The main purpose of panel studies is to analyze changes in values of studied variables over time. In micro panel research, a large number of elements are periodically observed within the relatively short time period of just a few years. Moreover, the number of repeated measurements is small. This dissertation deals with contemporary approaches to the regression and the clustering analysis of micro panel data. One of the approaches to the micro panel analysis is to use multivariate statistical models originally designed for crosssectional data and modify them in order to take into account the within-subject correlation. The thesis summarizes available tools for the regression analysis of micro panel data. The known and currently used linear mixed effects models for a normally distributed dependent variable are recapitulated. Besides that, new approaches for analysis of a response variable with other than normal distribution are presented. These approaches include the generalized marginal linear model, the generalized linear mixed effects model and the Bayesian modelling approach. In addition to describing the aforementioned models, the paper also includes a brief overview of their implementation in the R software. The difficulty with the regression models adjusted for micro panel data is the ambiguity of their parameters estimation. This thesis proposes a way to improve the estimations through the cluster analysis. For this reason, the thesis also contains a description of methods of the cluster analysis of micro panel data. Because supply of the methods is limited, the main goal of this paper is to devise its own two-step approach for clustering micro panel data. In the first step, the panel data are transformed into a static form using a set of proposed characteristics of dynamics. These characteristics represent different features of time course of the observed variables. In the second step, the elements are clustered by conventional spatial clustering techniques (agglomerative clustering and the C-means partitioning). The clustering is based on a dissimilarity matrix of the values of clustering variables calculated in the first step. Another goal of this paper is to find out whether the suggested procedure leads to an improvement in quality of the regression models for this type of data. By means of a simulation study, the procedure drafted herein is compared to the procedure applied in the kml package of the R software, as well as to the clustering characteristics proposed by Urso (2004). The simulation study demonstrated better results of the proposed combination of clustering variables as compared to the other combinations currently used. A corresponding script written in the R-language represents another benefit of this paper. It is available on the attached CD and it can be used for analyses of readers own micro panel data.

It is a dilemma problem to achieve both these two goals: a) to maintain a best indoor air quality and b) to use a most efficient energy for a house at the same time. One of the outstanding components involving these goals is a smart ventilation system in the house. Smart ventilation strategies, including demand-controlled ventilation (DCV), have been of great interests and some studies believe that DCV strategies have the potential for energy reductions for all ventilation systems. This research aims to improve smart ventilation system, in aspects of energy consumption, indoor CO2 concentrations and living comfortness, by analyzing long-term sensor data. Based on a case study on an experimental house -- Dalarnas Villa, this research investigates how the current two ventilations modes work in the house and improves its ventilation system by developing customized ventilation schedules. A variety of data analysis methods were used in this research. Clustering analysis is used to identify the CO2 patterns and hence determine the residents living patterns; correlation analysis and regression analysis are used to quantify a model to estimate fan energy consumption; a mathematical model is built to simulation the CO2 decreasing when the house is under 0 occupancy. And finally, two customized schedules are created for a typical workday and holiday, respectively, which show advantages in all aspects of energy consumption, CO2 concentrations and living comfortness, compared with the current ventilation modes.

No description available.

