This thesis aims to increase our understanding of mechanisms that influence larval dispersal in marine benthic invertebrates, particularly in the absence of strong oceanographic features (e.g. estuarine plumes, upwelling events, or markedly different water masses). Laboratory experiments identified behavioural mechanisms that regulate the vertical distribution of larvae in response to thermal stratification, and field studies in St. George’s Bay, Nova Scotia (NS), Canada, examined the relationship between larval abundance and physical variables (temperature, salinity, fluorescence, etc) and identified mechanisms that regulate larval distributions in situ. In the laboratory, I demonstrated that thermal stratification affects the vertical distribution of larvae by acting as a barrier to migration, or through temperature-dependent vertical swimming velocities. I also developed a random walk based model which highlighted that the key to successfully simulating larval response to temperature was 1) determining the temperature-dependent distribution of vertical swimming velocities and 2) the temporal autocorrelation in these velocities. In the field, the most striking pattern was that the larval distributions for species with similar swimming abilities were significantly correlated to one another at all scales (0.5 to 40 km). This suggests a common mechanism, related to larval swimming ability, which greatly influences the horizontal larval distribution. I found that the spatial scale of variability in larval distributions (~ 3 km) matches that in both the environmental variables and of coherent structures in current velocities (i.e. the tidal excursion). Results from an aggregation-diffusion model suggest that horizontal larval swimming could not be responsible for the observed level of aggregation in the larval horizontal distributions. I suggest that these horizontal patterns are the result of 1) an aggregative process (i.e. larvae swimming against a vertical current and maintaining their vertical position) and 2) a diffusive process which scales the aggregations to the scale of the coherent structures in current velocity (i.e. tidal excursion). In conclusion, this thesis increases our understanding of larval behaviour and its effects on larval dispersal. The results will be particularly useful to those who are interested in mechanisms regulate population connectivity, particularly those using bio-physical models to model dispersal trajectories.

The consequences of land use choices on soil water and gas transport properties are significant for gas and water flux in agricultural environments. Spatial and temporal patterns and associations of soil water and soil gas characteristics and processes in different land uses are not well understood. The objectives of this study were to 1) characterize soil structure under crop and grass systems, 2) quantify spatial patterns and associations of soil physical characteristics in crop and grass systems, and 3) quantify spatial and temporal patterns and associations of CO2 and N2O fluxes. The research was conducted in a 60 by 80 m field divided into grass and crop systems. Sixty sampling points were distributed in four transects with 5- and 1-m spatial intervals between measurement points. Gas fluxes were measured, at two-week time intervals, 22 times during a year. Pore size distribution was more homogeneous and more continuous pores were found in the grass than in the crop system. The spatial variability of most selected soil physical characteristics was more structured in the crop than in the grass system, which reflected the impact of land use and soil structure on their spatial patterns. CO2 flux was dependent for a longer distance in the grass than in the crop system, however, the two land-use systems exhibited similar spatial ranges of N2O flux. Gas fluxes were temporally dependent for a longer period in the grass than in the crop system. The spatial associations between CO2 and N2O fluxes and selected biochemical and physical factors depended on the flux sampling season and land use. Soil temperature was the dominant controlling factor on the temporal variability of CO2 and N2O fluxes but not on the spatial behavior. Considering the spatial and temporal ranges and dependency strength of soil variables helps identify efficient sampling designs that can result in better time and resource management. Spatial and temporal relationships between the selected soil variables also improve understanding soil management and sampling soil variables. This study provides the baseline and recommendations for future investigations specifically for sampling designs, soil management, and predictions of different soil processes related to gas fluxes.

Frequent use of phenylurea herbicide isoproturon (IPU) in agricultural fields has resulted not only in the contamination of the natural resources including soil and water but also in the adaptation of the soil microflora to its rapid degradation. However, up to now, the mechanisms underlying this microbial adaptation are not well elucidated. The aim of this study was to explore the processes and factors implicated in IPU degradation from the agricultural field to the genes coding for catabolic genes. The study carried out at the experimental field of Epoisses cropped with a winter wheat / barley / rape seed crop rotation indicated that as a result of its periodically repeated use, the soil microflora adapted to IPU mineralization activity. Further analysis using exploratory and geostatistical tools demonstrated the existence of spatial variability in IPU mineralization activity at the field scale which was correlated not only with several soil physico-chemical parameters like organic matter content, CEC and C/N ratio but also with the pesticide application plan over a three year crop rotation. In order to get further insight into underlying mechanisms, an IPU mineralizing bacterial culture and strain Sphingomonas sp. SH were isolated through enrichment cultures performed from two different adapted soils. Both had the catabolic activities highly specific for the mineralization of IPU and its metabolites but none of other structurally related phenylurea herbicides. IPU metabolic activity of both the mixed culture and the strain SH was found to be affected by pH with optimal activity taking place at pH 7.5. Based on the accumulation of different known metabolites during mineralization kinetics, IPU metabolic pathway was proposed to be initiated by two successive demethylations, followed by cleavage of the urea side chain resulting in the accumulation of 4-isopropylaniline, and ultimately the mineralization of the phenyl ring. In order to identify the genes involved in IPU degradation, BAC clone library was established from the genomic DNA of the bacterial culture. Although, the functional screening did not yield in identifying any BAC clone able to degrade IPU or its known metabolites, the PCR based screening led us to identify a cat gene cluster involved in ortho-cleavage of the phenyl ring of catechol through beta-ketoadipate pathway. Based on this finding, it was hypothesized that phenyl ring of 4-isopropylaniline formed during IPU transformation might be mineralized through ortho-cleavage of catechol. This finding allowed us to propose the lower IPU metabolic pathway which was not yet described.

