A soja e os seus derivados são uma das mercadorias agrícolas mais valiosas e comercializadas no mundo. O Brasil é o segundo maior produtor mundial de soja, cujo complexo é o principal gerador de divisas cambiais para o país. Porém, um dos principais fatores que podem limitar os rendimentos da cultura e a qualidade dos produtos é a presença de doenças. Para evitar maiores prejuízos é importante identificar rapidamente qual doença acomete a planta e determinar o método mais eficaz de manejo. Atualmente, uma doença recém descoberta que causa preocupação aos produtores é a infestação pelo nematoide Aphelenchoides besseyi, causador da, popularmente conhecida, Soja louca II. Ainda não existem métodos que proporcionem um diagnóstico preciso para a doença, sendo este realizado por inspeção visual da plantação. Sabendo que plantas sadias e doentes apresentam perfis nutricionais diferentes, este trabalho vem propor o uso da técnica analítica LIBS, associada ao método livre de calibração (CF), como uma ferramenta alternativa para diagnóstico da doença. A técnica LIBS permite análises multielementares simultâneas, de maneira rápida e de baixo custo. Já o método CF tem por principal vantagem dispensar o emprego de qualquer padrão de matriz compatível, ou material de referência, para as quantificações em LIBS, fazendo-o com base nas características físicas do plasma formado. Assim, a associação de ambos se torna absolutamente adequada para determinações analíticas em amostras desconhecidas. Neste trabalho, foram conduzidos estudos de caracterização qualitativos, por meio de um sistema LIBS de baixa resolução, e avaliações semiquantitativa e quantitativa por meio de medidas LIBS em um sistema com detecção de alta resolução (DP LIBS). O primeiro estudo demonstrou que, qualitativamente, as variações nos macroelementos Ca, Mg e K presentes nas folhas de soja, permitiu alcançar, em média, 90% de acurácia na diferenciação entre amostras doentes e sadias, e inferir que a doença provoca relativa diminuição da concentração de Ca e Mg, e um pequeno aumento nos níveis de K. As quantificações obtidas pelo método CF foram consideradas satisfatórias, além de pioneiras nas análises de amostras de folhas de soja puras. A partir dos perfis nutricionais quantificados foram construídos classificadores (classificação via regressão associada à regressão por mínimos quadrados parciais), cuja acurácia alcançada foi de 94%, 92% e 90%, respectivamente para os dados de concentração obtidos pela técnica de referência, e DP LIBS+CF nas geometrias colinear e ortogonal. Na tentativa de eliminar o emprego de qualquer reagente, e ainda assim obter uma ferramenta diagnóstica precisa, foi conduzido um estudo semiquantitativo, também a partir da associação DP LIBS+CF. Baseado nas relações elementares entre os analitos que compõem a amostra, este apresentou acurácia diagnóstica superior a 90% para a geometria ortogonal. Portanto, este estudo permitiu o desenvolvimento de uma metodologia para diagnóstico que respeita 100% dos preceitos da Química Verde, e agrega muito ao sistema atual, quando o que se deseja é uma avaliação diagnóstica rápida para tomada de decisão imediata no campo. / Soybeans and their by-products are one of the most valuable and traded commodities in the world. Brazil is the second largest soybean producer in the world, whose soy complex is the main generator of foreign exchange for the country. However, one of the main factors that can limit crop yields and product quality is the occurrence of disease. To avoid further damage, it is important to quickly identify which disease affects the plant and determine the most effective method of management. Currently, a disease newly discovered and of concern to producers is the infestation by the nematode Aphelenchoides besseyi, which causes the disease known as GSFR (Green Steam and Foliar Retention). There are still no methods to provide a precise diagnosis for the disease, which is, nowadays, performed by visual inspection of the plantation. Knowing that healthy and diseased plants have different nutrient profiles, this work proposes the use of the LIBS technique, associated with the calibration free method (CF), as an alternative tool for the disease detection. The LIBS technique allows simultaneous multi-element analysis, quickly and inexpensively. The CF method has the main advantage of avoiding the use of any compatible standard matrix, or reference material, for LIBS quantifications, based on the physical characteristics of the plasma formed. Thus, the association of both becomes absolutely suitable for the analytical determination of unknown samples. Qualitative characterization studies using a low-resolution LIBS system, and semiquantitative and quantitative analysis using LIBS measurements in a high-resolution detection system (DP LIBS), were performed. The first study showed that it was possible to achieve, on average, 90% accuracy in the differentiation between diseased and healthy samples, and to infer that the disease provokes relative decrease in Ca and Mg concentration, and a small increase in K levels. Quantifications achieved by the CF method were considered satisfactory since it is a pioneer study in the analysis of pure soybean leaves. Classifiers were constructed from the nutritional profiles quantified by CF, applying classification via regression associated with partial least square regression, with accuracy of 94%, 92% and 90%, respectively, for the concentration data obtained by the reference technique, and DP LIBS + CF colinear and orthogonal. In attempt to eliminate the use of any chemical reagent and still obtain an efficient diagnostic tool, a semiquantitative study was conducted, also from the DP LIBS + CF association and based on the elementary relations between the analytes that make up the sample, with accuracy diagnosis greater than 90% for DP LIBS in orthogonal geometry. Therefore, this study allowed the development of a diagnostic methodology that respects 100% of green chemistry principles, and adds a lot to the current system, if the desire is a rapid diagnostic evaluation for immediate decision making in the field.