In the context of decarbonizing the electricity generation of French non-interconnected territories, the knowledge of future electricity demand, in particular annual and peak demand in the long-term, is crucial to design new renewable energy infrastructures. So far, these territories, mainly islands located in the Pacific and Indian ocean, relies mainly on fossil fuels powered facilities. Energy policies envision to widely develop renewable energies to move towards a low-carbon electricity mix by 2028.  This thesis focuses on the long-term forecasting of hourly electricity demand. A methodology is developed to design and select a model able to fit accurately historical data and to forecast future demand in these particular territories. Historical data are first analyzed through a clustering analysis to identify trends and patterns, based on a k-means clustering algorithm. Specific calendar inputs are then designed to consider these first observations. External inputs, such as weather data, economic and demographic variables, are also included.  Forecasting algorithms are selected based on the literature and they are than tested and compared on different input datasets. These input datasets, besides the calendar and external variables mentioned, include different number of lagged values, from zero to three. The combination of model and input dataset which gives the most accurate results on the testing set is selected to forecast future electricity demand. The inclusion of lagged values leads to considerable improvements in accuracy. Although gradient boosting regression features the lowest errors, it is not able to detect peaks of electricity demand correctly. On the contrary, artificial neural network (ANN) demonstrates a great ability to fit historical data and demonstrates a good accuracy on the testing set, as well as for peak demand prediction. Generalized additive model, a relatively new model in the energy forecasting field, gives promising results as its performances are close to the one of ANN and represent an interesting model for future research.  Based on the future values of inputs, the electricity demand in 2028 in Réunion was forecasted using ANN. The electricity demand is expected to reach more than 2.3 GWh and the peak demand about 485 MW. This represents a growth of 12.7% and 14.6% respectively compared to 2019 levels. / I samband med utfasningen av fossila källor för elproduktion i franska icke-sammankopplade territorier är kunskapen om framtida elbehov, särskilt årlig förbrukning och topplast på lång sikt, avgörande för att utforma ny infrastruktur för förnybar energi. Hittills är dessa territorier, främst öar som ligger i Stilla havet och Indiska oceanen, beroende av anläggningar med fossila bränslen. Energipolitiken planerar att på bred front utveckla förnybar energi för att gå mot en koldioxidsnål elmix till 2028.  Denna avhandling fokuserar på den långsiktiga prognosen för elbehov per timme. En metod är utvecklad för att utforma och välja en modell som kan passa korrekt historisk data och för att förutsäga framtida efterfrågan inom dessa specifika områden. Historiska data analyseras först genom en klusteranalys för att identifiera trender och mönster, baserat på en k-means klusteralgoritm. Specifika kalenderinmatningar utformas sedan för att beakta dessa första observationer. Externa inmatningar, såsom väderdata, ekonomiska och demografiska variabler, ingår också.  Prognosalgoritmer väljs utifrån litteraturen och de testas och jämförs på olika inmatade dataset. Dessa inmatade dataset, förutom den nämnda kalenderdatan och externa variabler, innehåller olika antal fördröjda värden, från noll till tre. Kombinationen av modell och inmatat dataset som ger de mest exakta resultaten på testdvärdena väljs för att förutsäga framtida elbehov. Införandet av fördröjda värden leder till betydande förbättringar i exakthet. Även om gradientförstärkande regression har de lägsta felen kan den inte upptäcka toppar av elbehov korrekt. Tvärtom, visar artificiella neurala nätverk (ANN) en stor förmåga att passa historiska data och visar en god noggrannhet på testuppsättningen, liksom för förutsägelse av toppefterfrågan. En generaliserad tillsatsmodell, en relativt ny modell inom energiprognosfältet, ger lovande resultat eftersom dess prestanda ligger nära den för ANN och representerar en intressant modell för framtida forskning.  Baserat på de framtida värdena på indata, prognostiserades elbehovet 2028 i Réunion med ANN. Elbehovet förväntas nå mer än 2,3 GWh och toppbehovet cirka 485 MW. Detta motsvarar en tillväxt på 12,7% respektive 14,6% jämfört med 2019 års nivåer.