Masters Research - Master of Philospohy (MPhil) / It has been shown that masonry material properties, in particular, unit flexural bond strength (ft), vary significantly throughout masonry structures, despite the fact that often only one type of brick and mortar are used. Unit flexural bond strength was previously identified as one of the most important material parameters contributing to the strength of clay brick unreinforced masonry (URM) walls in flexure. It was the objectives of this research, in the context of clay brick URM walls subject to vertical bending, to examine how unit flexural bond strength varied spatially in a clay brick URM wall, determine a best fit probability distribution function which can describe expected variability in unit flexural bond strength and determine how this variability and other factors affect wall behaviour and failure load using 3D non-linear finite element analysis (FEA). It was hoped that modelling a full sized clay brick URM wall subject to vertical bending using a 3D non-linear FEA model would more accurately predict wall failure load (compared to current analytical methods) and allow the examination of crack pattern development as the wall progresses to failure upon being laterally loaded. The first part of the research project was to conduct an experimental program to examine unit-to-unit spatial strength correlation within six full sized clay brick URM walls and to characterise a unit flexural bond strength probability distribution. It was observed that although weak correlation in unit flexural bond strength exists in some courses and between courses, these locations were difficult to predict and didn����t follow any particular pattern relating to for example, mortar batch. Therefore, although somewhat counter-intuitive, the results indicate that statistically significant correlation between adjacent unit flexural bond strengths is not likely to be observed. It was also observed that clay brick wall unit flexural bond strengths obtained for all of the walls tested best fit a truncated Normal probability distribution. Strength of the brick/mortar interface appeared to be governed by factors relating to workmanship (and therefore mortar quality and moisture content), weather (which can affect material characteristics like brick suction rate) and inherent material variability. It would appear that brick suction rate can significantly affect the overall strength of a URM wall. v Stochastic analysis was conducted for walls with and without uncorrelated spatial variability in unit flexural bond strength and associated tensile fracture energy (GfI ). It was found that the TNO DIANA 9.2 FEA package could be used to implement spatial variability of various material parameters and reasonably accurately model failure of clay brick URM walls in vertical bending. From the non-linear FEA model development stage, it was observed that because the brick/mortar bond has significantly more strength capacity in compression, it appears that the lateral load resistance of the wall comes from a combination of the ability of the brick/mortar bond to tensile soften while providing significant compressive resistance at the compressive edge. It was found for a spatial stochastic analysis with spatial variability in bond strength (referred to from now on as a spatial stochastic analysis), with COVs of 0.1, 0.3 and 0.5, that COV of wall failure loads were relatively small, being 0.02, 0.04 and 0.06 respectively. For the non-spatially varying stochastic analysis with fully correlated bond strength (now referred to as non-spatial stochastic analysis), with COVs of 0.1, 0.3 and 0.5, COV of wall failure loads were 0.07, 0.20 and 0.32 respectively. For the spatial stochastic analysis, it was found that with a bond strength COV increase from 0.1 to 0.5 the mean wall failure load dropped from 2.25 kPa to 2.0 kPa (an 11% reduction). Despite the relatively small drop in magnitude of the mean wall failure load with increase in bond strength COV, the mean wall failure loads were statistically different to one another. For the non-spatial stochastic analysis, mean failure load stayed relatively constant at 2.24-2.25 kPa. These results could be explained by examining the 3D wall progression to failure. For walls with spatial variability in bond strength, it is expected that wall failure load COVs would be smaller because those walls would consistently be composed of smaller valued bond strengths which would consistently contribute to weakness in the wall. For the non-spatial wall simulations, this effect would not occur as failure load is determined by one uniform weak or strong bond strength. It was proposed that failure of a clay brick URM wall is not governed by one course only cracking, but rather, instability in the wall is governed by several courses in the vicinity of locations of large bending moment. It was shown that various current stochastic approximations which employ a unit failure hypotheses in combination with a linear/elastic approximation for first cracking load all underestimated wall capacity significantly. The reason for this is suggested as being vi because all hypotheses only assume failure is governed by one course and linear/elastic theory only considers the tensile capacity of a joint and neglects strength capacity available as a result of joint tension softening and the resistance to failure provided by compressive strength on the compression side of the wall. The hypotheses also don’t take into consideration factors which affect overall wall bond strength mean which result from influences such as workmanship, weather and material variability factors, such as (for example), variation in brick suction rate due to weather conditions which can make the overall strength of the wall stronger or weaker. Based upon a comparison in wall failure load COV for the spatial and non-spatial stochastic wall analysis results, a more realistic approach for future modelling attempts of spatial variability in masonry material properties is suggested. This would address the issue of external factors such as workmanship and weather on the overall strength of the wall, as well as the inherent bond strength variability due to material variability. For walls with spatial variability in bond strength, upon examination of numerous wall simulation results, several crack patterns were witnessed and are discussed.

