No description available.

Dissimilarity plays a very important role in object recognition. But, finding perceptual dissimilarity between objects is non-trivial as it is not equivalent to the pixel dissimilarity between the objects (For example, two white noise images appear very similar even when they have different intensity values at every corresponding pixel). However, visual search allows us to reliably measure perceptual dissimilarity between a pair of objects. When the target object is dissimilar to the distracter, visual search becomes easy and it will be difficult otherwise. Even though we can measure perceptual dissimilarity between objects, we still do not understand either the underlying mechanisms or the visual features involved in the computation of dissimilarities. For this thesis, I have explored perceptual dissimilarity in two studies – by looking at known simple features and understanding how they combine, and using computational models to understand or discover complex features. In the first study, we looked at how dissimilarity between two simple objects with known features can be predicted using dissimilarities between individual features. Specifically, we investigated how search for targets differing in multiple features (intensity, length, orientation) from the distracters is related to searches for targets differing in each of the individual features. We found that multiple feature dissimilarities could be predicted as a linear combination of individual feature dissimilarities. Also, we demonstrated for the first time that Aspect ratio of the object emerges as a novel feature in visual search. This work has been published in the Journal of Vision (Pramod & Arun, 2014). Having established in the first study that simple features combine linearly, we devised a second study to investigate dissimilarities in complex shapes. Since it is known that shape is one of the salient and complex features in object representation, we chose silhouettes of animals and abstract objects to explore the nature of dissimilarity computations. We conducted visual search using pairs of these silhouettes on humans to get an estimate of perceptual dissimilarity. We then used various computational models of shape representation (like Fourier Descriptors, Curvature Scale Space, HMAX model etc) to see how well they can predict the observed dissimilarities. We found that many of these computational models were able to predict the perceptual dissimilarities of a large number of object pairs. However, we also observed many cases where computational models failed to predict perceptual dissimilarities. The manuscript related to this study is under preparation.

Les recherches sur la justice organisationnelle ont prouvé que l’équité était un concept majeur pour appréhender l’expérience de travail des salariés. Pourtant, les chercheurs en ce domaine ont focalisé leur attention sur la relation verticale entre salariés et managers ou entre les salariés et les organisations auxquelles ils appartiennent, au détriment des rapports entre individus de même niveau hiérarchique. Cette thèse a pour but de présenter le concept de justice dans les équipes de travail en analysant la perception des salariés quant au traitement qui leur ait réservé, à titre individuel, sur leur lieu de travail. Afin de démontrer la pertinence théorique et pratique de la perception individuelle de l’équité dans les équipes de travail, le thème de la justice organisationnelle sera confronté au thème du conflit, au concept de diversité et aux valeurs socio-culturelles des salariés. Les problématiques de recherche se concentrent sur les conséquences de la perception de l´équité entre collègues, sur les relations entre les concepts d'injustice et de conflit, de dissemblance et de satisfaction et sur le rôle modérateur des valeurs socio-culturelles sur la perception de l'équité et du comportement de citoyenneté organisationnelle. Ces questions de recherche sont testées d´une part à travers une étude longitudinale portant sur 256 salariés de firmes allemandes et, d´autre part, à travers deux questionnaires portant sur 448 salariés répartis dans des équipes de travail multinationales en France. Ces données sont analysées grâce à une modélisation en équation structurelle sur Mplus et par la méthode de modération et médiation sur le logiciel macro SPSS de Hayes. Le résultat de la première enquête met en évidence un phénomène de réciprocité entre les différents types d'injustices et les catégories diverses de conflits sur le lieu de travail, cette corrélation se renforçant à mesure que le temps passe. La dissemblance des valeurs entre salariés d'une même équipe a un impact négatif sur la satisfaction globale du groupe. L'injustice distributive et procédurale est un vecteur de ce phénomène. Dans le même temps, la différence de classe d'âge entre salariés d'une même équipe a un effet positif sur la satisfaction du groupe. Dans ce cas, c'est la communication entre les membres du groupe qui sert de conducteur. Enfin, la seconde étude démontre que les effets du concept d'équité sur le comportement de citoyenneté organisationnelle ont plus d'impact