Submitted by (ana.araujo@ufrpe.br) on 2016-08-08T14:07:33Z No. of bitstreams: 1 Marystella Duarte Correia.pdf: 2726576 bytes, checksum: 4f62b9c923bf5dad5fab3bbed9657b6b (MD5) / Made available in DSpace on 2016-08-08T14:07:33Z (GMT). No. of bitstreams: 1 Marystella Duarte Correia.pdf: 2726576 bytes, checksum: 4f62b9c923bf5dad5fab3bbed9657b6b (MD5) Previous issue date: 2013-02-25 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Studies of some semiarid regions have pointed the beneficial effect of certain tree species in silvopastoral systems, by promoting the formation of soil fertility and increase the sustainability of agricultural productivity. In this study, we selected three tree species found in pastures of Grass buffel Cenchrus ciliaris, juazeiro Zyziphus joazeiro, umbuzeiro Spondias tuberosa and algaroba Prosopis juliflora caatinga, exclusively Brazilian biome, concentrated in the Northeast region of Brazil. The semivariogram is the central part of geostatistical studies, able to describe the spatial variation, besides being the key point in the data interpolation by kriging. Hence the importance of setting and selection of models. This study was conducted on an experimental farm in Custody silvopastoril PE, which we evaluated the differences in the chemical characteristics of the soil and herbaceous understory, between areas under and outside the canopy of these trees. Soil samples (0-15 cm) and the herbaceous layer were collected for evaluation of carbon storage and biomass of herbaceous and spatial dependence of these variables. Tree species were included in Algarobas cultivated fields, and the juazeiro umbuzeiro were already included on the site in order to explain whether the tree species or buffel grass can maintain and / or preserve soil carbon, also checking the growth of grass in region studies, classifying and mapping the spatial dependence is observed data with their probabilities. / Estudos de algumas regiões semiáridas têm apontado o efeito benéfico de certas espécies de árvores em sistemas silvopastoris, por promoverem a formação de fertilidade do solo e aumentarem a sustentabilidade da produtividade agrícola. Neste trabalho, foram selecionadas três espécies arbóreas encontradas em pastagens de Capim buffel (Cenchrus ciliaris), juazeiro (Zyziphus joazeiro), umbuzeiro (Spondias tuberosa) e algaroba (Prosopis juliflora) da caatinga, bioma exclusivamente brasileiro, concentrado na região Nordeste do Brasil. O semivariograma é a parte central dos estudos geoestatísticos, capaz de descrever a variação espacial, além de ser o ponto chave na interpolação dos dados por Krigagem. Daí a importância do ajuste e seleção dos modelos. O presente estudo foi conduzido em uma fazenda experimental silvopastoril em Custódia PE, onde foram avaliadas as diferenças nas características químicas do solo e do sub-bosque herbáceo, entre áreas embaixo e fora da copa destas árvores. Amostras de solo (0-15 cm) e do estrato herbáceo foram coletadas, para avaliação da conservação de carbono e biomassa de herbáceas, bem como a dependência espacial destas variáveis. As espécies arbóreas Algarobas foram incluídas nos campos cultivados, o juazeiro e umbuzeiro já estavam incluídas no local, a fim de explicar se as espécies arbóreas ou o Capim buffel consegue manter e/ou preservar o carbono no solo, verificando também o crescimento do capim na região de estudos, classificando-se a dependência espacial e mapeando-se os dados observados com suas probabilidades.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / O objetivo do trabalho foi caracterizar e determinar a variabilidade espacial de atributos fÃsicos de um Argissolo Amarelo, utilizando mÃtodos geoestatÃsticos e confecÃÃo de mapas de superfÃcie pelo mÃtodo da Krigagem, em duas camadas de solo. Foram realizadas coletas de solo em intervalos regulares de 2,5 m, numa Ãrea de 400 m2, sendo 81 pontos em cada uma das camadas (0 â 0,2 m e 0,2 â 0,4 m), totalizando 162 amostras que foram submetidas Ãs determinaÃÃes fÃsicas e os dados obtidos analisados pela estatÃstica descritiva para verificar a normalidade da distribuiÃÃo de freqÃÃncia e posteriormente submetidos a geoestatÃstica. O modelo que melhor se ajustou à curva do semivariograma, na maioria das variÃveis, foi o esfÃrico. No entanto, para a variÃvel densidade das partÃculas e agregados com diÃmetro 