Compreender como as espécies interagem e como a topologia de redes ecológicas influencia a dinâmica de populações e comunidades tem sido um dos principais objetivos de estudos ecológicos há mais de um século. Apesar desta longa tradição, o estudo de redes ecológicas tem aumentado drasticamente nas últimas duas décadas. No entanto, só recentemente ecólogos começaram a ir além da descrição de padrões topológicos e passaram a integrar outros dados biológicos importantes, como características funcionais e filogenia. Esta tese teve principalmente dois objetivos: i) desenvolver novas abordagens analíticas capazes de integrar informações funcionais e filogenéticas, a fim de descrever padrões estruturais em redes ecológicas, e ii) compreender a influência de dinâmicas eco-evolutivas na robustez de redes mutualísticas. No Capítulo I, desenvolvi uma abordagem analítica integradora para particionar os efeitos da filogenia e de características funcionais sobre a estrutura de redes de interação biótica. O método combina Teoria de Conjuntos Difusos e correlação matricial. Eu também desenvolvi um estudo de simulação para testar a acurácia da metodologia proposta em termos de Erro Tipo I. As simulações demonstram que o método é acurado, ou seja, rejeita incorretamente uma hipótese nula verdadeira em ~ 5% dos casos. No Capítulo II, investiguei como diferentes cenários de extinção afetam a robustez de uma rede de dispersão de sementes do sul do Brasil, incluindo cenários onde as espécies são eliminadas com base em sua distinção evolutiva e funcional. Os resultados indicam que a perda de espécies generalistas e diversidade funcional faz com que rede seja mais propensa a colapsar. No Capítulo III, desenvolvi uma investigação teórica sobre a influencia de diferentes modos de evolução de atributos sobre a robustez de redes mutualísticas que estão sofrendo um ataque funcional. Os resultados mostram que, apesar da pequena faixa de variação na robustez das redes, o modo de evolução dos atributos, e a interação entre modos de evolução de cada conjunto de espécies que interagem, influenciam a robustez de redes ecológicas, especialmente em casos extremos, onde os atributos das espécies apresentam sinal filogenético muito baixo ou muito forte. / Understanding how species interact and how the topology of ecological networks influences the dynamics of populations and communities has been mindboggling ecologists for over a century now. Despite this long tradition, the study of complex ecological networks has increased dramatically in the past two decades. Nonetheless, only recently ecologists have started to move beyond the description of topological patterns and started to integrate other important biological data, such as functional traits and phylogenies. Therefore, the aim of this dissertation was mainly two-fold: develop new analytical approaches capable to integrate functional and phylogenetic information in order to describe structural patterns in ecological networks, and to understand the effects of ecoevolutionary dynamics on network robustness. In Chapter I, I developed a new and integrative analytical framework to partition the effects of phylogenies and functional traits on the structure of ecological networks. The method combines fuzzy set theory and matrix correlation. I also developed a simulation study to test the accuracy of the framework I proposed. My simulation study demonstrates that the method is accurate, i.e., incorrectly rejecting a true null hypothesis in ~ 5%. In Chapter II, I investigated how different extinction scenarios affect the robustness of a seed dispersal network, from southern Brazil, including scenarios where species are eliminated based on their evolutionary and functional distinctiveness. The results indicate that loss of generalist species and functional diversity makes the system more likely to collapse. In Chapter III, I developed a theoretical investigation of the role of distinct trait evolution modes on the robustness of mutualistic networks undergoing functional trait extinctions. My results show that, despite the small range of variation in network robustness, the mode of trait evolution alone and the interaction between modes of evolution of each set of interacting species matter for network robustness, especially in extreme cases, where species traits present either very low or very strong phylogenetic signal.