本研究旨在探討國小班級關係氛圍的類型以及班級人際網絡結構的現況。 研究中抽樣調查了全國54個四到六年級的班級，依據問卷所得資料進行集群分析法(clustering method)，依同儕關係、師生關係兩向度將班級自然分類成三個集群。 研究進行了不同關係氛圍班級之差異分析，研究二檢驗不同類型班級在師生關係變項上的差別；研究三採用社會網絡分析法(social network analysis)描繪班級人際網絡結構的差別。研究四班級內部的結構，進行核心邊陲分析與班級塊模型分析，研究五進行人際互動課程的實驗研究。研究後建構出診斷班級小社會的訊息，未來可提供給實務現場教師採用。 本研究所得之研究結論如下： 一、國小班級內部關係品質，包含了學生之間的同儕關係、學生與教師的師生關係兩向度，共同建構班級關係氛圍，各班級區分類為低、中、高三種關係品質的班級。。 二、相較於低關係班級，高關係氛圍班級在班內師生關係的標準差較小、班級內師生關係的性別差異亦小。此外，高關係班級的學童的社交計量數與自身師生關係呈現顯著相關。 三、高關係與低關係班級，在人際網絡結構上並沒有顯著的差別，包含了網絡密度、EI指數、二方關係、派系數、成分數、孤離者數和比率。顯示無論班級關係品質為何，都會自然形塑而成班級小社會。 四、低關係班級有幾個值得關係的議題，包含核心學生之間的互惠關係、高密度的子群之間的關係，以及核心學生與高密度學生們對於自身班級的同儕關係與師生關係，都是教師在班級經營中須特別關注的。 五、進行班級氛圍的改變效果，短期課程對氛圍與人際網絡結構的改變效果並不明顯，推論需仰賴教師平時之互動與班務上的經營。 / The study was to investigated the primary school’s classes. The purpose is to understand the class climate and social network of the class. The forty-four classes were in the study by random sampling from Taiwan area.Teacher-student relationship and peer relationship are treated the classification variables. The classes were divided into three categories by clustering method. 18 classes were High-quality relationship; 11 calasses were Low-quality relationship. The results are the comparison of two types of class. High-quality relationship classes have some features: Standard deviation is smaller in teacher-student relationship, the same result of differences between boys and girls. The better the relationship between popular students and teacher. Second, the two-type classes were no differences in social network model. High-quality and Low-quality classes both forming a small community in nature. From the analysis of the class entrials, Mutually beneficial relationship and the relationship between small groups were the important issues. A six-week course experimented in a class. Only small impacts on the class climate and social network.

O estudo da diversidade genética e o conhecimento das relações entre cultivares melhoradas é fundamental para os programas de melhoramento de soja, pois auxiliam na seleção de genitores e recomendação de cultivares. Esta tese está subdividida em três capítulos, sendo que o primeiro traz o referencial teórico relacionado à cultura, à importância econômica e ao melhoramento da soja. O segundo capítulo, por sua vez, foi desenvolvido com os objetivos de: avaliar a diversidade genética a partir de caracteres fenotípicos de genótipos de soja convencional de ciclo precoce em ambientes distintos; determinar a importância de caracteres na divergência genética de soja; e selecionar genitores de ampla diversidade genética para programa de melhoramento, utilizando diferentes métodos de agrupamento multivariados. O experimento foi conduzido em dois locais distintos, Campo Novo dos Parecis - MT, safra 2010/2011 e Urutaí - GO, safra 2012/2013. Foram avaliados dez genótipos de soja convencional de ciclo precoce, em delineamento de blocos completos casualizados, nos quais foram mensurados oito caracteres agronômicos. Por meio de análises uni e multivariadas, foi possível concluir que os agrupamentos formados por todos os métodos multivariados, aliados às médias dos valores fenotípicos dos genótipos, permitiram inferir sobre as combinações promissoras para hibridações artificiais. Ao considerar os dois ambientes de cultivo, o número de dias para a floração, a altura da planta na maturidade e altura de inserção da primeira vagem foram os caracteres que mais contribuíram para a divergência genética em soja. As linhagens UFU 106 e UFU 108 são as mais recomendadas como parte das hibridações com genótipos divergentes, pois são complementares em produtividade de grãos e menor fase vegetativa. Recomenda-se hibridações entre os seguintes pares de genótipos: UFU 106 x UFU 112; UFU 106 x Emgopa 316; UFU 108 x Emgopa 316; UFU 112 x Emgopa 316, para a obtenção de populações segregantes com variabilidade genética superior. O terceiro capítulo foi elaborado para avaliar a interação genótipos por ambientes para o caráter produtividade de grãos em genótipos de soja convencional, de ciclo precoce. Assim, os genótipos foram cultivados em 15 ambientes, distribuídos em cinco estados brasileiros, para determinar sua adaptabilidade e estabilidade por intermédio de métodos paramétricos, não paramétricos e multivariados. O método Wricke (1965), Eberhart e Russel (1966) e AMMI identificaram as linhagens UFU 21 e UFU 22 como as mais estáveis, sendo que ambas apresentaram produtividade de grãos superior a 3800,00 kg ha-1. A linhagem UFU 06 obteve média