Both empirical and process-simulation models are useful for evaluating the effects of management practices on environmental quality and crop yield. The use of these models is limited, however, because they need many soil property values as input. The first step towards modelling is the collection of input data. Soil properties can be highly variable spatially and temporally, and measuring them is time-consuming and expensive. Efficient methods, which consider the uncertainty and cost of measurements, for estimating soil hydraulic properties form the main thrust of this study. Hydraulic properties are affected by other soil physical, and chemical properties, therefore it is possible to develop empirical relations to predict them. This idea quantified is called a pedotransfer function. Such functions may be global or restricted to a country or region. The different classification of particle-size fractions used in Australia compared with other countries presents a problem for the immediate adoption of exotic pedotransfer functions. A database of Australian soil hydraulic properties has been compiled. Pedotransfer functions for estimating water-retention and saturated hydraulic conductivity from particle size and bulk density for Australian soil are presented. Different approaches for deriving hydraulic transfer functions have been presented and compared. Published pedotransfer functions were also evaluated, generally they provide a satisfactory estimation of water retention and saturated hydraulic conductivity depending on the spatial scale and accuracy of prediction. Several pedotransfer functions were developed in this study to predict water retention and hydraulic conductivity. The pedotransfer functions developed here may predict adequately in large areas but for site-specific applications local calibration is needed. There is much uncertainty in the input data, and consequently the transfer functions can produce varied outputs. Uncertainty analysis is therefore needed. A general approach to quantifying uncertainty is to use Monte Carlo methods. By sampling repeatedly from the assumed probability distributions of the input variables and evaluating the response of the model the statistical distribution of the outputs can be estimated. A modified Latin hypercube method is presented for sampling joint multivariate probability distributions. This method is applied to quantify the uncertainties in pedotransfer functions of soil hydraulic properties. Hydraulic properties predicted using pedotransfer functions developed in this study are also used in a field soil-water model to analyze the uncertainties in the prediction of dynamic soil-water regimes. The use of the disc permeameter in the field conventionally requires the placement of a layer of sand in order to provide good contact between the soil surface and disc supply membrane. The effect of sand on water infiltration into the soil and on the estimate of sorptivity was investigated. A numerical study and a field experiment on heavy clay were conducted. Placement of sand significantly increased the cumulative infiltration but showed small differences in the infiltration rate. Estimation of sorptivity based on the Philip's two term algebraic model using different methods was also examined. The field experiment revealed that the error in infiltration measurement was proportional to the cumulative infiltration curve. Infiltration without placement of sand was considerably smaller because of the poor contact between the disc and soil surface. An inverse method for predicting soil hydraulic parameters from disc permeameter data has been developed. A numerical study showed that the inverse method is quite robust in identifying the hydraulic parameters. However application to field data showed that the estimated water retention curve is generally smaller than the one obtained in laboratory measurements. Nevertheless the estimated near-saturated hydraulic conductivity matched the analytical solution quite well. Th author believes that the inverse method can give a reasonable estimate of soil hydraulic parameters. Some experimental and theoretical problems were identified and discussed. A formal analysis was carried out to evaluate the efficiency of the different methods in predicting water retention and hydraulic conductivity. The analysis identified the contribution of individual source of measurement errors to the overall uncertainty. For single measurements, the inverse disc-permeameter analysis is economically more efficient than using pedotransfer functions or measuring hydraulic properties in the laboratory. However, given the large amount of spatial variation of soil hydraulic properties it is perhaps not surprising that lots of cheap and imprecise measurements, e.g. by hand texturing, are more efficient than a few expensive precise ones.

