<p>The aim of the research was to identify what female and male farmersthink are the best strategies to enhance the role of conservation farming as a buffer against social, economic and environmental hazards, and a means of ensuring livelihood sustainability and food security. The study also aimed at coming up with information useful to policy and other decision makers on how to improve adoption of these technologies. The empirical component of the research included a questionnaire survey of one hundred and fifty two(152) selected households in one identified ward in Insiza District, focus group discussions and semi-structured interviews with key informants and in-depth interviews of individual male and female members of a few selected farming households from the sampled population. The desktop portion of the study used secondary data from non-governmental organizations (NGOs), government and other stakeholders involved in conservation farming. Collected data was then disaggregated by gender and analyzed using the Statistical Package for the Social Sciences (SPSS). Perception statements that emerged as significant in chi-square tests of independence were be subjected to factor analysis and weighted factor scores from factor analysis were then used as independent variables in binary logistic regression analysis. The study concluded that both practising and non practising farmers were of the opinion that conservation farming was good though they indicated that information on conservation farming was not readily available The study found out that most farmers agreed on the possible positive effects of CA in addressing livelihood challenges effected by hazards such as HIV and AIDS and environmental hazards such as declining soil fertility but it had is labour intensive therefore is not suitable for people affected and infected by HIV and AIDS...</p>

Unintentional nitrogen (N) loss from agroecosystems produces greenhouse gases, induces eutrophication, and is costly for farmers; therefore, adoption of conservation agricultural management practices, such as no-till and cover cropping, has increased. This study assessed N loss via leaching, NH3 volatilization, N2O emissions, and N retention in plant and soil pools of corn conservation agroecosystems across a year. Three systems were evaluated: 1) an unfertilized organic system with cover crops Vicia villosa, Triticum aestivum, or a mix of the two; 2) an organic system with a Vicia cover crop employing three fertilization schemes (0 N, organic N, or a cover crop N-credit approach); 3) a conventional system with a Triticum cover crop and three fertilization techniques (0 N, urea N, or organic N). During cover crop growth, species affected N leaching but gaseous emissions were low across all treatments. During corn growth, cover crop and fertilizer approach affected N loss. Fertilized treatments had greater N loss than unfertilized treatments, and fertilizer type affected gaseous fluxes temporally and in magnitude. Overall, increased N availability did not always indicate greater N loss or yield, suggesting that N conserving management techniques can be employed in conservation agriculture systems without sacrificing yield.

O ambiente de terras baixas no sul do Brasil é predominantemente explorado com a cultura do arroz irrigado. Baseado em um sistema agrícola pouco diversificado, é comum o uso sucessivo das mesmas áreas ano após ano, com práticas de manejo que envolvem o revolvimento do solo. Isto acarreta em prejuízos tais como a degradação do solo e recursos naturais, além de recorrente baixa rentabilidade da atividade. O cultivo de espécies forrageiras hibernais, associado com o não revolvimento do solo, pode trazer melhorias ao sistema produtivo, seja com seu uso como plantas de cobertura ou com o pastejo animal. O objetivo deste trabalho foi avaliar os benefícios e potencialidades que a utilização destas espécies podem proporcionar e, para tanto, dois estudos principais foram realizados. O primeiro, conduzido no município de Cachoeirinha, RS, avaliou o estado de agregação e o incremento do teor de carbono orgânico do solo, comparando cultivos sucessivos de arroz irrigado em um experimento de longa duração. Os tratamentos foram um sistema manejado na forma convencional, com revolvimento do solo, e outros três sob semeadura direta, com coberturas de: azevém, cornichão e vegetação espontânea. Os resultados indicaram que, na camada de 0 a 5 cm de profundidade, os tratamentos sob semeadura direta apresentaram maiores teores de carbono orgânico e agregação do que o convencional. Já na camada de 5 a 10 cm, para os agregados, o tratamento azevém se destacou em relação aos demais. O segundo estudo foi conduzido no município de Cristal, RS. Foram avaliadas a produção animal e de forragem durante três anos no período hibernal, em diferentes combinações de sistemas integrados de produção agropecuária (SIPA). Os cultivos estivais trataram-se de arroz irrigado, soja, capim sudão e campo de sucessão, enquanto as pastagens hibernais consistiram de azevém e consórcios desta espécie com trevo branco e cornichão. Os resultados confirmaram a adaptabilidade destas espécies forrageiras ao ambiente de terras baixas e demonstraram as potencialidades que a exploração pecuária possui, sendo necessários, porém, novos estudos para o melhor ajuste do manejo das áreas. Por fim, foi realizado um terceiro estudo, comparando a relação entre a massa de forragem e altura de pastos hibernais em SIPA conduzidos em terras altas com SIPA em terras baixas. A relação no ambiente de terras baixas é menor, corroborando a necessidade de maiores