de produtividade de grãos superior a 4000,00 kg ha-1 e apresentou adaptação ampla pelos métodos Annicchiarico (1992) e Lin e Binns (1988) modificado por Carneiro (1998) e Centróide. / The study of genetic diversity and the knowledge of the relationships among improved cultivars are fundamental for soybean breeding programs, since they help the selection of breeders and recommendation of cultivars. This thesis is subdivided into three chapters, the first one deals with a theoretical reference regarding the culture, the economic importance and the improvement of soybean. The second chapter was developed with the objective of evaluating the genetic diversity from phenotypic traits, of conventional early maturity soybean genotypes in different environments, determining the importance of traits in soybean genetic divergence and selecting parents of broad genetic diversity for breeding programs, using different multivariate clustering methods. The experiment was conducted in two distinct locations, Campo Novo dos Parecis - MT, season 2010/2011 and Urutaí - GO, season 2012/2013. Ten genotypes of conventional early maturity soybean were evaluated in a randomized complete block design, in which eight agronomic characters were measured. By univariate and multivariate analyzes it was possible to conclude that the groupings formed by all the multivariate methods, with the means of the phenotypic values of the genotypes, allowed to infer about the promising combinations for artificial hybridizations. Number of days for flowering, plant height at maturity and height of insertion of the first pod were the characters that contributed the most to the genetic divergence in soybean when considering the two crop environments. UFU 106 and UFU 108 lines are the most recommended as part of the hybridizations with divergent genotypes, since they are complementary in grain yield and lower vegetative phase. Hybridizations between the following pairs of genotypes are recommended to obtain segregating populations with superior genetic variability: UFU 106 x UFU 112; UFU 106 x Emgopa 316; UFU 108 x Emgopa 316; UFU 112 x Emgopa 316. The third chapter was elaborated to evaluate the genotype interaction by environments for grain yield characteristics in conventional soybean genotypes of early maturity, grown in 15 environments distributed in five Brazilian states, to determine the adaptability and stability of the genotypes by parametric, non-parametric and multivariate methods. Wricke (1965), Eberhart and Russel (1966) and AMMI methods identified UFU 21 and UFU 22 lines as the most stable, both with grain yields higher than 3800,00 kg ha-1. The strain UFU 06 obtained an average grain yield of more than 4000,00 kg ha-1 and presented wide adaptation by Annicchiarico (1992), Lin and Binns (1988) modified by Carneiro (1998) and Centroid methods. / Tese (Doutorado)

La connaissance de la distribution spatiale des espèces et leurs déterminants constituent les principaux thèmes de la biogéographie et de l’écologie. Cette thèse a été réalisée sur les Fabaceae de l’Afrique Centrale :Burundi, République Démocratique du Congo et Rwanda. Composées de trois sous-familles (Caesalpinioideae, Faboideae et Mimosoideae), les Fabaceae sont présentes dans tous les biomes terrestres du monde. La présente étude a pour objectif d’évaluer et localiser la diversité spécifique, déterminer les zones de conservation des Fabaceae, identifier leur modèle de distribution spatiale et examiner l’impact potentiel de la déforestation sur les espèces indicatrices de cette famille. Les facteurs environnementaux déterminant la distribution des espèces endémiques de Fabaceae, les zones favorables à leur présence et l’évaluation de leur état de conservation ont également été analysés. L’étude a consisté à vérifier et à digitaliser tous les échantillons d’herbiers des Fabaceae conservés au Jardin Botanique National de Belgique et à l’Université Libre de Bruxelles. Les Systèmes d’Information Géographique ont été largement utilisés. Les analyses spatiales ont montré une distribution non uniforme de la diversité spécifique des Fabaceae. Les zones où la diversité spécifique coïncide avec celle des espèces endémiques ont été suggérées comme prioritaires pour la conservation. L’analyse de groupement appliquée sur le jeu de données des Mimosoideae a mis en évidence trois régions floristiques majeures. Ces trois régions correspondent aux régions phytogéographiques définies par White (1979, 1983). L’étude des