Modelos de ocorrência e abundância de espécies são importantes ferramentas para a elaboração de estrategias para a conservação da biodiversidade. Nestes artigos aplicamos, pela primeira vez, modelos de distribuição e abundância no entorno de uma Usina Hidroelétrica no Sul do Brasil, com o objetivo de mostrar possíveis aplicações destas técnicas no planejamento de ações de coleta e conservação, in situ e ex situ, destas espécies. Nossos objetivos secundários foram (i) modelar a ocorrência e abundância de plantas em um trecho da bacia do rio Pelotas e (ii) verificar o efeito da detectabilidade em modelos de ocorrência. Para tanto dividimos a tese em dois capítulos e cinco questões. No primeiro capítulo procuramos responder as seguintes questões: (1) existe relação entre a probabilidade de ocorrência e a abundância observada? (2) Modelos baseados na abundância são melhores para predizer a ocorrência de espécies do que modelos baseados apenas na ocorrência das espécies? No segundo capítulo, tentamos responder se (3) amostragens exaustivas, com presenças e ausências bem descritas, produzem modelos de ocorrência mais precisos? (4) Se a relação entre a ocorrência da espécie e as variáveis ambientais que descrevem sua distribuição é alterada de acordo com a detectabilidade? (5) Qual o impacto de falsos zeros na área ocupada estimada para a espécie? O trabalho foi realizado no entorno da UHE Barra Grande, erguida no rio Pelotas, entre os Estados de Santa Catarina e Rio Grande do Sul. Localizada na bacia do rio Pelotas, seu reservatório ocupa uma área de aproximadamente 90 km2 e seu entorno perfaz uma área aproximada de 4.600 km2, onde amostramos 388 parcelas georreferenciadas (10 x 50 m) distribuídas aleatoriamente. Em cada parcela contamos o número de indivíduos de nove espécies arborescente com altura ≥ 1,5 m. No primeiro artigo, modelamos, utilizando modelos lineares generalizados (GLM), a distribuição e abundância das nove espécies estudadas em relação a 15 variáveis ambientais, e demonstramos que modelos de ocorrência são mais precisos do que modelos de abundância. Demonstramos também que a probabilidade de ocorrência é positivamente correlacionada com a abundância observada, o que abre a possibilidade de utilizarmos a probabilidade de ocorrência como um indicador da abundância das espécies. No segundo capítulo, verificamos o impacto de falhas de detecção em modelos de distribuição de espécies, utilizando como espécie modelo A. angustifolia. Simulamos o impacto de falsos zeros na precisão dos modelos, na seleção das variáveis resposta (variáveis ambientais) e nas estimativas de área ocupada pela espécie. A redução da detectabilidade ocasionou alterações tanto na precisão dos modelos quanto na área estimada de ocorrência e nas variáveis ambientais que explicam a ocorrência de A. angustifolia. A área ocupada não apresentou padrão bem definido, porém mostrou tendência de elevação de acordo com a diminuição da detectabilidade, elevando com isso o erro de comissão dos modelos. A precisão dos modelos (AUC e correlação entre probabilidade de ocorrência e abundância observada), apresentou queda em relação ao redução da detectabilidade, enquanto apenas uma variável ambiental foi incluída em todos os modelos (pH do solo). Existe um grande possibilidade de uso destas técnicas como informação básica para planos de conservação, monitoramento e manejo da biodiversidade no entorno de empreendimentos hidroelétricos, porém, para que possamos tirar inferências fortes sobre as espécies e ecossistemas em questão, devemos observar sempre a necessidade de levar em conta os impactos da detecção imperfeita das espécies. / Species occurrence and abundance models are important tools in the development of biodiversity conservation strategies. This study presents, for the first time, the application of distribution and abundance models in the environment surrounding a hydropower dam in southern Brazil. The aim is to demonstrate the possible application of these techniques in planning collection and conservation strategies of the species inhabiting the area, in situ and ex situ. We also aimed at (i) modeling the occurrence and abundance of plants in a section of the Pelotas River basin and (ii) verifying the effect of detectability in occurrence models. This thesis was therefore divided into two chapters and five questions. The first chapter is an attempt to answer the following questions: (1) Is there a relationship between occurrence probability and abundance? (2) Are models based on abundance better to predict the occurrence of species as compared to models that utilize only species occurrence? The second chapter addresses the questions: (3) Do exhaustive sampling procedures, with well-sampled presences and absences, lead to more accurate occurrence models? (4) Does detectability influence the relationship between species occurrence and environmental variables? (5) What is the impact of false zeros in