) and exophthalmometry (p=0,015). Blood flow minimum speed presented correlation with muscle restriction (p<0,001) and clinical activity score (p=0,006). CONCLUSIONS: There are measurable difference in speed and direction of blood flow in the superior ophthalmic vein among patients of the different groups of this study