Compreender como as espécies interagem e como a topologia de redes ecológicas influencia a dinâmica de populações e comunidades tem sido um dos principais objetivos de estudos ecológicos há mais de um século. Apesar desta longa tradição, o estudo de redes ecológicas tem aumentado drasticamente nas últimas duas décadas. No entanto, só recentemente ecólogos começaram a ir além da descrição de padrões topológicos e passaram a integrar outros dados biológicos importantes, como características funcionais e filogenia. Esta tese teve principalmente dois objetivos: i) desenvolver novas abordagens analíticas capazes de integrar informações funcionais e filogenéticas, a fim de descrever padrões estruturais em redes ecológicas, e ii) compreender a influência de dinâmicas eco-evolutivas na robustez de redes mutualísticas. No Capítulo I, desenvolvi uma abordagem analítica integradora para particionar os efeitos da filogenia e de características funcionais sobre a estrutura de redes de interação biótica. O método combina Teoria de Conjuntos Difusos e correlação matricial. Eu também desenvolvi um estudo de simulação para testar a acurácia da metodologia proposta em termos de Erro Tipo I. As simulações demonstram que o método é acurado, ou seja, rejeita incorretamente uma hipótese nula verdadeira em ~ 5% dos casos. No Capítulo II, investiguei como diferentes cenários de extinção afetam a robustez de uma rede de dispersão de sementes do sul do Brasil, incluindo cenários onde as espécies são eliminadas com base em sua distinção evolutiva e funcional. Os resultados indicam que a perda de espécies generalistas e diversidade funcional faz com que rede seja mais propensa a colapsar. No Capítulo III, desenvolvi uma investigação teórica sobre a influencia de diferentes modos de evolução de atributos sobre a robustez de redes mutualísticas que estão sofrendo um ataque funcional. Os resultados mostram que, apesar da pequena faixa de variação na robustez das redes, o modo de evolução dos atributos, e a interação entre modos de evolução de cada conjunto de espécies que interagem, influenciam a robustez de redes ecológicas, especialmente em casos extremos, onde os atributos das espécies apresentam sinal filogenético muito baixo ou muito forte. / Understanding how species interact and how the topology of ecological networks influences the dynamics of populations and communities has been mindboggling ecologists for over a century now. Despite this long tradition, the study of complex ecological networks has increased dramatically in the past two decades. Nonetheless, only recently ecologists have started to move beyond the description of topological patterns and started to integrate other important biological data, such as functional traits and phylogenies. Therefore, the aim of this dissertation was mainly two-fold: develop new analytical approaches capable to integrate functional and phylogenetic information in order to describe structural patterns in ecological networks, and to understand the effects of ecoevolutionary dynamics on network robustness. In Chapter I, I developed a new and integrative analytical framework to partition the effects of phylogenies and functional traits on the structure of ecological networks. The method combines fuzzy set theory and matrix correlation. I also developed a simulation study to test the accuracy of the framework I proposed. My simulation study demonstrates that the method is accurate, i.e., incorrectly rejecting a true null hypothesis in ~ 5%. In Chapter II, I investigated how different extinction scenarios affect the robustness of a seed dispersal network, from southern Brazil, including scenarios where species are eliminated based on their evolutionary and functional distinctiveness. The results indicate that loss of generalist species and functional diversity makes the system more likely to collapse. In Chapter III, I developed a theoretical investigation of the role of distinct trait evolution modes on the robustness of mutualistic networks undergoing functional trait extinctions. My results show that, despite the small range of variation in network robustness, the mode of trait evolution alone and the interaction between modes of evolution of each set of interacting species matter for network robustness, especially in extreme cases, where species traits present either very low or very strong phylogenetic signal.

Compreender como as espécies interagem e como a topologia de redes ecológicas influencia a dinâmica de populações e comunidades tem sido um dos principais objetivos de estudos ecológicos há mais de um século. Apesar desta longa tradição, o estudo de redes ecológicas tem aumentado drasticamente nas últimas duas décadas. No entanto, só recentemente ecólogos começaram a ir além da descrição de padrões topológicos e passaram a integrar outros dados biológicos importantes, como características funcionais e filogenia. Esta tese teve principalmente dois objetivos: i) desenvolver novas abordagens analíticas capazes de integrar informações funcionais e filogenéticas, a fim de descrever padrões estruturais em redes ecológicas, e ii) compreender a influência de dinâmicas eco-evolutivas na robustez de redes mutualísticas. No Capítulo I, desenvolvi uma abordagem analítica integradora para particionar os efeitos da filogenia e de características funcionais sobre a estrutura de redes de interação biótica. O método combina Teoria de Conjuntos Difusos e correlação matricial. Eu também desenvolvi um estudo de simulação para testar a acurácia da metodologia proposta em termos de Erro Tipo I. As simulações demonstram que o método é acurado, ou seja, rejeita incorretamente uma hipótese nula verdadeira em ~ 5% dos casos. No Capítulo II, investiguei como diferentes cenários de extinção afetam a robustez de uma rede de dispersão de sementes do sul do Brasil, incluindo cenários onde as espécies são eliminadas com base em sua distinção evolutiva e funcional. Os resultados indicam que a perda de espécies generalistas e diversidade funcional faz com que rede seja mais propensa a colapsar. No Capítulo III, desenvolvi uma investigação teórica sobre a influencia de diferentes modos de evolução de atributos sobre a robustez de redes mutualísticas que estão sofrendo um ataque funcional. Os resultados mostram que, apesar da pequena faixa de variação na robustez das redes, o modo de evolução dos atributos, e a interação entre modos de evolução de cada conjunto de espécies que interagem, influenciam a robustez de redes ecológicas, especialmente em casos extremos, onde os atributos das espécies apresentam sinal filogenético muito baixo ou muito forte. / Understanding how species interact and how the topology of ecological networks influences the dynamics of populations and communities has been mindboggling ecologists for over a century now. Despite this long tradition, the study of complex ecological networks has increased dramatically in the past two decades. Nonetheless, only recently ecologists have started to move beyond the description of topological patterns and started to integrate other important biological data, such as functional traits and phylogenies. Therefore, the aim of this dissertation was mainly two-fold: develop new analytical approaches capable to integrate functional and phylogenetic information in order to describe structural patterns in ecological networks, and to understand the effects of ecoevolutionary dynamics on network robustness. In Chapter I, I developed a new and integrative analytical framework to partition the effects of phylogenies and functional traits on the structure of ecological networks. The method combines fuzzy set theory and matrix correlation. I also developed a simulation study to test the accuracy of the framework I proposed. My simulation study demonstrates that the method is accurate, i.e., incorrectly rejecting a true null hypothesis in ~ 5%. In Chapter II, I investigated how different extinction scenarios affect the robustness of a seed dispersal network, from southern Brazil, including scenarios where species are eliminated based on their evolutionary and functional distinctiveness. The results indicate that loss of generalist species and functional diversity makes the system more likely to collapse. In Chapter III, I developed a theoretical investigation of the role of distinct trait evolution modes on the robustness of mutualistic networks undergoing functional trait extinctions. My results show that, despite the small range of variation in network robustness, the mode of trait evolution alone and the interaction between modes of evolution of each set of interacting species matter for network robustness, especially in extreme cases, where species traits present either very low or very strong phylogenetic signal.