Both empirical and process-simulation models are useful for evaluating the effects of management practices on environmental quality and crop yield. The use of these models is limited, however, because they need many soil property values as input. The first step towards modelling is the collection of input data. Soil properties can be highly variable spatially and temporally, and measuring them is time-consuming and expensive. Efficient methods, which consider the uncertainty and cost of measurements, for estimating soil hydraulic properties form the main thrust of this study. Hydraulic properties are affected by other soil physical, and chemical properties, therefore it is possible to develop empirical relations to predict them. This idea quantified is called a pedotransfer function. Such functions may be global or restricted to a country or region. The different classification of particle-size fractions used in Australia compared with other countries presents a problem for the immediate adoption of exotic pedotransfer functions. A database of Australian soil hydraulic properties has been compiled. Pedotransfer functions for estimating water-retention and saturated hydraulic conductivity from particle size and bulk density for Australian soil are presented. Different approaches for deriving hydraulic transfer functions have been presented and compared. Published pedotransfer functions were also evaluated, generally they provide a satisfactory estimation of water retention and saturated hydraulic conductivity depending on the spatial scale and accuracy of prediction. Several pedotransfer functions were developed in this study to predict water retention and hydraulic conductivity. The pedotransfer functions developed here may predict adequately in large areas but for site-specific applications local calibration is needed. There is much uncertainty in the input data, and consequently the transfer functions can produce varied outputs. Uncertainty analysis is therefore needed. A general approach to quantifying uncertainty is to use Monte Carlo methods. By sampling repeatedly from the assumed probability distributions of the input variables and evaluating the response of the model the statistical distribution of the outputs can be estimated. A modified Latin hypercube method is presented for sampling joint multivariate probability distributions. This method is applied to quantify the uncertainties in pedotransfer functions of soil hydraulic properties. Hydraulic properties predicted using pedotransfer functions developed in this study are also used in a field soil-water model to analyze the uncertainties in the prediction of dynamic soil-water regimes. The use of the disc permeameter in the field conventionally requires the placement of a layer of sand in order to provide good contact between the soil surface and disc supply membrane. The effect of sand on water infiltration into the soil and on the estimate of sorptivity was investigated. A numerical study and a field experiment on heavy clay were conducted. Placement of sand significantly increased the cumulative infiltration but showed small differences in the infiltration rate. Estimation of sorptivity based on the Philip's two term algebraic model using different methods was also examined. The field experiment revealed that the error in infiltration measurement was proportional to the cumulative infiltration curve. Infiltration without placement of sand was considerably smaller because of the poor contact between the disc and soil surface. An inverse method for predicting soil hydraulic parameters from disc permeameter data has been developed. A numerical study showed that the inverse method is quite robust in identifying the hydraulic parameters. However application to field data showed that the estimated water retention curve is generally smaller than the one obtained in laboratory measurements. Nevertheless the estimated near-saturated hydraulic conductivity matched the analytical solution quite well. Th author believes that the inverse method can give a reasonable estimate of soil hydraulic parameters. Some experimental and theoretical problems were identified and discussed. A formal analysis was carried out to evaluate the efficiency of the different methods in predicting water retention and hydraulic conductivity. The analysis identified the contribution of individual source of measurement errors to the overall uncertainty. For single measurements, the inverse disc-permeameter analysis is economically more efficient than using pedotransfer functions or measuring hydraulic properties in the laboratory. However, given the large amount of spatial variation of soil hydraulic properties it is perhaps not surprising that lots of cheap and imprecise measurements, e.g. by hand texturing, are more efficient than a few expensive precise ones.

