With the explosion of social media, consumers are gaining control in social reach and can utilize online platforms to create and share misleading information when doing so helps to meet an end. This dissertation, consisting of three separate essays, represents an attempt to address how misleading information is created, how it is disseminated, and how it can be eliminated. Essay One (Chapter 2) uses a mixed-method approach to explore the Dark Triad, proactivity, and vigilantism in driving self-created misleading information sharing. Additionally, this essay introduces a dual-process model of inoculation theory to the marketing and consumer literature that shows how consumers autoinoculate when building justification to engage in malicious behavior. This process includes both automatic and analytical components that initiate a Negative Cascade. Without a larger number of posts, these initial messages may be overlooked. However, herd inoculation can develop when a message begins to sway larger groups. Essay Two (Chapter 3) determines that authentic messages from the original poster are most believable and most likely to initiate a Negative Cascade. This confirmation through mere exposure can then initiate herd inoculation as it flows to other consumers and develops further credibility. The implicit bystander effect is active when in the presence of larger groups. Findings suggest herd inoculation may go unbroken since posters exposed to a positive counter-cascade are less likely to both participate in a forum and post positive messages. Essay Three (Chapter 4) shows that when a consumer shares a message that develops into a Negative Cascade, additional effort is required to halt the consumer herd inoculation. The studies uncover the need for an overt response from the original poster to stop future sharing of misleading information and the role of brand-enacted quarantines in the prevention of the autoinoculation of consumer vigilantes. This dissertation shows how one message can become a much bigger problem for a brand when misinformation spreads. Insights within the dissertation provide numerous outlets for future research and numerous tools and recommendations for both academics and practitioners that hope to understand how misleading information is created, disseminated, and can be eliminated.

No description available.

Active seismicity and volcanism north of Washington state’s Mount St. Helens provide key ingredients for hydrothermal circulation at depth. This broad zone of seismicity defines the St. Helens Seismic Zone, which extends well north of the volcanic edifice below where several faults and associated fractures in outcrop record repeated slip, dilation, and alteration indicative of localized fluid flow. Candidate reservoir rocks for a geothermal system include marine metasediments overlain by extrusive volcanics. The colocation of elements comprising a geothermal system at this location is tested here by analysis of the structures potentially hosting a reservoir, their relationship to the modern stress state, and temperature logs to a depth of 250 m. Outcrop mapping and borehole image log analysis down to 244 m document highly fractured volcaniclastic deposits and basalt flows. Intervening ash layers truncate the vertical extent of most structures. However, large strike slip faults with well-developed fault cores and associated high fracture density cross ash layers; vein filling and alternation of the adjacent host rock in these faults suggest they act as vertically extensive flow paths. These faults and associated fractures record repeated slip, dilation, and healing by various dolomite, quartz, and hematite, as well as clay alteration, indicative of long-lived, localized fluid flow. In addition, where these rocks are altered by igneous intrusion, they host high fracture density that facilitated heat transfer evidenced by associated hydrothermal alteration. Breakouts in image logs indicate the azimuth of SHmax in the shear zone is broadly consistent with both the GPS plate convergence velocity field as well as seismically active strike slip faults and strike-slip faults mapped in outcrop and borehole image logs. However, the local orientation of SHmax varies by position relative to the edifice and in some cases with depth along the borehole making a simple regional average SHmax azimuth misleading. Boreholes within the seismic zone display a wider variety of fracture attitudes than those outside the shear zone, potentially promoting permeability. Temperature profiles in these wells all indicate isothermal conditions at average groundwater temperatures, consistent with rapidly flowing water localized within fractures. Together, these results indicate that the area north of Mount Saint Helens generates and maintains porosity and permeability suggesting that conditions necessary for a geothermal system are present, although as yet no modern heat source or hydrothermal circulation was detected at shallow depth. / Geology

