We conducted two experiments evaluating perennial ryegrass straw as a forage source for ruminants. Experiment 1 evaluated digestion and physiological variables in steers offered perennial ryegrass straw containing increasing levels of lolitrem B. Sixteen ruminally cannulated Angus X Hereford steers (231 ± 2 kg BW) were blocked by weight and assigned randomly to one of four treatments (TRT). Steers were provided perennial ryegrass straw at 120% of the previous 5-d average intake. Prior to straw feeding, soybean meal (SBM) was provided (0.1% BW; CP basis) to meet the estimated requirement for degradable intake protein. Low (L) and high (H) lolitrem B straws (<100 and 1550 ppb, respectively) were used to formulate TRT diets: LOW (100% L); LOW MIX (67% L:33% H); HIGH MIX (33% L:67% H); HIGH (100% H). Intake and digestibility of DM and OM, and ruminal pH, total VFA, and NH₃-N were not affected by increasing lolitrem B concentration (P>0.13). Ruminal indigestible ADF (IADF) fill increased linearly (P=0.01) and IADF passage rate (%/h) decreased linearly (P=0.04) as lolitrem B level increased. Experiment 2 evaluated performance and production of 72 Angus X Hereford cows (539 ± 5 kg BW) consuming perennial ryegrass straw containing increasing levels of lolitrem B during the last third of gestation. Cows were blocked by body condition score (BCS) and randomly assigned to one of three TRT. Cows were provided perennial ryegrass straw ad libitum and supplemented with SBM (0.1% BW; CP basis) to meet the estimated requirement for degradable intake protein. Mixtures of a L and H lolitrem B straw (467 and 2017 ppb, respectively) were used to formulate TRT diets: LOW (100% L); MIX (50% L:50% H); HIGH (100% H). Thirteen of 24 cows on the HIGH TRT exhibited signs of ryegrass staggers and were removed from the study. Dry matter intake was not affected (P>0.12) by increasing lolitrem B concentration; however, estimated DM digestibility decreased linearly (P<0.01) as lolitrem B concentration increased. Lolitrem B concentration did not influence pre- or post calving weight or BCS change (P>0.10). These data suggest that feeding perennial ryegrass straw containing up to 1550 ppb lolitrem B does not adversely affect nutrient digestion or physiological response variables in steers. However, providing straw with a lolitrem B concentration of approximately 2000 ppb resulted in 54% of cows exhibiting signs of ryegrass staggers. Blending of H and L straws appears to be a successful management practice. Keywords: Alkaloid, Beef Cattle, Endophyte, Lolitrem B, Perennial Ryegrass, Straw / Graduation date: 2004

Graduation date: 1993

This diplom thesis deals with the evaluation of quality machines work for soil cultivation used in minimalization soil cultivation technology and weed control in cereal crops. The machines were evaluated for incorporation of harvest residues, uniformity of depth and size of lumps. It was further performed an economic evaluation of selected cost per operation soil cultivation. In order to monitor the development of weeds was based small-plot experiment, where was monitored representation of individual weed species and the effectiveness of selected herbicide.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / The water evaporation (Es) is one of the main components of water balance and energy of ground surface. Precise measurements of the Es dynamics require the understanding of water and energy partition on the interface soil surface and atmosphere. The Es accumulated is a function of time, soil texture, and presence or absence of crop residues on soil surface, plants and its stage of development, rainfall and irrigation distribution, with strong influence of the atmosphere evaporative demand. This paper aims to quantify the effect of crop residues in no-tillage system on the components of soil water balance on irrigated and non-irrigated areas and evaluate the dynamic of retained water by the straw after the wetting cycles duet to raindall or irrigation events. Two experiments were carried out, on the experimental area of