A produção de café é uma atividade importante para a economia do Brasil, maior produtor e também principal exportador. O cafeeiro apresenta ciclo bienal de produção, cuja oscilação é acentuada no Brasil dada às condições climáticas e ao sistema de cultivo predominante a pleno sol. Pesquisas envolvendo avaliação do estado nutricional do cafeeiro em sistemas agroflorestais são raras, o que dificulta as recomendações de adubação para esta condição. Objetivou-se neste estudo, o melhor entendimento da fertilidade do solo e do estado nutricional do cafeeiro em relação aos teores de N e K no microclima gerado pelo sistema agroflorestal e em monocultivo. A pesquisa foi conduzida no período de março de 2006 a maio de 2008, no campo experimental pertencente ao Departamento de Produção Vegetal da Escola Superior de Agricultura Luiz de Queiroz da Universidade de São Paulo (ESALQ/USP) em Piracicaba SP, localizada nas coordenadas geográficas 220 42 20 S, 470 37 22 W e altitude 565 m. O experimento foi composto de seringueiras do clone PB 235, plantada em dezembro de 1991, no espaçamento de 8,0 x 2,5 m e cafeeiros cv. Obatã IAC 1669-20, plantado em janeiro de 2002, no espaçamento de 3,4 x 0,9 m, sob diferentes condições de sombreamento: no sub-bosque e interfaceando as árvores da seringueira e em monocultivo. O delineamento experimental adotado foi o inteiramente casualizado com quatro repetições. O experimento de avaliação da fertilidade do solo constou de seis tratamentos, constituídos pelas radiações sobre o solo: 1) 6,4%; 2) 7%; 3) 35,4%; 4) 47,5%; 5) 60,9%; 6) 64,9%. O experimento de avaliação dos teores foliares de N e K do cafeeiro constou de quatro tratamentos, constituídos pelas radiações sobre os cafeeiros: 1) 35%; 2) 45%; 3) 90%; 4) 100%. O estudo constou, adicionalmente, dos experimentos de deposição e avaliação da velocidade de decomposição da serapilheira, frações da matéria orgânica, crescimento, peso foliar específico, anatomia foliar, maturação dos frutos e produção do cafeeiro e qualidade da bebida do café. As variáveis analisadas no experimento de fertilidade do solo foram; pH, P, K, Ca, Mg, soma de bases, H+Al, CTC, V % e matéria orgânica; a avaliação do estado nutricional constou das análises das variáveis N e K foliar. Os atributos de fertilidade do solo (pH e V %), foram mais elevados nos tratamentos 6,4%, 7% e 35,4%, especialmente nas camadas superficiais (0-2 e 2-7 cm) e o teor de matéria orgânica na camada de 0-2 cm, proporcionados pelas seringueiras. Menores teores de K do solo ocorreram nos tratamentos 7% e 35,4%, devido a absorção pelas seringueiras. A característica de CTC elevada do solo não foi modificada pelos tratamentos. Os teores foliares de N e K do cafeeiro a pleno sol alcançaram valores mais elevados e mais baixos, respectivamente, do que aqueles dos cafeeiros à sombra. Os cafeeiros sob 45% de irradiância mantiveram estado nutricional adequado em relação aos teores foliares de N e K. / Coffee production is an important activity in the economy of Brazil, first world producer and exporter. Coffee trees present biennial production cycles, with oscillation accentuated due to the climatic conditions and the production system in monocrop under full natural radiation. Studies involving the evaluation of nutritional status of coffee in agroforestry systems are scarce, what difficult the fertilization recommendations for this conditions. The objective of the this study was of getting a better understanding of the soil fertility and nutritional status, specially for the N and K content, of coffee trees under agroforestry system and monocrop. The research was conducted from March 2006 to May 2008, at the experimental field of the Plant Production Department of the Escola Superior de Agricultura Luiz de Queiroz of the Universidade de São Paulo (ESALQ/USP), in Piracicaba SP (220 42 20 S, 470 37 22 W) at the altitude of 565 m. The experiment was composed of rubber trees, clone PB 235, planted in December 1991, in the spacing of 8.0 x 2.5 m and coffee trees cv. Obatã IAC 1669-20, planted in January 2002, in the spacing of 3.4 x 0.9 m, under different shading conditions: undertorey or interfacing the rubber trees and in monocrop. The experimental design was completely randomized with four replications. The soil fertility experiment had six treatments, done by the different radiation regimes reaching the soil surface: 1) 6.4%; 2) 7%; 3) 35.4%; 4) 47.5%; 5) 60.9%; 6) 64.9%. The coffee nutritional status, in terms of the N and K leaf content had four treatments, done by the different radiation regime reaching the coffee canopy: 1) 35%; 2) 45%; 3) 90%; 4) 100%. Additionally, the study involved the evaluation of the litter deposition and decomposition rates, the soil organic matter fractions, coffee trees growth, its specific leaf area and anatomy, beans maturation, coffee yield and beverage quality. The variable under analysis in the soil fertility experiment were; pH, P, K, Ca, Mg, bases sum, H+Al, CTC, V % and organic matter; in the nutritional status were the N and K leaf content. The soil fertility attributes of pH and V % were improved in the treatments 6.4%, 7% e 35.4%, specially in the upper soils layers (0-2 e 2-7 cm) and organic matter content in the top soil layer (0-2 cm), due to the effect of the rubber trees. Lower K contents in the soil occurred in the treatments 7% e 35.4%, due to the rubber root uptake. The natural high CTC of the soil was not modified by treatments. N and K leaf content of coffee trees under full natural radiation (monocrop) reached the highest and lowest levels, respectively, in comparison with coffee trees under shade. Coffee at 45% of natural radiation maintained appropriate nutritional status in terms of N and K leaf content.