Eco-evolutionary dynamics integrates the reciprocal interactions betweenecology and evolution. These two branches of biology traditionally assumethe other as static for simplicity. However, increasing evidence shows thatthis simplification may not always hold because ecology and evolution canoperate in similar timescales. This thesis theoretically explores how thereciprocal interactions may influence ecological and evolutionary outcomesin four different eco-evolutionary contexts.Many species of non-social animals live in groups. Aggregating ingroups often has both benefits and costs that depend on group size. Thanksto the benefits of aggregation, population growth likely depends positivelyon population density when it is small. This phenomenon, the Allee effect,has been hypothesized to explain the evolution of aggregation behavior. Ifind that the Allee effect alone does not lead to the evolution whenpopulation dynamics is explicitly accounted for. Some other mechanisms,such as frequent needs for colonizing new patches or anti-aggregation,should be invoked to explain why aggregation behavior could evolve.Phenotypic plasticity is the ability of a genotype to express distinctphenotypes when exposed to different environments. Although it is oftenshown to be adaptive and not costly, highly plastic organisms are rare. Paststudies demonstrated some potential reasons. I test another possibility; costsmay arise from sexual selection because highly plastic individuals may beless preferred as a mate. I show that, even in the absence of the direct cost ofplasticity, the level of plasticity remained low at intermediate strengths ofassortative mating. This pattern is robust across wide ranges of parametervalues.Ecological speciation occurs when ecologically divergent selectionbetween environments causes reproductive isolation between divergingsubpopulations. Several verbal models of ecological speciation emphasizethe roles of phenotypic plasticity in promoting speciation. The complexprocesses involved in speciation, however, are difficult to be evaluated byverbal accounts. I quantitatively test the proposed idea in a mechanisticmodel of ecological speciation in the presence and absence of plasticity. Ifind conditions under which plasticity can promote or hinder ecologicalspeciation. Plasticity facilitates speciation by producing a gap in thedistributions of expressed phenotypes, which serves as a barrier to gene flowin an assortatively mating population.Ecosystem ecology and evolutionary biology are the least integratedfields in ecology and evolution. Natural selection operating at the individuallevels on traits governing ecosystem functions may affect ecosystemproperties, which may feedback to individuals. I reviewed this idea anddemonstrate the feedback loop by using a simple consumer-resource model.

Sexual selection is a prevalent evolutionary force that prominently led to the evolution of elaborate and conspicuous traits. However, it remains an ongoing scientific debate if sexual selection has a net negative or positive effect on the general viability of a population (from here on ‘population fitness’). Sexual selection could reduce the fitness of a population, when sexual conflict leads to, for example, one sex harming the other during mating. By contrast, sexual selection could increase population fitness, if it increases selection against deleterious alleles in the gene pool. This could be the case if sexual selection acts on traits that capture the genetic quality of an individual. Indeed, recent meta-analytic evidence suggests that sexual selection typically increases population fitness. Therefore, the strength of sexual selection could prove important for the fitness of wild populations facing environmental stress. Importantly, the strength of sexual selection can differ fundamentally between the sexes and also among populations or species. One reason that might trigger this variation is the interdependence of evolutionary and ecological processes, also known as eco-evolutionary dynamics. Therefore, studying eco-evolutionary dynamics could help us to understand the vast variation in sexual selection and to predict the effect of sexual selection on population fitness. I studied different pathways of eco-evolutionary dynamics in diverse environments using experimental and comparative approaches: First, I investigated the effect of diet quality on the strength of selection in females and males in a model insect. I show that low diet quality increased the potential for selection in males and females, but surprisingly the effect of diet was particularly pronounced in females. This suggests that deteriorating diet quality might be predominantly stressful for females and could disproportionally increase selection on females compared to males. Overall, this study provides an example of how an ecological factor can influence the strength of selection. Secondly, I performed an experimental study to test the effect of a key demographic factor on sexual selection. To this end, I manipulated the density of populations of the red flour beetle by changing population size as well as habitat size. By quantifying mating behaviour and fitness, I demonstrated that an increase in density can have an impact on the potential for sexual selection. Especially in females, higher densities caused an increase in sexual selection, whereas in males, higher densities mainly increased the benefitted of additional matings. These effects were most profound when varying density through the number of individuals, whereas habitat size barely affected sexual selection. Collectively, this suggests that density dependent sexual selection mediates eco-evolutionary dynamics, which can be particularly important for the fitness of declining populations. Thirdly, I took a comparative approach to study how the