Ecological communities along gradients of environmental stress are thought to be structured by trade-offs between resisting biotic interactions in physically benign habitats and successfully exploiting physically stressful habitats. However, these trade-offs are likely to be affected by the predictability of abiotic stressors, and variation in the strength of biotic interactions. I investigated community assembly and food web interactions in ponds across an unpredictable gradient of water inundation (pond permanence) in Canterbury, New Zealand. Pond community composition and species richness were strongly influenced by pond permanence. However, species in temporary ponds were a nested subset of generalists that were also found in permanent ponds, rather than a unique assemblage of temporary pond specialists. Subsequent experiments indicated predator impact decreased with pond permanence, partially due to the foraging suppression of predatory invertebrates in permanent ponds by fish. Weak predation in permanent ponds combined with unpredictable drying regimes likely selected for generalist traits, and resulted in community assembly being driven by a gradient of drying stress rather than trade-offs between biotic interactions and drying. Furthermore, predator impact increased over time in temporary ponds. In predictable snow-melt ponds in the Colorado Rocky Mountains, seasonal windows of weak predation were exploited by vulnerable species, leading to increased diversity within habitats. However in unpredictable systems like Canterbury, temporal increases in predation risk that depend on drying history are likely to increase variability in the spatial arrangement of suitable habitats for particular species. This should further favour the evolution of generalist traits and reduce the importance of trade-offs between predation and drying in the assembly of communities. Considering the predictability of disturbance regimes and the spatial and temporal variation in biotic interactions will greatly enhance understanding and management of communities in heterogeneous landscapes.

Submitted by Valquíria Barbieri (kikibarbi@hotmail.com) on 2018-04-12T17:22:35Z No. of bitstreams: 1 DISS_2014_Thiago Henrique Lapa Oliveira Souza.pdf: 1606440 bytes, checksum: 145412aa311c87f2f916b4ff86921fda (MD5) / Approved for entry into archive by Jordan (jordanbiblio@gmail.com) on 2018-04-26T17:57:40Z (GMT) No. of bitstreams: 1 DISS_2014_Thiago Henrique Lapa Oliveira Souza.pdf: 1606440 bytes, checksum: 145412aa311c87f2f916b4ff86921fda (MD5) / Made available in DSpace on 2018-04-26T17:57:40Z (GMT). No. of bitstreams: 1 DISS_2014_Thiago Henrique Lapa Oliveira Souza.pdf: 1606440 bytes, checksum: 145412aa311c87f2f916b4ff86921fda (MD5) Previous issue date: 2014-05-02 / CAPES / O presente trabalho teve como objetivo avaliar se a variação na composição da comunidade de peixes em riachos é mais bem explicada por variáveis ambientais, por filtros espaciais, ou por ambos. Assim como verificar a relação da diversidade beta com a heterogeneidade ambiental e extensão espacial e a correlação da substituição de espécies com distâncias ambientais e espaciais. Espera-se que a variação na composição de espécies de peixes seja mais bem explicada por fatores ambientais dentro de cada sub-bacia (escala local) e que filtros espaciais ganhem maior importância na explicação da variação na comunidade de peixes em escala regional. Considerando a dissimilaridade na composição de espécies, espera-se que riachos mais distantes apresentem maior dissimilaridade na composição de espécies, enquanto que riachos ambientalmente mais similares possuam composição de espécies também mais similares. Em relação à diversidade beta, espera-se que seja maior em microbacias com maior heterogeneidade ambiental e espacial. Para determinação da estrutura espacial, a partir da distância geográfica, variáveis espaciais foram geradas a partir da Análise de Coordenadas Principais de Matrizes de Vizinhanças. Para determinação das variáveis ambientais e filtros espaciais mais importantes, foi utilizada a Análise de Redundância parcial, seguida da técnica de partição de variâncias para estimar as frações de explicação exclusivas de cada conjunto de variáveis, espaciais e ambientais, mais significantes. A diversidade beta, a heterogeneidade ambiental e a extensão espacial foram calculadas por meio da análise de homogeneidade multivariada de dispersão entre grupos. Para verificar a dependência espacial e ambiental da substituição de espécies, o teste de Mantel parcial foi utilizado. Foi estudado um total de 59 riachos e 15.101 indivíduos pertencentes a 93 espécies. Para toda bacia do Alto Rio Paraná, a variação na comunidade de peixes foi explicada por variáveis