Department of Rural Engineering of Federal University of Santa Maria. The experiment I was carried out from 2013 to 2014, under a mobile reinaout shelter, in 9 m² plots. A bi-factorial experimental design was used, with three replications, where the factor A was the three soil covering levels: 0, 2 e 4 t ha-1 of black oat residues. The factor B was the three irrigation levels. It was used a micro sprinkling system with an application rate of 8 mm h-1 and a service pressure of 100 kPa. After every irrigation event, a sample of 0,09 m² of the crop residues was collected and weighted, in a time interval of 0, 3, 6, 24 hours after the irrigation, in order to measure the residues retained water. The soil water content was monitored in all experimental units, to a depth of 85 cm, using a set of FDR sensors. The soil water balance was determined by the relation between irrigation depth applied, subtracted the infiltrated irrigation detph, the water retained by the residues and the evaporated irrigation depth, after every irrigation event. The experiment II was carried out during 2015/2016 crop season, using microlysimeters (ML), with dimensions of 10 cm diameter and 10 cm height. It was used the same treatment levels of the experiment I, with three replications. A set of ML was installed in a fallow area with natural rainfall and a second set was installed in an area with soybean, both submitted to natural rainfall. The MLs were weighted, daily, at 6 p.m., with the exception of the days with rainfalls. The Es of the ML was determined by the reason of the difference between the mass of ML measured at time intervals of 24h and ML area. The results indicated that the increase on the quantity of residues from 2 to 4 t ha-1 did not improve the water interception and storage by residues. The evaporation daily percentage was more influenced by atmosphere evaporative demand than by the amount of crop residues on soil surface. Crop residues are important to preserve the soil moisture and have direct impact on soil water availability and the soil water balance during the entire crop cycle duration. / A evaporação da água (Es) é um dos principais componentes do balanço de água e energia da superfície do solo. Medidas precisas da dinâmica da Es requerem o entendimento da participação da água e da energia na interface superfície e atmosfera. A Es acumulada é uma função do tempo, da textura do solo, da presença ou não de resíduos vegetais na superfície do solo, das plantas e estádios de desenvolvimento, da ocorrência de chuvas e irrigações, com forte influência da demanda evaporativa da atmosfera. Assim, o objetivo desse trabalho foi quantificar o efeito do resíduo vegetal do sistema plantio direto nos componentes do balanço hídrico do solo em áreas irrigadas e de sequeiro e avaliar a dinâmica da água retida pela palha após o umedecimento pela chuva ou irrigação. Dois experimentos foram conduzidos, em área experimental do Departamento de Engenharia Rural, da Universidade Federal de Santa Maria. O experimento I foi conduzido entre os anos de 2013 e 2014, sob uma cobertura móvel, em parcelas de 9 m2. Utilizou-se um delineamento bi-fatorial, com três repetições, onde o fator A foi constituído de três níveis de cobertura do solo: 0, 2 e 4 t ha-1 de resíduos de aveia preta. O fator B foi constituído de três lâminas de irrigação. Utilizou-se um sistema de micro aspersão, com taxa de aplicação de 8 mm h-1 e pressão de serviço de 100 kPa. Após cada evento de irrigação, uma amostra de 0,09 m2 de resíduos vegetais era coletada e pesada, em intervalos de 0, 3, 6 e 24 horas após a irrigação, à fim de medir a água retida pelos resíduos. O conteúdo de água no solo foi monitorado em cada unidade experimental, até a profundidade de 85 cm, utilizando-se um conjunto de sensores FDR. O balanço hídrico do solo foi determinado pela relação entre a lâmina aplicada, subtraída da lâmina infiltrada, a água retida pelo resíduo vegetal e da lâmina evaporada, após cada evento de irrigação. O experimento II foi conduzindo no ano agrícola de 2015/2016, utilizando