the estimated area occupied by one given species? This study was conducted in the area surrounding the hydropower plant “Barra Grande”, in the Pelotas River, on the border of the Rio Grande do Sul and Santa Catarina states. Located in the Pelotas River basin, the dam occupies an area of approximately 90 km². The surrounding area accounts for 4,600 km², where 388 plots with 10 x 50 m were randomly distributed and georeferenced. In each plot, individuals of nine selected tree species taller than 1.5 m were surveyed. In the first paper, we modeled the distribution and abundance of the species studied using generalized linear models (GLMs), utilizing 15 environmental variables. Occurrence models were proved to be more accurate than abundance models. The positive correlation between occurrence probability and observed abundance was also demonstrated, which opens the possibility to utilize occurrence probability as an indicator of species abundance. The second chapter reports the impact of failures in detection in species distribution models, using A. angustifolia as model species. We simulated the impact of false zeros in model accuracy, in the selection of response variables (environmental variables) and in estimates of the area occupied by A. angustifolia. The decrease in detectability caused changes in both model accuracy and in the environmental variables that explain the occurrence of A. angustifolia. The area occupied did not have a well-defined pattern, though it showed a trend towards elevation, in accordance with the decrease in detectability, thus raising the commission error in the model. The accuracy of models (AUC and correlation between occurrence probability and observed abundance) decreased with the decrease in detectability, while only one environmental variable was included in all models (soil pH). These techniques offer great potential in the collection of essential information in biodiversity conservation, monitoring and managements strategies in hydropower projects. However, robust inferences about species and ecosystems in question can only be drawn considering the need to evaluate the impact caused by defective species detection.

Aim of this work is to study the impact of inaccuracy in execution of required task (inaccuracy in subject’s behavioral response to experimental stimulation) by person who undergoes fMRI examination. The work is solved in several stages. First, theoretical analysis of inaccuracy in fMRI experiment was performed, and simulations with synthetic data were created. Several variables in general linear model and t-statistics were followed. We found that estimated effect size depends linearly on covariance between the corresponding columns of X and D matrices or their linear combination. The component of residual variance caused by inaccuracy is negligible at real-life noise levels. In such case, moreover, the dependence of t-statistics on inaccuracy becomes linear. Next, our theoretical results (dependencies/characteristics of variables) were verified using real data. All results were confirmed. Last, I focused on possible practical use of the uncovered characteristics and dependencies. Optimization of experimental design with respect to inaccuracy, correction of inaccurate results and reliability of inaccurate results are introduced and discussed. Especially, the calculation of maps of maximal tolerable inaccuracy can be useful to find robust or weak (tending to be not detected or to be significantly different from accurate value) activation in real fMRI experiments.

Factor score regression has recently received growing interest as an alternative for structural equation modelling. Two issues causing uncertainty for researchers are addressed in this thesis. Firstly, more knowledge is needed on how different approaches to calculating factor score estimates compare when estimating factor score regression models. Secondly, many applications are left without guidance because of the focus on normally distributed outcomes in the literature. This thesis examines how factor scoring methods compare when estimating regression coefficients in generalised linear factor score regression. An evaluation is made of the regression, correlation-preserving, total sum, and weighted sum method in ordinary, logistic, and Poisson factor score regression. In contrast to previous studies, both the mean and variance of loading coefficients and the degree of inter-factor correlation are varied in the simulations. A meta-analysis demonstrates that the choice of factor scoring method can substantially influence research conclusions. The regression and correlation-preserving method outperform the other two methods in terms of coefficient and standard error bias, accuracy, and empirical Type I error rates. Moreover, the regression method generally has the best performance. It is also noticed that performance can differ notably across the considered regression models.