ambientais (adjR2=0.08) e por filtros espaciais de escala mais ampla (adjR2=0.08). A substituição de espécies foi relacionada à dissimilaridade ambiental e a distância geográfica para maior escala entre todas as sub-bacias, porém essa a relação somente com a distância geográfica foi duas vezes mais importante para toda bacia do Alto rio Paraná. A maioria das sub-bacias e microbacias não apresentaram relação entre a dissimilaridade na composição de espécies com distâncias ambientais e geográficas. Além disso, a diversidade beta não foi relacionada à heterogeneidade ambiental e extensão espacial observada nas microbacias. A diversidade beta apresentou relação negativa e a heterogeneidade ambiental apresentou relação positiva com o efeito da distância geográfica sobre a dissimilaridade biológica entre as microbacias. No entanto, o efeito das distâncias geográfica ou ambiental não foi relacionado à extensão espacial de cada microbacia. Os dados estudados nesse trabalho demonstram a importância de padrões espaciais e ambientais sobre a composição e distribuição da ictiofauna de riachos da bacia hidrográfica do Alto Paraná. No entanto, a substituição de espécies de peixes esteve correlacionada mais com a distância geográfica do que ambiental. A heterogeneidade ambiental e a diversidade beta das microbacias foram importantes na formação de padrões na similaridade de espécies em relação à distância geográfica. / We aimed to evaluate whether variation in stream fish community composition is better explained by environmental factors, spatial factors or by both, over local and regional scales. We also analyzed whether fish beta diversity among basins is related to spatial extent and environmental heterogeneity and how species composition dissimilarity relates to geographic and environmental distances. It is hypothesized that the variation in species composition (raw data approach) would be more related to environmental factors within watersheds (at local scale) and that spatial filters would have increased importance when the analysis are run for the whole basin at scale (at regional scale). Considering the dissimilarity in species composition variation (distance data approach) it’s expected that the species dissimilarity would be higher among distant sites than among closer sites, while sites environmentally similar present similar species composition. For the beta diversity we expected the higher values to be found at microbasins with higher environmental heterogeneity and spatial extent. To obtain the spatial variables we used an eigenfunction analysis, called principal coordinates of neighbour matrices (PCNM) using overland distances. Most relevant spatial and environmental factors to explain fish composition variation were determined by the partial redundancy analyses (pRDA), followed by a variation partition method to estimate the exclusive fractions of each set of variables. Fish beta diversity, environmental heterogeneity and spatial extent values were calculated by the homogeneity of multivariate dispersions analysis between groups (PERMDISP). The partial Mantel's test was used to assess spatial and environmental dependency over species dissimilarity. The partition of variances demonstrated that fish community for the whole High Parana River basin was explained both by environmental factors (adjR2=0.08) and by spatial filters (adjR2=0.08). Stream fish species dissimilarity for the whole High Parana River basin was related to environmental distances, although geographic distances influence was two times more important than the effect caused only by the environment distances. Most sub-basins and microbasins biological distances had no relation to geographical or environmental distances. Also, there were no relation between fish beta diversity and environmental heterogeneity with spatial extent. Beta diversity presented a negative relation to the effect size (Mantel r) of environmental distances over composition dissimilarities, but the spatial extent and environmental heterogeneity had no influence over the effect of geographical or environmental distances over composition dissimilarity. Our study pointed out significant effects of the spatial structure (e.g dispersion, spatial distance and environmental control (e.g. species sorting) on stream fish assemblage distribution in Upper Paraná River basin. However, the geographical distance effect was more important to fish assemblage dissimilarities than the environmental distance only. Microbasins environmental heterogeneity and beta diversity were important to fish species similarity patterns in relation to geographical distance. Tough, fish beta diversity was not related to streams spatial extent and environmental heterogeneity.