microlisimetros (ML), com dimensões de 10 cm de diâmetro e 10 cm de altura. Utilizou-se os mesmos níveis de cobertura do experimento I, com três repetições. Um conjunto de ML foi instalado em uma área em pousio com chuva natural e um segundo conjunto foi instalado em uma área com soja, também com chuva natural. Os ML foram pesados em balança de precisão, diariamente, às 18 horas, com exceção dos dias onde se registou chuva. A Es da água nos ML foi determinada pela razão entre a diferença entre a massa dos mesmos num intervalos de 24 horas e a área do ML. Resultados indicaram que o aumento na quantidade de resíduos de 2 para 4 t ha-1 não incrementou a interceptação e armazenamento de água pelos resíduos vegetais. A taxa diária de evaporação foi mais influenciada pela demanda evaporativa da atmosfera do que pela presença de resíduos culturais na superfície do solo, que está relacionado com a quantidade de resíduo. Os resíduos vegetais tem importância quanto ao efeito na disponibilidade de água no solo em todo o ciclo vegetativo.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The use of alternative crops to rice (Oryza sativa L.), such as soybean (Glycine Maxl.) and sorghum (Sorghum bicolor L.) in lowland areas changes the chemical characteristics of the straw returned to soil. Straw quality and management type are among the main factors that control decomposition rates and straw-N transformations in soil. The objectives of this study were: 1) to compare N dynamics from rice straw decomposition incorporated in soil or on soil surface for two lowland soils; and 2) to assess the effects of straw quality (comparison of rice, soybean and sorghum straw decomposition) and of soil texture on N dynamics during straw decomposition on soil surface. A 180-days greenhouse incubation was carried out at the Federal University of Santa Maria, using rice (Oryza sativa L.), soybean (Glycine max L.) and sorghum (Sorghum bicolor L.) 15N-enriched straws on two lowland soils, one PLANOSOIL (PlaAr) and one Albaqualf Eutrophic gleysoil (PlaGl). Rice straw decomposition was followed both on soil surface and incorporated in soil, while sorghum and soybean straw decomposition was only studied on soil surface. Remaining 15N-straw, 15N-total, soil mineral 15N and soil mineral N were measured after 30, 90 and 180 days of incubation. Rice straw incorporation compared to its maintenance on soil surface increased the release of N from the straw and the amount of soil organic N , yet did not affect the amount of N recovered (straw + soil). For both management modes, soil type did not influence the release of N and C in the rice straw recovered. Regardless of the management mode, soil type only affected the amount of organic N in the soil, with higher value in PlaGl soil than PlaAr soil. Grasses straw (rice and sorghum), with lower N content (higher C: N), led to lower amount of N released, but provided similar amount of organic N in both soils compared to legume straw (soybean). After 180 days, the amount of N recovered (straw + soil) was affected by the interactions between straw and soil types, with higher value in PlaGl soil with grasses straw than with legume straw. For PlaAr soil, the amount of N recovered did not differ between the three straws. / O uso de culturas alternativas ao arroz (Oryza sativa L.), como a soja (Glycine max L.) e o sorgo (Sorghum bicolor L.), em áreas de várzea resultará na adição ao solo de diferentes palhas com características químicas distintas. A qualidade das palhas e o tipo de manejo adotado estão entre os principais fatores que controlam as taxas de decomposição e as transformações do N das palhas no solo. Os objetivos desses estudos foram: 1) avaliar a dinâmica do N durante a decomposição da palha de arroz incorporada ou na superfície de dois solos de várzea; e 2) avaliar o efeito da qualidade da palha de arroz, soja e sorgo e da textura do solo sobre a dinâmica do N durante a decomposição das palhas na superfície do solo. Foi realizada uma incubação com duração de 180 dias em casa de vegetação na Universidade Federal de