La spectroscopie représente un puissant outil en conservation grâce à la possibilité d’effectuer le suivi de la diversité végétale à travers de larges étendues géographiques. La réflectance spectrale montre un potentiel certain pour l’identification des espèces d’arbres et même des taxons inférieurs, mais ceci a rarement été testé sur un grand nombre d’espèces. J’examine la qualité de la classification de 45 espèces d’arbres des forêts tempérées du Québec à partir de plus de 3500 spectres de réflectance foliaires (400-2400 nm). Nous évaluons cette classification sur la base de la variation spectrale des espèces, de même qu’à partir des distances fonctionnelles et phylogénétiques mesurées. Nos résultats indiquent un taux de classification très satisfaisant (κ = 0.736, ±0.005). Nous observons des erreurs de classification plus fréquentes entre les espèces évolutivement proches, alors qu’il semble que la distance fonctionnelle établisse un seuil voulant qu'au-delà d’une certaine distinction fonctionnelle globale, il soit peu probable que deux espèces soient confondues. Ces résultats viennent renforcer le lien entre la diversité spectrale et l’organisation taxonomique des espèces, ajoutant à la valeur de substitution de la première pour la diversité phylogénétique. Cela suggère par contre que de fortes convergences fonctionnelles peuvent faire obstacle à l’identification des espèces à partir de la réflectance spectrale. Cette étude est prometteuse pour la classification de spectres foliaires non préalablement identifiés, et améliore notre compréhension du lien entre les données spectrales et la différenciation des espèces, d’une grande importance pour assurer la validité des estimations de la biodiversité à partir de données de télédétection. / Imaging Spectroscopy is a powerful tool for conservation due to its ability to monitor plant diversity over broad geographic areas. Increasing evidence suggests that spectral reflectance can be used to identify trees at the species level, and even below. However, most studies focus on only a few species. Here, we use foliar reflectance (400-2400 nm) to discriminate among 45 temperate forest tree species from southern Quebec, using over 3500 leaf-level spectra. Furthermore, we connect those classification results to functional and phylogenetic distinctiveness, as well as to intraspecific variation. We find that spectral reflectance shows a very good discriminatory power even with an extensive set of species (κ = 0.736, ±0.005). We find that close phylogenetic species get mistaken for one another more frequently than distantly related species, while functional variation acts as a threshold, beyond which misclassifications are unlikely. These results reinforce the link between spectral diversity and taxonomic organization or phylogenetic diversity, but also reiterate the potential confounding effects of functional convergences on species identification from hyperspectral reflectance. We believe these findings hold promise for the classification of unknown spectra and further improve the link between ground truth and remotely sensed data for biodiversity assessments.

No description available.