Este trabalho enfatizou as escalas espaciais de microhabitat e de mesohabitat para investigar como a fauna de macroinvertebrados aquáticos é influenciada pelas variáveis ambientais e espaciais. O primeiro capítulo trás uma visão geral da aplicabilidade das escalas de microhabitat e de mesohabitat ao avaliar a estrutura da fauna e, também, demonstra que a composição do substrato influencia a composição faunística. No segundo capítulo, realizamos uma avaliação da suficiência amostral e comparamos a diversidade alfa e beta entre métodos de coleta e entre diferentes mesohabitats. No terceiro capítulo avaliamos diferentes variáveis hidráulicas e suas relações com a abundância de indivíduos e a riqueza de táxons considerando as escalas espaciais de microhabitat e mesohabitat. Observamos que as condições hidráulicas foram importantes na estruturação da comunidade de macroinvertebrados, influenciando positivamente a abundância de indivíduos e negativamente a riqueza de táxons nos microhabitats. No quarto capítulo nos baseamos na teoria de metacomunidades para compreender a influência das variáveis ambientais e espaciais na estruturação da fauna em uma perspectiva de microescala. Além da esperada importância do ambiente para a microdistribuição dos organismos, identificamos uma forte influência do espaço na estruturação da fauna. Nossos resultados demonstraram que a fauna de macroinvertebrados aquáticos é estruturada pela composição do substrato, condições hidráulicas e fatores espaciais, todas estas condições perceptíveis na escala do microhabitat. Deste modo, a microescala pode ser considerada como uma perspectiva válida para o estudo de ecologia de riachos. / We focused on the microhabitat and the mesohabitat spatial scales to investigate the environmental and spatial features influence on aquatic macroinvertebrate fauna. The first chapter brings an overview about the microhabitat and mesohabitat applicability to evaluate the fauna structure. Likewise, it shows the substrate influence on the fauna composition. In the second chapter, we evaluate the sampling sufficiency and compare the alpha and beta diversity among the sampling methods and mesohabitats. In the third chapter we evaluated the hydraulic features and its relationships with abundance and richness considering the microhabitat and mesohabitat spatial scales. Our results show that the hydraulic conditions have influence on macroinvertebrate community structure, showing a positive relationship with abundance and negative with microhabitat richness. In the fourth chapter, we applied the metacommunity theory in the microhabitat scale to understand the hole of environmental and spatial features on the community structure. Further than the expected environmental influence on organisms microdistribution, we identified a strong spatial influence on the fauna structure. Our results showed that the aquatic macroinvertebrate fauna was structured by the substrate composition, hydraulic conditions and spatial features. All this conditions are noticeable on microhabitat spatial scale. Accordingly, the microhabitat scale can be considered as an effectual perspective to the stream ecology.

Hydroscape structure can play a critical role in animal behavior, abundance, and community structure dynamics. Hydroscape configuration can be dynamic and can change quickly in ephemeral systems. However, ephemeral freshwater wetlands are among the most impacted systems in the world and restoration efforts often rely on incomplete information when establishing management objectives. Further understanding how alterations in hydroscape structure in dynamic systems affect animals is critical for conservation and management success. To determine impacts that changing hydroscape conditions can have on consumers in freshwater wetlands, I examined the effects of a large-scale physical model on fish behavior, abundance, and community structure. The physical model incorporated the restoration of sheetflow, canal-fill treatments, and the removal of a decades-old levee that divided two water management areas in the central Everglades. Small fishes modified directional movement behaviors and speed of movement before and after alterations took place, though behavioral responses varied widely by species. Density and community structure of small fishes did change as a function of canal-fill and levee removal treatments. Behaviors of large fishes were also affected by hydroscape alterations, as well as hydroscape configuration beyond the limited footprint of the physical model. Large fish abundance was altered by hydroscape alteration, particularly among certain species. Composition of the large fish community changed before and after hydroscape alteration, though magnitude of responses were site-specific. Effects of hydroscape structure proximity on trophic dynamics were examined using exclosure cages that excluded large predators but allowed access for small consumers. Exclosures were stratified according to proximity to a deep-water canal. Predator avoidance behaviors in small consumers were limited but present. Differences in behavior between sites may also be caused by differences in structure across sites and limited differences in nutrient quality. Behavioral, population, and community responses to hydroscape alteration can be valuable metrics to assess the success of hydroscape restoration. While results can vary across individuals, species, and sampling sites, effects can still be detected even at the scale of the hydroscape. My research has detailed the potential effects of restoration plans across the greater Everglades and can be extended to other ephemeral wetland restoration programs.