strength of selection in general might alter the demography of populations, specifically their carrying capacity (i.e. the limit for population growth). Since population growth is typically limited by females, strong selection on males can affect the carrying capacity of a population. If selection is typically stronger in males compared to females, females could benefit from selection against deleterious alleles in males without the reduction in population growth imposed by strong selection on females. I compiled data on the genetic variance in fitness for females and males as a measure for the strength of selection. I found that selection is typically stronger in males compared to females across the animal kingdom. Sexual selection theory suggests that such a sex-difference in the strength of selection could be caused by stronger sexual selection on males. Indeed, the sex difference in selection strength was only significant in polygamous and not in socially monogamous species, with the latter presumably experiencing weaker sexual selection. In conclusion, these data suggest that in polygamous species selection is typically stronger on males compared to females and the sex-specific strength of selection could have an effect on the carrying capacity of populations. After exploring the relevance of sexual selection in the previous chapters, I fourthly, tested the robustness of a widely used and easily obtainable proxy for the strength of sexual selection across species: sexual size dimorphism. Theory predicts that sexual selection on males promotes the evolution of larger males relative to females. Indeed, my comparative study shows that the degree of sexual size dimorphism was significantly correlated with the strength of sexual selection, estimated by a diverse range of proxies for the strength of sexual selection from primary studies. Importantly, pre-copulatory sexual selection correlated positively with an increasing male-bias in sexual size dimorphism, while post-copulatory sexual selection was non-significantly negatively associated with the degree of male-bias in sexual size dimorphism. Overall, these data suggest that sexual size dimorphism can be a useful, albeit rough, proxy for the strength of pre-copulatory sexual selection across species. In the final chapter, I synthesise the results of my studies in the light of their implications for eco-evolutionary dynamics. While there is evidence that sexual selection could typically improve population fitness, the data I present suggest that the ecology of a population, here specifically diet quality and population density, could have a crucial impact on the strength of (sexual) selection. Hence, unravelling the eco-evolutionary dynamics of (sexual) selection could prove important for our understanding of their effect on population fitness. Importantly, selection seems to be typically stronger in males compared to females, at least in species with prevalent sexual selection on males. Overall, further exploration of the eco-evolutionary dynamics of sexual selection and their effects on population fitness promise to be exciting and profitable future endeavours.

A adaptação e a diversificação em sistemas multiespecíficos são crescentemente reconhecidas como processos relevantes para a compreensão da biodiversidade. Nosso objetivo foi investigar como a seleção natural relacionada a interações ecológicas influencia a estrutura, dinâmica e diversificação de assembleias mutualistas. Primeiro, modelamos como mutualismo e competição intraespecífica geram regimes seletivos antagônicos que definem padrões de diversificação. Nossos modelos preveem que em mutualismos de baixa intimidade, nos quais cada organismo têm muitos parceiros individuais, fenótipos extremos têm interações mutualísticas desajustadas em relação à complementaridade de traços, contrabalanceando efeitos diversificadores da competição intraespecífica e restringindo a especiação. Em sistemas de alta intimidade, nos quais mutualismos têm maior impacto adaptativo e cada organismo têm poucos parceiros, tal seleção estabilizadora imposta por mutualismos é reduzida, favorecendo a diversificação. Entretanto, mutualismos de baixa intimidade são mais ricos que mutualismos íntimos na natureza. Sob baixa intimidade de interações, adições de espécies não-aparentadas envolvidas em dinâmicas de convergência constituem explicação plausível para essa discrepância. Em sistemas de alta intimidade, restrições a adições de espécies impostas por histórias coevolutivas estreitamente relacionadas poderiam explicar menores riquezas, apesar do maior potencial de diversificação adaptativa. Em segundo lugar, avaliamos se reconfigurações adaptativas das interações ecológicas podem explicar a variação estrutural de redes mutualistas. Usando um modelo eco-evolutivo, mostramos que a seleção favorecendo trocas de mutualistas e maximizando a abundância das espécies altera propriedades das redes, aumentando seu aninhamento e diminuindo sua estabilidade. Nossos modelos superestimaram o aninhamento em mutualismos de alta intimidade, possivelmente porque não consideramos ligações proibidas impostas por morfologia ou fenologia. Entretanto, as redes simuladas reproduzem aninhamento e modularidade de mutualismos de baixa intimidade, cujas interações são mais flexíveis. Sob competição por mutualistas, as reconfigurações da rede continuam em um dinâmica sem fim, mesmo quando estrutura e a estabilidade atingem níveis assintóticos, o que pode explicar a variação empírica de interações em redes com estruturas temporalmente constantes. Em um terceiro estudo, modelamos como diferentes modos de especiação modificam propriedades de redes mutualistas. Se a especiação resulta em expansão do nicho e as espécies emergentes tornarem-se mais conectadas, o aninhamento aumenta e a modularidade diminui, frequentemente