Submitted by Bruna Rodrigues (bruna92rodrigues@yahoo.com.br) on 2017-10-04T13:00:24Z No. of bitstreams: 1 DissRSL.pdf: 2218443 bytes, checksum: 6c095cccc15f22c3fe747197514a5b16 (MD5) / Approved for entry into archive by Ronildo Prado (producaointelectual.bco@ufscar.br) on 2017-10-11T11:44:18Z (GMT) No. of bitstreams: 1 DissRSL.pdf: 2218443 bytes, checksum: 6c095cccc15f22c3fe747197514a5b16 (MD5) / Approved for entry into archive by Ronildo Prado (producaointelectual.bco@ufscar.br) on 2017-10-11T11:44:30Z (GMT) No. of bitstreams: 1 DissRSL.pdf: 2218443 bytes, checksum: 6c095cccc15f22c3fe747197514a5b16 (MD5) / Made available in DSpace on 2017-10-11T11:52:44Z (GMT). No. of bitstreams: 1 DissRSL.pdf: 2218443 bytes, checksum: 6c095cccc15f22c3fe747197514a5b16 (MD5) Previous issue date: 2017-07-05 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Throughfall (TF) is affected by different biotic and meteorological factors that can interact, resulting in high variability in estimates, increasing the uncertainty in hydrological models and the forecast of environmental impacts in watersheds. The knowledge of the relationship between rainfall (RF) and TF is a key factor for the definition of sampling strategies that result within commonly accepted error thresholds. Therefore, the main objective of this study was to analyze sources of TF variability in a secondary forest in SE Brazil, in order to understand the response of TF in function of different environmental drivers and sampling strategies. To attain these objectives, the influences of RF, sampling strategies, and stand structure on TF were analysed. The results indicated significant influence of RF on the response variables [TF volume (TFmm), TF fraction (TF%), and coefficient of variation of TF (CVTF)]. Whereas the linear model showed the best fit for TFmm, nonlinear models presented better fits for TF% and CVTF in response to RF. Regarding the influence of sampling strategies, there were no significant effects on average estimates of TFmm and CVTF. Based on the magnitude of the error, only treatments with 16 fixed gauges and nine roving gauges were able to provide estimates with errors smaller than 10%, both for all rainfall events and for those higher than 10.44 mm (median value). The results suggest that different vegetation structures affect TFmm at different spatial scales. For instance, a significant effect of canopy cover was observed on the point estimates of TFmm. However, no vegetation influences on this variable were observed at the plot scale. Regarding the temporal variability, significant effects were verified at two scales: distance to the nearest trunk at the point scale; and metrics of trees greater than 20 m at the plot scale. Finally, it was verified that combined effects of tree sizes and canopy cover influenced CVTF, indicating that the these variables reflect vegetation complexity. In general, the results suggest that RF is the main source of TF variability in the studied area. Therefore, it is necessary to consider both meteorological factors and characteristics of rainfall events to sample TF with adequate accuracy. Nevertheless, the interactions between biotic and meteorological factors affect their relative influences on TF, highlighting the need for further research in this area. / A precipitação interna (PI) é afetada por diferentes fatores bióticos e meteorológicos que podem interagir entre si, resultando numa elevada variabilidade em suas estimativas, aumentando a incerteza em modelos hidrológicos e na previsão de impactos ambientais em bacias hidrográficas. A compreensão da relação entre a precipitação em aberto (PA) e a PI é importante para a definição de estratégias de amostragem com limiares de erros aceitáveis. Portanto, o objetivo geral deste estudo foi analisar os fatores de variação da PI em uma floresta secundária, localizada no sudeste brasileiro, a fim de compreender o comportamento desse processo em função de diferentes fatores ambientais e de estratégias de amostragem. Para isto, foram analisadas as influências da PA, das estratégias de amostragem e da estrutura da vegetação sobre a PI. Os resultados indicaram influência significativa da PA sobre as variáveis resposta [volume de PI (PImm), fração de PI (PI%) e coeficiente de variação da PI (CVPI)]. Enquanto o modelo linear apresentou o melhor ajuste para PImm, os modelos não lineares tiveram melhores ajustes para PI% e CVPI em função de PA. Quanto à influência das estratégias de amostragem, não houve efeitos significativos sobre as estimativas médias de PImm e CVPI. Com base na magnitude do erro, apenas tratamentos com 16 pluviômetros fixos e nove rotativos foram capazes de fornecer estimativas com erros inferiores a 10%, tanto para o conjunto completo de dados quanto para chuvas superiores a 10,44 mm (valor mediano). Os resultados sugerem que diferentes estruturas da vegetação afetam a PImm em diferentes escalas espaciais. Por exemplo, foi verificado efeito significativo da cobertura do dossel sobre as estimativas pontuais de PImm. No entanto, não foram verificadas influências da vegetação sobre esta variável na escala da parcela. Em relação à variabilidade temporal, verificaram-se efeitos significativos nas duas escalas: da distância para o tronco mais próximo na pontual; e de métricas de árvores maiores que 20 m na escala parcela. Por fim, verificou-se que os efeitos combinados do tamanho das árvores e da cobertura do dossel influenciaram CVPI, indicando que essas variáveis refletem a complexidade da vegetação. Em geral, os resultados sugerem que PA é a principal fonte de variabilidade de PI na área estudada. Portanto, é necessário considera características meteorológicas e dos eventos de chuva para amostrar a PI com precisão adequada. Contudo, as interações entre fatores bióticos e meteorológicos afetam suas influências relativas sobre a PI, destacando a necessidade de mais pesquisas nessa área.