2,00-1,00 mm, na profundidade de 0 â 0,2 m e grau de floculaÃÃo, densidade do solo e umidade obtida com tensÃo 1,5 MPa, na profundidade de 0,2 â 0,4 m o modelo que melhor se ajustou foi o exponencial. O efeito pepita relativo para o grau de floculaÃÃo, densidade do solo e densidade das partÃculas foi de 100%, caracterizando o efeito pepita puro, confirmando a aleatoriedade dos dados. Os maiores alcances foram observados nas profundidades de 0,2 â 0,4 m para percentagem de silte (51,00 m) e percentagem de areia (35,06 m). Os valores da densidade do solo foram considerados relativamente altos sendo de 1,69 e 1,65 Mg.m-3 para 0 â 0,2 e 0,2 â 0,4 m, respectivamente. Houve uma maior retenÃÃo de umidade nas camadas inferiores. / The objective of this research was characterize and determine to spatial variability of physical attributes of a Alfisols, using methods geostatistical e and making of surface maps for the method of Kriging, in two soil depths. They were made soil collections in regular intervals of 2.5 m, in an area of 400 m2, being 81 points in each one of the layers (0 â 0.2 m and 0.2 â 0.4 m), to add 162 samples that were submitted to the physical determinations and the obtained data analyzed by the descriptive statistics to verify the normality of the frequency distribution and later on submitted to the geostatistical. The model that better it was adjusted to the curve of the semivariogram, in most of the variables, it was the spherical. However, for the variable density of the particles and aggregate with diameter 2.00-1.00 mm, in the depth of 0 â 0.2 m and flocculation degree, bulk density and soil-water content with tension 1.5 MPa, in the depth of 0.2 â 0.4 m the most appropriate model that better adjusted it was the exponential. The nugget ratio for the flocculation degree, bulk density and density of the particles was of 100%, characterizing ârandom varianceâ, confirming the independence of the data. The largest ranges were observed in the depths of 0.2 â 0.4 m for lime percentage (51.00 m) and percentage of sand (35.06 m). The values of the bulk density were considered relatively high being of 1.69 and 1.65 Mg.m-3 for 0 â 0.2 and 0.2 â 0.4 m, respectively. There was a larger humidity retention in the inferior layers.

Groundwater level models have an important role in the development and application of water management and policies. Understanding the temporal and spatial variations of groundwater levels in the Gorran water protection area in Nynäshamn is important for developing management strategies. Geostatistical analysis with several different methods was used to compare groundwater level records for 13 observation wells in Gorran from 1994 to 2012. The performance of the different methods was evaluated by using mean absolute error (MAE) and root mean square error (RMSE) metrics. The results showed that geostatistical methods had a higher general accuracy when utilizing the Empirical Bayesian kriging (EBK) method in Gorran. EBK was the best method with highest precision and lowest mean absolute error. Cross-validation was also applied to evaluate the best (smallest) root mean square error (RMSE). A predicted potentiometric groundwater level was estimated from the basis of the available digital elevation model to extend the observation area within the same geological specification. / Grundvattennivåns modeller har en viktig roll i utvecklingen och tillämpningen av vattenförvaltning och politik. Genom att förstå de tidsmässiga och geografiska variationer grundvattennivån i Gorran skyddat område är viktigt för att utveckla hanteringens strategier.Geostatistical analys med olika modeller som Inverse Distanse Weighted, Radial Basic Funktion och kriging modeller användes för att jämföra grundvattennivåns rekord under 13 observations brunnar 1994 till 2012. Prestandan hos metoder utvärderades med hjälp av medelabsolutfelet (MAE) och rot medelkvadratfelet (RMSE). Resultatet visade att geostatistiska metoder hade högre noggrannhet och Empirical Bayesian kriging (EBK) var den bästa metoden med högsta precision och lägst medelabsolutfel. Cross-validering applicerades också för att utvärdera de bästa minsta kvadratiska medelvärdet (RMSE). Förutsatt potentiometrisk grundvattennivå genomfördes från digital höjdmodell för att förlänga observations område inom samma geologiska specifikation.