resultando em redes instáveis. Se a especiação causar retração do nicho e as espécies emergentes tornarem-se menos conectadas, aninhamento e modularidade aumentam, promovendo estabilidade. Diferentes regras de sobreposição de nicho entre espécies emergentes não alteraram esses resultados. Assim, retrações de nicho via divergência adaptativa, como deslocamentos de caracteres na especiação simpátrica, devem gerar espécies incorporáveis às redes mutualistas sem desestabilização. Entretanto, expansões de nicho via adaptações a recursos complementares em alopatria devem desestabilizar as redes quando do contato secundário. Efeitos potencialmente amplos de um único evento de especiação mostram que estudos relacionando diversificação e dinâmica são relevantes para o debate sobre complexidade e estabilidade de redes ecológicas. Concluímos que a compreensão mecanística sobre as origens e a manutenção da biodiversidade depende da integração de teorias ecológicas e evolutivas com base em dados empíricos, como fizemos aqui ao modelar dinâmicas adaptativas de interações ecológicas usando informações sobre a organização e história natural de assembleias mutualistas / Adaptation and diversification in species-rich systems are increasingly recognized as relevant processes to improve our understanding on biodiversity. Our aim was to investigate how natural selection related to ecological interactions shapes the structure, dynamics and diversification of mutualistic assemblages. First, we modeled how mutualism and intraspecific competition generate antagonistic selective regimes that define patterns of diversification. Ours models predict that in low intimacy mutualisms, in which each organism has various individual partners, extreme phenotypes experience trait mismatches in mutualistic interactions that oppose the diversifying effects of intraspecific competition and constrain speciation. In high intimacy systems, in which mutualistic interactions have a higher impact on fitness and each organism has fewer partners, such stabilizing selection is reduced, favoring diversification. However, low intimacy mutualisms are richer than high intimacy mutualisms in nature. Under low interaction intimacy, additions of non-related species involved in trait convergence dynamics are a plausible explanation for such a discrepancy. In high intimacy systems, restrictions to species additions imposed by tight coevolutionary histories could explain lower richnesses despite of a higher potential for adaptive diversification. In a subsequent study, we evaluated whether the adaptive rewiring of ecological interactions explain the structural variation of mutualistic networks. Using an eco-evolutionary model, we show that selection favoring continuous interaction switching that maximizes species abundances changes network properties, increasing nestedness and decreasing stability. Our models overestimated nestedness in high intimacy mutualisms, probably because we did not consider forbidden links imposed by morphology or phenology. However, simulated networks reproduce nestedness and modularity of low intimacy mutualism, in which interactions are more flexible. Under competition for mutualists, rewires continue in an endless dynamics, even when the structure and stability reach asymptotic levels at the network level, which could explain the empirical variation of interactions in networks showing temporally constant structures. In a third study, we modeled effects of different modes of speciation on mutualistic network properties. If speciation results in niche width expansion and emerging species become more connected, nestedness increases and modularity decreases, often resulting in unstable networks. If speciation causes niche width retractions and emerging species become less connected, both nestedness and modularity increase, promoting stability. Different rules of niche overlap between emerging rules did not change these results. Therefore, niche retractions via adaptive divergence, such as character displacement in sympatric speciation, can generate species that will enter local networks without destabilizing them. However, niche width expansions due to adaptation to additional resources in allopatry should destabilize networks if secondary contact between emerging species occur. High magnitude potential effects of a single speciation event show that studies relating diversification and dynamics are relevant to the debate on complexity and stability of ecological networks. We concluded that the mechanistic understanding of biodiversity origins and maintenance relies on the integration between ecological and evolutionary theories based on empirical data, as wed did here by modeling the adaptive dynamics of ecological interactions using information on the structure and natural history of mutualistic assemblages

A ocorrência de distúrbios impacta a diversidade de estratégias de vida em comunidades e a evolução de estratégias de vida em populações. Na Ecologia, o distúrbio é estudado enquanto fator ambiental que altera a disponibilidade de recursos e a abundância das populações, ocasionando a exclusão competitiva de espécies menos favorecidas a depender da intensidade e da frequência de sua ocorrência. Na Biologia Evolutiva, o distúrbio é avaliado enquanto pressão que, dependendo de sua regularidade no ambiente, determina a intensidade da resposta evolutiva das espécies e, assim, a adaptação a estratégias de vida de maior aptidão. Ainda que haja separação entre as duas áreas, a dinâmica ecológica influencia a dinâmica evolutiva e vice-versa. Estudos que integram Ecologia e Evolução têm sido cada vez mais recorrentes, no entanto, poucos ou nenhum consideram o distúrbio. Neste trabalho, utilizamos um modelo baseado em indivíduo para criar cenários nos quais os processos de adaptação e exclusão