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-06-15T19:05:07Z No. of bitstreams: 1 ARMINDO BEZERRA LEÃO - DISSERTAÇÃO PPGEA 2006..pdf: 18036151 bytes, checksum: 3c7202f8b76aca90d252e37405bcede3 (MD5) / Made available in DSpace on 2018-06-15T19:05:07Z (GMT). No. of bitstreams: 1 ARMINDO BEZERRA LEÃO - DISSERTAÇÃO PPGEA 2006..pdf: 18036151 bytes, checksum: 3c7202f8b76aca90d252e37405bcede3 (MD5) Previous issue date: 2006-05 / O solo pode apresentar ampla variação dos seus atributos, estando esta relacionada a fatores ligados à sua formação ou manejo de práticas agrícolas, a exemplo da irrigação que tem, como efeito nocivo se mal empregada, problemas de salinidade. A estatística clássica considera que a variabilidade do solo ocorre de forma inteiramente aleatória; entretanto, vários estudos têm revelado que seus atributos apresentam grande dependência espacial necessitando, portanto, de uma análise geoestatística Neste trabalho se propôs como objetivo geral, estudar as variáveis que caracterizam a salinidade do solo através da estatística clássica e geoestatística e, como metas específicas avaliar a variabilidade espacial, a existência de dependência espacial entre amostras, a validação cruzada do método de interpolação utilizado (krigagen) e construir mapas de isolinhas das variáveis da área em estudo. O estudo foi conduzido no perímetro Irrigado Engenheiro Arcoverde, localizado no Município de Condado, PB, o qual se situa na Bacia do Alto Piranhas, cuja área experimental tinha 42 hectares e se constituía de solos Neossolos Flúvicos, onde se coletaram 53 amostras, nas profundidades de 0-20, 20-40 e 40-60 cm. A malha de amostragem foi irregular, os pontos de coleta se distanciaram de lOOm entre si. De início se utilizaram para avaliação da variabilidade espacial, métodos de estatística clássica a fim de se verificar medidas de posição, dispersão e geostatística, analisando-se os variogramas e se realizando, também, a validação cruzada como meio de conhecer os melhores ajustes para os modelos encontrados. Após a realização de interpolação por krigagem, construíram-se mapas de isolinhas revelando os problemas de salinidade e sodicidade, em que os resultados obtidos indicaram grande variabilidade justificando, assim, o uso da geoestatística, principalmente para os atributos CE e PST do solo, reconhecendo-se a geoestatística como uma poderosa ferramenta indicadora da dependência espacial entre amostras e a distância, a partir dos quais elas se tornam independentes. Os atributos estudados, indicadores de salinidade, apresentaram estrutura de dependência espacial, o que permitiu o seu mapeamento. A validação cruzada demonstrou forte relação entre os valores observados e os estimados, comprovando que a Krigagem ordinária é um bom interpolador. A identificação de compartimentos como sub-regiões da paisagem na área de estudo, mostrou-se eficiente na caracterização da salinidade do solo. A técnica de interpolação por krigagem permitiu verificar que a salinidade do solo aumentou em profundidade, além de distribuição diferenciada dos sais na área, com maiores teores em sub-região específica. / The soil atributes may present a great variability, due to formation and management factores, such as an inadequate irrigation that coutj produce salinity problems. The classical statistics consider that soil variability occurs in a randomized form, however several studies have shown that the attributes present an strong spatial dependence, nedding thus geoestatical analises. The objective of this work was to study the variables that characterize the soil salinity by classical statistics and geostatistics and also to evaluate the spatial variability, the spatial dependence amongst samples, to verify the crossed validation of the interpolation method (Kriging) and to build maps of isolines of the ali variables. The present work was conducted at the Engenheiro Arcoverde irrigated perimeter located at Condado, Paraíba, aiming to measure the soil parameters that define the soil salinity, to study the spatial variability of them throughout geoestatical analyses, to study the liability of the Kriging interpolation method and to construct salinity and sodicity maps, furnishing subsides for an adequate soil management which could allows their reclamation. For this, 159 soil samples were collected at 0-20, 20-40 and 40-60 cm depth intervals, on an irregular grid with the sampling points separated at approximately 100 m. To evaluate the spatial variability initially it was used classical statistics to verify position and dispersion measures and afterwards geoestatistics to analise the semivariograms, crossed validation and to obtain the best adjusted models. After interpolation by kriging, isoline maps were constructed showing the salinity and sodicity situation of the area. The results obtained allowed to observe a great spatial variability of eletrical conductivity and exchange sodium percentage. The pH variable showed a low spatial variability. The spatial dependence of the studied attributes permitted their mapping using kriging techniques. The crossed validation offered an excellent precision to estimate data. The compartments identification, sub-regions of the studied area, was excellent to characterize the soil salinity. Small variations on the soil relief, on the soil formation and management conditioned the variability found for the studied attributes. Finally, the geoestatistics showed the possibility to obtain a greater volume of information from a small volume of data, which means efficience, saving of time and resource