This investigation was conducted to determine how design variables affect the basic flow characteristics and performance of compressor cascades with stalled flow. The performance of stalled cascades is required for analyzing with computer models stall and post stall behavior of axial flow compressors. In this investigation, the unsteadiness of the stalled flow and the stalled cascade performance as indicated by the blade normal force and total pressure loss were evaluated. The investigation consisted of both experimental and numerical phases. The effects of stagger, angle of attack and Reynolds number were investigated experimentally using a two-dimensional cascade facility. Surface flow visualization, smoke flow visualization, velocity measurements and pressure measurements were used to evaluate the flow. The flow was modeled numerically by solving the Navier-Stokes equations for a cascade of flat plates. All of the results indicate that blade stagger is a key variable in determining the performance of a stalled cascade. The smoke flow visualization revealed that propagating stall occurred for the cascades with staggers of 36.5 and 45 degrees at all angles of attack greater than or equal to the angle of full stall inception. Propagating stall was never observed for the cascade with a 25 degree stagger. The flow in the passages of the 25 degree stagger cascade was characterized by two distinct regions of flow, a potential or inviscid region with no losses and a separated region with high losses. For the two higher stagger cascades, two distinct regions did not exist. The performance data for the cascades were consistent with the qualitative results obtained in the flow visualization. When presented as a function of angle of attack, the performance parameters indicate that the loss curve is steeper, the maximum value of the normal force coefficient is lower, and the maximum normal force occurred at a lower angle of attack for the higher staggered cascades. The numerical study revealed trends in cascade performance similar to those found in the experimental study and showed that the predicted losses continue to rise as the limiting inlet angle is approached. / Ph. D.

Given several related critical infrastructure (CI) networks, such as power grid, transportation, and water systems, one crucial question emerges: how to model the propagation of failed facilities and predict their spread over time to the whole system? Given digital surveillance data, can we predict the impact of Influenza-Like Illness (ILI), including the percentage of outpatient doctors visits, the season duration, and peak? These two questions are related to modeling and predicting the incidence of different types of contagions. In the case of CI, the contagions are the failures of facilities. In the case of flu spread, the contagions are the infective ILI. In this thesis, in the case of CI, we give a novel model of failure cascades and use it to identify key facilities in an optimization-based approach, called HotSpots. In the case of flu spread, we develop a deep neural network, EpiDeep, to predict multiple key epidemiology metrics. In both of these applications, we use the dynamics of propagation to develop better approaches. By collaborating with Oak Ridge National Laboratory (ORNL) and working on the real CI networks provided by them, we find that HotSpots helps solve what-if scenarios. By using the digital surveillance data reported by the Centers for Disease Control and Prevention (CDC), we carry on experiments and find that EpiDeep is better than non-trivial baselines and outperforms them by up to 40%. We believe the generality of our approaches, and it can be applied to other propagation-based scenarios in infrastructure and epidemiology. / M.S. / Critical Infrastructure Systems (CIS), including the power grid, transportation, and gas systems, are essential to national security, economy, and political stability. Moreover, they are interconnected and are vulnerable to potential failures. The previous event, like 2012 Hurricane Sandy, showed how these interdependencies can lead to catastrophic disasters among the whole systems. Therefore, one crucial question emerges: Given several related CIS networks: how to model the propagation of failed facilities and predict their spread over time to the whole system? Similarly, in the case of seasonal influenza, it always remains a significant health issue for many people in every country. The time-series of the weighted Influenza-like Illness (wILI) data are provided to researchers by the US Center for Disease Control and Prevention (CDC), and researchers use them to predict several key epidemiological metrics. The question, in this case, is: Given the wILI time-series, can we predict the impact of Influenza-Like Illness (ILI) accurately and efficiently? Both of these questions are related to modeling and predicting the incidence of different types of contagions. Contagions are any infective trend which can spread inside a network, including failures of facilities, illness of human, and popular news. In the case of CIS, the contagions are the failures of facilities. In the case of flu spread, the contagions are the infective ILI. In this thesis, in the case of CI, we present a novel model of failure cascades and use it to identify critical facilities in an optimization-based approach. In the case of flu spread, we develop a deep neural network to predict multiple key epidemiology metrics. In both of these applications, we use the dynamics of propagation to create better approaches. By collaborating with ORNL and working on the real CI networks provided by them, we find that F-CAS captures the dynamics of the interconnected CI networks. In the experiments using the wILI data from CDC, we find that EpiDeep is better than non-trivial baselines and outperforms them by up to 40%. We believe the generality of our approaches, and it can be applied to other propagation-based scenarios in infrastructure and epidemiology.