estudos neste ambiente, visto que as alturas ótimas preconizadas para o manejo do pasto se originam de estudos conduzidos em terras altas. / The lowland environment in Southern Brazil is predominantly explored with the irrigated rice crop. Based on a very low diversified system, in such environment is very common the successive use of the same areas year after year, with management practices that involve soil tillage. This leads to damages such as the degradation of soil and natural resources, in addition to recurrent profitability problems. The insertion of winter pasture species associated with the no-till, as cover crops or for animal grazing, may bring improvements to the system. The objective of the current work was to evaluate the benefits and potentialities that the utilization of such plants can provide. For this, two main studies were performed. The first, conducted in the Cachoeirinha County (Rio Grande do Sul State – RS), evaluated the soil aggregation status and its organic carbon content, by comparing successive irrigated rice croppings in a long-term experiment. The treatments consisted of one system under conventional soil tillage and three others under no-till, being the treatments the following cover crops: Italian ryegrass, birdsfoot trefoil and spontaneous vegetation. The results indicated that, in 0 to 5 cm soil layer, the treatments under no-till presented higher organic carbon contents and greater aggregation, as compared to the conventional system. In the 5 to 10 cm soil layer, the treatment with Italian ryegrass has its aggregation highlighted among the others. The second study was conducted in the Cristal County (RS). The animal and pasture production was evaluated during three years in the winter season, under different combinations of integrated crop-livestock systems (ICLS). The summer croppings were irrigated rice, soybean, Sudan grass and a succession field (native pasture), while the winter pastures consisted of Italian ryegrass and its mix with white clover and birdsfoot trefoil. The results confirmed the adaptability and potentiality of such pasture species and livestock exploration in the lowland environment. However, it is needed further studies for a better management adjustment of the areas. Finally, a third study was performed, comparing the relationship between forage mass and winter pasture height in ICLS conducted in highland and lowland. It was demonstrated that in lowland such relationship is lower, corroborating the necessity of more studies in these conditions, because the optimum heights preconized for pasture management are originated from studies conducted in highlands.

O ambiente de terras baixas no sul do Brasil é predominantemente explorado com a cultura do arroz irrigado. Baseado em um sistema agrícola pouco diversificado, é comum o uso sucessivo das mesmas áreas ano após ano, com práticas de manejo que envolvem o revolvimento do solo. Isto acarreta em prejuízos tais como a degradação do solo e recursos naturais, além de recorrente baixa rentabilidade da atividade. O cultivo de espécies forrageiras hibernais, associado com o não revolvimento do solo, pode trazer melhorias ao sistema produtivo, seja com seu uso como plantas de cobertura ou com o pastejo animal. O objetivo deste trabalho foi avaliar os benefícios e potencialidades que a utilização destas espécies podem proporcionar e, para tanto, dois estudos principais foram realizados. O primeiro, conduzido no município de Cachoeirinha, RS, avaliou o estado de agregação e o incremento do teor de carbono orgânico do solo, comparando cultivos sucessivos de arroz irrigado em um experimento de longa duração. Os tratamentos foram um sistema manejado na forma convencional, com revolvimento do solo, e outros três sob semeadura direta, com coberturas de: azevém, cornichão e vegetação espontânea. Os resultados indicaram que, na camada de 0 a 5 cm de profundidade, os tratamentos sob semeadura direta apresentaram maiores teores de carbono orgânico e agregação do que o convencional. Já na camada de 5 a 10 cm, para os agregados, o tratamento azevém se destacou em relação aos demais. O segundo estudo foi conduzido no município de Cristal, RS. Foram avaliadas a produção animal e de forragem durante três anos no período hibernal, em diferentes combinações de sistemas integrados de produção agropecuária (SIPA). Os cultivos estivais trataram-se de arroz irrigado, soja, capim sudão e campo de sucessão, enquanto as pastagens hibernais consistiram de azevém e consórcios desta espécie com trevo branco e cornichão. Os resultados confirmaram a adaptabilidade destas espécies forrageiras ao ambiente de terras baixas e demonstraram as potencialidades que a exploração pecuária possui, sendo necessários, porém, novos estudos para o melhor ajuste do manejo das áreas. Por fim, foi realizado um terceiro estudo, comparando a relação entre a massa de forragem e altura de pastos hibernais em SIPA conduzidos em terras altas com SIPA em terras baixas. A relação no ambiente de terras baixas é menor, corroborando a necessidade de maiores estudos neste ambiente, visto que as alturas ótimas preconizadas para o manejo do pasto se originam de estudos conduzidos em terras altas. / The lowland environment in Southern Brazil is predominantly explored with the irrigated rice crop. Based on a very low diversified system, in such environment is very common the successive use of the same areas year after year, with management practices that involve soil tillage. This leads to damages such as the degradation of soil and natural resources, in addition to recurrent profitability problems. The