En la minería actual, la mayor parte de los proyectos mineros contempla una etapa de exploración y evaluación de recursos. Se comienza con una toma de muestras, como lo es una campaña de sondajes para poder interpretar las litologías y alteraciones presentes, definir las zonas minerales más relevantes y describir la distribución de leyes de elementos de interés, de tal forma de llegar a un modelo de evaluación del yacimiento. Para determinar el potencial económico de un yacimiento existen herramientas y técnicas geoestadísticas de evaluación y modelamiento de recursos tales como el kriging o la simulación condicional. La primera permite hacer una estimación insesgada y precisa de las leyes, obteniéndose un modelo suavizado que no representa la variabilidad. La segunda posibilita realizar un estudio de la incertidumbre de los valores involucrados en los atributos geológico-minero-metalúrgicos y tener análisis de riesgos a considerar para la toma de decisiones. En esta memoria, se aplican simulaciones geoestadísticas para cuantificar los recursos minerales de un yacimiento perteneciente a Antofagasta Minerals SA (AMSA), ubicado en la Región de Antofagasta, Chile, el cual ha sido interpretado, recientemente en diez dominios minerales y evaluado por kriging a través de información de sondajes. Utilizando un algoritmo de simulación para variables categóricas, específicamente el algoritmo secuencial de indicador, se construye once modelos alternativos de dominios minerales. Luego se utiliza un algoritmo de simulación de variables continuas (secuencial gaussiano) para obtener once modelos de bloques para leyes de cobre total y cobre soluble, cada uno asociado con un modelo particular de dominios minerales. Se realiza la validación comparando las realizaciones con los datos originales, el modelo de kriging y el modelo geológico interpretado. Finalmente se construyen curvas tonelaje-ley para cuantificar los recursos para distintas leyes de corte. Considerando una ley de corte de 0.2% en cobre soluble, se obtiene para los modelos de simulaciones que, en promedio, hay 53 millones de toneladas con ley media de cobre soluble de 0.86%, en contraste con los 50 millones de toneladas de ley media 0.73% entregadas por el método de kriging, lo que entrega una diferencia aproximada de 90 mil toneladas de cobre fino. Así, existen zonas en el yacimiento donde la diferencia entre el modelo de kriging y los modelos simulados son relevantes. Debido a esto, es recomendable nuevas campañas de sondajes enfocadas en estas áreas para obtener información de mejor calidad que permita reducir la incertidumbre. También sería interesante considerar la mineralización de ganga como una de las variables a simular, para así tener modelos predictivos del comportamiento de cada bloque en los futuros procesos metalúrgicos.

Studies and surveys on climate phenomena serve to plan the geographical space. In this sense, research has mainly focused on spatialization of the drought in Rio Grande do Sul from 1981 to 2011 and the location of land reform settlements (PA). The social issue was viewed from the location of land reform settlements and their susceptibility to drought. The steps to performing the research consist of literature review, data surveys in primary and secondary sources, such as INCRA and INPE. For the treatment of information and preparation of final maps, the SPRING and the QGIS software were used, along with Ordinary Kriging geostatistical technique. A fieldwork was also carried out in selected settlements from the location in areas of high drought occurrence in order to verify the susceptibility of the settlers. The results are divided into two parts: use of GIS and geostatistical technique for obtaining the spatial maps of drought occurrence per decade and synthesis map of occurrence for the whole period of analysis and the PA location map per drought occurrence class. The second part is the analysis and discussion of the results obtained in maps and tables with information about the settlements, in order to support the territorial planning. / Estudos e levantamentos sobre os fenômenos climáticos servem para planejar o espaço geográfico. Nesse sentido, a pesquisa tem como objetivo central a espacialização da estiagem no Rio Grande do Sul no período de 1981 a 2011 e a localização dos assentamentos da reforma agrária. Abordou-se a questão social a partir da localização dos assentamentos da reforma agrária e sua vulnerabilidade a estiagem. As etapas de realização da pesquisa consistem em revisão bibliográfica, levantamentos de dados em fontes primárias e secundárias, como INCRA e INPE. Para o tratamento das informações e elaboração dos mapas finais utilizou-se os softwares SPRING, QGIS, e técnica geoestatística da Krigagem Ordinária. Também realizou-se trabalho de campo em assentamentos selecionados a partir da localização em área de alta ocorrência de estiagem com a finalidade de verificar a suscetibilidade dos assentados. Os resultados dividem-se em duas partes: utilização de SIG e técnica geoestatística para a obtenção dos mapas de espacialização de ocorrência da estiagem por década e mapa síntese da ocorrência em todo o período de análise e o mapa de localização dos PAs por classe de ocorrência da estiagem. A segunda parte consiste na análise e discussão dos resultados obtidos nos mapas e tabelas com informações sobre os assentamentos, com a finalidade de subsidiar o planejamento territorial.