sur les sujets à tendance individualiste, soumis à une forte hiérarchie et réceptif à l'incertitude. / Organizational justice research has shown that fairness is a key part of the employee’s work experience. However, justice scholars focused their considerations on the vertical relationship between employees and managers or organizations, neglecting horizontal relationships between peers. This dissertation advances justice research in teams by discussing and testing the individual peer justice perspective, which deals with the individuals’ perceptions of how fairly they are treated by teammates. To demonstrate the theoretical and practical relevance of the individual peer justice perspective, the organizational justice literature is integrated with the literatures of workplace conflict, workplace dissimilarity, and national cultural values. The research questions center around the predictors and outcomes of individual peer justice, the injustice-conflict interrelationship, the mechanisms of the dissimilarity-team satisfaction relationship, and the moderating role of cultural values on the relationship between justice perceptions and organizational citizenship behavior. The related hypotheses are tested through a longitudinal survey study with 256 employees from a variety of organizations in Germany and through a two-wave survey study with 448 multinational self-managed team members in France. The data is analyzed through structural equation modelling using Mplus and through moderation and mediation analyses using the SPSS macro of Hayes. The results of the first study indicate a reciprocal relationship between injustice dimensions and conflict types in the workplace, whereby this relationship becomes stronger over time. Distributive and procedural injustice further mediate the negative impact of perceived value dissimilarity (relative to teammates) on team satisfaction. The positive effects of perceived age dissimilarity (relative to teammates) on team satisfaction are mediated by information elaboration. Finally, the second study demonstrates that peer justice effects on organizational citizenship behavior are stronger for team members who score high in individualism, high in power distance, and low in uncertainty avoidance.

Les gestes de la main représentent un moyen naturel et intuitif de communication chez l'homme lui permettant d'interagir avec son environnement dans la vie de tous les jours. Ils permettent notamment de ponctuer et de renforcer l'expression orale d'un dialogue entre personnes. Outre la communication entre individus, les gestes de la main permettent de manipuler des objets ou encore d'interagir avec des machines. Avec le développement de la vision par ordinateur, on assiste à un véritable engouement pour de nouveaux types d'interactions qui exploitent le mouvement de la main et qui passent par une étape d'analyse et de reconnaissance du mouvement afin d'aboutir à l'interprétation des gestes de la main. La réalisation d'un tel objectif ouvre un large champ d'applications. C'est dans ce cadre que se positionne le travail réalisé au cours de cette thèse. Les objectifs visés étaient de proposer des méthodes pour: 1) permettre le transfert d'animation depuis une séquence réelle vers un modèle 3D représentant la main. Dans une telle perspective, le suivi permet d'estimer les différents paramètres correspondant aux degrés de liberté de la main. 2) identifier les gestes de la main en utilisant une base de gestes prédéfinie dans le but de proposer des modes d'interactions basés sur la vision par ordinateur. Sur le plan technique, nous nous sommes intéressés à deux types d’approches : le premier utilise un modèle 3D de la main et le deuxième fait appel à une base de gestes / Hand gestures take a fundamental role in inter-human daily communication. Their use has become an important part of human-computer interaction in the two last decades. Building a fast and effective vision-based hand motion tracker is challenging. This is due to the high dimensionality of the pose space, the ambiguities due to occlusion, the lack of visible surface texture and the significant appearance variations due to shading. In this thesis we are interested in two approaches for monocular hand tracking. In the first one, a parametric hand model is used. The hand motion tracking is first formulated as an optimization task, where a dissimilarity function between the projection of the hand model under articulated motion and the observed image features, is to be minimized. A two-step iterative algorithm is then proposed to minimize this dissimilarity function. We propose two dissimilarity functions to be minimized. We propose also in this thesis a data-driven method to track hand gestures and animate 3D hand model. To achieve the tracking, the presented method exploits a database of hand gestures represented as 3D point clouds. In order to track a large number of hand poses with a database as small as possible we classify the hand gestures using a Principal Component Analysis (PCA). Applied to each point cloud, the PCA produces a new representation of the hand pose independent of the position and orientation in the 3D space. To explore the database in a fast and efficient way, we use a comparison function based on 3D distance transform. Experimental results on synthetic and real data demonstrate the potentials of ours methods