insertion of winter pasture species associated with the no-till, as cover crops or for animal grazing, may bring improvements to the system. The objective of the current work was to evaluate the benefits and potentialities that the utilization of such plants can provide. For this, two main studies were performed. The first, conducted in the Cachoeirinha County (Rio Grande do Sul State – RS), evaluated the soil aggregation status and its organic carbon content, by comparing successive irrigated rice croppings in a long-term experiment. The treatments consisted of one system under conventional soil tillage and three others under no-till, being the treatments the following cover crops: Italian ryegrass, birdsfoot trefoil and spontaneous vegetation. The results indicated that, in 0 to 5 cm soil layer, the treatments under no-till presented higher organic carbon contents and greater aggregation, as compared to the conventional system. In the 5 to 10 cm soil layer, the treatment with Italian ryegrass has its aggregation highlighted among the others. The second study was conducted in the Cristal County (RS). The animal and pasture production was evaluated during three years in the winter season, under different combinations of integrated crop-livestock systems (ICLS). The summer croppings were irrigated rice, soybean, Sudan grass and a succession field (native pasture), while the winter pastures consisted of Italian ryegrass and its mix with white clover and birdsfoot trefoil. The results confirmed the adaptability and potentiality of such pasture species and livestock exploration in the lowland environment. However, it is needed further studies for a better management adjustment of the areas. Finally, a third study was performed, comparing the relationship between forage mass and winter pasture height in ICLS conducted in highland and lowland. It was demonstrated that in lowland such relationship is lower, corroborating the necessity of more studies in these conditions, because the optimum heights preconized for pasture management are originated from studies conducted in highlands.

International efforts to promote REDD+ (reducing emissions from deforestation and forest degradation, and the role of conservation, sustainable management of forests and enhancement of forest-carbon stocks) have enjoyed widespread support in climate negotiations. While proponents of this ‘payments for ecosystem services’ approach proclaim win-win benefits, others critique this commodification of forest carbon for contributing to social and environmental injustices that will undermine conservation and development in the longer-term. In this dissertation I respond to these concerns by critically examining how REDD+ initiatives emerge in the context of Lindi Region, Tanzania. I specifically investigate how REDD+ initiatives interact with local livelihood practices, local forest governance and the drivers of land use in order to interrogate the mechanism’s contribution to local development. I conducted ethnographic fieldwork in two villages, both characterised by relatively large forest areas and ‘shifting cultivation’, where different REDD+ projects are underway. In total I stayed in Tanzania for 11 months and applied qualitative and quantitative methods that resulted in 116 recorded interviews, one focus group discussion, innumerable journal entries from ethnographic interviewing and participant observation, 118 household surveys and data from document analysis. Drawing on debates within international development and neoliberalisation of nature I conceptualise REDD+ initiatives as processes promoting ‘inclusive’ neoliberal conservation. In doing so I point at the inherent contradictions of this mechanism that aims to combine a neoliberal conservation logic with inclusive development objectives. I empirically examine local livelihood practices to question popular notions of land use and argue that REDD+ initiatives must grapple with poverty, intra-village inequality and villagers’ dependence on land for crop production to contribute to inclusive economic development. I follow up on this argument by discussing the importance of material and discursive effects of REDD+ initiatives to the livelihoods of poor, middle income and wealthy households and to forest conservation. I then link these effects to an examination of how power and politics shape the implementation of REDD+ initiatives on the ground, specifically discussing the technically complex and politically contested process of territorialisation and the local practices of community-based forest management. I illustrate how seemingly technical REDD+ initiatives are inherently political, which gives them the potential to contribute to local empowerment. At the same time I question naïve assumptions over community conservation and good governance reforms by showing in detail how community-based forest management institutions are practiced on the ground and how this affects benefit distribution within the villages. My last empirical chapter examines how Conservation Agriculture is introduced in the villages as the best way to reconcile agricultural development with forest protection. I specifically discuss the role of social relations in shaping the dissemination and adoption of this new technology in rural Tanzania. Throughout this thesis I argue that local livelihood practices, power struggles and politics over land and people shape how REDD+ initiatives, as inherently contradictory processes of ‘inclusive’ neoliberal conservation, emerge on the ground and I empirically show what this means to different forest stakeholders.