Motivés par des besoins en industrie microélectronique, ces travaux apportent des contributions en modélisation probabiliste de données spatiales, et en maîtrise statistique de procédés.Le problème spatial a pour spécificité d’être posé sur un domaine circulaire. Il se représente par un modèle de krigeage dont la partie déterministe est constituée de polynômes orthogonaux et la partie stochastique de processus gaussiens. Traditionnellement définis avec la norme euclidienne et la mesure uniforme sur le disque, ces choix n’exploitent pas les informations a priori sur les procédés d’usinage.Pour tenir compte des mécanismes de rotation ou de diffusion à partir du centre, nous formalisons les processus gaussiens polaires sur le disque. Ces processus intègrent les corrélations radiales et angulaires dans le modèle de krigeage, et en améliorent les performances dans les situations considérées. Ils sont ensuite interprétés par décomposition de Sobol et généralisés en dimension supérieure. Des plans d’expériences sont proposés dans le cadre de leur utilisation. Au premier rang figurent les cylindres latins qui reproduisent en coordonnées polaires les caractéristiques des hypercubes latins.Pour intégrer à la fois les aspects spatiaux et temporels du problème industriel, la maîtrise statistique de procédé est abordée en termes d’application de cartes de contrôle aux paramètres des modèles spatiaux. Les séries temporelles suivies ont aussi la particularité de comporter des données atypiques et des changements structurels, sources de biais en prévision, et de fausses alarmes en suivi de risque. Ce problème est traité par lissage robuste et adaptatif. / Driven by industrial needs in microelectronics, this thesis is focused on probabilistic models for spatial data and Statistical Process Control. The spatial problem has the specificity of being defined on circular domains. It is addressed through a Kriging model where the deterministic part is made of orthogonal polynomials and the stochastic term represented by a Gaussian process. Defined with the Euclidean distance and the uniform measure over the disk, traditional Kriging models do not exploit knowledge on manufacturing processes. To take rotations or diffusions from the center into account, we introduce polar Gaussian processes over the disk. They embed radial and angular correlations in Kriging predictions, leading to significant improvements in the considered situations. Polar Gaussian processes are then interpreted via Sobol decomposition and generalized in higher dimensions. Different designs of experiments are developed for the proposed models. Among them, Latin cylinders reproduce in the space of polar coordinates the properties of Latin hypercubes. To model spatial and temporal data, Statistical Process Control is addressed by monitoring Kriging parameters, based on standard control charts. Furthermore, the monitored time – series contain outliers and structural changes, which cause bias in prediction and false alarms in risk management. These issues are simultaneously tackled with a robust and adaptive smoothing.