competitiva de espécies possam ocorrer, tanto conjunta quanto isoladamente, a fim de entender como o distúrbio determina as estratégias de vida presentes em comunidades sob dinâmicas ecológicas, evolutivas e eco-evolutivas. No modelo, as estratégias de vida são atributos herdáveis definidos por um trade-off entre longevidade e fecundidade. O cenário evolutivo foi composto por populações (apenas uma espécie) com mutação; o cenário ecológico, por diversas espécies sem mutação e o cenário eco-evolutivo, por diversas espécies com mutação. Observamos que o distúrbio esteve positivamente relacionado com a predominância de indivíduos fecundos em todos os cenários, mas o efeito do distúrbio sobre a diversidade de estratégias variou. Nos cenários evolutivo e eco-evolutivo, a diversidade de estratégias aumentou com a intensificação do distúrbio, enquanto no cenário ecológico a diversidade caiu. Isso indica a importância da mutação como fonte de novas variantes da estratégia quando há alta renovação de indivíduos da comunidade, condicionada pela mortalidade elevada. Apenas no cenário eco-evolutivo houve um pico de heterogeneidade de estratégias em níveis intermediários de distúrbio. Neste cenário, o isolamento reprodutivo das espécies, em contraposição à pan-mixia que ocorre dentro das populações, permite que as espécies difiram em relação à sua estratégia de vida média. Em paralelo, a entrada constante de variantes de estratégias por mutação previne a extinção definitiva das estratégias do sistema. Assim, quando o nível de distúrbio é intermediário, tanto espécies mais fecundas quanto espécies mais longevas conseguem coexistir na comunidade. Dado que os diferentes cenários resultaram em padrões distintos de frequência relativa de estratégias de vida, este trabalho evidencia a importância de estudarmos o efeito do distúrbio na estrutura e na dinâmica de comunidades unindo processos que são tipicamente estudados de forma isolada pela Ecologia e pela Biologia Evolutiva / Disturbance events impact life strategy diversity in communities and life strategy evolution in populations. In the field of Ecology, disturbance occurrence is studied while an environmental factor that alters resource availability and populations abundance, causing competitive exclusion of less favorable life strategies depending on disturbance frequency and intensity. In the field of Evolutionary Biology, disturbance is evaluated as a pressure, depending on its spatial and temporal regularity, that determines the intensity of species\' evolutionary response and, as a consequence, the adaptation towards the fittest life strategy. Although there is a separation between these two fields of knowledge, ecological dynamics influence evolutionary dynamics and the other way around. Studies that mix Ecology and Evolution are becoming more common, but few or none of them takes disturbance in consideration. Here, we use an individual-based model to generate contexts in which adaptation and competitive exclusion might act apart and together in order to understand how disturbance determines life strategies that occur in communities under ecological, evolutionary and eco-evolutionary dynamics. In the model, life strategy is an inheritable character defined for a trade-off between longevity and fecundity. Simulations from the evolutionary context were composed by one population under mutation acting, simulations from the ecological context by various species without mutation and simulations from the eco-evolutionary context by various species with mutation occurrence. We observed that disturbance was positively correlated with fecund individuals preponderance in all contexts but that disturbance effect on life strategy diversity varied between the different contexts. In the evolutionary and the eco-evolutionary contexts, life strategy diversity increased with disturbance raise, while in the ecological context diversity decreased. This result evidences the mutation role as a source of new life strategy variants when there is a high renovation of individuals given by raised mortality. Only in the eco-evolutionary context there was an interspecific heterogeneity peak on intermediate levels of disturbance. In this scenario, species reproductive isolation, in contrast to populations panmixy, allows species to differ in relation to its life strategies. In parallel to this, the constant input of new life strategy variants by mutation prevents the definite extinction of life strategies from the system. Therefore, when disturbance level is intermediate, productive species as well as long-lived ones are able to coexist. Given that the different contexts resulted in varied patterns of life strategies\' relative frequency, this study evidences the importance of studying disturbance effect on communities structure and dynamics unifying processes that are typically separated between Ecology and Evolution fields

A ocorrência de distúrbios impacta a diversidade de estratégias de vida em comunidades e a evolução de estratégias de vida em populações. Na Ecologia, o distúrbio é estudado enquanto fator ambiental que altera a disponibilidade de recursos e a abundância das populações, ocasionando a exclusão competitiva de espécies menos favorecidas a depender da intensidade e da frequência de sua ocorrência. Na Biologia Evolutiva, o distúrbio é avaliado enquanto pressão que, dependendo de sua regularidade no ambiente, determina a intensidade da resposta evolutiva das espécies e, assim, a adaptação a estratégias de vida de maior aptidão. Ainda que haja separação entre as duas áreas, a dinâmica ecológica influencia a dinâmica evolutiva e vice-versa. Estudos que integram Ecologia e Evolução têm sido cada vez mais recorrentes, no entanto, poucos ou nenhum consideram o distúrbio. Neste trabalho, utilizamos um modelo baseado em indivíduo para