Depuis le début du 20ième siècle, les changements d'usage des terres, la disparition des prédateurs et les régulations de la chasse ont provoqué une augmentation des populations de cervidés dans les forêts tempérés et boréales. Ce phénomène, qui est un grand succès de la conservation de ces espèces, a toutefois conduit à des surabondances qui ont entrainé des effets négatifs en cascades sur la végétation et les communautés animales qui en dépendent.J'ai utilisé l'expérience naturelle qu'est l'introduction du cerf à queue noire sur l'archipel d'Haïda Gwaii pour étudier les conséquences de sa surabondance sur un écosystème tempéré peu perturbé par les activités anthropiques. J'ai ainsi pu mettre en évidence les contrôles descendants directs et indirects qu'exerce l'herbivore sur les Bryophytes (positifs), les plantes vasculaires (négatifs) et sur l'avifaune (négatifs) quand il n'est pas limité par les prédateurs ou la chasse.J'ai complété ces résultats par une analyse régionale sur 20 ans pour montrer que la perte de biodiversité enclenchée par la surabondance de ces cerfs était un phénomène d'érosion continu se prolongeant bien au-delà de l'impact initial. A l'échelle de l'Amérique du Nord, j'ai ensuite pu montrer, conformément aux prédictions faite à partir des études locales, qu'il existait une relation entre surabondance des populations d'ongulés et déclin de l'avifaune du sous-bois du continent. Enfin, j'ai analysé les suivis d'une expérience de réduction des densités de cerfs entamée sur deux îles il y a 13 ans. La végétation et l'avifaune se sont partiellement reconstituées démontrant qu'il est possible (1) de restaurer les réseaux trophiques fortement modifiés et (2) qu'une telle restauration si elle est possible prendra du temps et ne convergera pas (rapidement) vers un état initial. Afin de limiter les conséquences dommageables provoquées par la surabondance des cervidés, la conservation des prédateurs et l'augmentation de la chasse doivent être favorisés. / Since the past century land use changes, elimination of predators and hunting regulations triggered an increase of deer populations in temperate and boreal forests. This remarkable conservation success, lead to deer overabundance and to a cascade of negative effects on vegetation and on animal communities depending on it. I used the natural experiment provided by the introduction of black-tailed deer to Haida Gwaii to analyze the consequences of overabundant herbivore populations on temperate ecosystem only slightly modified by human activities. This unique situation allowed me to demonstrate the direct and indirect top-down effects that herbivores uncontrolled by predation or hunting exert on Bryophytes (positive) and Vascular plants (negative) as well as on songbirds (negative).I used an analysis of regional trends spanning over 20 years to show that biodiversity erosion caused by overabundant deer extends far beyond the initial impact. At the North-American scale I was able to establish a relationship between expanding deer populations and continent wide declines in understory birds. Finally I analyzed the results of an experimental reduction in deer populations initiated 13 years ago on two islands. The understory plants and songbirds responded positively and we show that (1) it is possible to restore trophic networks even after their dramatically modification, (2) that such a restoration while possible takes time and does not (rapidly) converge towards an identified initial state. From a practical standpoint I recommend that to mitigate negative effects of overabundant deer, predator conservation and hunting have to be promoted.

