Les biomes herbacés africains intertropicaux devraient faire face, dans un proche futur, à des changements drastiques. Les modèles dynamiques de végétation (DGVM) ont des difficultés à simuler les limites actuelles de ces biomes, notamment parce qu’ils ne prennent pas en compte la diversité des couverts herbacés en C4. Il est donc nécessaire de caractériser cette diversité floristique et physionomique afin qu’elle puisse être facilement prise en compte dans les DGVMs, et que les comparaisons modèle/données (phytolithes) soient possibles.Dans cet objectif, les traits physionomiques des graminées en C4 dominantes au Sénégal et en Afrique du Sud ont été répertoriés. Quatre groupes physionomiques ont été statistiquement discriminés. Ils varient avec la distribution spatiale des biomes et les précipitations régionales. Deux groupes sont fortement corrélés à l’indice de surface foliaire (LAI) et à la biomasse herbacée. Au Sénégal ces deux groupes sont bien différenciés par l’indice phytolithique Iph qui est un proxy des couverts herbacés intertropicaux. En Afrique du Sud, les phytolithes n’ont pas permis de tracer l’ensemble de la transition savane/steppe. Ces deux groupes physionomiques remplissent les critères nécessaires à la caractérisation de types fonctionnels de plantes (PFT). L’intégration de ces PFTs dans le modèle LPJ-GUESS améliore la simulation des biomes herbacés actuels et permet de proposer des simulations pour l’horizon 2100. Ces simulations montrent que l’augmentation de la durée de la saison sèche et de la concentration en CO2 atmosphérique devraient favoriser l’expansion simultanée des steppes et des savanes fermées aux dépens des savanes ouvertes. / Intertropical african herbaceous biomes are expected to face drastic changes in a near future. However Dynamic Global Vegetation Models (DGVMs) simulate their modern boundaries with poor accuracy, especially at the regional scale. DGVMs fail to consider the diversity of their C4 grass cover. Efforts are thus needed to characterize this floristic and physiognomic diversity in a way that can be used for enhancing DGVMs simulations, and enabling model/data (phytoliths) comparisons. For that purpose, physiognomic traits of dominant C4 grass species settled in Senegal and South Africa were listed. Four grass physiognomic groups were statistically identified. The abundance of four of them significantly varied with biome distributions and regional precipitation. Two grass physiognomic groups were additionally strongly correlated with leaf area index (LAI) and grass biomass. In Senegal, those two groups were also well traced by the Iph phytolith index which is a tropical grass cover proxy. In South Africa the limited set of phytolith data did not allow to observe the full savanna/steppe transition. The two physiognomic groups finally fulfilled the criteria required for creating Plant Functional Types (PFTs). Those new PFTs, parameterized in the LPJ-GUESS DGVM, enhanced the simulation of modern herbaceous biomes distribution in Senegal and South Africa. Simulations were additionally performed for the 2100 horizon. They evidence that the increase of both length of the dry season and atmospheric CO2 concentration should favor the simultaneous spread of steppes and closed savannas at the expense of open savannas.

A central issue in climate change research is to identify what species will be most affected by variations in temperature, precipitation or CO2 and via which underlying mechanisms. Dynamic global vegetation models (DGVMs) have been used to address questions of habitat shifts, extinctions and changes in carbon and nutrient cycling. However, DGVMs have been criticized for assuming full migration and using the most generic of plant functional types (PFTs) to describe vegetation cover. My doctoral research addresses both of these concerns. In the first study, I added two new tropical PFTs to an existing regional model (LPJ-GUESS) to improve vegetation representation in Central America. Although there was an improvement in the representation of some biomes such as the pine-oak forests, LPJ-GUESS was still unable to capture the distribution of arid ecosystems. The model representations of fire, soil, and processes unique to desert vegetation are discussed as possible explanations. The remaining three chapters deal with the assumption of full migration, where plants can arrive at any location regardless of distance or physical barriers. Using LPJ-GUESS, I imposed migration limitations by using fat-tailed seed dispersal kernels. I used three temperate tree species with different life history strategies to test the new dispersal functionality. Simulated migration rates for Acer rubrum (141 m year-1) and Pinus rigida (76 m year-1) correspond well to pollen and genetic reconstructed rates. However, migration rates for Tsuga canadensis (85 m year-1) were considerably slower than historical rates. A sensitivity analysis showed that maturation age is the most important parameter for determining rates of spread, but it is the dispersal kernel which determines if there is any long distance dispersal or not. The final study demonstrates how northerly refugia populations could have impacted landscape recolonization following the retreat of the last glacier. Using three species with known refugia (Acer rubrum, Fagus grandifolia, Picea glauca), colonization rates were faster with a northerly refugia population present. The number of refugia locations also had a positive effect on landscape recolonization rates, which was most pronounced when populations were separated. The results from this thesis illustrate the improvements made in vegetation-climate models, giving us increasing confidence in the quality of future climate change predictions.

