Alvars are globally rare rock barren ecosystems on limestone pavement. This thesis focused on the quantitative classification of vegetation of Manitoba alvars, the relationships between vegetation patterns and environmental factors and the effects of grazing on vegetation. Vegetation plots were completed across twenty sites. Cluster analysis, indicator species analysis and PCA were used to describe eight vegetation types. A RDA revealed moisture regime, soil depth, bare rock cover and disturbance (grazing and browsing) are the most important factors affecting floristic composition. Grazing effects were studied at two sites using paired plots on either side of a fenceline dividing grazed and ungrazed areas. PCA and RDA showed significant difference between vegetation compositions based on grazing. A partitioning of species richness and diversity by introduced and native species revealed that both sites experienced significant replacement by introduced species. Current grazing levels on Manitoba alvars are severely impacting the vegetation of this ecosystem. / October 2016

Competition between species is believed to lead to patterns of either competitive exclusion or limiting similarity within ecological communities; however, to date the amount of support for either as an outcome has been relatively weak. The two classes of null model commonly used to assess co-occurrence and limiting similarity have both been well studied for statistical performance; however, the methods used to evaluate their performance, particularly in terms of type II statistical errors, may have resulted in the underreporting of both patterns in the communities tested. The overall purpose of this study was to evaluate the efficacy of the negative co-occurrence and limiting similarity null models to detect patterns believed to result from competition between species and to develop an improved method for detecting said patterns. The null models were tested using synthetic but biologically realistic presence-absence matrices for both type I and type II error rate estimations. The effectiveness of the null models was evaluated with respect to community dimension (number of species × number of plots), and amount of pattern within the community. A novel method of subsetting species was developed to assess communities for patterns of co-occurrence and limiting similarity and four methods were assessed for their ability to isolate the species contributing signal to the pattern. Both classes of null model provided acceptable type I and type II error rates when matrices of more than 5 species and more than 5 plots were tested. When patterns of negative co-occurrence or limiting similarity were add to all species both null models were able to detect significant pattern (β > 0.95); however, when pattern was added to only a proportion of species the ability of the null models to detect pattern deteriorated rapidly with proportions of 80% or less. The use of species subsetting was able to detect significant pattern of both co-occurrence and limiting similarity when fewer than 80% of species were contributing signal but was dependent on the metric used for the limiting similarity null model. The ability of frequent pattern mining to isolate the species contributing signal shows promise; however, a more thorough evaluation is required in order to confirm or deny its utility.

L’étude de l’écologie des communautés et de la ségrégation écologique revêt une importance particulière, notamment pour ses apports en biologie évolutive mais aussi pour ses applications dans le domaine de la conservation. La présente étude s’intéresse à la ségrégation écologique des delphinidés de l’île de Mayotte (Canal de Mozambique, sud-ouest de l’océan Indien) à deux échelles : la communauté d’espèces (approche interspécifique) et les communautés d’individus (approche intra-spécifique). Autour de cette île, une communauté diversifiée de delphinidés se partage l’espace et les ressources, et ce à de très faibles échelles spatiales. Sur treize espèces observées, au moins cinq sont observées régulièrement et ont donc été étudiées : le grand dauphin de l’Indo-Pacifique (Tursiops aduncus), le dauphin à long bec (Stenella longirostris), le dauphin tacheté pantropical (Stenella attenuata), le péponocéphale (Peponocephala electra) et le dauphin de Fraser (Lagenodelphis hosei). Chez les espèces vivant en sympatrie, des mécanismes de ségrégation devraient s’observer selon les trois dimensions principales de la niche écologique : l’espace, la ressource et le temps. A l’échelle interspécifique, les analyses de l’habitat défini par les caractères physiographiques associés à chaque observation, de l’utilisation des ressources exprimée par les isotopes stables du carbone et de l’azote et des budgets d’activités montrent que les delphinidés occupent des niches écologiques distinctes. Parallèlement, certaines espèces jumelles peuvent constituer des associations poly-spécifiques, les mettant apparemment en situation de compétition pour les ressources et les habitats. Il a été montré que ces associations, notamment chez les delphinidés du genre Stenella, n’avaient pas de signification trophique, mais constituaient plutôt une stratégie de vigilance contre les prédateurs. Le dernier volet de l’étude s’intéresse à la ségrégation intra-spécifique et à la structure de population à fine échelle, notamment chez le grand dauphin de l’Indo-Pacifique, la principale espèce de delphinidé à vivre dans les eaux intérieures du lagon. La combinaison d’approches dont les échelles sont emboitées : de l’échelle évolutive populationnelle (structure génétique) à l’échelle de la vie de l’individu (domaine vital), a permis de démontrer que l’unique groupe panmictique de Mayotte se segmentait en communautés (définies par des ensembles d’individus ayant un domaine vital commun). Au moins deux communautés ont été identifiées, utilisant des domaines distincts, formant des groupes sociaux stables mais non constitués d’individus apparentés. L’ensemble de l’étude montre que la ségrégation écologique s’observe aux échelles inter- et intra-spécifiques chez les delphinidés, et que celle-ci ne peut être mis en évidence que par des approches multi-échelles et transdisciplinaires. / The study of communities and ecological segregation is of primary importance, especially for its contribution to the field of evolutionary biology and its applications to conservation. The present study aims to assess ecological segregation of tropical delphinids of the island of Mayotte (Mozambique Channel, SW Indian Ocean) at two scales: interspecies and intra-species level. Around Mayotte, a diverse community of delphinids can be found within a restricted range. Thirteen species have been recorded, among which five species regularly occur: the Indo-Pacific bottlenose dolphin (Tursiops aduncus), the spinner dolphin (Stenella longirostris), the pantropical spotted dolphin (Stenella attenuata), the melon-headed whale (Peponocephala electra) and the Fraser’s dolphin (Lagenodelphis hosei). In species living in sympatry, segregation mechanisms should be observed along the three main dimensions of the ecological niche: habitat, resource and time. In inter-specific comparisons, analyses of habitat defined by physiographic variables at sighting locations, resource use expressed as carbon and nitrogen stable isotope signatures, and activity budgets revealed that each species occupies its own ecological niche. At the same time, sibling species form mixed-species associations that place them in situation of apparent competition for habitats. It has been shown that these mixed-species associations, especially among dolphins of the genus Stenella, do associate for anti-predator vigilance rather than for trophic benefits. Finally, the last approach of this study aims to investigate fine scale population structure and intra-species segregation in the Indo-Pacific bottlenose dolphin, the main delphinid occurring in the inner lagoon. The combination of approaches over temporal scales: from the genetic population structure to the community characteristics at individual life-scale (home range), showed that a single panmictic group of Indo-Pacific bottlenose dolphins was segmented into communities (defined by a common home range). At least two bottlenose dolphin communities were found, using a common home range and within which genetically unrelated individuals are preferentially associated. The whole study underlines that ecological segregation can be observed inter- and intra-specifically in delphinids, and this segregation can only be assessed by using multi-scale and multidisciplinary approaches.