Submitted by Gabriela Lopes (gmachadolopesufpel@gmail.com) on 2017-09-22T14:46:47Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) DISSERTAÇÃO_BRUNA_RIGTH_CORRIGIDA+Catalog.pdf: 1361981 bytes, checksum: 172bb73af16db094ed298ba7457443d2 (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2017-09-25T19:40:41Z (GMT) No. of bitstreams: 2 DISSERTAÇÃO_BRUNA_RIGTH_CORRIGIDA+Catalog.pdf: 1361981 bytes, checksum: 172bb73af16db094ed298ba7457443d2 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-09-25T19:40:41Z (GMT). No. of bitstreams: 2 DISSERTAÇÃO_BRUNA_RIGTH_CORRIGIDA+Catalog.pdf: 1361981 bytes, checksum: 172bb73af16db094ed298ba7457443d2 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-07-18 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / O trigo (Triticum aestivum L.) está entre os cereais mais produzidos no mundo devido a sua capacidade de adaptação a diferentes condições edafoclimáticas, além do valor nutritivo do grão. Aumentar a qualidade e a produtividade do trigo no Brasil tem sido um desafio para todos os segmentos envolvidos no processo produtivo. Nas últimas décadas, ocorreram avanços na produtividade, porém, a quantidade do trigo produzida no país ainda é inferior à necessária para o abastecimento interno. Para isto, é necessário incrementar o potencial de rendimento em condições de lavoura, onde os cultivares precisam interagir de forma benéfica com diferentes situações de ambiente e de manejo. Diferentes estratégias são utilizadas para composição do rendimento de grãos em cultivares de trigo. Dessa forma, o presente trabalho objetivou estudar diferentes cruzamentos entre genótipos de trigo e seu potencial para rendimento de grãos. O experimento foi conduzido junto ao campo experimental do Centro de Genômica e Fitomelhoramento (CGF), que está localizado no Centro Agropecuário da Palma, pertencente à Universidade Federal de Pelotas, no município de Capão do Leão, Rio Grande do Sul. O delineamento experimental foi em blocos casualizados, com três repetições. A unidade experimental utilizada para avaliação foi de 1m² para os anos de 2015 e 2016. Foram avaliados 51 genótipos de trigo na geração F6 e F7, além de cinco testemunhas comerciais, onde avaliou-se altura da planta (AP) e calculada a média de vários pontos da unidade experimental e número de afilhos férteis (AF). Após, foram coletadas 10 espigas de cada parcela, nas quais foram avaliados os seguintes caracteres do rendimento dos genótipos em estudo: massa da espiga (ME), número de grãos por espiga (NGE), massa de grãos da espiga (MGE), número de espiguetas (NE), comprimento da espiga (CE), medida do diâmetro (DG) e comprimento do grão (CG), massa de mil grãos (MMG) e massa da parcela inteira para avaliação do rendimento por área (MP). A análise da variância apresentou diferenças significativas entre os caracteres avaliados. Através de análises de dissimilaridade genética, houve formação de grupos distintos para os dois anos de estudo onde os genótipos 12 (ITAIPÚ/BRS 220) e 21 (MARFIM/FRONTANA) são genótipos promissores para serem utilizados em cruzamentos dirigidos entre si, tendo em vista sua superioridade para os caracteres comprimento de espiga e massa de grãos por espiga. As correlações obtidas entre os caracteres evidenciaram a importância das mesmas para seleção de genótipos superiores onde os caracteres MGE, NGE e ME apresentam as maiores correlações nos dois anos de cultivo. Através das análises de médias entre os anos, podemos observar que as variações climáticas tem forte influência sobre os caracteres que compõe o rendimento de grãos. O cruzamento FRONTANA x BRS 220 apresentou o maior potencial para incremento no rendimento de grãos. Dessa maneira, este estudo mostrou diferenças genéticas entre os genótipos avaliados e a importância e contribuição dos caracteres envolvidos no rendimento de grãos. / Wheat (Triticum aestivum L.) is one of the most produced cereals in the world due to its ability to adapt to different soil and climatic conditions, as well as the nutritive value of the grain. Increasing yield and quality of wheat in Brazil has been a challenge for all segments involved in the production process. In the last decades, there have been advances in yield, however, a quantity of wheat produced in the country is still lower than the domestic demand. Therefore, it is necessary to increase yield potential in the crop systems, where cultivars need to positively interact with different environmental and management conditions. The cultivars of wheat use different strategies for the composition of grain yield. Thus, the present work aimed to study different crosses between wheat genotypes and their potential for grain yield. The experiment was carried out at the Experimental Field of the Centro de Genômica e Fitomelhoramento (CGF), which is located at the Centro Agropecuário da Palma, of the Federal University of Pelotas, in the municipality of Capão do Leão, Rio Grande do Sul