Santa Maria, utilizando palha de arroz (Oryza sativa L.), soja (Glycine max L.) e sorgo (Sorghum bicolor L.) enriquecidas com 15N em dois solos de várzea, um Planossolo Hidromórfico Eutrófico arênico (PlaAr) e um Planossolo Hidromórfico Eutrófico gleissólico (PlaGl). A palha de arroz foi avaliada na superfície e incorporada aos solos e as palhas de sorgo e soja foram avaliadas apenas na superfície dos solos. Aos 30, 90 e 180 dias foram avaliadas as quantidades de 15N remanescente nas palhas, 15N total e 15N mineral no solo e N mineral no solo. A incorporação da palha comparado a sua manutenção na superfície do solo promove maior liberação de N da palha, maior quantidade de N orgânico no solo e quantidade semelhante de N recuperado (palha + solo). Independente da modalidade de uso, o tipo de solo não influenciou na liberação de N e na quantidade de N recuperado da palha de arroz. Independente da modalidade de uso, o tipo de solo afetou apenas quantidade de N orgânico no solo, sendo os maiores valores encontrados no PlaGl comparado ao PlaAr. As palhas de gramíneas (arroz e sorgo) com menor teor de N (maior C:N) apresentaram menor quantidade de N liberada, mas promoveram quantidade semelhante de N orgânico nos dois solos em relação a palha da leguminosa (soja). Após 180 dias, a quantidade de N recuperado (palha + solo) foi afetada pela interação entre palha e solo, sendo os maiores valores recuperados no solo PlaGl com as palhas de gramíneas do que com a leguminosa. No solo PlaAr a quantidade de N recuperada não diferiu entre as três palhas.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Crop residues (CR) composed by shoots and roots of plants are the main source of C for the agricultural soils. During the decomposition of these CR in the soil, part of C is evolved as CO2 and nutrients are mineralized, with emphasis on the N. Among the factors that affect decomposition of CR, the biochemical composition has great importance. The objectives of this study were: a) evaluate the dry matter production (DM) and biochemical composition of CR of shoot and roots from commercial and summer cover crops; and b) evaluate the influence of biochemical composition of the transformations of C and N, when CR are kept on the surface (shoot), or incorporated to soil (roots). It was realized two studies, in the first one, it was quantified the DM and biochemical composition of CR from shoot (leaves and stalks) and roots. In the second study, it was evaluated under laboratory conditions for 120 days, the mineralization of C and N and the results were correlated with the biochemical composition of the CR studied. The DM yield of roots varied from 0.54 to 2.86 Mg ha-1 in commercial crops and from 0.56 to 1.66 Mg ha-1 in cover crops. Rice and millet were the species that had the highest DM production of shoots and roots between the commercial and cover crops, respectively. Most of the DM of roots was found in the topsoil (0-10cm) and near the sowing row. The relation root/shoot average of cover crops was less than commercial crops (0.10 vs. 0.17). The CR show wide variation in their biochemical composition. The N content in the DM of CR ranged from 2.8 g kg- 1 (sunflower stalks) to 46.3 g kg- 1 (velvet leaves). The soluble fraction of Van Soest ranged from 108 g kg-1 (soybean stalks) to 774 g kg-1 DM (bean leaves). While the values of cellulose in DM ranged from 138 g kg-1 (bean leaves) to 612 g kg-1 (sunflower stalks), those of lignin ranged only from 38 g kg-1 in the leaves of Crotalaria juncea to 197 g kg- 1 in roots of soybean. The highest levels of polyphenols were found in the CR from shoot (14.1 g kg-1 DM) and roots (22.2 g kg-1 DM) of velvet beans. The CR with N content less than 15 g kg-1 (C/N > 29) cause net immobilization of N in the soil. All the roots of valued commercial crops caused immobilization of N in the soil, and the predominance of net mineralization of N was observed only with the roots of cover crops legume. In general, the mineralization of C from CR decreased in the following order: leaf > stalk+leaf > stalk. Among the