Caesalpinioideae a confirmé l’existence d’espèces indicatrices de territoires phytogéographiques de Ndjele (1988). L’étude de cas réalisée au Katanga (R.D. Congo) a montré que la déforestation constitue une menace potentielle pour ces espèces, les plus vulnérables étant inféodées aux habitats forestiers. La distribution potentielle des espèces endémiques de Fabaceae (Caesalpinioideae) a montré une convergence entre les cartes de la distribution actuelle et celles issues de la distribution potentielle. Des sites où les espèces n’ont jamais été signalées ont été identifiés. La distribution potentielle a démontré le rôle des déterminants environnementaux dans la distribution des espèces endémiques. Cette étude a également prouvé que d’importantes proportions des zones de diversité des espèces endémiques ne sont pas couvertes par la conservation actuelle de l’Afrique Centrale. Cette thèse confirme que l’Afrique Centrale est une zone importante de conservation de la biodiversité. L’étude devrait être poursuivie en étendant la distribution potentielle aux Fabaceae non endémiques afin de définir leurs aires de distributions, critère indispensable pour évaluer le degré de vulnérabilité d’une espèce. La création de nouvelles aires protégées renforcerait le système actuel de conservation dans cette région./Knowledge of the spatial distribution of species and its determinants constitutes a principal theme in biogeography and ecology. This dissertation focused on the Fabaceae family of Central Africa: Burundi, the Democratic Republic of the Congo and Rwanda. Composed of three subfamilies (Caesalpinioideae, Faboideae and Mimosoideae), Fabaceae species are present in all terrestrial biomes of the world. The current study aimed to assess and locate Fabaceae species diversity, to determine potential conservation zones, to identify the spatial distribution pattern and to examine the potential impact of deforestation on some indicator species of this family. The environmental factors driving the distribution of endemic Fabaceae have been identified, as well as the potential areas of occurrence; their conservation status in the region has also been studied. The current study has verified and digitized all the herbarium samples kept at the National Botanical Garden of Belgium and the Université Libre de Bruxelles regarding the Fabaceae family. Geographic Information Systems have been used frequently. Spatial analysis showed an uneven distribution of Fabaceae species diversity. The zones where species richness depends on endemic species diversity are suggested to have priority for conservation. A cluster analysis applied to the dataset of Mimosoideae has highlighted three major floristic regions. These three regions correspond to existing phytogeographical regions defined by White (1979, 1983). A study on Caesalpinioideae species of Central Africa has confirmed the existence of indicator species of phytogeographic territories of Ndjele (1988). A case study conducted in Katanga (D.R Congo) showed that deforestation constitutes a potential threat for these species, the most vulnerable being those strictly confined to forest habitats. The potential distribution of endemic Fabaceae (Caesalpinioideae) showed a clear convergence between the current distribution maps and the potential distibutions. Sites where those species had not been reported before have been identified. The potential distribution enabled to identify the importance of the different environmental factors for each endemic species’ distribution. This study also showed that large regions characterized by important endemic species diversity are not covered by current zones of conservation in Central Africa. This dissertation confirms Central Africa as an important zone for biodiversity conservation. The current research should be completed by the potential distribution of non endemic Fabaceae species of Central Africa, an important criterion to evaluate their degree of vulnerability. Creation of new protected areas would reinforce the current status of conservation in the region. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