State. The experimental design used was random blocks with three replications. The experimental unit used for evaluation was 1m² for the years 2015 and 2016. In this experiment, the following characters were evaluated: plant height (AP) and calculated on average of several points of the experimental unit and number of fertile tiller (AF). Then, ten ears of each plot were collected for the measurement of yield components as follows: ear mass (ME), number of grains per ear (NGE), ear grain mass (MGE), number of spikelets (NE), ear length (CE), grain diameter (DG), grain length (CG), mass of a thousand grains (MMG) and mass of the whole plot (MP) for yield per area evaluation. An analysis of variance showed significant differences among the evaluated charcaters. Through genetic dissimilarity analyses, distinct groups were formed for the two years of study, where genotypes 12 (ITAIPÚ / BRS 220) and 21 (MARFIM / FRONTANA) are promising to be used in crosses directed towards each other, aiming at superiority for the characters of ear length and mass of grains per ear. The correlations obtained between the characters evidenced an importance for the selection of superior genotypes where MGE, NGE and ME presented the highest correlations in the two years of cultivation. Moreover, means analyzes between both years demonstrates that the climatic variations deeply influence the yield grains traits. The FRONTANA x BRS 220 cross showed the greatest potential for increment of grain yield. Thus, this study showed genetic differences among the studied genotypes and the importance and contribution of the characters involved in grain yield.

The learning process consists of different steps: building a Training Set (TS), training the system, testing its behaviour and finally classifying unknown objects. When using a distance based rule as a classifier, i.e. 1-Nearest Neighbour (1-NN), the first step (building a training set) includes editing and condensing data. The main reason for that is that the rules based on distance need many time to classify each unlabelled sample, x, as each distance from x to each point in the training set should be calculated. So, the more reduced the training set, the shorter the time needed for each new classification process. This thesis is mainly focused on building a training set from some already given data, and specially on condensing it; however different classification techniques are also compared.The aim of any condensing technique is to obtain a reduced training set in order to spend as few time as possible in classification. All that without a significant loss in classification accuracy. Somenew approaches to training set size reduction based on prototypes are presented. These schemes basically consist of defining a small number of prototypes that represent all the original instances. That includes those approaches that select among the already existing examples (selective condensing algorithms), and those which generate new representatives (adaptive condensing algorithms).Those new reduction techniques are experimentally compared to some traditional ones, for data represented in feature spaces. In order to test them, the classical 1-NN rule is here applied. However, other classifiers (fast classifiers) have been considered here, as linear and quadratic ones constructed in dissimilarity spaces based on prototypes, in order to realize how editing and condensing concepts work for this different family of classifiers.Although the goal of the algorithms proposed in this thesis is to obtain a strongly reduced set of representatives, the performance is empirically evaluated over eleven real data sets by comparing not only the reduction rate but also the classification accuracy with those of other condensing techniques. Therefore, the ultimate aim is not only to find a strongly reduced set, but also a balanced one.Several ways to solve the same problem could be found. So, in the case of using a rule based on distance as a classifier, not only the option of reducing the training set can be afford. A different family of approaches consists of applying several searching methods. Therefore, results obtained by the use of the algorithms here presented are compared in terms of classification accuracy and time, to several efficient search techniques.Finally, the main contributions of this PhD report could be briefly summarised in four principal points. Firstly, two selective algorithms based on the idea of surrounding neighbourhood. They obtain better results than other algorithms presented here, as well as better than other traditional schemes. Secondly, a generative approach based on mixtures of Gaussians. It presents better results in classification accuracy and size reduction than traditional adaptive algorithms, and similar to those of the LVQ. Thirdly, it is shown that classification rules other than the 1-NN can be used, even leading to better results. And finally, it is deduced from the experiments carried on, that with some databases (as the ones used here) the approaches here presented execute the classification processes in less time that the efficient search techniques.

