Bien que le feu soit reconnu comme un facteur d’influence dans la dynamique de végétation des savanes, son rôle n’est pas clairement défini. Cette thèse aborde le problème de l’étude de la relation entre l’usage des feux et la dynamique de végétation. L’approche choisie repose sur l’analyse de séries temporelles d’images de télédétection à moyenne résolution spatiale. Les savanes étudiées sont situées sur le bassin versant de Marovoay au nord-ouest de Madagascar. Dans la mesure où il n’existe pas de consensus quant aux méthodes à utiliser, les savanes de Madagascar offrent un contexte particulier, en raison de la dégradation très prononcée du couvert végétal et des changements recherchés, pour tester les méthodes existantes et en proposer des nouvelles. Le premier objectif de ce travail est d’identifier le régime des feux à travers le suivi des variations spatio-temporelles des surfaces brûlées en milieu de savane. Pour cela, une méthode de cartographie des surfaces brûlées a été développée : elle est basée sur le calcul d’un indicateur annuel indiquant le passage d’un feu pendant la saison sèche et d’un indicateur saisonnier traduisant la période de passage du feu. Cette méthode, appliquée au site d’étude, a permis de produire une série temporelle de données utilisées pour caractériser le régime des feux à partir de deux paramètres, la période d’occurrence et la fréquence de passage du feu. En parallèle, le deuxième objectif consiste à caractériser la dynamique de végétation par l’analyse des variations spatio-temporelles de l’activité végétale. Deux approches de détection des changements, basées sur le traitement de série temporelle de NDVI, ont été testées. La première repose sur l’analyse des variations inter annuelles d’un indicateur phénologique traduisant l’activité végétale pendant la phase de croissance des savanes. La deuxième utilise une technique de décomposition temporelle pour extraire la tendance d’une série de NDVI. Dans les deux cas, les résultats ont permis de caractériser la dynamique de végétation à travers trois classes d’évolution de l’activité végétale (séries progressive, régressive ou stable). Ces résultats ont été évalués par comparaison avec ceux issus de techniques de détection des changements basées sur l’analyse diachronique d’images à haute résolution spatiale. Enfin, dans la dernière étape du travail, nous avons étudié les relations entre les informations relatives aux régimes des feux et à la dynamique de végétation en utilisant des modèles de régression multivariée. L’objectif est d’estimer l’importance et le rôle du feu dans la dynamique de végétation. Les résultats ont amené à trois conclusions : a) Le feu est un facteur de maintien des savanes ; b) Dans les situations où la pression liée aux activités anthropiques est faible, le feu, en particulier par la fréquence de son usage, est un facteur déterminant de la dynamique de végétation ; c) Dans les autres situations, l’interprétation des résultats est complexe et difficile, très certainement en raison de l’interaction de multiples facteurs anthropiques. / Fire is recognized to be an essential factor that explains savanna vegetation dynamics. But its role is not clearly defined. This work investigates the problem of studying the relation between fire usage and vegetation dynamic. This is addressed through the analysis of time series of medium spatial resolution remotely sensed images. We studied the savanna localized on the Marovoay watershed, on the northwest part of Madagascar. As no consensus exists on the adapted methods, the savanna of Madagascar offers a particular context to test existing methods or develop new ones, because of the advanced vegetation cover degradation and the nature of change to be detected. The first objective of this work is to identify fire regime by analyzing the spatio-temporal variations of burned areas in savanna areas. To this end, a burned area mapping method was developed. It is based on the definition of an annual indicator, which indicates the occurrence of a fire during the dry season, and a seasonal indicator, which gives the information on the date of the fire event. Applied to the study site, this method has lead to the production of a time series of data used for the characterization of the fire regime through two parameters, the period of occurrence and the frequency of fire. In parallel, the second objective consists on characterizing vegetation dynamic by monitoring spatiotemporal variations of vegetation activity. Two change detection approaches based on NDVI time series, were tested. The first consists on analyzing the inter annual variations of a phenological indicator related to vegetation activity during the active growth season. The second uses a temporal decomposition technique to extract the trend from an NDVI time series. In both cases, the vegetation dynamic is characterized through three classes linked to the evolution of the vegetation activity (progressive, regressive or stable series). Results were evaluated by a comparison with the results obtained from a diachronic change detection technique based of high spatial resolution images. Finally, in the last part of this work, we investigated the relation between fire regimes and vegetation activity change classes using multivariate regression models. The objective is to analyze to determine the importance and the role of fire into the vegetation dynamic. Results leaded to three conclusions: a) Fire is a factor that maintains savanna; b) In areas where pressure due to anthropogenic activities is low, fire frequency represents a determinant factor to explain vegetation dynamic; c) In others areas, the interpretation of results appears to be complex and difficult, probably because of the high level of interactions between multiple environmental factors.