Cette thèse a été effectuée dans le cadre du projet ANCRES. Son rôle est d'assurer la compréhension physique du milieu récepteur, le sol, alors qu'une autre équipe étudie l'impact physico-chimique des eaux usées traitées sur le sol. Pour cela, il faut connaître ses propriétés texturales et structurales contrôlant ses fonctions d'épuration et d'évacuation. La problématique est donc de comprendre le processus de l'infiltration imposée par un dispositif d'ANC dans un sol caillouteux. D'abord, suite au positionnement du site pilote d'ANC au pied d'un versant au cœur du plateau portlandien dans le département de l'Yonne en France, on a identifié là un type de sol hétérogène et complexe, caillouteux, une colluvion qui n'était pas cartographiée sur la carte géologique BRGM de cette région et sur laquelle on focalise la thèse. Ensuite, dans le sol complexe du site pilote, on a d'abord mesuré la conductivité hydraulique du sol par l'appareil Guelph sur 15 m² de la fouille de l'ANC à une profondeur de 120 cm : au total, on a fait 15 essais d'infiltration. De plus, on a récupéré 15 échantillons du sol accompagnant ses essais de Guelph pour la caractérisation physique au laboratoire. On a mesuré la texture, l'humidité résiduelle (HR), la teneur en cailloux (Rw) et la matière organique (MO) des échantillons au laboratoire. Ensuite, pour l'étude du processus d'infiltration, on a instrumenté ce site par un dispositif de surveillance hydrique (tensiométrie, teneur en eau, piézométrie) et de prélèvement d'eau interstitielle du sol. De plus, la masse volumique (densité) apparente d'un sol hétérogène a été déterminée. Parmi les paramètres mesurés, certains sont choisis comme indicateurs pour caractériser un sol comme milieu récepteur potentiel des eaux usées traitées en ANC. Puis, en utilisant la MO et la texture des échantillons, on a estimé la masse volumique apparente de la matrice du sol du site piloteau moyen de fonctions de pédotransfert appelées BD-FPTs et on a testé la relation entre la conductivité hydraulique à saturation Ks et la texture de ce sol complexe. Pour cela, une méthodologie en quatre phases a été développée pour évaluer la capacité prédictive des fonctions Ks-FPTs. Cette méthodologie de sélection n'a pas été trouvée dans la littérature mais est élaborée pour les besoins de la thèse. On a déduit les meilleures Ks-FPTs pour ce type de sol. Enfin, avec les données d'humidité volumique et du potentiel matriciel du sol, acquises par une centrale d'acquisition des données, le régime hydrodynamique du sol sous le massif filtrant de l'ANC a été étudié et on a mis en évidence l'écoulement préférentiel dans un sol caillouteux. Les résultats ont montré que dans la colluvion, bien que la matrice du sol soit fine, la conductivité hydraulique mesurée est plus élevée qu'attendu. Cela démontre que la fraction des cailloux dans le sol joue un rôle essentiel en accélérant l'évacuation des eaux usées traitées et aussi l'eau pluviale vers les couches sous-jacentes, et finalement vers la nappe. Cette propriété serait un point fort pour la fonction de transfert du sol et on peut en déduire une méthode pour améliorer la capacité de transfert de l'eau des sols lourds dans les projets d'aménagement urbain ou périurbain : l'ajout de graviers et graves calcaires par mélange au sol en place. On conclut que cette expérimentation, unique en son genre, a été utile pour évaluer la fonction de rétention / transfert de l'eau dans le sol recevant les eaux usées traitées. On a identifié les paramètres prédicteurs pertinents et les relations empiriques qui permettent de faire l'économie de nombreux essais in situ d'eau / The soil is an essential compartment in hydrologic cycle of water in the nature. Therefore, it is clear that taking into account the properties and organization of the soil is essential to the understanding and management of flows involved in the development of the quality of groundwater and surface water. The on-site sanitation (ANC) is a management method of domestic wastewater, by which the water is sent into the soil after settling and filtration liquefaction / aerobic degradation. The second step was often provided by the top soil in place himself, and this continues to this day on many plots of on-site sanitation.This thesis was carried out under ANCRES project. Its role was to ensure the physical understanding of the receptor medium, soil, while another research team was investigating the physico-chemical impact of treated wastewater on the sol. To ensure it, we have to understand its textural and structural properties controlling its purifying power and hydrodynamic processes. So, the problematic is to understand the process of infiltration imposed by an ANC in a heterogeneous stony soil. At first, due to the positioning of the pilot ANC site, at the foot of a slope in the Yonne department in France, a type of complex and heterogeneous soil, a colluvion, was identified. This soil has not been mapped on