In recent years, the responsibility for the desegregation of American public schools has transitioned from federal court mandates to school board programs and policies. There is widespread belief that this has resulted in the resegregation of schools across the country. One popular policy that is purported to provide the opportunity for voluntary integration, along with accountability for academic quality, is school choice. The purpose of this study was to consider the implications of such a policy in one large school district. There is an extensive body of research exploring who participates in school choice, how they make their choices, and why they choose the schools their children attend. In contrast, this study was designed to investigate the actual choices made by parents and the impact of those choices on the elementary schools in the district. This quantitative descriptive study examined the racial and socioeconomic composition of students in one district’s elementary schools during the 2009-2010 school year, and explored the extent to which the student populations in these schools would differ if all students had attended their attendance area schools, rather than participating in the district’s voluntary choice plan. The actual 2009-2010 demographics were compared to “counterfactual” demographics for each school. The researcher generated the counterfactual data by removing the students who chose to attend the school and adding back the students who chose to exit the school. These actual and counterfactual demographics for each school were used to compare dissimilarity indices calculated for the district’s elementary schools as they actually were, and as they theoretically would have been without the school choice program. Additionally, the quality of the schools parents chose was investigated. The results showed that, in this district, the school choice plan did not impact the level of integration in the elementary schools. The schools were moderately segregated with the school choice plan in place, but were also moderately segregated based on the counterfactual demographics that represented the district without school choice. Most parents (60%) chose high quality schools, as identified by the state’s accountability plan. However, parents who chose low achieving schools were disproportionately black and poor. Further research is warranted to determine if the mechanics of the school choice plan could be manipulated to improve the level of integration in the district, and to better understand the decisions made by some parents to send their children to low performing schools.

Submitted by Franciele Moreira (francielemoreyra@gmail.com) on 2018-12-04T12:12:31Z No. of bitstreams: 2 Dissertação - Ana Letycia Basso Garcia - 2017.pdf: 3161693 bytes, checksum: 8ea7726402d9acbdddc870e75a0b1cef (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-12-05T09:52:37Z (GMT) No. of bitstreams: 2 Dissertação - Ana Letycia Basso Garcia - 2017.pdf: 3161693 bytes, checksum: 8ea7726402d9acbdddc870e75a0b1cef (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-12-05T09:52:37Z (GMT). No. of bitstreams: 2 Dissertação - Ana Letycia Basso Garcia - 2017.pdf: 3161693 bytes, checksum: 8ea7726402d9acbdddc870e75a0b1cef (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-05-30 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Improving rice yield has been a big challenge for rice breeding programs around the world. One alternative to identify inbred lines with high yield potential, and discover genes related to yield and its components is to explore the genic pool of the population originated from crosses between cultivars not regularly used, as those introduced from another country. The objective of this study was characterizing a recombinant inbred lines population from the cross between Maninjau and Epagri 108. There were evaluated 296 RILs in experiments conducted in Goianira (GO), Boa Vista (RR) and Pelotas (RS), in 2016. In the experiments in GO and RR, the alpha lattice design (17x18) with two replications were used, and in Pelotas was applied the BAF design. Data were collected for yield (PD) and plant height (AP) in the three places, days to flowering (DF), in RR and GO, 100-grain weight (PG), in RS and GO and leaf blast resistance (BS), in GO. The data were analyzed by a mixed model with the deviance analysis. Variance components were estimated by REML/BLUP and the genetic parameters and correlation coefficients were calculated. The statistics parameters as CV e , CV g , CV r and selective accuracy were also estimated. The G x E interaction analysis was processed by the MHPRVG method. Also, the genetic distances between the progenies that had highest breeding values in each place and their relatives (Maninjau and Epagri 108), was estimated, using a 24 SSR markers panel. For Boa Vista and Goianira, most of the highest yield RILs were like the parental Epagri 108. Most of the random effects of the statistic model used in this study were significant. The RILs population showed genetic variability inside (σ g2 significant). The experimental precision in RR and GO was from good to excellent with accuracy over 90% and in RS it was moderate (~50%), probably because of the environmental effect action. Yield showed moderate heritability (0,67) and the characters DF, AP and PG showed high heritability (>0,90). Positive significant correlation was observed between the characters PD and PG, and DF and AP, however the last one showed negative correlation with yield. Nine RILs had the best performance by the MHPRVG, and it was above 30% of the general mean. They are recommended for the breeding program use. The RIL 105 were ranked as the best for stability, adaptability and yield, simultaneously. These results suggest that there is a significant genetic variation between the RILs evaluated. Therefore, this population might be used either in selection of high yield performance genotypes or for QTL mapping foragronomic traits in many environments. / O aumento da produtividade em arroz é um desafio para os programas de melhoramento do mundo todo. Uma alternativa para identificar linhagens mais produtivas, ou mesmo descobrir genes correlacionados à produtividade e seus componentes, é conhecer e explorar o pool gênico de populações provenientes de cruzamentos entre cultivares ainda pouco utilizadas, como materiais introduzidos. O objetivo desse trabalho foi caracterizar uma população de linhas puras recombinantes (RILs), provenientes do cruzamento entre Maninjau x Epagri 108. Foram avaliadas 296 RILs em experimentos conduzidos em Goianira (GO), Boa Vista (RR) e Pelotas (RS), no ano de 2016. Em GO e RR os ensaios foram implantados em delineamento alfa-látice (17x18) e em Pelotas foi utilizado BAF. Foram coletados dados referentes à produtividade (PD) e altura de plantas (AP) nos três locais, além de dias até o florescimento (DF), em RR e GO, peso de 100 grãos (PG), em RS e GO. Os dados foram analisados via modelos mistos, através da análise de deviance. Os componentes de variância foram estimados via REML/BLUP e foram estimados os parâmetros genéticos e coeficientes de correlação entre caracteres, bem como os parâmetros estatísticos CV e , CV g , CV r e acurácia seletiva. A análise de interação G x E foi feita com base no método da MHPRVG. Foram estimadas as distâncias genéticas entre as linhagens de maior valor genotípico em cada local e os parentais do cruzamento, através de um painel de 24 marcadores SSR, e para Boa Vista e Goianira, a maioria das linhagens mais produtivas foram mais similares ao genitor Epagri 108. A maioria dos efeitos aleatórios do modelo estatístico adotado foi significativa. A população de RILs apresentou variabilidade genética (σ g2 significativo). A precisão dos experimentos de RR e GO foi de boa à ótima, com acurácia maior que 90%, enquanto do RS foi moderada (~50%), provavelmente devido à maior ação do efeito ambiental. A produtividade se revelou com h2 moderada (0,67) e os caracteres DF, AP e PG apresentaram alta herdabilidade (>0,90). Foi verificada correlação positiva significativa entre os caracteres PD e PG e DF e AP, porém esses dois últimos têm correlação negativa significativa com a produtividade. Nove RILs se destacaram pela MHPRVG, com desempenho superior a 30% em relação à média geral. Elas são, portanto, recomendadas para uso do programa de melhoramento. A RIL 105 foi classificada como a de melhor estabilidade, adaptabilidade e produtividade, simultaneamente. Os resultados indicam que a população tem alta variabilidade genética e, pode ser utilizada tanto na seleção de genótipos de bom desempenho produtivo, quanto na detecção de QTLs para caracteres de interesse agronômico em múltiplos locais.

Submitted by Jaqueline Silva (jtas29@gmail.com) on 2016-10-18T15:32:55Z No. of bitstreams: 2 Dissertação - Carolina Moreno do Santos -2012.pdf: 1999516 bytes, checksum: 98e22a43a499ffbde4e49ed07b291788 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Jaqueline Silva (jtas29@gmail.com) on 2016-10-18T16:28:36Z (GMT) No. of bitstreams: 2 Dissertação - Carolina Moreno do Santos -2012.pdf: 1999516 bytes, checksum: 98e22a43a499ffbde4e49ed07b291788 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-10-18T16:28:36Z (GMT). No. of bitstreams: 2 Dissertação - Carolina Moreno do Santos -2012.pdf: 1999516 bytes, checksum: 98e22a43a499ffbde4e49ed07b291788 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2012-05-17 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / The Cerrado Biome presents a mosaic of vegetation ranging from open areas with no shrub component (grassland) to areas with a high density of tall trees (forests formations). This mosaic of vegetation that occurs in the Cerrado harbors a high biodiversity of different groups of animals and enables to this biome presents high beta diversity in the region. However, data of species occurrence are lacking for most groups of animals, especially for the invertebrates. In this work, Arctiinae moths were sampled in different vegetation formations of the Cerrado. Arctiinae are one of the richest subfamilies of Lepidoptera. About 11.000 species were described in worldwide, 6.000 in the Neotropics, 1.400 in Brazil and 720 in the Cerrado. Most larvae of Arctiinae are polyphagous. In addition to eating the green angiosperm and gymnosperm tissues, their ranks include species that consume algae, lichens, liverworts and