La zone économique exclusive Mauritanienne est le siège d’upwelling et constitue une zone de transition où cohabitent des espèces d’affinités tempérée et tropicale. Pour comprendre le comportement spatio-temporel des assemblages démersaux du point de vue de leur composition, structuration, distribution de probabilité et diversité face aux préoccupations écologiques. Les facteurs abiotiques contribuent à la structuration des assemblages démersaux persistants au cours du temps. Les effets de la pêche étaient relativement faibles. Les trajectoires temporelles entre les assemblages et les conditions environnementales ont été mises en évidence pour certaines années et des zones. Dans les types d’habitats, un groupe minoritaire d’espèces très agrégatives obéissant au modèle de distribution en log-séries de Fisher et un autre majoritaire peu ou pas du tout agrégatives obéissant au modèle de distribution binomiale négative tronquée ont été identifiés. La diversité spécifique peut être divisé en deux groupes distincts et complémentaires : la richesse spécifique et l'autre associé à l’équitabilité. Un seul composant de la diversité ne peut donc pas représenter la diversité des poissons démersaux de la zone étudiée. Les GLM des indices complémentaires ont montré essentiellement un effet temporel et l’interaction Année-Strates bathymétriques. Aucun effet de l’effort de pêche n’a été observé sur la richesse spécifique, ni de la concentration en chlorophylle sur l’équitabilité. Ce travail pourrait fournir aux gestionnaires et aux scientifiques des connaissances complémentaires sur la dynamique spatio-temporelle des assemblages démersaux exploités dans des écosystèmes d’upwelling. / The Mauritanian exclusive economic zone is the seat of an upwelling phenomenon and constitutes a transition zone where species of temperate and tropical affinities coexist. To understand the spatio-temporal behavior of demersal assemblages from the point of view of their composition, structure, distribution of probability and diversity faced to ecological concerns. Abiotic factors contribute in the structuring of persistent groundfish assemblages over time. The fishing effects were relatively low, although significant in some years and in some specific geographic areas. Temporal trajectories between groundfish assemblages and environmental conditions have been highlighted for some years and in some specific areas. In each type habitats, two species groups were identified: a minority group of species very aggregative well fitted by Fisher’s log-series distribution and another majority of species little or not aggregative well fitted by the truncated negative binomial distribution. Diversity indices analyzed reveal that this set can be split into two distinct and complementary groups: a group associated with the species richness and another group associated with evenness. One component of diversity may not represent the diversity of the groundfish in the study area. GLMs of complementary indices showed essentially a temporal effect and Bathymetric strata-Year interaction. No effect of fishing effort was observed on the species richness and neither was the concentration of chlorophyll a on the evenness. This work could provide managers and scientists to further knowledge on the spatio-temporal dynamics of groundfish species assemblages exploited in upwelling ecosystems.

In der Arbeit wird eine Methodik einer zusammenhängenden Analyse und modellhaften Beschreibung der Verkehrssicherheit in städtischen Hauptstraßennetzen am Beispiel der Stadt Dresden entwickelt. Die dabei gewonnenen Modelle dienen der Abschätzung von Erwartungswerten von Unfallhäufigkeiten mit und ohne Personenschaden unter Berücksichtigung der Verkehrsbeteiligungsart. Die Grundlage bilden multivariate Regressionsmodelle auf Basis verallgemeinerter linearer Modelle (GLM). Die Verwendung verallgemeinerter Regressionsmodelle erlaubt eine Berücksichtigung von Verteilungen, die besser geeignet sind, den Unfallentstehungsprozess wiederzugeben, als die häufig verwendete Normalverteilung. Im konkreten Fall werden hierzu die Poisson-Verteilung sowie die negative Binomialverteilung verwendet. Um Effekte im Hauptverkehrsstraßennetz möglichst trennscharf abbilden zu können, werden vier grundsätzliche Netzelemente differenziert und das Netz entsprechend zerlegt. Unterschieden werden neben Streckenabschnitten und Hauptverkehrsknotenpunkten auch Annäherungsbereiche und Anschlussknotenpunkte. Die Kollektive der Knotenpunkte werden ferner in signalisierte und nicht-signalisierte unterteilt. Es werden zunächst Modelle unterschiedlicher Unfallkollektive getrennt für alle Kollektive der vier Netzelemente berechnet. Anschließend werden verschiedene Vorgehensweisen für eine