Magister Philosophiae - MPhil / The aim of the research was to identify what female and male farmersthink are the best strategies to enhance the role of conservation farming as a buffer against social, economic and environmental hazards, and a means of ensuring livelihood sustainability and food security. The study also aimed at coming up with information useful to policy and other decision makers on how to improve adoption of these technologies. The empirical component of the research included a questionnaire survey of one hundred and fifty two(152) selected households in one identified ward in Insiza District, focus group discussions and semi-structured interviews with key informants and in-depth interviews of individual male and female members of a few selected farming households from the sampled population. The desktop portion of the study used secondary data from non-governmental organizations (NGOs), government and other stakeholders involved in conservation farming. Collected data was then disaggregated by gender and analyzed using the Statistical Package for the Social Sciences (SPSS). Perception statements that emerged as significant in chi-square tests of independence were be subjected to factor analysis and weighted factor scores from factor analysis were then used as independent variables in binary logistic regression analysis. The study concluded that both practising and non practising farmers were of the opinion that conservation farming was good though they indicated that information on conservation farming was not readily available The study found out that most farmers agreed on the possible positive effects of CA in addressing livelihood challenges effected by hazards such as HIV and AIDS and environmental hazards such as declining soil fertility but it had is labour intensive therefore is not suitable for people affected and infected by HIV and AIDS. / South Africa

Although, the potential of management induced changes of soil organic matter, soil hydraulic properties (SHPs) and soil physical quality has been studied particularly in relation to tillage, few studies have evaluated combined effect of tillage, crop residue retention and cropping sequence, which are essential components of conservation agriculture (CA), on stratification and storage of soil organic matter, its effect on near-saturated soil hydraulic properties and soil physical quality in intensive cereal based irrigated cropping systems. Hence, the present study critically analyses the effects of CA on organic matter and hydro-physical properties of soil in a long-term CA field trial in NWIGP, India, which is one of the most fragile agro-ecosystems in the world. The objectives were (I) to investigate the stratification of soil organic carbon (SOC), total nitrogen (TN), C/N ratio and evaluate SR as an indicator of storage of SOC and TN and soil quality for different CA practices, (II) to assess the long-term effect of CA practices and short-term effect of crops on near-saturated soil hydraulic conductivity and water transmission properties, and (III) to assess the effect of CA practices on soil physical quality using capacitive and dynamic indicators. There were four treatments: (1) conventionally tilled rice-wheat cropping system (CT-RW), (2) reduced till CA-based rice-wheat-mungbean system (RT-RWMB), (3) no-till CA-based rice-wheat-mungbean system (NT-RWMB), and (4) no-till CA-based maize-wheat-mungbean system (NT-MWMB). To achieve these objectives, soil bulk density, SOC and TN were measured in an increment of 5 cm up to 30 cm soil depth. Furthermore, the effects of CA were also evaluated in terms of soil hydro-physical properties. Soil physical properties such as bulk density and soil aggregate distribution were evaluated in two cropping seasons along with near saturated hydraulic properties. Steady state infiltration rates were obtained at four pressure heads by hood infiltrometer consecutively over two cropping seasons, i.e. during harvest season of rice/maize (October 2017) and maximum crop growth stage of wheat (February 2018). Data were analysed in terms of soil hydraulic conductivity, k(h), flow weighted mean pore radius (r0), hydraulically active