mosses. Despite being polyphagous, the most species of Arctiinae feed mainly on the plant species that provide secondary metabolities (mainly pyrrolizidine alkaloids) that offer to them chemical protection against natural enemies. Thus, variations in the host plant species composition reflect at least partially in the variations in the moths species composition. In the first chapter of this dissertation, the Arctiinae moth fauna was characterized in four vegetation formations (dirty grassland, closed grassland, cerrado sensu stricto and semideciduous forest). Moths species richness was significantly higher in vegetation formations more structurally complex. The moth species composition in semideciduous forest was significantly dissimilar from the others vegetation formations. There was a positive correlation between the moths similarity patterns and the plants similarity patterns. In the first chapter was also tested the premise that structurally simple habitats presents a higher intensity of predation because they available less quantity of refuges for prey. This premise was corroborated because the predation intensity on artificial caterpillars was twice higher in the dirty grassland than in the structurally complex vegetation. In the second chapter of this dissertation was tested which environmental variable best determines the variation in Arctiinae species richness and composition between the vegetation formations of the cerrado sensu lato (dirty grassland, closed grassland and cerrado sensu stricto). The trees richness was the most determinant of moths species richness. The moth species composition was influenced by the trees richness, followed by herbs density, predation intensity and by the variation coefficient of tree height. The results that the vegetation characteristics are the best predictors of the variations in the moths species richness and composition indicate that it is necessary to invest in conservation measures of all vegetation formations that occurs in the Cerrado. / O Bioma Cerrado apresenta um mosaico de formações vegetais que varia desde áreas abertas sem nenhum elemento arbustivo (campo limpo) até áreas que apresentam uma grande densidade de árvores altas (formações florestais). Este mosaico de vegetação que ocorre no Cerrado abriga uma alta biodiversidade de animais e contribui para a alta diversidade beta observada no bioma. Contudo, dados de ocorrência de espécies são escassos para a maioria dos grupos animais, principalmente para os invertebrados. Neste trabalho foram amostradas mariposas Arctiinae em diferentes formações vegetais do Cerrado. As mariposas Arctiinae representam uma das subfamílias de Lepidoptera mais ricas em espécies. Já foram descritas aproximadamente 11 mil espécies em todo o mundo, das quais cerca de 6000 ocorrem nos neotrópicos, 1400 no Brasil e 720 no Cerrado. Além de se alimentarem de angiospermas e gimnospermas, algumas espécies de Arctiinae também se alimentam de algas, liquens e musgos. Apesar de serem polífagas, muitas espécies de Arctiinae se alimentam de plantas que fornecem compostos secundários (principalmente alcaloides pirrolizidínicos), o que as tornam impalatáveis aos inimigos naturais. Desta forma, as variações na composição de espécies das plantas hospedeiras devem, pelo menos em parte, refletir a mudança da composição das espécies de mariposas. No primeiro capítulo desta dissertação a fauna das mariposas Arctiinae foi caracterizada em quatro formações vegetais (campo sujo, campo cerrado, cerrado sensu stricto e mata estacional semidecídua). As formações vegetais mais estruturalmente complexas apresentaram maior riqueza de espécies. A mata estacional semidecídua apresentou uma composição de espécies dissimilar com as demais formações vegetais. Houve uma correlação positiva entre o padrão de similaridade das mariposas e o padrão de similaridade da vegetação. Também foi testada a hipótese de que os hábitats estruturalmente simples apresentam maior intensidade de predação porque disponibilizam menor quantidade de refúgios para as presas. Esta hipótese foi corroborada porque a fitofisionomia menos estruturalmente complexa (campo sujo) apresentou uma intensidade de predação nas lagartas artificiais duas vezes maior do que as formações vegetais estruturalmente complexas. No segundo capítulo desta dissertação, foi testado qual variável ambiental melhor determina a variação na riqueza e na composição de espécies de mariposas Arctiinae entre as formações vegetais do cerrado sensu lato (campo sujo, campo cerrado e cerrado sensu stricto). A riqueza de arbóreas e arbustivas foi a variável que mais explicou a variação da riqueza de espécies. A composição das espécies de mariposas Arctiinae também foi influenciada pela riqueza de arbóreas e arbustivas, seguida pela cobertura de herbáceas, intensidade de predação e pelo coeficiente de variação da altura de plantas arbóreas e arbustivas. Os resultados obtidos sugerem que as características da vegetação são os principais preditores dos padrões de riqueza e composição das espécies de mariposas Arctiinae no Cerrado, o que reforça a importância de se investir em medidas de conservação de todas as formações vegetais que ocorrem neste Bioma.