Zusammenfassung zu Netzmodellen entwickelt. Neben der Verwendung verkehrstechnischer und infrastruktureller Größen als erklärende Variable werden in der Arbeit auch Kenngrößen zur Beschreibung der Umfeldnutzung ermittelt und im Rahmen der Regression einbezogen. Die Quantifizierung der Umfeldnutzung erfolgt mit Hilfe von Korrelations-, Kontingenz- und von Hauptkomponentenanalysen (PCA). Im Ergebnis werden Modelle präsentiert, die eine multivariate Quantifizierung erwarteter Unfallhäufigkeiten in Hauptverkehrsstraßennetzen erlauben. Die vorgestellte Methodik bildet eine mögliche Grundlage für eine differenzierte Sicherheitsbewertung verkehrsplanerischer Variantenabschätzungen. / A methodology is developed in order to predict the number of accidents within an urban main road network. The analysis was carried out by surveying the road network of Dresden. The resulting models allow the calculation of individual expectancy values for accidents with and without injury involving different traffic modes. The statistical modelling process is based on generalized linear models (GLM). These were chosen due to their ability to take into account certain non-normal distributions. In the specific case of accident counts, both the Poisson distribution and the negative binomial distribution are more suitable for reproducing the origination process than the normal distribution. Thus they were chosen as underlying distributions for the subsequent regressions. In order to differentiate overlaying influences, the main road network is separated into four basic elements: major intersections, road sections, minor intersections and approaches. Furthermore the major and minor intersections are additionally subdivided into signalised and non-signalised intersections. Separate models are calculated for different accident collectives for the various types of elements. Afterwards several methodologies for calculating aggregated network models are developed and analysed. Apart from traffic-related and infrastructural attributes, environmental parameters are derived taking into account the adjacent building structure as well as the surrounding land-use, and incorporated as explanatory variables within the regression. The environmental variables are derived from statistical analyses including correlation matrices, contingency tables and principal components analyses (PCA). As a result, a set of models is introduced which allows a multivariate calculation of expected accident counts for urban main road networks. The methodology developed can serve as a basis for a differentiated safety assessment of varying scenarios within a traffic planning process.

In der Arbeit wird eine Methodik einer zusammenhängenden Analyse und modellhaften Beschreibung der Verkehrssicherheit in städtischen Hauptstraßennetzen am Beispiel der Stadt Dresden entwickelt. Die dabei gewonnenen Modelle dienen der Abschätzung von Erwartungswerten von Unfallhäufigkeiten mit und ohne Personenschaden unter Berücksichtigung der Verkehrsbeteiligungsart. Die Grundlage bilden multivariate Regressionsmodelle auf Basis verallgemeinerter linearer Modelle (GLM). Die Verwendung verallgemeinerter Regressionsmodelle erlaubt eine Berücksichtigung von Verteilungen, die besser geeignet sind, den Unfallentstehungsprozess wiederzugeben, als die häufig verwendete Normalverteilung. Im konkreten Fall werden hierzu die Poisson-Verteilung sowie die negative Binomialverteilung verwendet. Um Effekte im Hauptverkehrsstraßennetz möglichst trennscharf abbilden zu können, werden vier grundsätzliche Netzelemente differenziert und das Netz entsprechend zerlegt. Unterschieden werden neben Streckenabschnitten und Hauptverkehrsknotenpunkten auch Annäherungsbereiche und Anschlussknotenpunkte. Die Kollektive der Knotenpunkte werden ferner in signalisierte und nicht-signalisierte unterteilt. Es werden zunächst Modelle unterschiedlicher Unfallkollektive getrennt für alle Kollektive der vier Netzelemente berechnet. Anschließend werden verschiedene Vorgehensweisen für eine Zusammenfassung zu Netzmodellen entwickelt. Neben der Verwendung verkehrstechnischer und infrastruktureller Größen als erklärende Variable werden in der Arbeit auch Kenngrößen zur Beschreibung der Umfeldnutzung ermittelt und im Rahmen der Regression einbezogen. Die Quantifizierung der Umfeldnutzung erfolgt mit Hilfe von Korrelations-, Kontingenz- und von Hauptkomponentenanalysen (PCA). Im Ergebnis werden Modelle präsentiert, die eine multivariate Quantifizierung erwarteter Unfallhäufigkeiten in Hauptverkehrsstraßennetzen erlauben. Die vorgestellte Methodik bildet eine mögliche Grundlage für eine differenzierte Sicherheitsbewertung verkehrsplanerischer Variantenabschätzungen. / A methodology is developed in order to predict