CR from commercial crops, those of maize (leaf+stalk) were which presented least amount of C mineralized (39%). The mineralization of C from roots presents values similar to those observed for CR from shoot. The mineralization of N in the soil was significantly correlated with the total content of N in the DM and in the Van Soest soluble fraction of shoot and roots (P < 0.01 and r2 ranging from 0.90 to 0.98). The values of correlation between biochemical composition of CR and C mineralization were lower than those observed for N mineralization, but there was a greater number of significant correlations, especially for ratio C/N, polyphenols and NDF from Van Soest. / Os resíduos vegetais (RV) compostos pela parte aérea e raízes das plantas constituem a principal fonte de C para os solos agrícolas. Durante a decomposição desses RV no solo, parte do C é evoluído como CO2 e nutrientes são mineralizados, com destaque para o N. Dentre os fatores que afetam a decomposição dos RV, a composição bioquímica exerce grande importância. Os objetivos do presente trabalho foram os seguintes: a) avaliar a produção de matéria seca (MS) e a composição bioquímica dos RV da parte aérea e de raízes de culturas comerciais e de plantas de cobertura de solo de verão; e b) avaliar a influência da composição bioquímica sobre as biotransformações do C e do N, quando os RV são mantidos na superfície (parte aérea) ou incorporados ao solo (raízes). Foram realizados dois estudos, sendo que no primeiro quantificou-se a MS e a composição bioquímica dos RV da parte aérea (folhas e talos) e de raízes e no segundo estudo, avaliou-se em condições de laboratório por 120 dias, a mineralização do C e do N e os resultados obtidos foram correlacionados com a composição bioquímica dos RV estudados. A produção de MS de raízes variou de 0,54 a 2,86 Mg ha-1 nas culturas comerciais e de 0,56 a 1,66 Mg ha-1 nas plantas de cobertura de solo. O arroz e o milheto foram as espécies que apresentaram as maiores produções de MS da parte aérea e de raízes entre as culturas comerciais e as plantas de cobertura, respectivamente. A maior parte da MS de raízes foi encontrada na camada superficial do solo (0-10 cm) e próximo à linha de semeadura. A relação raiz/parte aérea média das plantas de cobertura foi inferior aquela das culturas comerciais (0,10 vs 0,17). Os RV apresentam ampla variação na sua composição bioquímica. O teor de N na MS dos RV variou de 2,8 g kg-1 (talos girassol) a 46,3 g kg-1 (folhas mucuna). A fração solúvel de Van Soest variou de 108 g kg-1 (talos de soja) a 774 g kg-1 MS (folhas de feijão). Enquanto que os valores de celulose na MS variaram de 138 g kg-1 (folhas feijão) a 612 g kg-1 (talos girassol) os de lignina variaram de apenas 38 g kg-1 nas folhas de crotalária juncea a 197 g kg-1 nas raízes de soja. Os maiores teores de polifenóis foram encontradas nos RV da parte aérea (14,1 g kg-1 MS) e raízes (22,2 g kg-1 MS) da mucuna cinza. Os RV com teores de N inferior a 15 g kg-1 (C/N > 29) provocam imobilização líquida de N no solo. Todas as raízes das culturas comerciais avaliadas causaram imobilização de N no solo, sendo que a predominância de mineralização líquida de N somente foi observada com as raízes das plantas de cobertura leguminosas. De maneira geral, a mineralização do C dos RV decresceu na seguinte ordem: folha > talo+folha > talo. Entre os RV das culturas comerciais, aqueles do milho (folha + talo) foram os que apresentaram menor quantidade de C mineralizado (39%). A mineralização do C das raízes apresenta valores semelhantes àqueles observados para os RV da parte aérea. A mineralização do N no solo foi significativamente correlacionada com os teores de N total na MS e na fração solúvel de Van Soest dos RV da parte aérea e raízes (P < 0.01 e r2 variando de 0,90 a 0,98). Os valores de correlação entre a composição bioquímica dos RV e a mineralização do C foram menores do que aqueles observados para a mineralização do N, porém observou-se maior número de correlações significativas, com destaque para a relação C/N, polifenóis e FDN de Van Soest.