The objective was to model on a scale spatiotemporal dynamics of rangeland vegetation types characterized by functional when subjected to different disorders, aiming to predict patterns in the vegetation inventory for future analysis of the fractal dimension associated with the model of cellular automata. So, initially evaluated the use of fractal analysis as a tool to determine patterns in a reduced scale of grazing on natural pastures under different managements. The experiment evaluated the effect of grazing and burning disturbances applied in two relief positions in a natural grassland located in Santa Maria-RS. Another area located in Bagé-RS assesses the improvement of natural grassland with the introduction of ryegrass, white clover and birdsfoot trefoil. The diversity of the pasture was derived by Shannon and Equitabilidade indices. The information fractal dimension was calculated by regression between the Shannon index(ε) and ε natural algorithm, and the slope of the fractal dimension information. The results showed that the fractal dimension can be a tool for determining patterns in plant community in small-scale grazing. The limitation of this technique to small scales is the need for a greater number of repetitions. Secondly, we validated a model is spatially explicit empirical predictive vegetation dynamics of a natural grassland subjected to different disorders based on cellular automata using the approach of plant functional types by comparing the model predictions with actual observations of experiments. The simulations were performed in two temporal and spatial scales, small and large. The temporal resolution of monthly simulation was considered when the spatial dimension was 0.0625 square meters, featuring a reduced scale. A wide range was considered when the temporal resolution was annual and the spatial dimension was 306.25 m². The factors were: grazing, absence and presence, he burns the absence and presence and position of relief, slope and lowland. The model consists of a three-dimensional array, the first being characterized by the temporal dimension and the second and third, representing flat two-dimensional space, each cell being a PFT, which changes with each step in time as the composition of cell itself and the eight neighboring cells. The model evaluated proved to be efficient in predicting the dynamics of natural grassland vegetation can be used to simulate different scenarios. However, the limitation of the model is correct in determining the level of disturbance. Finally, we introduced the fractal dimension to the previously mentioned model and simulated the effect of grazing management of the biennial burns or every three years, with each simulation corresponded to 100 iterations with a spatial resolution of 306.25 m² and temporal equal to one year. The level of disturbance adopted to characterize both the grazing and burning was 0.18. In the treatments with burns, the years when it was used the level of disorder was 0.36, corresponding to the sum of grazing disturbance and burning. The treatments were evaluated in relation to two situations of environmental resources (and with great restraint). The evaluation of fractal dimension in spatially explicit model based on cellular automata has proven to be a viable tool to determine patterns and dynamics of plant communities regardless of scale. The model does not adequately simulated the colonization of species, this being a key topic for future studies that use this approach. The use of only two functional types allowed better visualization of results. The determination of the amount of functional types to be used in the simulation will depend on further investigations. / O objetivo deste trabalho foi