Pour d'évidentes raisons agronomiques et environnementales, être en mesure de concevoir et mettre en place des systèmes de culture dans lesquels les plantes accèdent aux ressources de manière optimale revêt une importance cruciale pour tous les intervenants impliqués dans la production agricole. Les techniques d'intensification telles que la mise en place de cultures d'inter-rang et l'agro-foresterie visent à accroître la productivité globale des terres tout en assurant la durabilité des agro-écosystèmes, via une optimisation de l'utilisation des ressources environnementales (lumière, eau et nutriments) par les plantes et une préservation des cycles géochimiques. En théorie, les moyens d'atteindre ces objectifs sont nombreux mais en pratique, les interactions souterraines sont complexes et difficiles à mesurer, de sorte que les progrès réalisés dans la conception d'agro-écosystèmes améliorés et durables demeurent modestes. Dans ce contexte, l'objectif de ce travail a été d'évaluer, au travers de mesures racinaires detaillées en rhizotron et au champ, les effets des cultures d'inter-rang sur la croissance des jeunes hévéas. La dynamique des interactions souterraines a été étudiée, tant au niveau de la racine individuelle qu'à celui du système racinaire entier, sur la base, notamment, de descriptions numériques détaillées. Une telle approche a permis de proposer, en outre, une voie novatrice pour l'analyse de la dynamique racinaire à l'echelle du systeme racinaire entier, à savoir l'analyse des trajectoires de croissance. Dans le cas de l'association maïs-hévéa, les expérimentations en rhizotron ont permis de mettre en evidence que les interactions souterraines entre ces deux plantes peuvent induire des modifications de la croissance de leurs racines, à la fois à l'échelle de la racine individuelle et à celle du système racinaire entier. Toutefois, une telle coordination des dynamiques racinaires des plantes associées n'a pas pu être confirmée dans le cas des traitements manioc-hévéa et arachide– hévéa. Les expérimentations au terrain ont fourni, de manière assez prévisible, une image complexe des interactions souterraines entre hévéa et cultures d’inter-rang. Toutefois, un premier résultat obtenu par le biais de la mise en place de 'pièges à racines' dans un traitement niébé-hévéa, a été de montrer que ces deux plantes n'avaient pas un comportement compétitif marqué l'une vis-à-vis de l'autre. Il est également apparu que les hévéas paraissent ‘investir’ dans des racines ‘coûteuses’, car de faible longueur spécifique, probablement pour assurer une certaine durabilité de ces organes, tandis que les cultures d’inter-rang favorisent l'allocation des assimilâts vers des racines de longueur spécifique élevée, de construction moins ‘coûteuses’, probablement en réponse a un impératif de croissance plus rapide (suggéré par les taux d’élongation racinaire mesurés au cours des expérimentations en rhizotron). Enfin, excepté le cas du manioc, l’introduction de cultures d’inter-rang telles que le maïs et l'arachide n'a pas eu d'impact significatif sur le développement des jeunes hévéas, comme en attestent l'évolution de leur circonférence, hauteur et développement foliaire. Ce résultat de terrain est compatible avec les résultats des expérimentations en rhizotron qui n'ont démontré aucun effet inhibiteur des cultures d’inter-rang sur le développement de la partie aérienne des hévéas. Bien que les travaux présentés dans ce rapport, ne permettent pas, à eux seuls de conclure de manière définitive sur la façon dont les espèces cultivées en association peuvent se compléter mutuellement sur le plan fonctionnel, ils apportent des éléments de réponse préliminaires à cette question complexe ainsi que des méthodes permettant de les obtenir. Au total, ce travail représente donc une contribution à la conception des agro-écosystèmes durables qui deviennent de plus en plus indispensables dans le contexte d'une demande mondiale croissante en produits alimentaires et matières premières. En outre, certains des résultats obtenus dans le cadre de cette thèse ouvrent des perspectives pour des recherches plus approfondies, avec une finalité agronomique appliquée / For obvious agronomic and environmental reasons, being able to design and implement agro-ecosystems in which crops have optimal access to resources is of pivotal importance to all stakeholders involved in agricultural production. Intensification techniques such as agro-forestry or the introduction of inter-crops aim to increase land productivity while conserving geochemical cycles, to ensure the sustainability of agro-ecosystems through an optimized use of environmental resources (light, water and nutrients). In theory, there are many ways of achieving such a goal, but in practice, below-ground interactions between plants are complex and difficult to measure, so that progress with the development of sustainable agro-ecosystems has been slow and remains modest. In this context, the objective of this work was to assess the effects of inter-crops on the growth of young rubber trees, based on a detailed analysis of below-ground interactions between the associated plants. The dynamics of below-ground interactions has been studied in rhizoboxes, at both the scale of individual roots and that of the whole root system, using detailed numerical descriptions of root architecture. Such an approach resulted in the design of an innovative method for the analysis of the entire root system dynamics, namely, the analysis of growth trajectories. In