L'objectif de la thèse est de caractériser des populations bactériennes susceptibles de dégrader la lignine des résidus de culture afin d'optimiser la biodisponibilité du compartiment holocellulosique. L'originalité du sujet réside dans la recherche de bactéries ligninolytiques en utilisant des approches combinant à la fois de l'isolement de souches, des techniques de biologie moléculaire et de la mise en œuvre de ces micro-organismes à l'échelle de réacteur pilote, dans le cadre de prétraitements des résidus de culture. Les travaux de recherche ont été réalisés pour mettre en évidence et identifier des micro-organismes potentiellement intéressants et caractériser leurs systèmes enzymatiques susceptibles d'intervenir dans la dégradation de la lignine. Suite à l'isolement de plusieurs souches d'intérêts et à la mise en place de réacteurs, des expérimentations ont été réalisées afin de suivre l'effet des prétraitements biologiques sur la dégradation de la biomasse et notamment par l'étude des transformations biochimiques, de l'accessibilité à la cellulose ainsi que de l'évolution des communautés microbiennes. Les résultats de ces travaux ont montré que la communauté endogène joue un rôle primordial dans la biodégradation de la paille de colza. Bien que l'inoculation de bactéries exogènes ait eu peu d'influence, il est possible d'améliorer l'accessibilité à cellulose par un mélange de lignine et des sels minéraux permettant d'enrichir la communauté endogène active. Cette thèse s'articule autour de plusieurs analyses multi-échelles et complémentaires permettant de développer un prétraitement biologique de résidu de culture, d'améliorer sa mise en œuvre et son utilisation pour la digestion anaérobie par voie sèche ou la production de bioéthanol. / The aim of the PhD project is to characterize bacterial populations potentially capable of degrading lignin within crop residues to optimize the bioavailability of holocellulose compartment to be valorized by anaerobic digestion. The originality of the subject lies in finding ligninolytic bacteria using approaches combining the isolation of strains and technique of molecular biology, and the application of these microorganisms on a reactor scale, as part of pretreatment of crop residues. Firstly, research work was carried out to find and identify potentially interesting micro-organisms and characterize their specific enzyme systems. Secondly, studies have been conducted on the application of micro-organisms isolated for a biological pretreatment of rape straw. Pretreatments were carried out in a series of laboratory reactors dedicated to follow the effect in real time. A method allowing the quantification of the cellulose accessibility for cellulases in substrate was established. In parallel, high-throughput sequencing analysis was performed to monitor and/or characterize the bacterial and archaeal communities in the reactors. The results obtained have shown that the main factor influencing the degradation of the straw was not the presence of ligninolytic strains which could not influence the development of endogenous community, but the supply of some nutrients included in a mixture with lignin and mineral salts which could enrich a particularly active community. The data obtained in the thesis allowed to capitalize knowledge on the biodegradation of lignocellulosic biomass. This prospective study allowed the consideration of using the biological pretreatments developed at the upstream of biogas and biofuel sectors.

Fusarium graminearum est un champignon pathogène des plantes, responsable de la fusariose de l'épi (plus connue sous le nom de Fusarium Head Blight : FHB) sur céréales, notamment sur le blé et le maïs. En interaction avec la plante, le champignon produit des mycotoxines, parmi lesquellse le déoxynivalénol (DON), dont la finalité pour le champignon producteur est méconnue mais qui sont toxiques pour les humains et les animaux. Ainsi la qualité des grains contribue fortement aux pertes de rendement observées et les résidus contaminés restent au champ. Une première revue bibliographique (Leplat et al 2012) a mis en évidence l'importance des résidus de culture (habitat écologique) pour la survie saprophyte du champignon, pour sa reproduction sexuée et pour l'établissement de l'inoculum primaire susceptible d'infecter la prochaine culture. Une seconde revue bibliographique a souligné les lacunes en ce qui concerne le rôle que les mycotoxines pourraient jouer dans la survie de F. graminearum dans un cet habitat. L'objectif principal de cette thèse était donc de vérifier si la présence de mycotoxines dans les résidus de récolte donne un avantage compétitif à F. graminearum vis-à-vis des composantes biotiques du sol et des résidus et notamment les champignons, les bactéries, les protozoaires, les nématodes et les vers de terre. L'impact du DON sur ces différentes communautés a été évalué dans des résidus de maïs et de blé, au champ et en microcosmes, en condition de labour et de travail superficiel du sol. Le développement de la maladie et ses conséquences sur le rendement ont été observés dans l'expérience de terrain à l'Unité Expérimentale de l'INRA de Dijon.Au cours de cette étude, la survie et les dynamiques de développement de la souche modèle d'étude F. graminearum MIAE00376 et des communautés fongiques et bactériennes ont été