La variabilité spatiale des formations géologiques rend difficile la détermination des paramètres géotechniques nécessaires pour l'évaluation des aléas naturels (sismiques et gravitaires). Les méthodes d'imagerie géophysique, non-destructives et rapides, sont de plus en plus utilisées pour la reconnaissance de telles structures hétérogènes. Une campagne d'essais géophysiques et géotechniques a été réalisée dans la plaine alluviale de Beyrouth (Liban), ville soumise à un fort risque sismique, en vue de caractériser la variabilité des couches alluviales. Les résultats combinés ont permis de caractériser la structure 3D du site et de détecter la présence d'une couche d'argile molle, peu profonde et d'épaisseur variable. Cette couche de faible compacité, qui a rendu complexe l'interprétation des courbes de dispersion des ondes de surface, a une influence importante sur la réponse sismique du site. Les incertitudes reliées à la variabilité spatiale des propriétés géotechnique (N60) et géophysiques (Vs, ρ) ont pu être quantifiées au sein des couches rencontrées et les fonctions de distribution de ces paramètres ont été déterminées dans chaque couche à partir de ces essais, à la fois verticalement et horizontalement. Les valeurs de distance d'autocorrélation verticale (Vs, N60) et horizontale (ρ) obtenues ainsi que les valeurs de coefficient de variation se situent dans la gamme de valeurs trouvées dans la littérature. La réponse dynamique (amplification spectrale) de la plaine alluviale de Beyrouth a été simulée avec des modèles probabilistes unidimensionnels, et l'effet des trois paramètres statistiques (loi d'autocorrélation, distance d'autocorrélation et coefficient de variation) décrivant les variabilités des propriétés élastiques du sol (Vs) a été quantifié. Pour obtenir des réponses sismiques réalistes, un critère de sélection des profils Vs générés de façon probabiliste a été introduit afin de ne retenir que les profils compatibles (dans une gamme d'incertitude) avec la courbe de dispersion établie. Les modélisations probabilistes ont montré des différences significatives par rapport aux modélisations déterministes. Le principal paramètre probabiliste contrôlant l'amplification spectrale est le coefficient de variation, suivi de la distance d'autocorrélation, alors que le type de loi a peu d'influence. Enfin, nous avons vu que l'activité humaine peut avoir une influence significative sur l'application des méthodes géophysiques en site urbain. La compréhension d'un milieu complexe dans ce contexte nécessite de combiner toutes les méthodes géotechniques et géophysiques d'investigation afin d'obtenir un modèle robuste 2D/3D de la structure du sol / The spatial variability of geological formations makes it difficult to determine the geotechnical parameters necessary for the evaluation of natural hazards (seismic and gravity). The geophysical imaging methods, non-destructive and fast, are now increasingly used for heterogeneous structures of sub-surface recognition. Geophysical and geotechnical tests were carried out in the alluvial plain of Beirut (Lebanon), city with high seismic risk, to characterize the variability in the alluvial layers. Analyses of these tests were used to characterize the 3D structure of the site and to detect the presence of a shallow soft clay layer of variable thickness. This layer of low compactness, which made the interpretation of dispersion curves of surface waves complex, could be of prime importance for seismic response of the site. Using all the collected data, the uncertainties related to the spatial variability of geotechnical (N60) and geophysical (Vs, ρ) properties of soil were quantified in the layers encountered and the distribution functions of these parameters were determined in each layer, in both directions (vertical and horizontal). The autocorrelation distance in the vertical (Vs, N60) and horizontal (ρ) directions and the coefficient of variation are within the range of values founded in the literature. The dynamic response (spectral amplification) of the alluvial plain of Beirut was modeled by one dimensional probabilistic model and we quantified the effect of the three statistical parameters (autocorrelation function, autocorrelation distance and coefficient of variation) describing the elastic variability properties of soil (Vs). To obtain realistic seismic responses, we proposed a probabilistic (Vs) profile selection criteria in order to retain only compatible profiles (in a range of uncertainty) with the obtained dispersion curve. Probabilistic modeling showed significant differences from the deterministic modeling. It appeared that the main factor controlling probabilistic spectral amplification is the coefficient of variation (COVVs) followed by the autocorrelation distance, while the type of autocorrelation function has little influence. Finally, Human activity was also found to have a significant influence on the application of geophysical prospecting at this urban site. This case illustrates the need of combining investigation methods in order to understand the geophysical measurements in a complex medium and to reach a robust 2D/3D model.