AlNiCo magnets are known for high-temperature stability and superior corrosion resistance and have been widely used for various applications. Reported magnetic energy density ((BH) max) for these magnets is around 10 MGOe. Theoretical calculations show that ((BH) max) of 20 MGOe is achievable which will be helpful in covering the gap between AlNiCo and Rare-Earth Elements (REE) based magnets. An extended family of AlNiCo alloys was studied in this dissertation that consists of eight elements, and hence it is important to determine composition-property relationship between each of the alloying elements and their influence on the bulk properties. In the present research, we proposed a novel approach to efficiently use a set of computational tools based on several concepts of artificial intelligence to address a complex problem of design and optimization of high temperature REE-free magnetic alloys. A multi-dimensional random number generation algorithm was used to generate the initial set of chemical concentrations. These alloys were then examined for phase equilibria and associated magnetic properties as a screening tool to form the initial set of alloy. These alloys were manufactured and tested for desired properties. These properties were fitted with a set of multi-dimensional response surfaces and the most accurate meta-models were chosen for prediction. These properties were simultaneously extremized by utilizing a set of multi-objective optimization algorithm. This provided a set of concentrations of each of the alloying elements for optimized properties. A few of the best predicted Pareto-optimal alloy compositions were then manufactured and tested to evaluate the predicted properties. These alloys were then added to the existing data set and used to improve the accuracy of meta-models. The multi-objective optimizer then used the new meta-models to find a new set of improved Pareto-optimized chemical concentrations. This design cycle was repeated twelve times in this work. Several of these Pareto-optimized alloys outperformed most of the candidate alloys on most of the objectives. Unsupervised learning methods such as Principal Component Analysis (PCA) and Heirarchical Cluster Analysis (HCA) were used to discover various patterns within the dataset. This proves the efficacy of the combined meta-modeling and experimental approach in design optimization of magnetic alloys.

La biologie de la conservation est un domaine de recherche en pleine expansion en raison de la perte accélérée de la biodiversité à l’échelle mondiale. Pour mieux comprendre les processus et les menaces au maintien des populations de petite taille et les effets des facteurs anthropiques sur la biodiversité, la génétique est fréquemment utilisée en conservation. Des analyses génétiques peuvent, par exemple, nous informer sur les tendances à long terme, la diversité des populations et les stratégies de reproduction d’une espèce. La tortue des bois (Glyptemys insculpta) est une espèce endémique à l’Amérique du Nord qui est en danger d’extinction selon l’Union internationale pour la conservation de la nature. Dans le cadre de ma thèse, j’avais comme objectif de caractériser la diversité génétique de cette espèce menacée au Canada. À cet effet, j’ai analysé la génétique des populations de tortues des bois à plusieurs échelles spatiales et temporelles, afin de mieux cerner les processus ayant un impact sur la diversité des populations. Dans un premier temps, les relations de parentalité ont été reconstruites au sein d’une population de tortues des bois pour estimer la fréquence de paternité multiple et de paternité répétée. Les résultats de mes travaux suggèrent que l’emmagasinement de sperme chez la femelle et la reproduction multiple avec les mêmes partenaires pour plus d’une saison de reproduction pourraient expliquer ces phénomènes. Ces stratégies de reproduction pourraient dans ce cas être induites par la faible densité de la population à l’étude, ou encore par la fidélité au site d’hibernation où la majorité des évènements de copulation ont lieu. Par la suite, je me suis intéressée à la diversité génétique des populations de tortues de bois. J’ai voulu comprendre les effets de la configuration spatiale des éléments du paysage et les évènements de dispersion géographique sur la diversité des populations. À l’aide d’une approche de génétique du paysage, mes analyses montrent que la division des populations par bassins versants explique une large fraction de la diversité génétique interpopulations. Ces résultats confirment également que les bassins versants représentent des unités de gestion propices à la protection des populations de tortues des bois. Finalement, des analyses de réseaux ont été utilisées pour mieux cerner la dynamique de flux génique entre les populations de la rive nord et de la rive sud du fleuve