mesurées en utilisant la réaction de polymérisation en chaîne en temps réel (Q-PCR) ainsi que par comptage sur boîtes. Dans le même temps, l'évolution des structures des communautés microbiennes a été déterminée par analyse du polymorphisme de longueur des fragments de restriction terminaux (T-RFLP). Les nématodes et les vers de terre ont été quantifiés par extraction et observations à l'œil ou a la loupe binoculaire. Le DON introduit dans le sol et les résidus a été extrait et quantifié au cours du temps par chromatographie liquide haute performance (CLHP). Des dynamiques de population de la souche MIAE00376 associée à différents microorganismes isolés de paille en décomposition et sélectionnés pour leur résistance au DON, à des bactéries fixatrice d'azote et à des Fusarium sp. appartenant au complexe fongique du FHB ont été mesurées en microcosmes de paille en présence ou non de DONLes résultats suggèrent que le DON dans les résidus de culture a une incidence sur les composantes biotiques du sol, mais l'impact dépend des communautés et de la localisation des résidus (en surface ou incorporés dans le sol). La biomasse moléculaire montre que les densités bactériennes et fongiques ont été significativement affectées par la présence de DON. La présence de DON a joué un rôle significatif sur la structure des communautés bactériennes et protozoaires, plus faible sur les communautés fongiques et nul sur les nématodes voire positif sur les vers de terre.Il est conclu que le DON est rapidement inaccessible en profondeur et un peu moins rapidement en surface (immobilisation ou dégradation), qu'il ne confère pas d'avantage compétitif au champignon producteur et que la gestion de l'habitat privilégié que constituent les résidus de culture pour F. graminearum peut être envisagée par le travail du sol en favorisant la décomposition rapide des résidus, par le labour ou l'utilisation d'organismes décomposeurs indigènes ou introduits. / Fusarium graminearum is a plant pathogenic fungus, causing devastating disease “Fusarium head blight” (FHB) in cereals including wheat and maize. It also contaminates the grains with mycotoxins including deoxynivalenol (DON) which are toxic to human and animals. This disease has resulted in the serious losses in grain yield and quality. We established through a first bibliographic review that during off season fungus survives saprophytically on the crop residues (ecological habitat) and serves as primary inoculum for the next season crop. However, we noticed also that the literature was poor about the role mycotoxins could play in the establishment of F. graminearum in such a habitat. The main aim of this thesis was therefore to test whether the presence of mycotoxins in the crop residues gives an advantage to F. graminearum to survive and develop a primary inoculum in the presence of the whole soil biota including fungi, bacteria, protozoa, nematodes and earthworms. We studied the impact of DON on the soil communities in the field as well as in microcosms, in wheat as well as in maize residues under tillage and no-tillage conditions. The disease development and the yield were noted in the field experiment. Some DON resistant active fungal decomposers and nitrogen fixing bacteria were picked and the dynamics of F. graminearum was observed by accelerating decomposition of crop residue in their presence, in the presence or absence of DON.During this study, the dynamic and survival of F. graminearum and total fungal and bacterial communities were examined by using quantitative real time polymerase chain reaction (qPCR) as well as by plate counting. At the same time, the structures of microbial communities were determined by using terminal restriction fragment length polymorphism analysis (T-RFLP). The DON resistance of isolated fungal decomposers and nitrogen fixers was tested by using minimal inhibitory concentration test (MIC). Nematodes and earthworms were quantified through binocular observations. The fate of DON was determined by quantifying the mycotoxin by high performance liquid chromatography (HPLC).The results suggested that DON in crop residues showed an impact on the biotic components of the soil but the impact depended on the communities and on the location of the residues (on surface or incorporated in the soil). The molecular biomass shows that the fungal and bacterial densities were significantly affected by the presence of DON. The presence of DON played significant role on the structure of bacterial and protozoan community while the nematodes and fungal communities remained unaffected. MIC results showed that the susceptibility of some competitive fungal strains towards DON was dependent on the dose of mycotoxin. The earthworms (Lumbricus terrestris) were not affected by the presence of mycotoxin. The degradation of DON in the residues was dependent on the time, the location of residues and the soil biota. The quantification of F. graminearum suggested that the presence of DON gave no advantage in the survival and development of primary inoculum during the decomposition of crop residues in the soil. We conclude that fungal decomposers can be selected on their enzymatic potential towards organic matter more than on the DON resistance to increase the degradation of the straw left at the surface and limit the subsequent development of F. graminearum.