criar cenários nos quais os processos de adaptação e exclusão competitiva de espécies possam ocorrer, tanto conjunta quanto isoladamente, a fim de entender como o distúrbio determina as estratégias de vida presentes em comunidades sob dinâmicas ecológicas, evolutivas e eco-evolutivas. No modelo, as estratégias de vida são atributos herdáveis definidos por um trade-off entre longevidade e fecundidade. O cenário evolutivo foi composto por populações (apenas uma espécie) com mutação; o cenário ecológico, por diversas espécies sem mutação e o cenário eco-evolutivo, por diversas espécies com mutação. Observamos que o distúrbio esteve positivamente relacionado com a predominância de indivíduos fecundos em todos os cenários, mas o efeito do distúrbio sobre a diversidade de estratégias variou. Nos cenários evolutivo e eco-evolutivo, a diversidade de estratégias aumentou com a intensificação do distúrbio, enquanto no cenário ecológico a diversidade caiu. Isso indica a importância da mutação como fonte de novas variantes da estratégia quando há alta renovação de indivíduos da comunidade, condicionada pela mortalidade elevada. Apenas no cenário eco-evolutivo houve um pico de heterogeneidade de estratégias em níveis intermediários de distúrbio. Neste cenário, o isolamento reprodutivo das espécies, em contraposição à pan-mixia que ocorre dentro das populações, permite que as espécies difiram em relação à sua estratégia de vida média. Em paralelo, a entrada constante de variantes de estratégias por mutação previne a extinção definitiva das estratégias do sistema. Assim, quando o nível de distúrbio é intermediário, tanto espécies mais fecundas quanto espécies mais longevas conseguem coexistir na comunidade. Dado que os diferentes cenários resultaram em padrões distintos de frequência relativa de estratégias de vida, este trabalho evidencia a importância de estudarmos o efeito do distúrbio na estrutura e na dinâmica de comunidades unindo processos que são tipicamente estudados de forma isolada pela Ecologia e pela Biologia Evolutiva / Disturbance events impact life strategy diversity in communities and life strategy evolution in populations. In the field of Ecology, disturbance occurrence is studied while an environmental factor that alters resource availability and populations abundance, causing competitive exclusion of less favorable life strategies depending on disturbance frequency and intensity. In the field of Evolutionary Biology, disturbance is evaluated as a pressure, depending on its spatial and temporal regularity, that determines the intensity of species\' evolutionary response and, as a consequence, the adaptation towards the fittest life strategy. Although there is a separation between these two fields of knowledge, ecological dynamics influence evolutionary dynamics and the other way around. Studies that mix Ecology and Evolution are becoming more common, but few or none of them takes disturbance in consideration. Here, we use an individual-based model to generate contexts in which adaptation and competitive exclusion might act apart and together in order to understand how disturbance determines life strategies that occur in communities under ecological, evolutionary and eco-evolutionary dynamics. In the model, life strategy is an inheritable character defined for a trade-off between longevity and fecundity. Simulations from the evolutionary context were composed by one population under mutation acting, simulations from the ecological context by various species without mutation and simulations from the eco-evolutionary context by various species with mutation occurrence. We observed that disturbance was positively correlated with fecund individuals preponderance in all contexts but that disturbance effect on life strategy diversity varied between the different contexts. In the evolutionary and the eco-evolutionary contexts, life strategy diversity increased with disturbance raise, while in the ecological context diversity decreased. This result evidences the mutation role as a source of new life strategy variants when there is a high renovation of individuals given by raised mortality. Only in the eco-evolutionary context there was an interspecific heterogeneity peak on intermediate levels of disturbance. In this scenario, species reproductive isolation, in contrast to populations panmixy, allows species to differ in relation to its life strategies. In parallel to this, the constant input of new life strategy variants by mutation prevents the definite extinction of life strategies from the system. Therefore, when disturbance level is intermediate, productive species as well as long-lived ones are able to coexist. Given that the different contexts resulted in varied patterns of life strategies\' relative frequency, this study evidences the importance of studying disturbance effect on communities structure and dynamics unifying processes that are typically separated between Ecology and Evolution fields