the case of the maize-rubber tree association, the experiments in rhizoboxes showed that the below-ground interactions between these two plants can induce changes in root growth, at both the individual root and the whole root system levels. However, such a coordination of rooting patterns could not be confirmed in the case of the cassava-rubber tree and groundnut-rubber tree associations. Not unexpectedly, field experiments provided a rather complex picture of the underground interactions between rubber trees and inter-crops. However, initial results obtained using 'root traps' in a cowpea-rubber tree treatment indicated that these two plants were unlikely to have a marked underground competitive behavior relative to each other. Results of field experiments also indicated that, in general, rubber trees seem to 'invest' in 'expensive roots' of low specific root length, presumably to confer some degree of durability to these organs, while inter-crops favoured the allocation of assimilates to 'cheaper' roots, i.e. roots of much higher specific length, probably in response to a 'fast growth imperative' (an hypothesis supported by the root elongation rate values measured during the rhizobox experiments). Finally, to the possible exception of cassava, inter-crops were found to have no significant impact on the development of young rubber trees, as evidenced by measured changes in tree circumference, height and leaf development. This result is consistent with results from the rhizobox experiments which also showed no inhibitory effect of inter-crops on the above-ground development of rubber trees. Although the work presented in this report does not allow, in itself, to conclude firmly on the issue of the functional complementarity between plants grown in association in inter-cropping systems, they provide preliminary answers to this complex issue together with methods to obtain such answers. Overall, this work therefore represents a contribution to the design of sustainable agro-ecosystems which are becoming increasingly needed in the context of a growing global demand for food and raw materials. In addition, some of the results generated by this work open up prospects for future research for the development of sustainable agro-ecosystems

Dès la conquête de la Cochinchine en 1862, l’Administration coloniale et des particuliers français exploitèrent l’agriculture locale et y développèrent l’économie. Ils tentèrent de faire l’essai et d’introduire diverses cultures, en particulier des arbres à caoutchouc. L’année 1897 marqua le début de l’hévéaculture de Cochinchine, lorsqu’on planta avec succès près de deux mille hévéas brasiliensis. La superficie de l’hévéaculture en Cochinchine se développait prodigieusement, allant de cent hectares à la fin du XIXème siècle à près de cent mille hectares au début des années trente, grâce d’une part à des capitaux provenant de la Métropole et, d’autre part à des mesures d’encouragement du Gouvernement colonial. Les plantations d’hévéa attirèrent les travailleurs locaux, surtout en provenance du Tonkin et de l’Annam, à raison d’une dizaine de mille, parfois une vingtaine de mille par an.Parallèlement à l’extension des superficies plantées, la production du caoutchouc de la colonie s’accrut rapidement, allant d’un peu plus d’une tonne en 1908 à plus de soixante mille tonnes en 1939. Les plantations d’hévéa devinrent l’une des cultures les plus importantes de Cochinchine à l’époque coloniale française. Non seulement elles apportèrent la fortune aux planteurs de la colonie, mais elles assurèrent une partie, et depuis 1938, la totalité des besoins de caoutchouc de l’industrie métropolitaine. Les plantations d’hévéa de Cochinchine représentaient un symbole de la colonisation agricole française, mais aussi hélas l’une des pages noires de l’histoire du colonialisme français au Vietnam par l’exploitation brutale des planteurs envers les travailleurs vietnamiens. / As early as the conquest of Cochinchina in 1862, the colonial administration and French individuals exploited the local agriculture and developed the economy there. They tried to experiment and introduce various crops, especially rubber trees. The year of 1897 marked the beginning of the rubber plantation of Cochinchina, when two thousand rubber trees brasiliensis were successfully planted. The area of rubber tree plantation in Cochinchina grew tremendously, ranging from one hundred hectares at the end of the 19th century to nearly one hundred thousand hectares in the early 1930s, because of, on the one hand, the capital invested from the metropolis, and, on the other hand, the measures of encouragement taken by the colonial Government. The rubber plantations attracted local workers, mainly from Tonkin and Annam, at a rate of about 10.000, sometimes 20.000 persons a year. In parallel with the extension of the area of rubber plantation, the colonial rubber production rapidly increased from just over one tonne in 1908 to more than 60.000 tons in 1939.The rubber tree plantation became one of the most important crops of Cochinchina during the French colonial era. Not only they brought fortune to the planters of the colony, but they secured a part, and since 1938, the whole of the rubber demands of the metropolitan industries. The Cochinchina rubber plantations represented a symbol of French agricultural colonization and, unfortunately, one of the black pages of the history of French colonialism in Vietnam by the brutal exploitation of Vietnamese workers by rubber planters.