Le terme variabilité spatiale des mouvements sismiques (SVGM en anglais pour Spatial Variability of Ground Motion) désigne les différences entre deux mesures du mouvement du sol effectuées à différents endroits, généralement en surface. La modélisation de SVGM ainsi que son effet sur la réponse dynamique des barrages est nécessaire pour l'intégration du phénomène dans les codes parasismiques. L'étude actuelle présente une mesure et une enquête approfondie sur SVGM à l'interface barrage voûte - fondation rocher. Des mesures in situ sont utilisées pour sa quantification et des simulations numériques pour la compréhension plus approfondie des phénomènes physiques qui contribuent à SVGM notamment à l'interface, c-à-d la topographie de la voûte et l'interaction sol-structure. Les données in situ provient d'une campagne sismologique qui a eu lieu sur et autour le barrage voûte du Saint Guérin pendant six mois. Le sous-ensemble d'événements consiste des événements de faible à moyenne magnitude, locaux et régionaux. Ainsi, des analyses linéaires sont permises. Tout d'abord, l'analyse dynamique du barrage voûte est réalisée; les fréquences de vibration, le coefficient d'amortissement et l'amplification de crête sont estimées profitant des mesures continues de bruit ambiant et des enregistrements sismiques. Ensuite, le SVGM est quantifiée au moyen de la phase et de la variabilité d'amplitude en utilisant des estimations de cohérence et de l'écart type de la différence des amplitudes des spectres de Fourier respectivement. Forte variabilité est observée à la fois en phase et en amplitude à l'interface de barrage-fondation rocher. Une fois que les estimateurs de la variabilité sont obtenus à partir des données, les modèles paramétriques existants sont comparés avec eux. Accentuation est mis sur deux observations: 1) les mouvements sismiques au champ libre semblent être légèrement moins variable par rapport aux mouvements à l'interface barrage-fondation rocher et 2) à l'interface barrage-fondation rocher, il y a une variabilité plus forte autour des fréquences de vibration du barrage. Ces observations suggèrent que la présence de la structure ainsi que la topographie du canyon augmentent SVGM. Cette hausse semble cependant être faible étant donné que les observations sont montés par des modèles paramétriques satisfaisants basés sur des données provenant de réseaux sismiques plates (sans topographie et structure). Des simulations numériques dans le code SPECFEM3D, basé sur les éléments spectraux, sont utilisées pour étudier plus profondément les observations in situ par le découplage des différentes causes de SVGM et évaluer l'impact de chacune. Une étude paramétrique en utilisant une topographie du canyon simplifié tente d'identifier l'effet de la topographie du canyon local sur SVGM tandis qu'un modèle géométrique précis du barrage voûte à Saint Guérin et sa topographie du canyon nous donne une idée plus précise sur l'impact de l'interaction barrage-fondation rocher sur SVGM. Les résultats de cette recherche devraient contribuer à l'amélioration de notre compréhension de SVGM à l'interface barrage-fondation rocher et de proposer des modèles de variabilité utilisés dans la conception de barrages voûtes. / Spatial variability of seismic ground motions (SVGM) denotes the differences between two time histories of the ground motion recorded at different locations, generally at the ground surface. The modeling of SVGM and the understanding of its influence on the dam's response are necessary so as design codes start to incorporate its effects in their provisions. This study presents a measure and a profound investigation of SVGM at the dam-foundation rock interface of an arch dam. In-situ measurements are used to quantify SVGM and numerical simulations to deeper understand the particular physical phenomena that contribute to SVGM at the interface, i.e. local canyon topography and rock-structure interaction. The in-situ data comes from a seismological experimental campaign that has taken place on and around Saint Guérin arch dam over the period of six months. The campaign was held in the framework of the present thesis. The subset of events consists of low to moderate magnitude local and regional earthquakes. Thus, analysis is allowed in the linear range. Firstly, dynamic analysis of the arch dam is conducted; the frequencies of vibration, the damping coefficient and the crest amplification are estimated based on ambient noise and seismic records. Then, SVGM is quantified by means of phase and amplitude variability using coherency estimates and standard deviation of difference of Fourier amplitudes respectively. High variability is observed both in phase and amplitude at the dam-foundation rock interface. Once the estimators of variability are obtained from the data, parametric models are fitted to them. Focus is given on two observations : 1) the ground motions in the free field appear to be slightly less variable with respect to the motions at the dam-foundation rock interface and 2) at the dam-foundation rock interface, there is higher variability around the frequencies of vibration of the dam. These observations suggest that the presence of the structure along with the canyon topography increase SVGM. This increase though seems to be small given that the observations are satisfactory fitted by parametric models based on data coming from flat seismic arrays. Numerical simulations in the SPECFEM3D code, based on the spectral element method are used to deeper investigate the in-situ observations by decoupling the various causes of SVGM and evaluating the impact of each one. A parametric study using a simplified canyon topography attempts to identify the effect of local canyon topography on SVGM while a geometrically accurate model of the Saint Guérin arch dam and its canyon topography gives us a better insight on the dam-foundation rock interaction impact on SVGM. Although both features are found to increase SVGM, their impact remains secondary. The findings of the present research are expected to contribute in enhancing our understanding of SVGM at the dam-foundation rock interface and proposing variability models used in arch dams' design.