A central issue in climate change research is to identify what species will be most affected by variations in temperature, precipitation or CO2 and via which underlying mechanisms. Dynamic global vegetation models (DGVMs) have been used to address questions of habitat shifts, extinctions and changes in carbon and nutrient cycling. However, DGVMs have been criticized for assuming full migration and using the most generic of plant functional types (PFTs) to describe vegetation cover. My doctoral research addresses both of these concerns. In the first study, I added two new tropical PFTs to an existing regional model (LPJ-GUESS) to improve vegetation representation in Central America. Although there was an improvement in the representation of some biomes such as the pine-oak forests, LPJ-GUESS was still unable to capture the distribution of arid ecosystems. The model representations of fire, soil, and processes unique to desert vegetation are discussed as possible explanations. The remaining three chapters deal with the assumption of full migration, where plants can arrive at any location regardless of distance or physical barriers. Using LPJ-GUESS, I imposed migration limitations by using fat-tailed seed dispersal kernels. I used three temperate tree species with different life history strategies to test the new dispersal functionality. Simulated migration rates for Acer rubrum (141 m year-1) and Pinus rigida (76 m year-1) correspond well to pollen and genetic reconstructed rates. However, migration rates for Tsuga canadensis (85 m year-1) were considerably slower than historical rates. A sensitivity analysis showed that maturation age is the most important parameter for determining rates of spread, but it is the dispersal kernel which determines if there is any long distance dispersal or not. The final study demonstrates how northerly refugia populations could have impacted landscape recolonization following the retreat of the last glacier. Using three species with known refugia (Acer rubrum, Fagus grandifolia, Picea glauca), colonization rates were faster with a northerly refugia population present. The number of refugia locations also had a positive effect on landscape recolonization rates, which was most pronounced when populations were separated. The results from this thesis illustrate the improvements made in vegetation-climate models, giving us increasing confidence in the quality of future climate change predictions.

Aim: Identifying how climate change, habitat loss, and corridors interact to influence species survival or extinction is critical to understanding macro-scale biodiversity dynamics under changing environments. In North America, the ice-free corridor was the only major pathway for northward migration by megafaunal species during the last deglaciation. However, the timing and interplay among the late Quaternary megafaunal extinctions, climate change, habitat structure, and the opening and reforestation of the ice-free corridor have been unclear. Location: North America. Time period: 15–10 ka. Major taxa studied: Woolly mammoth (Mammuthus primigenius). Methods: For central North America and the ice-free corridor between 15 and 10 ka, we used a series of models and continental-scale datasets to reconstruct habitat characteristics and assess habitat suitability. The models and datasets include biophysical and statistical niche models Niche Mapper and Maxent, downscaled climate simulations from CCSM3 SynTraCE, LPJ-GUESS simulations of net primary productivity (NPP) and woody cover, and woody cover based upon fossil pollen from Neotoma. Results: The ice-free corridor may have been of limited suitability for traversal by mammoths and other grazers due to persistently low productivity by herbaceous plants and quick reforestation after opening 14 ka. Simultaneously, rapid reforestation and decreased forage productivity may have led to declining habitat suitability in central North America. This was possibly amplified by a positive feedback loop driven by reduced herbivory pressures, as mammoth population decline led to the further loss of open habitat. Main conclusions: Declining habitat availability south of the Laurentide Ice Sheet and limited habitat availability in the ice-free corridor were contributing factors in North American extinctions of woolly mammoths and other large grazers that likely operated synergistically with anthropogenic pressures. The role of habitat loss and attenuated corridor suitability for the woolly mammoth extinction reinforce the critical importance of protected habitat connectivity during changing climates, particularly for large vertebrates.