La gestion des déchets constitue un problème majeur au niveau mondial. En agriculture, le retour au sol des résidus de culture est une pratique courante et constitue une opportunité intéressante pour maintenir la fertilité du sol et/ou pour stocker du carbone. La décomposition des matières végétales dans le sol est influencée par plusieurs facteurs (tels que la composition microbienne, la nature et qualité des matières végétales), et a pour acteurs principaux les microorganismes telluriques. Au-delà du retour au sol, d’autres voies de valorisation de la matière végétale non récoltée peuvent être envisagées, telle que sa transformation pour des usages non alimentaires. Dans le domaine du bâtiment, l’intérêt de l’utilisation de matériaux biosourcés (destinés à l’isolation thermique) est croissant et encouragé par les pouvoirs publics dans la construction ou la rénovation. Toutefois, à notre connaissance, la question de la gestion de la fin de vie de ces matériaux après déconstruction des bâtiments n’est pas encore abordée. Afin de mieux comprendre le retour aux sols agricoles de matières végétales de natures diversifiées (paille de blé, paille de colza et tiges de tournesol), nous avons combiné différentes caractérisations biochimiques/physicochimiques tels que le fractionnement biochimique, l’analyse thermogravimétrique et la spectroscopie infra rouge, avec l’étude de la dynamique microbienne (abondance, diversité, fonction), complétées du suivi de minéralisation du carbone et de l’azote durant 3 mois d’incubation, en microcosmes. La minéralisation des résidus de grandes cultures est principalement régie par des microbiodiversités initialement différentes, issues de la prairie permanente ou la grande culture, et dans une moindre mesure par leur qualité biochimique. Quant à la dynamique microbienne, elle est impactée par ces deux types de sols et la nature de apports. Dans le cas de coproduits contrastés de la tige de tournesol (moelle et écorce), la minéralisation du C est principalement dictée par leur qualité initiale. Concernant la mise en oeuvre d’un matériau biosourcé à base de moelle de tige de tournesol, elle semble favoriser sa minéralisation dans le sol. Le retour au sol de ce type de matériau pourrait donc constituer un moyen intéressant de gestion de sa fin de vie. / Waste management is a major problem worldwide. In agriculture, the return of crop residues to the soil is a common practice and represents an interesting opportunity to maintain soil fertility and / or to store carbon. The decomposition of plant materials in soils is influenced by several factors (such as microbial composition, plant material’s nature and quality), and soil microorganisms are its main actors. Moreover, other ways of valorization of non-harvested plant materials are possible, such as their transformation for non-food applications. In the building industry, the interest in using biobased materials (for thermal insulation) is growing and encouraged by the public authorities in construction or renovation projects. However, to our knowledge, the issue of management of these materials end-of-life is not yet addressed, after deconstructing the buildings. In order to better understand different crop residues (wheat straw, rapeseed straw and sunflower stems) decomposition in agricultural soils, we have combined different biochemical / physicochemical characterizations such as biochemical fractionation, thermogravimetric analysis and infrared spectroscopy, with microbial dynamics monitoring (abundance, diversity, function), supplemented by carbon and nitrogen mineralization measures during 3 months incubation in microcosms. Crop residues mineralization is mainly governed by initially different microbiodiversities (derived from permanent grassland or conventional cropping system), and to a lesser extent by their biochemical quality. However, microbial dynamics are influenced by both, types of soils and nature of inputs. As for the sunflower stem coproducts (pith and bark), C mineralization is mainly dictated by their initial quality. Concerning the manufacturing process of a sunflower pith biobased material, it seems to favor its mineralization in soil. The return of this type of material to soil could thus constitute an interesting means of managing its end-of-life.

No description available.