Made available in DSpace on 2016-08-31T13:24:13Z (GMT). No. of bitstreams: 1 CarlosGFN_DISSERT.pdf: 1398522 bytes, checksum: 652d67d0830366c62f6d4ccfcedab473 (MD5) Previous issue date: 2008-12-10 / Rice crop area in Apodi River Valley, RN, Brazil, had a significant increase due to construction of Santa Cruz dam, which improved water availability downstream of the reservoir. Rice crop has been adversely affected by problems like disorganization of farmers, inefficient commercialization, product without added value, indiscriminate use of chemical inputs, loss of yield, decrease in plant vigor, and lower tillering. In this context, this work had as objectives to study causes of yield loss in irrigated red rice crop at Apodi River Valley; specifically, to study the effects of irrigation water used in that region and of soil salinity on red rice crop yield, besides to study spatial variability of physical and chemical soil characteristics of soil irrigated and cropped with red rice. The work was carried out in three small farms, where a georeferenced sampling grid was delimited with regular spacing, in which points were taken samples of soil and plants to determinate soil physical and chemical characteristics and, in two areas, yield components of red rice, allowing the use of geostatistics to study spatial variability. Rice yield and its components showed spatial dependence, which allowed the drawing of isovalue maps of the studied areas. Soil physical and chemical characteristics also showed spatial variability and correlated with rice yield components. The two studied areas showed different behavior in relation to spatial dependence and correlations. In the area of Francisco rice yield showed correlations mainly with distance to water inlet and with calcium to magnesium ratio, while in the area of Márcio the better correlation was with potassium / A cultura do arroz irrigado no vale do rio Apodi, RN, teve um aumento significativo na área plantada devido à construção da barragem de Santa Cruz, que melhorou a disponibilidade hídrica à jusante deste reservatório. Esta cultura vem sofrendo uma série de problemas, passando desde a desorganização dos plantadores, comercialização deficiente, inexistência de valor agregado ao produto, uso indiscriminado de insumos químicos, perda da produtividade, diminuição do vigor das plantas e menor perfilhamento. Neste sentido, este trabalho teve como objetivo estudar as causas das perdas de produtividade na cultura arroz vermelho irrigado no vale do rio Apodi; em particular estudar os efeitos da água de irrigação utilizada e da salinização do solo sobre a produtividade da cultura e estudar a variabilidade espacial das propriedades físicas e químicas do solo, irrigado e cultivado com arroz vermelho. O trabalho foi realizado em três pequenas propriedades, em cada uma das quais foi delimitada uma malha de pontos de amostragem georeferenciados com espaçamento regular e, nestes pontos, foram realizadas as coletas de solo e de planta para determinar atributos químicos e físicos do solo e, em duas delas, os componentes de produção do arroz vermelho, permitindo assim o uso da geoestatística para estudar a variabilidade espacial dos atributos. A produtividade do arroz e seus componentes apresentaram dependência espacial sendo possível a construção de mapas de isolinhas das áreas em estudo; constatou-se também variabilidade dos atributos físicos e químicos dos solos estudados e a existência de correlações entre os componentes de produção e alguns destes atributos. As duas áreas em que foram estudados os componentes de produção apresentaram comportamento diferenciado quanto à dependência espacial e as correlações, sendo que na área de Francisco a produtividade apresentou correlações principalmente com a distância da entrada de água na área e com a relação cálcio:magnésio, enquanto que na área de Márcio as correlações mais evidenciadas foram com potássio

Les méthodes de contrôle non destructif (CND) sont essentielles pour estimer les propriétés du béton (mécaniques ou physiques) et leur variabilité spatiale. Elles constituent également un outil pertinent pour réduire le budget d'auscultation d'un ouvrage d'art. La démarche proposée est incluse dans un projet ANR (EvaDéOS) dont l'objectif est d'optimiser le suivi des ouvrages de génie civil en mettant en œuvre une maintenance préventive afin de réduire les coûts. Dans le cas du travail de thèse réalisé, pour caractériser au mieux une propriété particulière du béton (ex : résistance mécanique, porosité, degré de Saturation, etc.), avec des méthodes ND sensibles aux mêmes propriétés, il est impératif de développer des outils objectifs permettant de rationaliser une campagne d'essais sur les ouvrages en béton armé. Dans ce but, premièrement, il est proposé un outil d'échantillonnage spatial optimal pour réduire le nombre de points d'auscultation. L'algorithme le plus couramment employé est le recuit simulé spatial (RSS). Cette procédure est régulièrement utilisée dans des applications géostatistiques, et dans d'autres domaines, mais elle est pour l'instant quasiment inexploitée pour des structures de génie civil. Dans le travail de thèse, une optimisation de la méthode d'optimisation de l'échantillonnage spatial (MOES) originale inspirée du RSS et fondée sur la corrélation spatiale a été développée et testée dans le cas d'essais