It is well established that acquiring large amounts of range data with vision sensors can quickly lead to important data management challenges where processing capabilities become saturated and pre-empt full usage of the information available for autonomous systems to make educated decisions. While sub-sampling offers a naïve solution for reducing dataset dimension after acquisition, it does not capitalize on the knowledge available in already acquired data to selectively and dynamically drive the acquisition process over the most significant regions in a scene, the latter being generally characterized by variations in depth and surface shape in the context of 3D imaging. This thesis discusses the development of two formal improvement measures, the first based upon surface meshes and Ordinary Kriging that focuses on improving scene accuracy, and the second based upon probabilistic occupancy grids that focuses on improving scene coverage. Furthermore, three selection processes to automatically choose which locations within the field of view of a range sensor to acquire next are proposed based upon the two formal improvement measures. The first two selection processes each use only one of the proposed improvement measures. The third selection process combines both improvement measures in order to counterbalance the parameters of the accuracy of knowledge about the scene and the coverage of the scene. The proposed algorithms mainly target applications using random access range sensors, defined as sensors that can acquire depth measurements at a specified location within their field of view. Additionally, the algorithms are applicable to the case of estimating the improvement and point selection from within a single point of view, with the purpose of guiding the random access sensor to locations it can acquire. However, the framework is developed to be independent of the range sensing technology used, and is validated with range data of several scenes acquired from many different sensors employing various sensing technologies and configurations. Furthermore, the experimental results of the proposed selection processes are compared against those produced by a random sampling process, as well as a neural gas selective sensing algorithm.

The broader research area investigated during this programme was conceived from a goal to contribute towards solving the challenge of food security in the 21st century through the reduction of crop loss and minimisation of fungicide use. This is aimed to be achieved through the introduction of an empirical approach to agricultural disease monitoring. In line with this, the SYIELD project, initiated by a consortium involving University of Manchester and Syngenta, among others, proposed a novel biosensor design that can electrochemically detect viable airborne pathogens by exploiting the biology of plant-pathogen interaction. This approach offers improvement on the inefficient and largely experimental methods currently used. Within this context, this PhD focused on the adoption of multidisciplinary methods to address three key objectives that are central to the success of the SYIELD project: local spore ingress near canopies, the evaluation of a suitable model that can describe spore transport, and multivariate analysis of the potential monitoring network built from these biosensors. The local transport of spores was first investigated by carrying out a field trial experiment at Rothamsted Research UK in order to investigate spore ingress in OSR canopies, generate reliable data for testing the prototype biosensor, and evaluate a trajectory model. During the experiment, spores were air-sampled and quantified using established manual detection methods. Results showed that the manual methods, such as colourimetric detection are more sensitive than the proposed biosensor, suggesting the proxy measurement mechanism used by the biosensor may not be reliable in live deployments where spores are likely to be contaminated by impurities and other inhibitors of oxalic acid production. Spores quantified using the more reliable quantitative Polymerase Chain Reaction proved informative and provided novel of data of high experimental value. The dispersal of this data was found to fit a power decay law, a finding that is consistent with experiments in other crops. In the second area investigated, a 3D backward Lagrangian Stochastic model was parameterised and evaluated with the field trial data. The bLS model, parameterised with Monin-Obukhov Similarity Theory (MOST) variables showed good agreement with experimental data and compared favourably in terms of performance statistics with a recent application of an LS model in a maize canopy. Results obtained from the model were found to be more accurate above the canopy than below it. This was attributed to a higher error during initialisation of release velocities below the canopy. Overall, the bLS model performed well and demonstrated suitability for adoption in estimating above-canopy spore concentration profiles which can further be used for designing efficient deployment strategies. The final area of focus was the monitoring of a potential biosensor network. A novel framework based on Multivariate Statistical Process Control concepts was proposed and applied to data from a pollution-monitoring network. The main limitation of traditional MSPC in spatial data applications was identified as a lack of spatial awareness by the PCA model when considering correlation breakdowns caused by an incoming erroneous observation. This resulted in misclassification of healthy measurements as erroneous. The proposed Kriging-augmented MSPC approach was able to incorporate this capability and significantly reduce the number of false alarms.