the BRGM (Bureau de recherches géologiques et minières) geological map of the area. So, the thesis was focalised on this soil. Then, in this soil complex at a depth of the 120 cm of the excavation of the ANC, the soil hydraulic conductivity was first measured by a Guelph apparatus on 15 m². We totally made 15 infiltration tests. Furthermore, we have collected 15 soil samples with each Guelph test for physics laboratory physical characterization. Then, in order to study the soil functions, this site was instrumented by the water monitoring devices (tensiometers, water content probes, and piezometer) and the interstitial water sampling device from the soil. In addition, the bulk of this heterogeneous soil was determined. Among the measured parameters, a series of indicators chosen to characterize the soil as a potential receptor medium of treated wastewater of the ANC. Then, the bulk density of the soil matrix using pedotransfer functions called BD-FPTs was estimated and the relationship between saturated hydraulic conductivity Ks and the texture of the soil complex (by Ks-FPT function) was tested. For this aim, a four-phase methodology was developed to assess the predictive ability of Ks- FPTs functions. This methodology in four phases is not found in the literature but was prepared for the purposes of the thesis. We concluded the best Ks- FPTS for this type of soil. Finally, with the volumetric water content and soil matrix potential data acquired by a the data loggers, the hydrodynamic regime of the soil under the sand pack of the ANC has been studied and demonstrated the preferential flow in a stony soil was demonstrated. The results showed that in the colluvion, although the soil matrix is fine, the measured hydraulic conductivity is higher than expected. This shows that the stone fraction in the soil plays an important role in accelerating the evacuation of treated wastewater and also rainwater to the underlying layers, and finally to the water table. This property would be a strong point for the transfer function of the soil and we can deduce a method to improve the transfer function of heavy soils in urban or peri-urban development projects. This could be possible by adding the gravel and -limestone to soil matrix and mixing them. It has been concluded that this experiment, unique in its kind, has been useful in evaluating the function of retention / transfer of water in the soil receiving treated wastewater. In addition, the relevant predictor parameters and empirical relationships that make the economy of many water tests were identified

Modeling the seedling recruitment microsite involves characterization of the soil environment of the shallow profile from which weed seedlings recruit. Understanding the environment of the seedling recruitment microsite is the prelude to weed emergence studies. Because of spatial and temporal heterogeneity of the recruitment microsite, sufficient measurements are often not feasible. An experiment was established in 2003 and 2004 across topography within an annually cropped field in south-central Manitoba to determine the effect that hillslope aspect and position, and soil residue and depth would have on microsite environment within the shallow seedling recruitment zone. Microclimatic, topographic, soil surface and soil properties were assessed in the context of the weed recruitment microsite. The soil water retention characteristic was measured by pressure plate to determine water availability to germinating seeds at the various topographic positions. The soil water characteristic was evaluated across topography and soil depth. Evaluation of the soil water characteristic by pedotransfer function indicated that a single soil water characteristic is representative of the recruitment zone. Field and laboratory experimental measurements were used as parameterization for the simultaneous heat and water (SHAW) model to generate continuous water and temperature profiles for the recruitment zone. Soil temperature and temperature fluctuation decreased with depth in the recruitment zone. Despite differences of texture, bulk density, and organic matter across topography and soil depth, the soil water characteristic differed only across topography. Soil water potential fluctuated considerably at the soil surface due to numerous precipitation events and direct evaporation. Implications for germinating seeds is that the seedling recruitment zone is influenced by spatial effects of topography and the vertical location of the seed microsite. Physical process based modeling used in this study to predict temperature and water within the seedling recruitment zone enables better understanding of interactions between above-ground microclimate and the recruitment microsite. Such interactions enable linkage between atmospheric models and recruitment models that can enhance our ability to evaluate crop management decisions.