Malgré des ressemblances frappantes dans les processus qui les contrôlent, la diversité génétique et la diversité spécifique ont été majoritairement étudiées de manière disjointe, d'une part par la génétique des populations, d'autre part par l'écologie des communautés. L'objectif de cette thèse est de quantifier les processus qui contrôlent la diversité génétique au sein des espèces et la diversité spécifique au sein des communautés. Mon étude porte sur la métacommunauté de mollusques d'eau douce présente sur l'île de Grande-Terre (800 km²) en Guadeloupe. Celle-ci se compose de 29 espèces qui vivent dans une mosaïque de milieux d'eau douce soumis à une forte instabilité temporelle (à-secs fréquents), supposée induire une dynamique en extinction-colonisation des espèces. Cependant, tout à-sec n'entraîne pas nécessairement une extinction, car certaines espèces ont la capacité de rester en vie ralentie dans le sol (estivation). Je me suis tout d'abord focalisé sur le gastéropode pulmoné Drepanotrema depressissimum en combinant trois approches complémentaires : (1) une étude de la distribution spatiale et temporelle de la diversité génétique, (2) un modèle de dynamique de l'occupation des sites qui prend en compte les formes de résistance (estivation) et (3) une expérience en laboratoire pour évaluer si cette espèce peut utiliser l'autofécondation lors de la recolonisation de sites vides. Ces approches révèlent que les extinctions ne sont pas aussi fréquentes que présumées au départ, et qu'elles ne sont pas liées aux à-secs. Au contraire, les sites instables sont les moins touchés. En effet, D. depressissimum surmonte très efficacement les périodes d'à-sec en estivant, et recourt très peu à l'autofécondation. Les extinctions sont en revanche plus importantes dans les sites stables qui correspondent aussi aux communautés les plus riches en espèces. Enfin, à l'échelle d'étude, la diversité génétique dépend des variations de taille et de connectivité des sites plutôt que de l'âge supposé de la population, suggérant un impact relativement mineur de la dynamique d'extinction-colonisation dans ce système. Dans la dernière partie de ma thèse, je mesure l'impact des processus neutres sur l'assemblage des communautés locales. Pour ce faire, j'analyse la corrélation entre la diversité moléculaire (microsatellites) des deux espèces les plus fréquentes, diversité considérée comme une référence reflétant uniquement les processus neutres d'échantillonnage, avec la diversité spécifique des communautés locales. Cette corrélation se révèle très significativement positive, et résulte de l'action parallèle d'une même variable (la connectivité des sites au réseau hydrographique lors de la saison des pluies) sur la migration et la colonisation à la fois des allèles et des espèces, révélant ainsi un rôle important de processus neutres dans la régulation de la diversité spécifique locale. / Striking similarities underlies those processes involved in the dynamics of genetic diversity and species diversity. However these diversities have been considered separately as part of the agenda of population genetics and community ecology respectively. My work aims at quantifying processes that govern genetic diversity within species and species diversity within communities. To address this question I focus on the freshwater snail metacommunity from Grande-Terre Island (800 km²) in Guadeloupe. Twenty-nine snail species inhabit a network of highly unstable freshwater habitats that frequently dry out. This instability probably triggers extinction-colonization cycles of local populations. However, drying-out periods do not necessarily lead to extinction as some species are able to aestivate. First, I focused on the pulmonate gastropod Drepanotrema depressissimum. I used three complementary approaches: (1) an analysis of the spatial and temporal distribution of genetic diversity, (2) a patch occupancy model that take into account the presence of resistance forms such as aestivating snails and (3) a laboratory experiment in order to infer whether this species can rely on selfing to recolonize empty sites. These analyses reveal that extinction is not as frequent as previously thought and is not positively related to instability. Indeed, unstable sites are less prone to extinction. D. depressissimum overcome efficiently drying-out events aestivating in the ground and do not rely on selfing. Besides, extinction is more frequent in stable environments that encompass more speciose communities. Finally, at the scale of investigation, genetic diversity depends much more on patch size and connectivity than on apparent population age, suggesting that extinction-colonization cycles play a minor role in the species dynamics. In the last chapter, I quantify the impact of neutral processes on community assemblage. To this aim, I analyze the correlation between genetic diversity (microsatellites) of the two most-commonly encountered species – this diversity reflects neutral sampling process, with species diversity of local communities. This correlation is highly positive and relies on the parallel effect of a single site characteristic (connection to the local hydrographic network during the rainy season) on migration and colonization of both alleles and species. This suggests that neutral processes play an important role on the regulation of both genetic diversity and local species occurrence.