porosity (ε) and threshold pore radius (rbp), a new pore measure indicative of macropore stability derived by substituting soil’s bubble pressure in the capillary equation. Finally, the effects of CA on soil physical quality in terms of both capacitive and dynamic indicators, derived from soil moisture retention curve and field measured hydraulic conductivity, respectively, were assessed and related with crop yield to infer which indicator better represented the soil physical quality and its effect on crop yield under irrigated intensive cereal based cropping systems. Results showed that CA had profound impacts on distribution of SOC and TN in the soil profile. Significantly higher proportion of both SOC and TN were observed in the top soil in the CA-based treatments as compared with conventional intensive tillage-based treatment. The mean stratification ratio of both SOC and TN were found > 2 in CA-based treatments whereas the same was < 2 in intensive tillage-based treatment. Storage of SOC and TN in the 0-30 cm were found higher in CA-based treatments as compared with the intensive tillage-based treatment. These results on vertical distribution and storage of SOC and TN indicated a relatively better soil carbon sequestration and soil quality in CA-based treatment. The higher concentrations and storage of soil organic matter in CA-based treatments were, however, not translated into significantly (p < 0.05) lower bulk density due to probable compaction effect of no-tillage and harvest machinery and hydraulic pressure exerted by the flooded irrigation water. However, the increased soil organic matter in the top soil in CA-based treatments improved the soil aggregation significantly which helped in enhancing soil structural quality. Improvement in soil structure was reflected in relatively higher near saturated hydraulic conductivity in CA-based treatments. Irrespective of crop seasons, higher k(h) was observed under CA due to formation of macropores with better continuity, greater size and numbers as compared with conventional intensive tillage treatment. Moreover, higher r0 values were observed for a given k(h) for CA treatments suggesting that interaggregate pores are the dominant pathways of infiltration in CA. A relatively smaller temporal variation of rbp was indicative of a more stable macropore system established by rice-based CA as compared with maize-based CA. CA also enhanced hydraulically active macropores as compared with intensive tillage based conventional agriculture. Results also indicated that crops play an important role in relative distribution of the hydraulically active macropores in the root zone. The impact of CA on soil organic matter stratification and soil hydraulic properties were found to be expressed in terms of changes in soil physical quality. Soil moisture retention curves and pore size distributions under different treatments suggested higher soil water storage in structural pores in CA as compared with intensive tillage-based conventional agriculture. The impact of CA on soil physical quality and consequent effect on crop yield was found to be more expressed through dynamic indicators such as hydraulically active porosity rather than capacitive indicators derived from soil moisture retention curve. Overall, this study reveals that conservation agriculture has great potentials to reverse the intensive tillage induced degradation of soil resources in Indo-Gangetic Plains of India by improving the soil hydro-physical properties and soil physical quality.:Table of Contents Declaration i Declaration of Conformity ii Acknowledgements iii Table of Contents v List of Figures vii List of Tables xi List of Symbols, Abbreviations and Acronyms xiv Abstract xvii 1 Introduction and Background 1 1.1 General Overview 1 1.2 Statement of the Research Problem 5 1.3 Objectives 6 1.4 Research Flow and Chapter Description 7 2 Materials and Methods 9 2.1 Study Area Description 9 2.1.1 Study site 9 2.1.2 Climate 9 2.1.3 Soil 10 2.1.4 Treatments 10 2.1.5 Field Campaigns and Measurement/Analysis 14 2.2 Methods and Theoretical Considerations 14 2.2.1 Soil Sampling and Analysis 14 2.2.1.1 Calculation of Stratification Ratio 15 