Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2018-04-12T11:15:36Z No. of bitstreams: 2 Dissertação - Cleber Ten Caten - 2018.pdf: 2396519 bytes, checksum: f33219129581f393738906ad6922cecf (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-04-12T11:30:33Z (GMT) No. of bitstreams: 2 Dissertação - Cleber Ten Caten - 2018.pdf: 2396519 bytes, checksum: f33219129581f393738906ad6922cecf (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-04-12T11:30:33Z (GMT). No. of bitstreams: 2 Dissertação - Cleber Ten Caten - 2018.pdf: 2396519 bytes, checksum: f33219129581f393738906ad6922cecf (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2018-03-14 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / Traditionally, ecologists and biogeographers have been interested in ecogeographical patterns with increasing demand over the last years. Bergmann´s and Rapoport´s rules are two of the most debated ecogeographical patterns, which propose increasing in species body size and range size, respectively, with latitudes. However, whether such rules widely apply to reptiles remains unclear. Moreover, there might be uncertainty regarding the method used to obtain species geographical range that might change our perception of such patterns. Here we tested different hypotheses regarding Bergmann’s and Rapoport’s rules using the New World Pit vipers (Viperidae: Crotalinae) as biological model, as well as analyzed the robustness of different methods to obtain species geographical range and evaluated both ecogeographical patterns using different approaches. We gathered occurrence data for the 136 Crotalinae species and generated geographical ranges by building polygons from Alpha Hull method and Ecological Niche Modelling. We assessed both rules using a ‘cross-species’ and an ‘assemblage’ approach. The former considers each species as an independent data, whereas the latter consider each assemblage (i.e. a grid cell) to be an independent data. We used Phylogenetic Least Squares (PGLS) and Generalized Least Squares (GLS) to evaluate the cross-species and the assemblage pattern, respectively. The former considers the phylogenetic independence of the data as the latter the geographic autocorrelation and both provide unbiased coefficients and significance levels. Our results show that Bergmann’s rule did not occur in the cross-species level, whereas it was statistically significant in the assemblage level regardless of the method used to obtain range size. We found support for Rapoport’s rule in the cross-species level regardless of the method used to generate range size. Meanwhile, the assemblage analysis was not robust methodologically, revealing different ecogeographical patterns depending on the method used to generate species geographical range. Our findings point that there are inconsistences between the patterns observed in the cross-species and the assemblage analysis, which could indicate that different processes producing these patterns in the cross-species and assemblage levels. Finally, our results highlight that this sensibility is especially evident in Rapoport’s rule assemblage analysis and that when evaluating this pattern in assemblage level the method that will be used to obtain species geographical range should be carefully chosen. / Tradicionalmente, ecólogos e biogeógrafos tem demonstrado grande interesse por padrões ecogeográficos, especialmente durante as ultimas décadas. As regras de Bergmann e de Rapoport são dois dos padrões ecogeográficos mais debatidos, os quais propõe um aumento no tamanho corporal e na área de distribuição geográfica das espécies, respectivamente, com aumento na latitude. No entanto, ainda é incerto se repteis seguiriam essas regras. Além disso, ainda há duvida se o método utilizado para obter a área de distribuição geográfica das espécies pode afetar nossa percepção de tais padrões. No presente trabalho, testamos diferentes hipóteses em relação as regras de Bergmann e Rapoport usando os viperídeos do Novo Mundo (Viperidae: Crotalinae) como modelo biológico, além de analisarmos a robustez de diferentes métodos de obter a distribuição geográfica das espécies e avaliamos os dois padrões ecogeográficos usando diferentes abordagens. Nós reunimos dados de ocorrência para as 136 espécies de Crotalineos e geramos suas distribuições geográficas construindo polígonos utilizando o método Alpha Hull e através de modelos de nicho ecológico. Nós avaliamos ambas as regras utilizando uma abordagem interespecífica e uma de assembleia. A primeira considera cada espécie como um dado independente enquanto a segunda considera cada assembleia (i.e. célula de grid) como um dado independente. Nós utilizamos Quadrados mínimos generalizados filogenéticos (PGLS) e Quadrados mínimos generalizados (GLS) para avaliar os padrões interespecíficos e de assembleias, respectivamente. O primeiro considera a independência filogenética do dado enquanto o segundo a autocorrelação espacial e ambos fornecem coeficientes e níveis de significância não enviesados. Nossos resultados demonstram que a regra de Bergmann não ocorre a nível interespecífico, ao passo que é estatisticamente significativa em nível de assembleia e independente de como foi obtido a distribuição geográfica das espécies. Nós encontramos suporte para regra de Rapoport a nível interespecífico independente de como geramos a distribuição geográfica das espécies. Por outro lado, a analise de assembleia não foi robusta metodologicamente, revelando diferentes padrões ecogeográficos dependendo do método usado para gerar a distribuição geográfica das espécies. Nossos resultados demonstram inconsistência entre os padrões observados nas análises interespecíficas e de assembleia. Finalmente, nossos resultados ressaltam que essa sensibilidade é especialmente evidente na análise de assembleia da regra de Rapoport e que ao avaliar esse padrão em nível de assembleia o método que será usado para obter a distribuição geográfica das espécies deveria ser escolhido cuidadosamente.