La question de la coexistence entre espèces est cruciale dans le contexte des changements globaux, où certaines espèces colonisent de nouveaux écosystèmes, tel le cas des espèces invasives. Dans les habitats discontinus, la coexistence entre espèces invasives et natives gagne à être considérée à la lumière de la théorie des métacommunautés, et en particulier en examinant séparément les processus de dispersion et les interactions locales afin de déterminer leurs effets relatifs. Les approches observationnelles en milieux naturels permettent par ailleurs la prise en compte de la complexité des relations possibles. C’est le travail abordé dans cette thèse à travers l’exemple de l’effet de l’invasion de l’écrevisse de Louisiane sur les amphibiens natifs dans des réseaux de mares. La densité des larves et la distribution des amphibiens ont montré la conccurrence avec l’écrevisse dans les mares. Grâce à la génétique du paysage, la dispersion de l’écrevisse s’est révélée fortement contrainte par le caractère discontinu des habitats, contrairement à celle des amphibiens, bien qu’elle ait été variable parmi les trois espèces étudiées. Les analyses d’isotopes stables ont montré que malgré sa position centrale dans les réseaux trophiques, l’écrevisse n’avait pas de fortes interactions trophiques avec les amphibiens, tout au plus un effet indirect par la modification de l’habitat. Sur la base des mécanismes considérés ici, la coexistence entre les amphibiens natifs et l’écrevisse invasive semble ainsi envisageable. Des suites de ce travail sont envisageables à l’échelle des communautés grâce à l’ADN environnemental. / The question of species coexistence is crucial in the context of global changes, where some species colonize new ecosystems, such as the case of invasive species. In patchy habitats, the coexistence between invasive and native species gains at being considered in the light of the metacommunity theory, and particularly by examining separately the processes of dispersal and local interactions to determine their relative effects. Besides, observational approaches in natural habitats allow to account for the complexity of the possible relationships. It is the work tackled in this thesis through the example of the effect of the red swamp crayfish on native amphibians in pond networks. The density of larvae and the distribution of native amphibians showed the co-occurrence with the crayfish in ponds. Owing to landscape genetics, the dispersal of crayfish was shown to be highly constrained by the patchy nature of habitats, contrary to that of amphibians, although it was variable among the three species considered. The stable isotope analyses showed that, despite its central position in the food webs, the crayfish did not have strong trophic interactions with amphibians, all the more an indirect effect through habitat modification. On the basis of the mechanisms considered here, the coexistence between native amphibians and the invasive crayfish seem possible. Perspectives from this work are envisaged at the community scale using environmental DNA.

Using metacommunity theory, I investigated the mechanisms of meta-assemblage structure and assembly among medium- to large-bodied mammals in North Texas. Mammals were surveyed with camera-traps in thirty property units of the LBJ National Grasslands (LBJNG). In Chapter II the dispersal and environmental-control based processes in community assembly were quantified within a metacommunity context and the best-fit metacommunity structure identified. A hypothesis-driven modelling approach was used in Chapter III to determine if the patterns of species composition and site use could be explained by island biogeography theory (IBT) or the habitat amount hypothesis (HAH). Islands were defined as the LBJNG property unit or the forest patch bounded by the property unit. Forest cover was selected as the focal habitat for the HAH. Seasonal dynamics were explored in both chapters. Metacommunity structure changed with each season, resulting in quasi-nested and both quasi and idealized Gleasonian and Clementsian structures. Results indicated that the anthropogenic development is, overall, not disadvantageous for this assemblage, that community assembly receives equal contributions from spatial and environmental factors, and that the metacommunity appears to operate under the mass effects paradigm. The patterns of species composition and site use were not explained by either IBT or HAH. Likely because this assemblage of generalist, dispersal-capable mammals are utilizing multiple habitat types both in the protected land and in the private land. This research highlights the versatility of these species and the potential value of rural countryside landscapes for wildlife conservation.