Modeling the seedling recruitment microsite involves characterization of the soil environment of the shallow profile from which weed seedlings recruit. Understanding the environment of the seedling recruitment microsite is the prelude to weed emergence studies. Because of spatial and temporal heterogeneity of the recruitment microsite, sufficient measurements are often not feasible. An experiment was established in 2003 and 2004 across topography within an annually cropped field in south-central Manitoba to determine the effect that hillslope aspect and position, and soil residue and depth would have on microsite environment within the shallow seedling recruitment zone. Microclimatic, topographic, soil surface and soil properties were assessed in the context of the weed recruitment microsite. The soil water retention characteristic was measured by pressure plate to determine water availability to germinating seeds at the various topographic positions. The soil water characteristic was evaluated across topography and soil depth. Evaluation of the soil water characteristic by pedotransfer function indicated that a single soil water characteristic is representative of the recruitment zone. Field and laboratory experimental measurements were used as parameterization for the simultaneous heat and water (SHAW) model to generate continuous water and temperature profiles for the recruitment zone. Soil temperature and temperature fluctuation decreased with depth in the recruitment zone. Despite differences of texture, bulk density, and organic matter across topography and soil depth, the soil water characteristic differed only across topography. Soil water potential fluctuated considerably at the soil surface due to numerous precipitation events and direct evaporation. Implications for germinating seeds is that the seedling recruitment zone is influenced by spatial effects of topography and the vertical location of the seed microsite. Physical process based modeling used in this study to predict temperature and water within the seedling recruitment zone enables better understanding of interactions between above-ground microclimate and the recruitment microsite. Such interactions enable linkage between atmospheric models and recruitment models that can enhance our ability to evaluate crop management decisions.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Soil water retention stems from the action of attractive forces manifested by adsorption and capillary phenomena. Matric potential (ψ) expresses the soil water retention energy, which varies according to water content and is described by soil water retention curve (SWRC). The experimental obtainment of SWRC is difficult and costly, therefore, the use of pedotransfer functions (PTF) is preferred, because PTF estimate SWRC (or part of it) from other soil properties related to water retention. The use of available PTFs is limited for being calibrated from a particular set of soils and/or for not being able to represent the effect of structural changes (e. g. bulk density (Bd)) on water retention. Thus, the aim of this study is to propose and evaluate a mathematical model to estimate the volumetric moisture based on the specific surface area (SSA) and on the integration of the effect of the Bd with a physicoempirical parameter (fa). SWRCs were obtained experimentally from soils with different particle size and Bd variations were simulated by samples prepared in metal cylinders. The SSA was estimated by particle size distribution curve. Dependence of fa on ψ, SSA and Bd was described by Van Genuchten (1980) sigmoidal model, while the effect of SSA and Bd was incorporated in parameters of this same sigmoidal model by graphical and sensibility analysis. The estimates of the model were compared with the literature FPTs estimates for the same data set. Initially, the results showed that samples prepared in the laboratory did not ensure consistent experimental data of SWRC. This finding masked the model predictive ability. However, it is clear that the proposed FPT coherently described the effect of SSA and Bd with adsorption and capillarity phenomena, complying with the study purpose. In addition, the physicoempirical consistency of SSA and Bd with parameters of the Van Genuchten (1980) sigmoidal model allowed a better predictive performance compared to literature PTFs. In a broader sense, results reinforce the idea that models which mechanistically explore the relationship between predicted and predictor variables are more efficient in estimating the magnitude of the natural soil processes. and Bd was incorporated in parameters of this same sigmoidal model by graphical and sensibility analysis. The estimates of the model were compared with the literature FPTs estimates for the same data set. Initially, the results showed that samples prepared in the laboratory did not ensure consistent experimental data of SWRC. This finding masked the model predictive ability. However, it is clear that the proposed FPT coherently described the effect of SSA and Bd with adsorption and capillarity phenomena, complying with the study purpose. In addition, the physicoempirical consistency of SSA and Bd with parameters of the Van Genuchten (1980) sigmoidal model allowed a better predictive performance compared to literature PTFs. In a broader sense, results reinforce the idea that models which mechanistically explore the relationship between predicted and predictor variables are more efficient in estimating the magnitude of the natural soil processes. / A retenção de água no solo decorre da atuação de forças atrativas que se manifesta por meio dos fenômenos de adsorção e capilaridade. O potencial matricial (ψ) expressa a energia com que a água está retida no solo, que varia com a umidade volumétrica (Uv) e é descrita pela curva de retenção de água no solo (CRAS). A obtenção experimental da CRAS é difícil e onerosa, o que motiva o uso de funções de pedotransferência (FPT), que estimam a CRAS (ou parte dela) a partir de propriedades do solo ligadas à retenção de água. As FPTs disponíveis são limitadas quanto à aplicação para solos diferentes dos quais foram calibradas ou por não conseguirem representar o efeito de mudanças estruturais (variação na densidade do solo (Ds)) sobre a retenção de água. Portanto, o objetivo deste estudo é propor e avaliar um modelo matemático para estimar a umidade volumétrica, baseado na área superficial específica (ASE) e na integração do efeito da Ds a partir de um parâmetro físico-empírico (fa). Experimentalmente, CRAS foram obtidas a partir de solos com diferente granulometria, para os quais variações de Ds foram simuladas por meio da confecção de amostras em cilindros metálicos. A ASE foi estimada por meio da curva granulométrica. A dependência de fa em relação às variáveis ψ, ASE e Ds, foi descrita pelo modelo sigmoidal de Van Genuchten (1980), enquanto o efeito da ASE e Ds foi incorporado aos parâmetros desse modelo sigmoidal por meio de análise gráfica e de sensibilidade. As estimativas do modelo proposto foram comparadas com estimativas de FPTs da literatura para o mesmo conjunto de dados. Inicialmente, verificou-se que, mesmo utilizando amostras confeccionadas em laboratório, a obtenção experimental da CRAS demonstrou-se, por vezes, inconsistente, o que mascarou a capacidade preditiva do modelo. Contudo, é possível afirmar que a FPT proposta foi capaz de descrever o efeito da ASE e da Ds de forma fisicamente coerente com os fenômenos de adsorção e capilaridade, cumprindo com a proposta deste estudo. Além disso, a consistência física-empírica entre ASE e Ds e os parâmetros de ajuste do modelo sigmoidal de Van Genuchten (1980) permitiram melhor desempenho preditivo comparado às FPTs da literatura. De forma mais ampla, os resultados reforçam a ideia de que modelos que explorem mecanisticamente a relação entre variáveis preditas e preditoras são mais eficientes na estimativa da magnitude dos processos naturais ligados ao solo.

This thesis deals with direct and indirect determination of soil hydraulic characteristics (retention curve and hydraulic conductivity) in the locality of Bohaté Málkovice. For laboratory determination of soil moisture retention curve is used sand tank and pressure relief device. For the indirect determination of soil hydraulic characteristics is used computer software Rosetta, which includes 5 models of pedotransfer functions. For estimatiton of retention curves are applied previously derived pedotransfer function (continuous parametric PTF and point PTF).