Saint-Laurent. Plus spécifiquement, la rive nord se caractérise par un réseau robuste de populations isolées, alors que les populations de la rive sud présentent plutôt une structure de métapopulation. En utilisant les réseaux construits à partir de données génétiques, des scénarios hypothétiques furent comparés pour explorer la sélection de populations à l’aide du logiciel BRIDES. Les résultats de ces analyses ont permis de cibler l’importance de certaines populations de tortues des bois pour la connectivité du réseau. L’importance de ces populations n’aurait pu être prédite par les résultats de la diversité et de la différenciation génétique, les indices de centralité et les analyses d’élimination de nœuds. Grâce à la génétique, cette thèse apporte de nouvelles connaissances sur la tortue des bois, les stratégies de reproduction des différents sexes, le flux génique, la connectivité et l’influence du réseau hydrographique sur la diversité des populations. Ces résultats nous permettent d’avoir une meilleure compréhension des processus affectant la diversité génétique de cette espèce afin de mieux la protéger. Toutes les analyses réalisées pour cette thèse sont directement applicables à l’ensemble des autres espèces longévives avec des générations chevauchantes. / Conservation biology is a rapidly expanding field of research due to the accelerating loss of global biodiversity. To better understand the processes and threats to the persistence of small populations and the effects of anthropogenic factors on biodiversity, genetic approaches are frequently used in conservation. Genetic analyzes can, for example, inform us about long-term trends, population diversity and reproductive strategies of a species. The wood turtle (Glyptemys insculpta) is a species endemic to North America that is endangered according to the International Union for the Conservation of Nature. As part of my thesis, my objective was to characterize the genetic diversity of this threatened species in Canada. In order to better understand the impact of reproductive strategy and landscape structure on population diversity, I analyzed the genetics of wood turtle populations at several spatial and temporal scales. First, parentage relationships were reconstructed in a population of wood turtles to estimate the frequency of multiple and repeated paternity. The results of my work suggest that sperm storage in females and multiple reproduction with the same partners for more than one breeding season could explain these phenomena. These reproduction strategies could in this case be induced by the low density of the study population, or by fidelity to the overwintering site where the majority of copulation events take place. Subsequently, I assessed the genetic diversity of wood turtle populations. I wanted to understand the effects of the landscape configuration and geographic dispersion events on the diversity of populations. Using a landscape genetics approach, my analyzes show that the division of populations by watershed explains a large fraction of the genetic diversity between populations. These results also confirm that watersheds represent management units conducive to the protection of wood turtle populations. Finally, network analysis was used to better understand the gene flow dynamics among populations located on the north and south shores of the St. Lawrence River. More specifically, the north shore is characterized by a robust network of isolated populations, whereas the populations on the south shore present more of a metapopulation structure. Using population graphs, hypothetical scenarios were compared to explore the node selection process using the BRIDES algorithm. The results of these analyzes made it possible to point out specific populations of wood turtles, considering their importance for network connectivity. This could have not been predicted by using genetic diversity and distinctiveness estimates, node-based metrics, and node removal analysis for these populations. Thanks to genetics, this thesis brings new knowledge on the wood turtle, the reproductive strategies of both sexes, the gene flow, the connectivity and the influence of the hydrographic network on population diversity. These results allow us to have a better understanding of the processes affecting the genetic diversity of this species in order to better protect it. All analyses performed for this thesis are directly applicable to other long-lived species with overlapping generations.