With an increasing worldwide infrastructure more habitats are fragmented by roads and buildings, which can cause a reduction in biodiversity up to 75%. One way to counteract this is by predicting the outcome, with the help of theoretical models, before it happens. In this study I used a Bayesian network model on a fragmented landscape, to test how well trophic metacommunities are able to persist habitat loss, when increasing dispersal between the fragments in the landscapes by implementing corridors. By implementing just three corridors, the species with the highest trophic level went extinct at a considerable later stage, and by just implementing 10 corridors, the metapopulation capacity for all species in all trophic levels increased. Similar results were obtained when changing the way the species extinction probabilities react to their resources being extinct, which further strengthen the efficacy of corridors. The results from this study suggests that increasing connectivity between landscape fragments, and therefore promoting dispersal of organisms, would help the conservation of biodiversity.

Metacommunity connectivity, i.e., multi-species dispersal events, is vital to metapopulation persistence in patchy landscapes. Assessments of metacommunity connectivity are not trivial. However, a relationship between trophic rank and the species-area relationship has been found in previous studies, allowing for the use of the predator species-area relationship to act as a surrogate measure of actual metacommunity connectivity of prey species in some systems. For this study, avian species were selected as they are generalist top predators within the study system. Predator species richness within geographically isolated freshwater marshes is influenced by a number of factors. I explore the relative roles of patch area, seasonality, hydroperiod, isolation, and vegetation structure on habitat use in the isolated freshwater marshes embedded within the dry prairie ecosystem of Central Florida. Predator species richness was surveyed in 50 sites for three seasons: fall 2005, winter 2005/06, and spring 2006 and the observed avian assemblage measures were subdivided into foraging guilds for analysis. Wading guild (e.g., egrets, herons, bitterns) species richness was correlated with hydroperiod and vegetation structural variables while perching guild (e.g., blackbirds, sparrows, meadowlarks) species richness was correlated with isolation, hydroperiod, and area annually. Overall predator and all guild species richness measures were also correlated with patch area for all seasons. These results suggest that while a complex mixture of patch area, hydroperiod and isolation influence habitat utilization that varies by season and at the community, guild and individual species level, the underlying predictors that define habitat use in wetlands annually includes hydroperiod, and is not exclusively patch area. Additionally, seasonal differences in predator species richness were found to be significant in some cases indicating that future avian population studies may benefit by sampling outside of the normally studied spring breeding season. Results of this study support the use of predator species richness as a suitable assay of metacommunity connectivity of prey species. Applications and implications of this approach toward future conservation efforts are discussed.

Emergence of the metacommunity concept has explicitly recognized the interplay of local and regional processes. The metacommunity concept has already made a substantial contribution to the better understanding of the community composition and dynamics in a regional context. However, long-term field data for testing of available metacommunity models are still scarce and the extent to which these models apply to the real world remains unknown and some of their assumptions untested. Tests conducted so far have largely sought to fit data on the entire regional set of species to one of several metacommunity models, implicitly assuming that all species, members of the metacommunity, can be modelled in the same manner (using a single model). However, species differ in their habitat use to the extent that such uniform treatment may be inappropriate. Furthermore, in testing the metacommunity models, all metacommunity studies relay on snapshots of species distribution to assess the relative importance of local and regional processes. However, snapshot patterns may be insufficient for producing a reliable picture of metacommunity dynamics and processes shaping it. I hypothesised that the relative importance of local (competition, predation or abiotic filtering or constraints) and regional (interaction of populations with landscape, migration or dispersal) processes may vary with species' traits, including habitat specialization. The perception of importance of local and regional processes in structuring community composition obtained via static approach may vary from that obtained by considering the temporal dynamics of component species. My general approach used a model system comprising natural rock pools microcosms. I have also employed experimental approach in the laboratory. I found that different metacommunity models suit for different groups of habitat specialization. I also found this to be true whether the analyses are based on snapshot data or describing temporal dynamics of species populations. These results suggest that a metacommunity system exhibits an internal differentiation of structuring processes. Specifically, from the metacommunity perspective, the dynamics of habitat specialists are best explained by a combination of species sorting and mass effects models, while that of habitat generalists is best explained by patch dynamics and neutral models. / Thesis / Doctor of Philosophy (PhD)