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The preconsolidation pressure (σp) is obtained from the curve of the soil compression visual graphical form, mathematical or associating both and is an estimate of its load-bearing capacity. In order to remove the subjectivity of the visual graphical method is need for mathematical description (models using) of the compression curve. However, research results do not provide subsidies for the choice of a model whose σp represents well the resulting load-bearing capacity of soil, but if some model to represent, the σp should be a function of density and water content soil, other properties remain constant. The objective of this study was to understand the variability in the preconsolidation pressure caused by different models and options to adjust these models to the compression curve and evaluate the relationship of preconsolidation pressure with bulk density (Ds), volumetric water content (θ) and macroporisity (Ma) . For this, we performed the uniaxial compression test of 300 samples from a Hapludox, clayey, with wide variation Ds and θ. The preconsolidation pressure was determined by strictly mathematical, from the set of ten models, where seven of them consisted of variations of the model van Genuchten (1980) and others have been proposed by Friton (2001), Assouline (2002) and Gregory et al. (2006). The model proposed by Gregory and Friton possible to obtain the σp for only 62 and 56% of the curves, respectively, the variations of the model van Genuchten were enable the calculation of σp at least 90% of the curves. In three of the variations of the van Genuchten model, it was possible to get σp in 100% of the curves. The range of σp produced by the various options described in compression curve affects the physical significant of σp to represent the load bearing capacity of the soil. Likewise, Ds, θ and Ma, which influence the same load bearing capacity of the soil, not explained more than 58% of the variation of σp produced using different models. With the criteria used, it is not possible to choose one model over the other which allows to estimate more adequately by σp, the load bearing capacity of the soil, which limits the use of this parameter as an indicator for soil management agricultural. Given this result, it was investigated whether the percentage of deformation of the sample as a function of the applied loads would be more closely related to Ds, θ, and Ma was considered a safe limit of deformation of the physical viewpoint, that Ma deformation where the remains higher than 10%. The use of maintenance limit of 80% of voids index (Mεi) displaces the deformation to the plastic region of the curves of all the density ranges and therefore cannot be used for soil with structure more affected by compaction. The load to achieve a certain percentage (Mεi) is best explained by the variation of Ds and θ. The pedotransfer functions to load in Mεi between 85 and 87.5% resulted in higher coefficients of determination (0.74), having potential use of such values as critical limits to maintenance of porosity without degradation of soil physical quality. / A pressão de preconsolidação (σp) é obtida da curva de compressão do solo de forma gráfica visual, matemática ou associando ambas e é uma estimativa de sua capacidade de suporte de carga. A fim de se retirar a subjetividade do método gráfico visual, há necessidade de descrição matemática (uso de modelos) da curva de compressão. Porém, os resultados de pesquisa não fornecem subsídios para a escolha de um modelo cuja σp resultante represente, de maneira confiável, a capacidade de suporte de carga do solo. Entretanto, se algum modelo a representar, a σp deverá ser uma função da densidade e conteúdo de água do solo, mantidas constantes outras propriedades. O objetivo deste trabalho foi conhecer a variabilidade na pressão de preconsolidação provocada por diferentes modelos e opções de ajuste desses modelos à curva de compressão e avaliar a relação da pressão de preconsolidação com a densidade do solo (Ds), conteúdo volumétrico de água (θ) e macroporosidade do solo (Ma). Para isso, foi realizado o ensaio de compressão uniaxial para 300 amostras provenientes de um Latossolo Vermelho Distroférrico típico, com textura argilosa, com ampla variação Ds e θ. A pressão de preconsolidação foi determinada por meio estritamente matemático, a partir do ajuste de dez modelos, onde sete deles consistiram em variações do modelo de van Genuchten (1980) e os demais foram propostos por Friton (2001), Assouline (2002) e Gregory et al. (2006). Os modelos propostos por Gregory e Friton possibilitaram a obtenção da σp para apenas 62 e 56 % das curvas, respectivamente, já as variações do modelo de van Genuchten viabilizam o cálculo da σp para no mínimo 90 % das curvas. Em três das variações do modelo de van Genuchten, foi possível obter a σp em 100% das curvas. A variação da σp produzida pelas diferentes opções de descrição da curva de compressão compromete o significado físico da σp para representar a capacidade de suporte de carga do solo. Da mesma forma, Ds, θ e Ma, mesmo que influenciem a capacidade de suporte de carga do solo, não explicaram mais que 58% da variação da σp produzida com o uso de diferentes modelos. Com os critérios utilizados, não é possível eleger um modelo em relação aos demais que possibilite estimar de maneira mais adequada, através da σp, a capacidade de suporte de carga do solo, o que limita o uso desse parâmetro como um indicador para o manejo dos solos agrícolas. Diante desse resultado, foi investigado se a porcentagem de deformação da amostra em função das cargas aplicadas estaria melhor relacionada com Ds, θ e Ma. Considerou-se um limite de deformação prudente do ponto vista físico, aquela deformação onde a Ma permaneça maior que 10%. O uso do limite de manutenção de 80 % do índice de vazios inicial (Mεi) desloca as deformações para a região plástica das curvas de todas as faixas de densidade e, por isso, não pode ser utilizado para solos com estrutura mais comprometida pela compactação. A carga para atingir determinada porcentagem de Mεi é melhor explicada pela variação da Ds e θ. As funções de pedotransferência da carga para a Mεi entre 85 e 87,5% resultaram em maiores coeficientes de determinação (0,74), havendo potencial da utilização destes valores como limites críticos de manutenção da porosidade sem haver degradação da qualidade física do solo.