modelar em uma escala espaço-temporal a dinâmica vegetacional de pastagens naturais caracterizada por tipos funcionais quando submetidas a diferentes distúrbios, visando predizer padrões na vegetação de inventários futuros pela análise da dimensão fractal associada ao modelo de autômatos celulares. Assim, inicialmente avaliou-se o emprego da análise fractal como ferramenta na determinação de padrões em uma escala reduzida de pastejo sobre pastagens naturais submetidas a diferentes manejos. O experimento avaliou o efeito dos distúrbios pastejo e queima, aplicados em duas posições de relevo, em uma pastagem natural localizada em Santa Maria-RS. Outra área avaliada localizou-se em Bagé-RS onde foram introduzidos azevém, trevo branco e cornichão na pastagem natural. A diversidade da pastagem foi obtida pelos índices de Shannon e Equitabilidade. A informação da dimensão fractal foi calculada pela regressão entre o índice de Shannon(ε) e o algoritmo natural de ε, sendo a inclinação da linha a informação da dimensão fractal. Os resultados demonstraram que a dimensão fractal pode ser uma ferramenta na determinação de padrões na comunidade vegetal na escala reduzida de pastejo. A limitação do emprego desta técnica para escalas reduzidas é a necessidade de um número maior de repetições. Num segundo momento, validou-se um modelo espacialmente explícito empírico e preditivo da dinâmica vegetacional de uma pastagem natural submetida a diferentes distúrbios. Esse foi baseado em autômatos celulares, utilizando a abordagem de tipos funcionais de plantas, através da comparação das predições do modelo com observações reais de experimentos. As simulações foram realizadas em duas escalas temporais e espaciais, reduzida e ampla. A resolução temporal da simulação foi mensal quando a dimensão espacial considerada foi de 0,0625 m², caracterizando a escala reduzida. A escala ampla foi considerada quando a resolução temporal foi anual e a dimensão espacial foi de 306,25 m². Os fatores testados foram: pastejo, ausência e presença; queima, ausência e presença e posição de relevo, encosta e baixada. O modelo é formado por uma matriz representando duas dimensões espaciais planas, sendo cada célula um PFT, a qual se altera a cada passo no tempo conforme a composição da própria célula e de oito células vizinhas. O modelo avaliado demonstrou eficiência na predição da dinâmica da vegetação da pastagem natural podendo ser utilizado para simular diferentes cenários. Entretanto, a limitação do modelo está na correta determinação do nível de distúrbio. Por fim, avaliou-se a dimensão fractal ao modelo citado anteriormente e simulou-se o efeito do pastejo no manejo da queima bianual ou a cada três anos, sendo que cada simulação correspondeu a 100 iterações com resolução espacial de 306,25 m² e temporal igual a um ano. O nível de distúrbio adotado para caracterizar tanto o pastejo como a queima foi de 0,18. Nos tratamentos com queima, nos anos em que ela foi utilizada, o nível de distúrbio foi de 0,36, correspondendo a soma dos distúrbios pastejo e queima. Os tratamentos foram avaliados em relação a duas situações de recurso ambiental (ótimo e com restrição). A avaliação da dimensão fractal em modelo espacialmente explícitos baseados em autômatos celulares demonstrou ser uma ferramenta viável na determinação de padrões e da dinâmica da comunidade vegetal independente da escala. O modelo não simulou adequadamente a colonização de espécies, sendo este um tema fundamental para futuros trabalhos que utilizarem esta abordagem. A utilização de apenas dois tipos funcionais permitiu uma melhor visualização dos resultados. A determinação da quantidade de tipos funcionais a serem utilizados na simulação dependerá de futuras investigações.