La thématique centrale de cette thèse, située à l’interface entre macroécologie et macroévolution, consiste en l’analyse des dynamiques spatio-temporelles de distribution des ammonites du Pliensbachien inf. de la Téthys de l’Ouest. Deux problématiques sont abordées : comment la similarité compositionnelle des assemblages décroît-elle avec la distance géographique? Quel est le lien entre la taille des aires de répartition des espèces et leur phylogénie, leur durée de vie, et leur position latitudinale?Grâce à une analyse des relations de SDD, nous montrons que plus l’environnement est homogène, plus les ammonites se dispersent librement ; et que leur dispersion sur de grandes distances n’est pas liée à leur morphologie, ce qui suggère qu’elle se faisait essentiellement via une dérive planctonique aux cours des stades juvéniles.Notre étude de l’étendue de la répartition des ammonites révèle que cette caractéristique n’est ni stable, ni propre à un clade. Son héritabilité phylogénétique dépend de la stabilité spatio-temporelle de l’environnement et varie selon l’estimateur de répartition et l’intervalle de temps considérés, et entre des espèces d’un même groupe contemporaines les unes des autres. De plus, l’étendue de répartition n’a pas d’effet sur la durée de vie des espèces. Cette absence de corrélation semble prévaloir chez les ammonoïdes, ce groupe étant connu pour présenter de grandes aires de répartition associées à des courtes durées de vie. Enfin, les étendues latitudinales des ammonites étudiées tendent à être maximales aux basses latitudes et augmentent des moyennes vers les hautes latitudes, conformément à la Règle de Rapoport. Ceci suggère qu’elles étaient vraisemblablement température-dépendantes et qu’un gradient de température des eaux existait au Pliensbachien inf. dans la province NW européenne. / The present work, which lies between macroecology and macroevolution, focuses on the spatio-temporal dynamics of early Pliensbachian ammonites of the western Tethys. Two main questions are addressed: how does inter-assemblage compositional similarity decrease with geographical distance? What is the relationship between species range size and phylogeny, species duration and latitudinal position?Ammonite dispersal dynamics are studied through a multi-scale Similarity Distance Decay analysis. We show that their long-distance dispersal is facilitated when the environment is more homogeneous and that it is not related to their morphology. This suggests that their long-distance dispersal occurred essentially through a passive planktonic drift during the first juvenile stages.We also show that ammonite geographical range size is neither stable, nor clade-specific. Indeed, its phylogenetic heritability depends on the spatio-temporal stability of the environment and it varies according to the range variable chosen, the temporal interval considered, and among contemporary species of the same group. Furthermore, the size of their geographical range has no effect on species duration. In fact, this lack of correlation seems to prevail among ammonoids, as this group is well-known for showing both large geographical distributions and short species durations. Finally, species latitudinal ranges tend to be maximal at low latitudes and to increase from middle to high latitudes during the early Pliensbachian, in accordance with the Rapoport’s rule. This suggests that these ammonites were probably temperature-dependent and that a latitudinal gradient of sea-surface temperature occurred in the NW European province during the early Pliensbachian.