No description available.

Long-term food security and environmental quality are closely linked to maintaining soil quality. Therefore, the assessment of the effect of agricultural management practices on soil chemical, physical and biological parameters provide fundamental information about sustainability. An agricultural management practice which has received much attention in the last decade is tillage. The loss of topsoil due to erosion and a reduction of soil organic matter under conventional tillage practices, together with escalating fuel prices, have lead to the increased implementation of conservation tillage practices. However, the response of soil to a reduction in tillage is dependent on the inherent soil properties, environmental conditions, crop type and the land management practices. The successful implementation of conservation tillage practices is thus site specific. Furthermore, the effect of fertilizer application on soil quality is affected by tillage regime and therefore has important implications for recommendations of fertilizer application rates. The objectives of this study were to investigate the effect of tillage regime at three rates of nitrogen fertilization on soil microbial activity and selected soil physical properties in the Loskop area of KwaZulu-Natal, South Africa. Based on the outcomes of these investigations, recommendations regarding sustainable tillage practice and nitrogen fertilizer application rate are made. A field trial was initiated in 2003 on Gourton Farm in the Loskop area of KwaZulu-Natal on an area that was previously under annual conventional tillage and is currently planted to dry-land maize. The trial was arranged as a split plot experimental design with tillage regime (whole plots) replicated three times, and fertilizer type and application rate forming randomized subplots within the whole plots. The trial was on a clay loam soil type (Hutton soil form). The effects of annual conventional tillage (CT1) and no-till (NT) at three rates of nitrogen (N) fertilizer (as limestone ammonium nitrate (LAN)) applied at rates of 0 kg N ha-1 annum-1 (0N), 100 kg N ha-1 annum-1 (100N) and 200 kg N ha-1 annum-1 (200N) were evaluated for their effects on soil organic carbon (SOC), microbial activity, bulk density (ñb), water retention characteristics, saturated hydraulic conductivity (Ks), micro-aggregate stability and soil penetration resistance. Undisturbed soil cores were taken from three inter-rows in triplicate from each sub-plot for the A horizon (0 to 20 cm) and from three inter-rows in duplicate for the B horizon (20 to 40 cm). These undisturbed soil cores were used to determine the ñb, water retention characteristics and Ks. Bulk soil samples were collected from three inter-rows in triplicate from each sub-plot for the A (0 to 20 cm) and B (20 to 40 cm) horizons. The bulk samples from each horizon in each sub-plot were thoroughly mixed and halved. One half was used to determine microbial activity as measured by the hydrolytic and cellulolytic activity and the other half was used to determine SOC content, particle size distribution and aggregate stability. Penetration resistance was taken in duplicate in three rows in each sub-plot at 1 cm increments to a depth of 50 cm or until an instrument limiting penetration resistance of 5000 kPa was reached. Tillage regime and N application rate considerably affected soil microbial and physical properties in the A horizon (0 to 20 cm). The SOC, hydrolytic activity and ñb are significantly greater (P 0.05) effect on the soil microbial activity and physical properties except for Ks, where the Ks is significantly (P 0.05) in the B horizon on the measured soil microbial activity and physical properties except for the penetration resistance. Increasing levels of fertilizer resulted in increased penetration resistance throughout the soil profile under NT. Under CT1, this same trend is evident from below the plough layer. These results indicate that the microbial activity, as measured by hydrolytic and cellulolytic activity, is improved under NT compared to CT1. Furthermore, the soil under NT retains more plant available water (PAW) and although the ñb and penetration resistance are greater there was no obvious adverse effect on maize growth. In addition, a high rate of LAN fertilizer adversely affected soil microbial and physical properties, especially under NT. Therefore, it is proposed that NT is the preferred tillage practice in providing long-term sustainability and soil health without causing negative soil structural properties for crop productivity in the short-term. In addition, it is recommended that although increased levels of nitrogen fertilizer results in higher yielding maize plants it is unsustainable to apply high applications of LAN due to the negative effect on the soil microbial and physical properties and thus there is a need to re-evaluate the sustainability of using high rates of LAN to increase crop yields, especially under NT systems. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2010.

CREAMS-NT, a modified version of the field-scale model for Chemicals, Runoff, and Erosion from Agricultural Management Systems (CREAMS) model, was developed to simulate the nitrogen (N) transformations and subsequent nutrient transport processes which occur in the soil following organic waste applications. CREAMS-NT accounts for nutrient addition through fertilization and rainfall and losses of N by volatilization, denitrification, plant uptake, leaching, and overland flow. Data required by CREAMS-NT includes runoff volume, sediment yield, percolation, and soil environmental conditions which is generated by the hydrologic and erosion components of the original CREAMS model. The mineralization, nitrification, denitrification, ar1d volatilization processes are simulated using firstâ order knetic equations adjusted for the effects of soil environmental conditions including temperature, moisture, pH, soil/sludge contact, and soil cation exchange capacity. Prior to a runoff event, soil ammnonium-N (NH4 â N) is partitioned between adsorbed and desorbed phases. CREAMS-NT predicts the transport of organic-N, NH4 â N, and sediment=bound phosphorus (P3b) in runoff using enrichment ratios. The transport of soluble nutrients, nitrate-N (NO3 â N), NH4 â N, and soluble P (PTF), is estimated using extraction coefficients. / Master of Science