sur site avec deux fonctions objectifs complémentaires : l'erreur de prédiction moyenne et l'erreur sur l'estimation de la variabilité. Cette méthode est décomposée en trois parties. Tout d'abord, la corrélation spatiale des mesures ND est modélisée par un variogramme. Ensuite, la relation entre le nombre de mesures organisées dans une grille régulière et la fonction objectif est déterminée en utilisant une méthode d'interpolation spatiale appelée krigeage. Enfin, on utilise l'algorithme MOES pour minimiser la fonction objectif en changeant les positions d'un nombre réduit de mesures ND et pour obtenir à la fin une grille irrégulière optimale. Des essais destructifs (ED) sont nécessaires pour corroborer les informations obtenues par les mesures ND. En raison du coût ainsi que des dégâts possibles sur la structure, un plan d'échantillonnage optimal afin de prélever un nombre limité de carottes est important. Pour ce faire, une procédure utilisant la fusion des données fondée sur la théorie des possibilités et développée antérieurement, permet d'estimer les propriétés du béton à partir des ND. Par le biais d'un recalage nécessitant des ED réalisés sur carottes, elle est étalonnée. En sachant qu'il y a une incertitude sur le résultat des ED réalisés sur les carottes, il est proposé de prendre en compte cette incertitude et de la propager au travers du recalage sur les résultats des données fusionnées. En propageant ces incertitudes, on obtient des valeurs fusionnées moyennes par point avec un écart-type. On peut donc proposer une méthodologie de positionnement et de minimisation du nombre des carottes nécessaire pour ausculter une structure par deux méthodes : la première, en utilisant le MOES pour les résultats des propriétés sortis de la fusion dans chaque point de mesure et la seconde par la minimisation de l'écart-type moyen sur la totalité des points fusionnés, obtenu après la propagation des incertitudes des ED. Pour finir, afin de proposer une alternative à la théorie des possibilités, les réseaux de neurones sont également testés comme méthodes alternatives pour leur pertinence et leur simplicité d'utilisation. / Non-destructive testing methods (NDT) are essential for estimating concrete properties (mechanical or physical) and their spatial variability. They also constitute an useful tool to reduce the budget auscultation of a structure. The proposed approach is included in an ANR project (EvaDéOS) whose objective is to optimize the monitoring of civil engineering structures by implementing preventive maintenance to reduce diagnosis costs. In this thesis, the objective was to characterize at best a peculiar property of concrete (e.g. mechanical strength, porosity, degree of saturation, etc.), with technical ND sensitive to the same properties. For this aim, it is imperative to develop objective tools that allow to rationalize a test campaign on reinforced concrete structures. For this purpose, first, it is proposed an optimal spatial sampling tool to reduce the number of auscultation points. The most commonly used algorithm is the spatial simulated annealing (SSA). This procedure is regularly used in geostatistical applications, and in other areas, but yet almost unexploited for civil engineering structures. In the thesis work, an original optimizing spatial sampling method (OSSM) inspired in the SSA and based on the spatial correlation was developed and tested in the case of on-site auscultation with two complementary fitness functions: mean prediction error and the error on the estimation of the global variability. This method is divided into three parts. First, the spatial correlation of ND measurements is modeled by a variogram. Then, the relationship between the number of measurements organized in a regular grid and the objective function is determined using a spatial interpolation method called kriging. Finally, the OSSM algorithm is used to minimize the objective function by changing the positions of a smaller number of ND measurements and for obtaining at the end an optimal irregular grid. Destructive testing (DT) are needed to corroborate the information obtained by the ND measurements. Because of the cost and possible damage to the structure, an optimal sampling plan to collect a limited number of cores is important. For this aim, a procedure using data fusion based on the theory of possibilities and previously developed is used to estimate the properties of concrete from the ND. Through a readjustment bias requiring DTs performed on carrots, it is calibrated. Knowing that there is uncertainty about the results of DTs performed on carrots, it is proposed to take into account this uncertainty and propagate it through the calibration on the results of the fused data. By propagating this uncertainty, it is obtained mean fused values with a standard deviation. One can thus provide a methodology for positioning and minimizing the number of cores required to auscultate a structure by two methods: first, using the OSSM for the results of fused properties values in each measuring point and the second by the minimization of the average standard deviation over all of the fused points obtained after the propagation of DTs uncertainties. Finally, in order to propose an alternative to the possibility theory, neural networks are also tested as alternative methods for their relevance and usability.