Le déclin actuel de la biodiversité a poussé un grand nombre d’études à s’intéresser aux relations entre la biodiversité et le fonctionnement des écosystèmes. La diversité intraspécifique est une facette centrale de la biodiversité qui permet aux espèces de s’adapter aux variations environnementales et dont l’importance écologique est de plus en plus examinée. A l’aide d’une méta-analyse synthétisant des données sur plusieurs espèces et écosystèmes, nous avons démontré qu’un changement de la diversité intraspécifique peut avoir des conséquences écologiques aussi fortes qu’un changement de la diversité spécifique. Ensuite, au travers d’études empiriques nous avons étudié la diversité fonctionnelle et trophique de populations sauvages de vairons (Phoxinus phoxinus), une espèce de poisson abondante en rivière. Une forte variabilité fonctionnelle et trophique a été montrée entre populations de vairons. Cette diversité était structurée de manière complexe du fait de l’action jointe des conditions environnementales, de facteurs évolutifs et des multiples liens existants entre les traits et la niche trophique. Par ailleurs, nous avons montré expérimentalement que les différences fonctionnelles entre populations de vairons pouvaient engendrer des conséquences écologiques aussi fortes qu’une augmentation de la température. De plus, des populations plus diversifiées génétiquement soutenaient des diversités spécifiques élevées puis augmentaient l’intensité des processus écosystémiques. Ces travaux mettent en évidence que la diversité intraspécifique est une composante essentielle de la biodiversité qui doit être considérée comme un facteur majeur affectant la structure des communautés et le fonctionnement des écosystèmes / Current decline of biodiversity has led to many investigations on the relationship between biodiversity and ecosystem functioning. Diversity within species is a pivotal facet of biodiversity that allows species adapting to environmental conditions, and can have ecological effects. Using a meta-analysis synthetizing data across species and ecosystems, we showed that changes in intraspecific diversity can have ecological consequences as strong as changes in species diversity. Then, we empirically studied the functional and trophic diversity among wild populations of European minnows (Phoxinus phoxinus), a highly abundant freshwater species. We observed a high functional and trophic variability among populations, which was shaped by environmental conditions, evolutionary factors, and covariations linking traits and trophic niche among each other. Additionally, we experimentally demonstrated that functional variation among populations led to ecological consequences as strong as those of warming ecosystem. Finally, we uncovered that high genetic diversity within population sustained higher species diversity and higher rates of ecosystem processes. Overall, this thesis reveals that intraspecific diversity is a fundamental component of biodiversity that should be considered as a strong driver of community and ecosystem dynamics.

Les pressions anthropiques agissent en synergie du gène à l’écosystème, des régions polaires aux régions tropicales et induisent une érosion biologique telle, qu'elle est qualifiée de sixième extinction de masse.Ce constat pose la question de l’évolution de la biodiversité face aux changements environnementaux en cours. C’est dans ce contexte que s’inscrit ce travail de thèse focalisé sur les populations de surface (5 -60 m) de corail rouge, Corallium rubrum (Linnaeus, 1758), en Méditerranée Nord-Occidentale.Le corail rouge (Octocorallia, Coralliidae) est une espèce sessile à phase larvaire caractérisée par sa longévité importante, sa dynamique de population lente et son rôle structurant au sein des communautés de substrats durs de Méditerranée. Soumis à une importante pression de récolte, Corallium rubruma récemment subi deux évènements de mortalité massive concordant avec des anomalies thermiques positives, potentiellement liées au réchauffement climatique. La combinaison de ces deux pressions environnementales est donc susceptible d’affecter fortement l’évolution des populations de surface de cette espèce. Basé sur une approche intégrant génétique des populations et écologie de terrain, ce travail a pour objectif principal d’étudier les processus microévolutifs en jeu chez le corail rouge, de l’inter- à l’intra- populationel, afin de contribuer à l'amélioration de nos connaissances sur la biologie de cette espèce dans le contexte environnemental actuel.Ce travail de thèse élargit le champ de connaissances relatif à l’écologie de Corallium rubrum et fournit un ensemble d’outils et de données pour sa conservation face aux perturbations environnementales en cours.Il contribue, par la même occasion, à une avancée significative dans la compréhension de la biologie des organismes marins sessiles à phase larvaire.Il illustre notamment la pertinence d'approches mises place à des échelles spatiales réduites, pour répondre à des questions fondamentales sur l’évolution de ces organismes, souvent clés au sein de communautés sous pression / Anthropic pressures act synergistically from gene to ecosystems and from polar to tropical regions, inducing a strong biological loss, which is considered by many as the sixth mass extinction. The evolution of biodiversity facing the ongoing global change is thus an open question.The present study is focused on the shallow populations (5 - 60 m) of Corallium rubrum (Octocorallia,Coralliidae) in the North Western Mediterranean Sea. The red coral is a sessile and long-lived species with a larval phase, a slow population dynamics and an important structuring role in the Mediterranean hard substrates communities. This species faces a strong harvesting pressure, and recently underwent two massmortality events linked to positive thermal anomalies putatively due to ongoing climate change. These two pressures may have deep implications on the evolution of the shallow populations of this species. Using population genetics and field ecology, the main objective of this study was to define microevolutionary processes acting between and within red coral populations, to enhance our knowledge on the biology of this species facing the environmental changes. This work extends our knowledge concerning the ecology of Corallium rubrum, and provides new toolsand data for its conservation in the context of the ongoing global change. Moreover, this work improves our understanding in the biology of sessile marine organisms with a larval phase, illustrating for example the relevance of approaches conducted at fine geographical scales to address questions regarding the evolution of these organisms.