L’empreinte des activités humaines sur les milieux naturels commence à se faire sentir avec la révolution néolithique, il y a quelques 10.000 ans, aux débuts de l’Holocène. Dans le cadre de l’ANR PaléoSyr/PaléoLib et à partir d’archives sédimentaires prélevées dans différents contextes environnementaux (côtier, lacustre et fluviatile/marais), le but de ce travail est de mettre en évidence l’évolution des changements paléoenvironnementaux pour chaque site et d’en évaluer la part relevant respectivement des fluctuations climatiques et des activités humaines, afin de mieux comprendre la relation Homme- Environnement-Climat à l’Holocène. L’étude de ces différents contextes paléoenvironnementaux permet de contraindre temporellement le développement des premières influences humaines sur leur milieu naturel, et de préciser leur mode d’action, à l’aide de marqueurs biotiques (pollen) et abiotiques du sédiment. Les principales fluctuations environnementales enregistrées soulignent les hétérogénéités régionales de réponse aux forçages climatiques et permettent ainsi d’affiner nos connaissances aux échelles locale et régionale / The first sizeable consequences of human activities on the natural environment are thought to have taken place during the Neolithic revolution, at the beginnings of the Holocene, about 10,000 years ago. The aim of this study is to assess the respective role of climate fluctuations and human activities on the evolution of different environmental contexts (coastal, lake and fluvial / swamp) during this interval using sedimentary and palynological records, and hence improve our understanding of Human-Environment-Climate relationships during the Holocene. The study of these different paleoenvironmental settings through the use of both biotic (pollen) and abiotic (sediment geochemistry) markers allow to constrain the temporal development of the first human influences on the natural environment, and to determine their various modes of action (e.g. pastoralism, cultivation). The sequence of environmental changes reconstructed in this study hence highlights profound regional heterogeneities in the response of the studied areas to climatic and anthropic forcing and thus have important implications for our knowledge of the environment-human relationships at the local and regional scales

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Plant functional types defined from morphological attributes are fundamental to understanding functioning of grasslands communities. Thus, to build a classification based solely on grasses (family of greater biomass contribution), it is necessary for increasing the knowledge about complex ecosystems. This work aims to characterize vegetation dynamics by species and by functional types of grasses of two RS physiographic regions (Depressão Central-Santa Maria and Campanha-Bagé) under different managements. The evaluated treatments were: natural pasture and natural pasture overseeded with cool season species (Bagé) and natural pasture submitted to burning and grazing treatments (Santa Maria). For this, permanent transects were used in order to represent the different communities at paddock level. Vegetation dynamics was evaluated using procedures of BOTANAL method. Tillers of grasses with contributions exceeding 3% of total aboveground biomass were collected for subsequent measurement of the attributes specific leaf area (AFE) and leaves dry matter content (TMS). Burned treatments presented greater contribution of Andropogon lateralis, while in grazed treatments there was a greater species diversity. Introduction of cool season species combined with fertilizer promotes an increase of species characterized by resources´ capture. TMS was more stable for species linked to capturing resources strategy and AFE was more robust for species characterized by resources conservation. Therefore, until a regional data basis were developed, it is recommended to use both attributes for future research. / Tipos funcionais de plantas definidos a partir de atributos morfológicos são fundamentais para o entendimento do funcionamento de comunidades campestres. Desta maneira, a construção de uma tipologia baseada apenas em gramíneas (família de maior contribuição), torna-se necessária para o avanço do conhecimento sobre ecossistemas complexos. Este trabalho tem por objetivo caracterizar duas regiões fisiográficas do RS (Depressão Central-Santa Maria e Campanha-Bagé) sob distintos manejos quanto à dinâmica vegetacional por espécies e por tipos funcionais de gramíneas. Os tratamentos avaliados foram: pastagem natural e pastagem natural com introdução de espécies hibernais (Bagé) e pastagem natural submetida a tratamentos de queima e pastejo (Santa Maria). Para isto, foram dispostas transectas fixas de modo a representar as diferentes comunidades em nível de potreiro. Foi avaliada a dinâmica vegetacional utilizando os procedimentos do método BOTANAL. Foram coletados afilhos das gramíneas com contribuição superior a 3% da biomassa aérea total para posterior medida dos atributos área foliar específica (AFE) e teor de matéria seca da folha (TMS). Os tratamentos queimados tiveram maior contribuição de Andropogon lateralis, enquanto nos tratamentos pastejados houve uma maior diversidade de espécies. A introdução de espécies de estação fria, aliada à adubação, promove um acréscimo das espécies caracterizadas por captura de recursos. O TMS se mostrou mais estável para espécies ligadas à estratégia de captura de recursos e a AFE mais robusta para as espécies caracterizadas pela conservação de recursos. Por tanto, até que se desenvolva uma base regional de dados, recomenda-se que futuras pesquisas continuem a utilizar os dois atributos.