Orientadores: Edson Eiji Matsura, Jarbas Honorio de Miranda / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Agricola / Made available in DSpace on 2018-08-09T21:31:09Z (GMT). No. of bitstreams: 1 Souza_AndersonLuizde_D.pdf: 4529373 bytes, checksum: 402972ab3e735c73df6c25a04dbbd1e9 (MD5) Previous issue date: 2007 / Resumo: Conhecer a dinâmica da água e de solutos no solo é condição imprescindível para a produção agrícola sustentável. O sucesso destes modelos numéricos em descrever a dinâmica da água e de solutos no solo está intimamente ligado à precisão com que os parâmetros físicohídricos, envolvidos nestes processos são determinados. De uma maneira geral, a determinação dos parâmetros de transporte de íons no solo esta associado ao ajuste de modelos teóricos a dados associados a ensaios experimentais de deslocamento miscível. Tal procedimento além dos erros numéricos associados ao ajuste destes parâmetros, também é susceptível a erros experimentais. Sendo assim, mostra-se necessário a proposição de técnicas de modelagem que incorporem, de maneira apropriada as imprecisões intrínsecas ao estudo experimental da dinâmica da água e de soluto no solo. Neste sentido, este trabalho teve como objetivo avaliar o potencial do uso da lógica fuzzy, sobretudo de um modelo baseado em regras fuzzy (MBRF), na simulação numérica destes processos. A utilização da teoria dos conjuntos fuzzy é vantajosa à medida que permite que as incertezas envolvidas na dinâmica da água e de solutos no solo sejam absorvidas pela utilização de conjuntos de regras. A base de regras foi construída a partir de um conjunto de dados teóricos, denominado conjunto de treinamento, obtido a partir da simulação numérica das Equações Diferenciais Parciais que regem as dinâmicas da água no solo. O conjunto de treinamento foi obtido utilizando-se o modelo computacional Hydrus-2D. A construção do MBRF, utilizado para descrever a dinâmica da água no solo na direção vertical foi feita utilizando-se o ambiente computacional MATLAB 6.5. Paralelamente a modelagem da dinâmica da água, o estudo do transporte de potássio no solo foi feito considerando-se apenas o fluxo de massa. Os resultados obtidos pelo MBRF foram comparados com resultados obtidos no estudo experimental da dinâmica da água e de solutos em colunas verticais de solo equipadas com sondas de TDR e extratores de solução. Face aos resultados obtidos, pode-se concluir que, neste trabalho, muito embora os resultados apresentados pelo MBRF na simulação da dinâmica de água e de solutos no solo tenham sido apenas razoáveis, o uso da teoria de conjuntos fuzzy aplicada ao estudo destes processos, devido a sua simplicidade de formulação e interpretação, apresentou-se como uma importante ferramenta para auxiliar na tomada de decisão de processos que busquem o uso racional da água e do solo nas atividades agrícolas / Abstract: To know the water and solute dynamics is a indispensable condition for the sustainable agricultural production. The numeric models success in escribe those processes is intimately linked to the precision with the physical- hydric arameters involved in these processes are determineted. In a general way, the determination of the ions transport parameters are obtained by the adjustment of theoretical models to experimental data of miscible displacement studies. Such procedure, besides numerical mistakes, associated to the adjustment of these parameters, is also associated to experimental errors. In this way, the proposition of modeling techniques that incorporate, in an appropriate way, those intrinsic imprecisions are necessary. For this reason, this work¿s goal is evaluate the rule based fuzzy model¿s (RBFM) potential in the numeric simulation of these processes. The use of the fuzzy theory is advantageous once it allows to incorporate the uncertainties involved in the soil water and solutes dynamics using a set of rules. The base of rules was built using a set of theoretical data, denominated training set, obtained by the numeric simulation of the Equations that govern the water dynamics in the soil profile. The training set was obtained using the computational model Hydrus-2D. The construction of the RBFM, used to describe the vertical water dynamics, was done the computational software MATLAB 6.5. Parallel to the water dynamics modeling, a potassium transport study was made considering just the mass flow process. The results obtained with RBFM were compared with experimental data obtained in study of the water and solute dynamics in a vertical soil column, equipped with TDR probes and solution extractors. Face to the obtained results, it can be concluded that, in this work, the presented RBFM wasn¿t very efficient in describe water and solute dynamics the dynamics. In the other hands, apart of those results, the development of this modeling technique showed a great potential in the soils water and solute dynamics, reinforcing the importance of more investigations in this way / Doutorado / Agua e Solo / Doutor em Engenharia Agrícola