2.2.1.2 Calculation of SOC and TN Storage 15 2.2.1.3 Aggregate Size Distribution 16 2.2.2 Infiltration Measurements 16 2.2.3 Soil Moisture Retention Experiments 17 2.2.4 Derivation of Hydraulic Properties from Steady State Infiltration Rates 18 2.2.4.1 Near-Saturated Hydraulic Conductivity 18 2.2.4.2 Flow Weighted Mean Pore Radius 20 2.2.4.3 Equivalent Threshold pore Radius 21 2.2.4.4 Hydraulically Active Porosity 21 2.2.5 Determiation of Soil Moisture Charachtristics and Pore Size Distribution 22 2.2.6 Derivation of Soil Physical Quality Indicators 23 2.3 Statistics 25 3 Results and Discussion 26 3.1 Stratification and Storage of Soil Organic Matter 26 3.1.1 Bulk Density 26 3.1.2 Concenrations of SOC 27 3.1.3 Concentrations of TN 28 3.1.4 C/N Ratio 29 3.1.5 Stratification Ratio of SOC, TN and C/N Ratio 30 3.1.6 Storage of SOC and TN 33 3.1.7 Discussion 34 3.1.8 Summary of Results 39 3.2 Soil Hydro-Physical Properties 40 3.2.1 Soil Physical Properties 40 3.2.2 Near-Saturated Hydraulic Conductivity 43 3.2.3 Soil Pore Characteristics-Conductivity Relationship 47 3.2.4 Hydrailically active Porosity 51 3.2.5 Summary of Results 54 3.3 Soil Physical Quality (SPQ) 56 3.3.1 Soil Moisture Retention Curve (SMRC) 56 3.3.2 Soil Pore Size Distribution (SPSD) 58 3.3.3 Capacitive Indicators 59 3.3.4 Dynamic Indicators 60 3.3.5 Relationship between capacitive indicators of SPQ with dynamic indicators of SPQ and long-term crop yield 60 3.3.6 Relationship between dynamic indicator of SPQ (hydraulically active porosity) and Long-term Crop Yield 62 3.3.7 Summary of Results 64 4 Synthesis and Conclusions 65 5 Implications and Outlook 69 References 71

Meloidogyne graminicola est un nématode à galle du genre Meloidogyne spp. (RKN) qui est prépondérant dans les rizières d’Asie du Sud-Est et qui provoque des pertes significatives chez le riz. Malgré son impact négatif sur le rendement, les moyens de control et les méthodes d’identification de cette espèce sont encore limités. Dans cette étude, nous avons évalué l'incidence et la prévalence des RKN dans les rizières du Cambodge, du Laos et du Vietnam grâce à des marqueurs moléculaires développés dans cette étude. Nous avons de plus évalué, sur une séquence de deux ans, les effets de l'Agriculture de conservation (CA) sur la présence et la dynamique des nématodes parasites des racines (RPN) du riz, en relations avec le rendement et la qualité du sol, dans un champ d’expérimentation sableux de bas fond au Cambodge. Nous avons parallèlement évalué la susceptibilité à M. graminicola de 13 plantes adventices couramment utilisées en CA, ainsi que celle de trois variétés de riz, et identifié des plantes réservoir poussant dans les rizières Cambodgiennes. Dans une autre partie, nous avons validé des marqueurs intraspécifiques pour M. graminicola et étudié sa diversité génétique en Asie du Sud-Est. Par des approches de génomique comparative nous avons discuté des relations évolutives qu’il y aurait pu potentiellement y avoir entre M. graminicola et les espèces génétiquement proches. / The rice Root-knot nematode (RKN), Meloidogyne graminicola, reduces significantly rice yield in Southeast (SE) Asia and is prevalent in rice fields in this region. Despite its negative impact on yields, suitable methods of controls and current molecular markers used for M. graminicola identification are still limited. In this study, we evaluated the current occurrence of RKN in rice fields of Cambodia, Laos and Vietnam, and identified those RKN species using molecular markers developed in this study. We then evaluated, during two years sequences, the effects of the Conservation agriculture (CA) on the occurrence and dynamics of root-parasitic nematodes (RPN) of rice, in relation with yields and soil nutrients, in a lowland-sandy field in Cambodia. Host response to M. graminicola of 13 selected cover crops commonly used in CA, of three rice cultivars as well as rice weeds as reservoirs for M. graminicola, were evaluated. Furthermore, we validated intraspecific markers for M. graminicola, studied its genetic diversity in SE Asia, and discussed the potential evolutionary relations of Meloidogyne species using comparative genomics approaches.