Made available in DSpace on 2014-08-20T14:06:14Z (GMT). No. of bitstreams: 1 Tese_Ariano_Martins_Magalhaes.pdf: 871444 bytes, checksum: 19f2ef1b5a219328a4f47b41290ddc87 (MD5) Previous issue date: 2007-04-17 / The worldwide adaptability of rice, allied to constant research efforts, reassures its grain a place among the major staple foods for humankind. The influence of natural selection resulted in a wide diversity in the Oryza genus, which is currently composed by 23 species. On the other hand, only O. sativa L. (asian cultivated rice) and O. glaberrima (african cultivated rice) have been used for human consumption, therefore, subjected to artificial selection. The process of domestication in rice has selected important agronomic characters for better plant performance in specific environments, resulting in a bottleneck effect on the genetic diversity. This effect makes a change in the initial gene pool, keeping some selected genes and eliminating others. Thus, the objective of this work was to amplify the genetic basis of rice in southern Brazil, through collections and introductions, characterization of accessions and estimation of genetic distances between accessions. The results revealed a limited number of rice landraces in the state of Rio Grande do Sul. These landraces are found in small farms that still keep these genotypes for their own consumption. A large area of the state is cultivated with modern phillipine-type varieties. However, the rescued landrace genotypes present genetic variability for a large number of the evaluated characters, and may serve as source of genes for breeding programs. The descriptors used for the multivariate analysis were efficient to distinguish the collected accessions, to identify duplicated genotypes in the collection as well as for phenotyping the whole collection. The estimate of genetic distance was obtained using Mahalanobis (D2), distance, which indicated a tight clustering of cultivars released by Embrapa Temperate Climate. The genetic distance matrices based on field- and greenhouse-measured qualitative characters as well as on the joint analysis showed high degree of association, indicating that many characters can be selected based on only one of these criteria. Similar results were obtained for quantitative characters. / A ampla adaptabilidade do arroz, aliada ao continuado esforço da pesquisa no mundo, assegura que o seu grão permaneça sendo um importante produto de consumo pelo homem. A influência da seleção natural resultou em uma ampla diversidade encontrada no gênero Oryza, que, atualmente, é composta por 23 espécies. No entanto, apenas O. sativa L. (arroz cultivado asiático) e O. glaberrima (arroz cultivado africano) vem sendo exploradas na alimentação humana, portanto, sob forte efeito da seleção artificial. O processo de domesticação do arroz resultou na seleção de caracteres agronômicos importantes para o melhor desempenho das plantas em determinado ambiente. Isto resultou no chamado efeito de afunilamento em termos de diversidade genética, ou seja, a partir de um background genético bastante rico, alguns grupos de genes de interesse vão sendo mantidos na população e outros eliminados. Assim sendo, o objetivo deste trabalho foi ampliar a base genética do arroz no Sul do Brasil, através de coletas e introduções, caracterização dos acessos e estudo da distância genética existente entre eles. Os resultados revelaram um número limitado de coletas de genótipos de arroz no Rio Grande do Sul, sendo esta, basicamente, restrita a pequenos produtores que ainda mantém genótipos antigos para sua subsistência. Grande área do Estado é cultivada com variedades modernas do tipo filipina. No entanto, os genótipos resgatados apresentaram variabilidade genética para um grande número de caracteres avaliados, podendo servir como fonte de genes aos programas de melhoramento genético da cultura. Os descritores utilizados na análise multivariada foram eficientes para separar os acessos coletados, identificar duplicações na coleção, bem como na fenotipagem da coleção. O método de estimativa da distância genética utilizado, baseado na matriz de distâncias de Mahalanobis (D2), indicou um agrupamento das cultivares lançadas pela Embrapa Clima Temperado, sendo revelada pouca variabilidade genética entre estas. As matrizes de distância genética entre os genótipos com base nos caracteres qualitativos aferidos a campo, em casa-devegetação e a conjunta de ambos os ambientes, evidenciaram associação elevada entre si, indicando que possivelmente a avaliação destes caracteres poderá ser realizada efetivamente sob apenas um dos critérios. Esta mesma afirmativa foi observada para os caracteres quantitativos.