Submitted by Alison Vanceto (alison-vanceto@hotmail.com) on 2017-02-16T13:09:22Z No. of bitstreams: 1 DissACMC.pdf: 3790593 bytes, checksum: cb1417be4a2fe538b7b730f90132e43e (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-03-20T18:46:36Z (GMT) No. of bitstreams: 1 DissACMC.pdf: 3790593 bytes, checksum: cb1417be4a2fe538b7b730f90132e43e (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-03-20T18:47:04Z (GMT) No. of bitstreams: 1 DissACMC.pdf: 3790593 bytes, checksum: cb1417be4a2fe538b7b730f90132e43e (MD5) / Made available in DSpace on 2017-03-20T18:53:16Z (GMT). No. of bitstreams: 1 DissACMC.pdf: 3790593 bytes, checksum: cb1417be4a2fe538b7b730f90132e43e (MD5) Previous issue date: 2016-08-11 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / In recent years, Brazil has proposed policies to reduce emissions of greenhouse gases (GEE). In this context, the aim of this study was to estimate the carbon stock (EC) from different Brazilian soils under different agricultural uses and propose strategies that contribute to mitigation of GEE emissions. The research was conducted in four stages: i) organization of a soil database; ii) development of a pedotransfer function (PTF) for the estimation of bulk density (DS) and evaluate the effect on estimate of EC; iii) estimation of the EC; and iv) evaluation of potential EC by Brazilian agriculture. Data from 38.456 soil samples were performed and, after standardization, they formed a database with 10.445 data samples corresponding to 5.823 data for the 0-30 cm layer. These data covered all Brazilian states, all classes of the Brazilian System of Soil Classification and nine types of land use: annual crop in no-tillage system (SPD), annual crop in conventional tillage system, perennial crop, planted forest, integrated crop-livestock (ILP) system, integrated crop-livestock-forest (ILPF) system, pasture, uncovered soil and native vegetation. Many samples had no DS record, then 12 PTF for DS estimation were developed using 974 soil samples. The performance of PTFs was assessed by R2, and in the validation, the accuracy of prediction was measured based on the mean error (ME), the mean absolute error (MAE), and the root mean squared error (RMSE). All functions overestimated DS values and one of them (PTF 5) presented the best performance. The evaluation of the estimated EC was made with 926 samples layer 0-30 cm, using observed data DS (ECobs), estimated data DS from the PTF5 (ECest) and estimated data DS from the null model (ECnull), in which the DS was given by the mean value of DS observed. Based on the calculation of ME, MAE, the RMSE and comparison with ECobs values, it was found that the ECnull values were overestimated and dispersed. It was concluded at this stage that the null model was not a reliable alternative and PTF5 was applied in 4.540 samples from 0-30 cm layer with missing DS. The estimated DS values ranged from 0,10 kg dm-3 at 1,92 kg dm-3 with a mean of 1,39 kg dm-3 and standard deviation of 0,19 kg dm-3. The coefficient of variation was less than 15% configuring a homogeneous data. It was made the EC calculation for 5.823 EC data for the layer 0-30 cm and these values were grouped by types of land use and soil types. In grouping by type of land use, the lowest EC values were observed in annual crop and native vegetation (0,10 Mg ha-1) and the largest maximum values were observed in annual crop, pasture and native vegetation, with 297,3 Mg ha-1, 259,9 Mg ha-1 and 253,6 Mg ha-1 respectively. In grouping by type of soil, it was observed that a minimum value of 0,10 Mg ha-1 in an Argisol while maxima were observed in a Cambisol (297,3 Mg ha-1) and an Argisol (265,8 Mg ha-1). Three scenarios were developed to estimate the potential increase of EC, from changes in management practices and land use. In a scenario where 18% of the areas of annual crops adopt SPD, the increase in EC was 73,6 Gg. If 15 million hectares of degraded pastures were recovered with ILP system in 20% of the area, and ILPF system in 10% of the area, the potential for increasing the EC would be 88.13 Gg. These values could represent an addition of 59,6 Gg of CO2, which could offset the 472,1 Gg CO2 emitted by the Brazilian agricultural sector in 2010, according to estimates by the Ministry of Science, Technology and Innovation. It concludes that: i) the absence of DS data, using a simple PTF is more appropriate than the use of the null model; ii) the detected errors in the estimation of DS by PTF not propagate the EC calculation; iii) the proposed changes would offset emissions from the Brazilian agricultural sector; and iv) the estimates presented highlight the role of the agricultural sector in mitigating GEE emissions. / Nos últimos anos, o Brasil tem proposto políticas para diminuir as emissões de gases de efeito estufa (GEE). Nesse contexto, o objetivo geral deste trabalho foi estimar o estoque de carbono (EC) de diferentes solos brasileiros com diferentes usos agrícolas e propor estratégias que contribuam para mitigação das emissões de GEE. A pesquisa foi