La complexité des mondes virtuels ne cesse d'augmenter et les techniques de modélisation classiques peinent à satisfaire les contraintes de quantité nécessaires à la production de telles scènes. Les techniques de génération procédurale permettent la création automatisée de mondes virtuels complexes à l'aide d'algorithmes, mais sont généralement contre-intuitives et par conséquent réservées à des artistes expérimentés.En effet, ces méthodes offrent peu de contrôle à l'utilisateur et sont rarement interactives.De plus, il s'agit souvent pour l'utilisateur de trouver des valeurs pour leurs nombreux paramètres en effectuant des séries d'essais et d'erreurs jusqu'à l'obtention d'un résultat satisfaisant, ce qui est long et fastidieu.L'objectif de cette thèse est de combiner la puissance créatrice de la génération procédurale avec un contrôle utilisateur intuitif afin de proposer de nouvelles méthodes interactives et intuitives de modélisation de mondes virtuels.Nous proposons plusieurs méthodes combinant génération procédurale et contrôle intuitif, en accentuant graduellement leur aspect interactif. Ainsi, nous présentons :- une méthode de génération procédurale de villages sur des terrains accidentés, dont les éléments sont soumis à de fortes contraintes de l'environnement,- une méthode interactive de modélisation de cascades, basée sur un contrôle utilisateur fin et la génération automatisée d'un contenu cohérent en regard de l'hydrologie et du terrain,- une méthode d'édition de terrains par croquis, où les éléments caractéristiques du terrain sont analysés et déformés pour correspondre aux lignes de crêtes tracées par l'utilisateur,- une méthode de peinture de mondes virtuels, où des techniques de synthèse d'éléments vectoriels sont utilisées pour automatiser la déformation et l'édition de la scène tout en préservant sa cohérence. / The complexity of virtual worlds is increasing and conventional modeling techniques are struggling to meet the constraints and amount of time required for the production of such scenes.Procedural generation techniques allow automated virtual worlds creation using algorithms, but are often non-intuitive and therefore reserved to experimented artists.Indeed, these methods offer few controls to users and are rarely interactive.Moreover, the user often needs to find values of several parameters by performing a series of trial and error, until obtaining a satisfactory result, which is a long and tedious task.The objective of this thesis is to combine the power of procedural modeling with intuitive user control to propose intuitive and interactive methods for modeling virtual worlds.We propose several methods combining procedural modeling and intuitive control, gradually increasing their interactive aspect.Thus we present:- a technique for procedural modeling of villages over arbitrary terrains, whose elements are subject to strong environmental constraints,- an interactive technique for procedural modeling of waterfall sceneries, based on an intuitive user control and the automated generation of a consistent content, in regard of hydrology and terrain constraints,- an sketch-based technique for editing terrains, where terrain features are extracted and deformed to fit the user sketches,- an interactive method for virtual world painting, where example-based synthesis of vectorial elements methods are used to automate deformation and editing of the scene while maintaining its consistency.

Nous présentons une étude expérimentale de l’effet d’une rotation d’ensemble sur les écoulements turbulents statistiquement stationnaires. Dans une première expérience, l’écoulement est entretenu à l’aide de générateurs de tourbillons contrarotatifs agissant de manière périodique dans une cuve en rotation remplie d’eau. Des mesures résolues en temps des trois composantes de la vitesse sont réalisées, dans des plans horizontaux et verticaux, à l’aide d’un dispositif de vélocimétrie stéréoscopique par images de particules embarqué dans le référentiel tournant. L’écoulement étudié présente, conformément à la littérature, une forte anisotropie et montre l’émergence d’un mode 2D énergétique. Pour la première fois expérimentalement, nous décrivons le bilan global d’énergie entre échelles d’une turbulence en rotation à travers la mesure des termes de l’équation de Kármán-Howarth-Monin généralisée au cas inhomogène. Nous mettons ainsi en évidence la présence d’une double cascade d’énergie : directe à petite échelle et inverse à grande échelle, l’échelle de renversement des cascades étant décroissante avec le taux de rotation. Nous évaluons ensuite la puissance injectée qui est intrinsèquement liée au caractère inhomogène de l’écoulement. L’injection d’énergie provient de l’auto-advection des structures turbulentes traversant les frontières de la zone de contrôle. Elle est large bande en échelles et s’étale à mesure que la rotation croît. Nous nous intéressons ensuite à la pertinence des modèles de turbulence d’ondes d’inertie. Nous réalisons tout d’abord une analyse spatio-temporelle qui révèle la présence d’ondes d’inertie linéaires à grande échelle spatiale et grande fréquence temporelle. En revanche, nous montrons que la signature spatio-temporelle des structures turbulentes associées aux échelles et fréquences faibles est brouillée par le processus linéaire de balayage stochastique par le mode 2D énergétique. Dans une seconde expérience, l’écoulement est engendré par une hélice constituée de quatre pales rectangulaires dans une cuve fermée en rotation. Nous évaluons le taux de dissipation d’énergie à travers la mesure de la puissance injectée par le moteur qui entraîne l’hélice. Nous fournissons alors, pour la première fois, une preuve directe de la loi d’échelle du taux de dissipation d’énergie prédite par la turbulence d’onde d’inertie qui est diminuée d’un facteur Rossby par rapport à la loi d’échelle de la turbulence 3D homogène et isotrope. / We present an experimental study of the effect of global rotation on statistically stationary turbulent flows. In a first experiment, the flow is generated with counter-rotating vortex generators acting in a periodic motion in a rotating tank filled with water. Resolved in time measurements of the three component of the velocity are performed, in both horizontal and vertical planes, thanks to a stereoscopic particle image velocimetry system embarked in the rotating frame. The flow has, in accordance with the bibliography, a strong anisotropy and shows the emergence of an energetic 2D flow. For the first time experimentally, we describe the global scale by scale energy budget of a rotating turbulence through the measure of the terms of the inhomogeneous generalization of Kármán-Howarth-Monin equation. We thus reveal a double energy cascade: direct at small scale and inverse at large scale, the scale of cascade reversal decreasing with the rotation rate. Then, we evaluate the injected power into the system which is intrinsically linked to the inhomogeneities of the flow. The energy input comes from auto-advection of turbulent structures through the boundaries of the area considered. It is broadband in scales and spreads as ration increases. We then focus on relevance of inertial wave turbulence models. We first perform a spatiotemporal analysis which reveals the presence of linear inertial waves at large frequencies and scales. However, we show that the spatiotemporal signature of small frequencies and scales are scrambled by the linear process of stochastic sweeping by the 2D energetic mode. In a second experiment, the flow is created thanks to a four-rectangular-blade impeller in a closed rotating tank. We estimate the energy dissipation rate through the measure of the injected power by the motor that drives the impeller. We then bring, for the first time, a direct evidence of the scaling law predicted by inertial wave turbulence models which is fallen by a factor Rossby compared to the scaling law of 3D homogeneous isotropic turbulence.

A highly regio- and diastereoselective approach towards complex 6-membered lactones was developed using allyl/propargyl benzyl ethers and delta keto esters. Crucially, the classical ET reagent SmI2 gave unsatisfactory results and it was necessary to develop and screen new Sm(II) cyclopentadienyl complexes to deliver high selectivity in the transformation. The methodology was extended to a one-pot approach to complex cycloheptanols using SmI2-H2O in a second stage of the process. Samarium(II) folding cascades were developed where simple, linear starting materials are converted to complex polycyclic architectures bearing multiple stereocentres. It was found that, depending on the sidechain in the starting material, it was possible to achieve four different pathways from the common radical intermediate. Crucially, transannular 1,5-HAT from tertiary and benzylic positions was observed to give diverse products. A proposed 1,5-HAT facilitated SmI2-mediated 6-membered lactone radical cyclisations for the first time without an activating proton donor additive. Enantioselective samarium(II) mediated cyclisation cascades were achieved, where simple beta keto esters are converted to complex polycyclic architectures bearing up to five contiguous stereocentres with high diastereo- and enantiocontrol. In the process, a simple and easy to prepare chiral aminodiol was employed which could be recycled after the reaction. Unprecedented, enantioselective transannular radical cascades allowed access to unique 3- dimensional scaffolds inaccessible by other synthetic methods.

Cette thèse cible l'étude de la structure thermique de la croûte supérieure (<10km) dans les arcs magmatiques continentaux, et son influence sur l'enregistrement thermochronologique de leur exhumation et de leur évolution topographique. Nous portons notre regard sur deux chaînes de montagne appartenant aux Cordillères Américaines : Les Cascades Nord (USA) et la zone de faille Motagua (Guatemala). L'approche utilisée est axée sur l'utilisation de la thermochronologie (U-Th-Sm)/He sur apatite et zircon, couplée avec la modélisation numérique de la structure thermique de la croûte. Nous mettons en évidence la variabilité à la fois spatiale et temporelle du gradient géothermique, et attirons l'attention du lecteur sur l'importance de prendre en compte la multitude des processus géologiques perturbant la structure thermique dans les chaînes de type cordillère, c'est à dire formées lors de la subduction océanique sous un continent. / This thesis focuses on the influence of the dynamic thermal structure of the upper crust (<10km) on the thermochronologic record of the exhumational and topographic history of magmatic continental arcs. Two mountain belts from the American Cordillera are studied: the North Cascades (USA) and the Motagua fault zone (Guatemala). I use a combined approach coupling apatite and zircon (U-Th-Sm)/He thermochronology and thermo-kinematic numerical modelling. This study highlights the temporal and spatial variability of the geothermal gradient and the importance to take into account the different geological processes that perturb the thermal structure of Cordilleran-type mountain belts (i.e. mountain belts related to oceanic subduction underneath a continent).