the number of accidents within an urban main road network. The analysis was carried out by surveying the road network of Dresden. The resulting models allow the calculation of individual expectancy values for accidents with and without injury involving different traffic modes. The statistical modelling process is based on generalized linear models (GLM). These were chosen due to their ability to take into account certain non-normal distributions. In the specific case of accident counts, both the Poisson distribution and the negative binomial distribution are more suitable for reproducing the origination process than the normal distribution. Thus they were chosen as underlying distributions for the subsequent regressions. In order to differentiate overlaying influences, the main road network is separated into four basic elements: major intersections, road sections, minor intersections and approaches. Furthermore the major and minor intersections are additionally subdivided into signalised and non-signalised intersections. Separate models are calculated for different accident collectives for the various types of elements. Afterwards several methodologies for calculating aggregated network models are developed and analysed. Apart from traffic-related and infrastructural attributes, environmental parameters are derived taking into account the adjacent building structure as well as the surrounding land-use, and incorporated as explanatory variables within the regression. The environmental variables are derived from statistical analyses including correlation matrices, contingency tables and principal components analyses (PCA). As a result, a set of models is introduced which allows a multivariate calculation of expected accident counts for urban main road networks. The methodology developed can serve as a basis for a differentiated safety assessment of varying scenarios within a traffic planning process.

L’Organització Mundial de la Salut (OMS) estima que fins a un 3% de la població mundial ha estat infectada pel virus de l’hepatitis C i és la causa més important d’hepatitis crònica, cirrosi i de malaltia hepàtica terminal, que finalment acaba conduint a un transplantament de fetge. La relació entre la variabilitat en la seqüència del virus de l’hepatitis C i el desenvolupament de la malaltia hepàtica és de tipus multifactorial. La infecció crònica causa fibrosi hepàtica, fet que es veu accelerat per mecanismes desconeguts en el cas de pacients coinfectats amb VIH. La progressió de la malaltia produïda pel VHC en pacients coinfectats, està influenciada no només per factors demogràfics, epidemiològics o pels antecedents clínics dels pacients, si no també per diferències genètiques entre els diferents virus i els hostes.

The area of land farmed in Africa is predicted to double by the year 2050 yet very few African studies have investigated the impact of different farming intensities and regimes on bird communities. This study examined avian species richness and diversity along with the densities of some common bird species on the Jos Plateau, Nigeria, in relation to habitat features on farmland over a gradient of differing farming intensities. The study area exhibited a variety of different levels of farming which differed in the habitat available for birds. Birds normally associated with savanna woodland were more associated with less intensive farming, and open-country birds were more associated with more intensive sites, with more species of birds observed where farming was less intensive. Common species of birds using cultivated land associated with different crops, with acha and millet being the most commonly used. Tree density was the most important variable predicting avian species richness and diversity, with medium tree density predicting the highest species richness and diversity. The densities of two common farmland birds were predicted best by tree density, but varied in their responses to the habitat variables, with common bulbul, a savanna generalist, associating more with less intensive, wooded areas and red-cheeked cordon-bleu, an open savanna granivore, associating with medium intensity, more open farmland. Whinchats were common in open, intensively farmed areas with few trees and good herbaceous vegetation cover. The data presented indicates the importance of retaining natural features of savanna habitat in farmland in order to maintain high avian diversity on farmland. More detailed studies are needed in order to determine the mechanisms involved in the associations observed and collaborations between ecologists and social scientists will be necessary to develop effective policies to limit the impact of the intensification of agriculture in Africa on avian biodiversity.