L'activité agricole peut entraîner un déséquilibre des cycles du carbone (C) et de l'azote (N) dans les écosystèmes terrestres naturels et entrainer une diminution des stocks de C et N dans le sol, une augmentation de la lixiviation du nitrate et des pertes d'azote par voie gazeuse. Pour réduire ces impacts environnementaux, la mise en place de systèmes agricoles innovants et durables est encouragée, tels que les systèmes à bas niveau intrants, l'agriculture de conservation ou l'agriculture biologique. Les objectifs de cette thèse sont i) de quantifier l'impact à long terme des différents systèmes de culture sur le devenir du carbone et de l'azote dans le système sol-plante-atmosphère et ii) de simuler la dynamique de ces éléments avec le modèle agro-environnemental STICS. À cette fin, nous avons étudié trois essais de longue durée : l'essai de La Cage (France) établi en 1998, l'essai DOK (Suisse) débuté en 1978 et l'essai Foulum (Danemark) créé en 1998. Alors que l'essai de La Cage a permis une quantification in situ du stockage du carbone et de l'azote organiques du sol, de la lixiviation de l'azote, des émissions de protoxyde d’azote (N2O) et de la balance des gaz à effet de serre pour des systèmes de culture alternatifs, les essais danois et suisses ont permis l'estimation in silico du devenir du C et N en agriculture biologique, après adaptation du modèle STICS pour simuler de nouvelles pratiques culturales. Après 16 années d’expérimentation, une accumulation annuelle significative de SOC et de SON a été observée en agriculture et en agriculture biologique à La Cage, alors qu'aucun changement significatif n'a été observé dans les systèmes conventionnels et bas intrants. La minéralisation spécifique de SOC et SON des quatre systèmes, simulée sur AMG et mesurée lors d’incubation des sols pendant quatre mois, s’est montrée équivalente entre systèmes. Le stockage de C et N observé dans les systèmes de conservation et biologiques s’explique principalement par l'augmentation des résidus de cultures plutôt que par l'effet du non-labour en agriculture de conservation. De plus, le surplus azoté (différence entre apports et exportations d’azote) a été calculé pour chaque système de culture. Le devenir de l’excédent d’azote a été estimé entre stockage de N dans le sol, pertes gazeuses et lixiviation de l'azote. Les émissions cumulatives de N2O mesurées en continu pendant plus de trois ans sont fortement corrélées avec les pertes totales calculées de N par voix gazeuse (volatilisation et dénitrification), ces pertes étant les plus importantes dans le système de conservation. Enfin, la réalisation d’un bilan complet des émissions de GES a montré de fortes différences entre système et des phénomènes de compensation entre stockage et perte de C et N. Le modèle sol-culture STICS a ensuite été utilisé pour simuler le devenir de l’azote dans les essais DOK et Foulum. Après une adaptation du modèle, sa calibration et son évaluation ont été réalisées permettant de simuler de façon satisfaisante les rendements, l'absorption de N, le surplus de N et l’évolution des stocks de SON dans les systèmes conventionnels et biologiques. Les simulations suggèrent que le devenir de l’azote dans ces systèmes peut être contrasté en fonction de la fertilisation et de la gestion des cultures et que les pertes d'azote ne sont pas systématiquement réduites en agriculture biologique par rapport au conventionnel. Cette thèse remet en question les appréciations simplistes qui associent systématiquement systèmes de culture alternatifs et diminution des impacts environnementaux liés aux cycles de C et N / Agricultural activities can lead to imbalanced carbon (C) and nitrogen (N) dynamics compared to natural terrestrial eco-systems, causing potential damages for soil, water and air quality. Among these prejudices, decreased soil C and N stocks, increased nitrate leaching in waters and gaseous N emissions towards the atmosphere are of a major concern. To reduce these environmental impacts, innovative and sustainable farming systems are promoted, such as low inputs cropping systems, “conservation” agriculture or organic farming. The objectives of this work were i) to quantify the long term impact of different alternative cropping systems on the fate of C and N in the soil-plantatmosphere system and ii) to simulate C and N dynamics with the agro-environmental model STICS. For this purpose, we studied three long-term field trials: the experiment of La Cage (France) established in 1998, the DOK (Switzerland) started in 1978 and the Foulum Organic (Denmark) established in 1998. The methodological approach combined experimentation and modelling. While La Cage trial enabled an in situ quantification of soil organic C and N storage, N leaching, nitrous oxide (N2O) emissions and greenhouse gas (GHG) balance in alternative cropping systems compared to conventional, the Swiss and Danish experiments were used for in silico estimation of the C and N fates in organic cropping systems, after adaptation of the STICS model, followed by calibration and evaluation of the model. Significant annual SOC and SON accumulation was found under conservation agriculture and organic farming at La Cage, whereas no significant change was observed in the low input and the conventional systems. No difference of specific SOC and SON mineralization rates was found between systems in vitro or in silico : we conclude that the higher C and N storage in soil observed in the conservation and organic systems was mainly driven by increased crop residues, rather than by the effect of no tillage practiced in conservation agriculture. The N surplus, i.e. the difference between N inputs and N exports at the field scale, varied widely between treatments. The fate of this N surplus also varied between systems with wide variations in SON storage and gaseous losses but no differences in N leaching. The cumulative N2O emissions measured continuously for three years were highly correlated with the calculated gaseous N losses (volatilization and denitrification), with higher losses in the conservation system. These calculations allowed establishing a full GHG balance. Therefore the four agricultural systems dissimilarly impacted the N fate, which could not be predicted by the N surplus alone. The GHG balance is a much better indicator of the environmental impact of cropping systems relative to C and N fluxes. In the Danish and Swiss experiments, the soil-crop model STICS was used to mimic crop production, N uptake and N surplus. The model was first adapted and evaluated to simulate organic farming systems. The model could satisfactorily simulate crop production, N uptake, N surplus and SON storage in the organic and conventional systems of these two longterm experiments. Model outputs suggested that the N fate could be contrasted according to fertilization and crop management, and that N losses were not systematically reduced in organic compared to conventional cropping systems. This study challenges the frequent belief that alternative cropping systems systematically improve the global C and N environmental impacts of agriculture.

Submitted by Eduardo Cenci (educenci@gmail.com) on 2015-04-13T17:11:52Z No. of bitstreams: 1 Cenci_thesis_final.pdf: 7470303 bytes, checksum: 3000068146f7a5672bc2107b428b9466 (MD5) / Approved for entry into archive by Vera Lúcia Mourão (vera.mourao@fgv.br) on 2015-04-14T18:02:56Z (GMT) No. of bitstreams: 1 Cenci_thesis_final.pdf: 7470303 bytes, checksum: 3000068146f7a5672bc2107b428b9466 (MD5) / Made available in DSpace on 2015-04-14T18:06:32Z (GMT). No. of bitstreams: 1 Cenci_thesis_final.pdf: 7470303 bytes, checksum: 3000068146f7a5672bc2107b428b9466 (MD5) Previous issue date: 2015-04-01 / This paper investigates the adoption of the Direct Planting System (DPS) in Brazil, a rural technology whose adoption is widely promoted because of benefits, both private and social, it allegedly delivers. This study relies on recent evidence on the determinants of adoption of this technology to investigate its impacts using a large panel of microdata of farm units in the state of Sao Paulo, Brazil. We explore variations in geographic locations and soil characteristics to construct instruments for adoption in order to identify the impacts of this adoption in agricultural outputs, in land allocations and in other production decisions. Although evidence of productivity gains remains uncertain, we do find significant impacts of DPS adoption in land allocations and production decisions. / Este trabalho investiga a adoção do Sistema de Plantio Direto (SPD) no Brasil, uma tecnologia rural bastante promovida devido aos benefícios, tanto privados como sociais, que alegadamente traz. O trabalho baseia-se em evidências recentes sobre os determinantes da adoção desta tecnologia para investigar seus impactos utilizando-se de um grande painel de microdados de unidades de produção agrícola no estado de São Paulo, Brasil. Nós exploramos variações em localizações geográficas e características de solos para construir instrumentos para a adoção e assim identificar os impactos dessa adoção na produção agrícola, nas alo- cações de terras e em outros indicadores da organização produtiva das fazendas. Embora as evidências de ganhos de produtividade permaneçam inconclusivas, nós encontramos impactos significativos da adoção de SPD na alocação de terras e nas decisões de produção.