desenvolvida em quatro etapas: i) organização de um banco de dados de solos; ii) desenvolvimento de função de pedotransferência (PTF) para a estimativa de densidade do solo (DS) e avaliação do efeito na estimativa de EC; iii) estimativa do EC; e iv) potencial de EC pela agricultura brasileira. Foram levantados dados de 38.456 amostras de solos que, após padronização, constituíram um banco com dados de 10.445 amostras, que corresponderam a 5.823 dados para a camada 0-30 cm. Esses dados recobriram todos os estados brasileiros, todas as classes do Sistema Brasileiro de Classificação de Solos e nove tipos de uso do solo: cultura anual em sistema de plantio direto (SPD), cultura anual sem SPD, cultura perene, floresta plantada, integração lavoura-pecuária (ILP), integração lavoura-pecuária-floresta (ILPF), pastagem, solo descoberto e vegetação nativa. Como muitas amostras não apresentavam registro de DS, 12 PTF foram desenvolvidas em um subconjunto de 974 amostras do banco de dados, sendo 664 amostras utilizadas para o desenvolvimento das PTFs e 310 utilizadas para teste. O desempenho de cada PTF foi avaliado através do R2, e na validação, a acurácia da predição foi medida pela comparação dos valores estimados com os observados, com base no cálculo de erro médio (ME), do erro absoluto médio (MAE) e da raiz quadrada do erro quadrático médio (RMSE). Todas as PTFs testadas superestimaram os valores de DS, porém, a PTF5 apresentou o melhor desempenho. A avaliação da estimativa de EC foi feita com 926 amostras da camada 0-30 cm, usando-se dados de DS observados (ECobs), dados de DS estimados a partir da PTF5 (ECest) e dados de DS estimados a partir do modelo nulo (ECnulo), no qual a DS é dada pelo valor médio das DS observadas. Com base no cálculo do ME, da MAE, da RMSE e da comparação com os valores de ECobs, constatou-se que os valores de ECnulo foram superestimados e dispersos. Concluiu-se nesta etapa que o modelo nulo não era uma alternativa confiável e a PTF5 foi aplicada em 4.540 amostras de camada 0- 30 cm com DS faltante. Os valores de DS estimados variaram de 0,10 kg dm-3 a 1,92 kg dm-3, com média de 1,39 kg dm-3 e desvio padrão de 0,19 kg dm-3. O coeficiente de variação foi menor que 15%, configurando dados homogêneos. Foi feito o cálculo de EC para 5.823 dados de EC referentes à camada 0-30 cm e esses valores foram agrupados por tipos de uso do solo e por classes de solo. No agrupamento por tipos de uso do solo, os menores valores de EC foram observados em cultura anual e vegetação nativa (0,10 Mg ha-1) e o maiores valores máximos foram observados em cultura anual, pastagem e vegetação nativa, com 297,3 Mg ha- 1, 259,9 Mg ha-1 e 253,6 Mg ha-1 respectivamente. No agrupamento por tipos de solo, observou-se que um valor mínimo de 0,10 Mg ha-1 em um Argissolo enquanto que valores máximos foram observados em um Cambissolo (297,3 Mg ha-1) e em um Argissolo (265,8 Mg ha-1). Foram elaborados três cenários para estimar o potencial de aumento de EC, a partir de mudanças nas práticas de manejo e de uso do solo. Em um cenário de adoção de SPD em 18 % das áreas de culturas anuais, o aumento do EC seria de 73,6 Gg. Se 15 milhões de hectares de pastagens degradadas forem recuperados com ILP, em 20% da área, e com ILPF, em 10% da área, o potencial de aumento do EC seria de 88,13 Gg. Esses valores somados representariam 593,6 Gg de CO2, que poderiam compensar os 472,1 Gg de CO2 emitidos pelo setor agropecuário brasileiro em 2010, conforme estimativas do Ministério da Ciência, Tecnologia e Inovação. Conclui-se que: i) na ausência de dados de DS, o uso de uma PTF simples é mais indicado do que o uso do modelo nulo; ii) os erros detectados na estimativa de DS por meio de PTF não se propagam no cálculo de EC; iii) as mudanças propostas compensariam emissões do setor agrícola brasileiro; e iv) as estimativas apresentadas ressaltam o papel do setor agropecuário na mitigação das emissões de GEE.

The most important characteristics of each soil is its hydro-physical properties, which include retention curve. Retention curve expresses the relationship between humidity and pressure head. It is course affected by many factors such as grain size and mineralogical composition, humus content, structure and bulk density of dry soil. This thesis deals with direct and indirect determination of one of the basic hydraulic characteristics of the soil, soil moisture retention curve. For laboratory determination of moisture retention curves were used for sand tank and pressure instruments. Measured body retention curves were optimized RETC program.For indirect determination of retention curves were used Pedotransfer function by Wösten et . al .(1998 ) , derived from the European soils and under HYPRES Šútora and Štekauerová (2000). As in the project Wösten et al . (1998), PTF were processed for arable land in the area. The accuracy of both models PTF was assessed by the correlation coefficient R and standard error RMSE. Simple models PTF provide estimates of retention curves with sufficient accuracy for practical use in the locality.