Orientador: Alexandre Campane Vidal / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Geociências / Made available in DSpace on 2018-08-24T10:58:04Z (GMT). No. of bitstreams: 1 Costanzo_CaetanoPontes_M.pdf: 3384236 bytes, checksum: 7b4516eef162c87ff3e03f0d56f08215 (MD5) Previous issue date: 2014 / Resumo: Em face da crescente demanda de execução de projetos de remediação de aquíferos contaminados, a técnica de bombeamento e tratamento está sendo utilizada de forma significativa. Esta ferramenta de recuperação ambiental consiste no bombeamento da água subterrânea em pontos estrategicamente posicionados de maneira a gerar uma barreira hidráulica que impeça o avanço da extensão de uma pluma de contaminação. Em áreas com alta heterogeneidade litológica, a organização espacial dos valores de condutividade hidráulica (K) em subsuperfície apresenta incertezas associadas à variabilidade na distribuição das fácies. No presente estudo foram analisadas estas incertezas por meio de métodos geoestatísticos (krigagem da indicatriz e simulações estocásticas) e também a partir de simulações numéricas de fluxo e transporte de contaminantes. Estas ferramentas permitiram a análise das possibilidades de deslocamentos espaciais das plumas de contaminação, bem como possíveis cenários de bombeamento e tratamento numa área contaminada, em função de distintos campos aleatórios de condutividade locais, devido à complexidade geológica. A partir da krigagem da indicatriz foi possível elaborar um modelo geoestatístico de fácies que foi utilizado como ferramenta para validar o uso da condutividade hidráulica em função de cada litologia, definindo assim os possíveis grupos de hidrofácies presentes na área. Por meio das simulações estocásticas foram gerados vinte campos distintos de condutividade hidráulica, os quais configuraram distintas distribuições espaciais das áreas das plumas de contaminação, como também cenários otimista, intermediário e pessimista para o deslocamento das mesmas. Foram simulados três cenários de bombeamento e tratamento resultando nove situações desta técnica de remediação. Dessa forma, os efeitos da heterogeneidade atrelada à condutividade hidráulica evidenciaram que, para a execução de um projeto de remediação por bombeamento e tratamento em áreas com certa heterogeneidade geológica, é necessária a análise de incertezas atreladas ao diagnostico hidrogeológico. Ao considerar modelos com homogeneidade na condutividade e mais simplistas, o projeto de remediação pode vir a ser ineficiente aumentando o tempo para a recuperação ambiental local e consequentemente aumentando os custos / Abstract: Due to the growing demand for project execution remediation of contaminated aquifers, the pump-and-treat is being used significantly. This remediation technique consists in groundwater pumping in strategically placed at points to generate a hydraulic barrier to prevent the advance of a contamination plume. In areas with high lithological heterogeneity, the spatial organization of the hydraulic conductivity (K) values in the subsurface introduces uncertainties associated with the variability in the distribution of facies. In this study these uncertainties were evaluated using geostatistical methods (indicator kriging and stochastic simulations) and also from numerical simulations of flow and contaminant transport. These tools enable the analysis of the contamination plumes areas, as well as scenarios for pump-and-treat, due to different random fields of local conductivity according to the geological complexity. By indicator kriging, was possible to provide a geostatistical facies model which was used as a tool to validate the hydraulic conductivity as a function of each lithology, thus defining the possible hydrofacies groups in the area. The stochastic simulations generated twenty distinct hydraulic conductivity fields, which represented different spatial distributions of contamination plumes, as well as scenarios optimistic, intermediate and pessimistic to offset these. Three scenarios for pump-and-treat were simulated resulting nine situations of this remediation technique. Therefore the effects of heterogeneity associated with hydraulic conductivity showed that, for the remediation design in areas with certain geological heterogeneity, the analysis of uncertainties linked to hydrogeological studies is needed. When considering models with homogeneous conductivity and most simplistic, the remediation project may prove to be inefficient increasing the time to the project and consequently increasing costs / Mestrado / Geologia e Recursos Naturais / Mestre em Geociências

Continuous and reliable monitoring of contaminants in drinking water, which adversely affect human health, is the main goal of the Broward County Well Field Protection Program. In this study the individual monitoring station locations were used in a yearly and quarterly spatiotemporal Ordinary Kriging interpolation to create a raster network of contaminant detections. In the final analysis, the raster spatiotemporal nitrate concentration trends were overlaid with a pollution vulnerability index to determine if the concentrations are influenced by a set of independent variables. The pollution vulnerability factors are depth to water, recharge, aquifer media, soil, impact to vadose zone, and conductivity. The creation of the nitrate raster dataset had an average RMS Standardized error close to 1 at 0.98. The greatest frequency of detections and the highest concentrations are found in the months of April, May, June, July, August, and September. An average of 76.4% of the nitrate intersected with cells of the pollution vulnerability index over 100.