Space-borne remote sensing datasets have the potential to allow us to progress towards global scale flood prediction systems. However, these datasets are limited in terms of space-time resolution and accuracy, and the best use of such data requires understanding how uncertainties propagate through hydrological models. An unbiased investigation of different datasets for hydrological modeling requires a parsimonious calibration-free model, since calibration masks uncertainties in the data and model structure. This study, which addresses these issues, consists of two parts: 1) the development and validation of a multi-scale distributed hydrological model whose parameters can be directly linked to physical properties of the watershed, thereby avoiding the need of calibration, and 2) application of the model to demonstrate how data uncertainties propagate through the model and affect flood simulation across scales. I based the model development on an interactive approach for model building. I systematically added processes and evaluated their effects on flood prediction across multiple scales. To avoid the need for parameter calibration, the level of complexity in representing physical processes was limited by data availability. I applied the model to simulate flows for the Cedar River, Iowa River and Turkey River basins, located in Iowa. I chose this region because it is rich in high quality hydrological information that can be used to validate the model. Moreover, the area is frequently flooded and was the center of an extreme flood event during the summer of 2008. I demonstrated the model's skills by simulating medium to high-flow conditions; however the model's performance is relatively poor for dry (low flow) conditions. Poor model performance during low flows is attributed to highly nonlinear dynamics of soil and evapotranspiration not incorporated in the model. I applied the hydrological model to investigate the predictability skills of satellite-based datasets and to investigate the model's sensibility to certain hydro-meteorological variables such as initial soil moisture and bias in evapotranspiration. River network structure and rainfall are the main components shaping floods, and both variables are monitored from space. I evaluated different DEM sources and resolution DEMs as well as the effect of pruning small order channels to systematically decreasing drainage density. Results showed that pruning the network has a greater effect on simulated peak flow than the DEM resolution or source, which reveals the importance of correctly representing the river network. Errors on flood prediction depend on basin scale and rainfall intensity and decrease as the basin scale and rainfall intensity increases. In the case of precipitation, I showed that simulated peak flow uncertainties caused by random errors, correlated or not in space, and by coarse space-time data resolution are scale-dependent and that errors in hydrographs decrease as basin scale increases. This feature is significant because it reveals that there is a scale for which less accurate information can still be used to predict floods. However, the analyses of the real datasets reveal the existence of other types of error, such as major overall bias in total volumes and the failure to detect significant rainfall events that are critical for flood prediction.

Two novel personal calibration procedures that do not require active user participation are presented. These procedures, in conjunction with a state-of-the-art remote eye-gaze tracking (REGT) technology, allow estimation of the angle between the optical and visual axes (angle kappa) automatically without explicit/active user involvement. The first algorithm for the binocular estimation of angle kappa (BEAK) is based on the assumption that at each time instant both eyes look at the same point on a surface with a known geometry (e.g., a computer monitor). The sensitivity of the BEAK procedure to the geometry of the observation surface and to the noise in the estimates of the optical axis is studied both analytically and in computer simulations. Experimental results with 4 adult subjects suggest that with the current REGT technology angle kappa can be estimated with an RMS error of 0.5°. The second personal calibration algorithm (‘calibrate and validate’, CaVa) adopts a probabilistic approach to the estimation of angle kappa in infants. Even though the presentation of visual stimuli at known positions is part of the procedure, the CaVa algorithm does not require/assume continuous fixation on the presented targets. If an infant attends to roughly half of the presented targets, angle kappa can be estimated accurately and with high confidence. In experiments with five babies, the average difference between repeated measurements of angle kappa was 0.04 ± 0.31°. The second part of the thesis describes two methods for automated measurement of eye misalignment in adults and infants. These methods are based on the user-calibration-free (UCF) technology presented in the first part of the thesis. The first method is based on the clinical Hirschberg test. It is shown that the UCF-REGT technology can improve significantly the accuracy of the Hirschberg test by enabling the estimation of subject-specific parameters (the Hirschberg ratio and angle kappa) in infants. The maximum error in the estimation of the horizontal and vertical components of eye misalignment in five orthotropic infants was shown to be less than 1°, which is significantly better than the accuracy of a standard clinical Hirschberg test. Finally, a novel Eye-Tracker Based Test (ETBT) for the estimation of the maximum (manifest + latent) angle of deviation is presented. ETBT is based on the UCF-REGT system. ETBT allows free head movements and does not require continuous fixation on specific targets. Experiments with 22 adult subjects demonstrated a good agreement of 0.7 ± 1.7° between ETBT and the gold-standard clinical procedure—the altenate prism and cover test. A pilot study with 5 orthotropic infants and one infant with strabismus demonstrated that the ETBT can be used in infants.

Two novel personal calibration procedures that do not require active user participation are presented. These procedures, in conjunction with a state-of-the-art remote eye-gaze tracking (REGT) technology, allow estimation of the angle between the optical and visual axes (angle kappa) automatically without explicit/active user involvement. The first algorithm for the binocular estimation of angle kappa (BEAK) is based on the assumption that at each time instant both eyes look at the same point on a surface with a known geometry (e.g., a computer monitor). The sensitivity of the BEAK procedure to the geometry of the observation surface and to the noise in the estimates of the optical axis is studied both analytically and in computer simulations. Experimental results with 4 adult subjects suggest that with the current REGT technology angle kappa can be estimated with an RMS error of 0.5°. The second personal calibration algorithm (‘calibrate and validate’, CaVa) adopts a probabilistic approach to the estimation of angle kappa in infants. Even though the presentation of visual stimuli at known positions is part of the procedure, the CaVa algorithm does not require/assume continuous fixation on the presented targets. If an infant attends to roughly half of the presented targets, angle kappa can be estimated accurately and with high confidence. In experiments with five babies, the average difference between repeated measurements of angle kappa was 0.04 ± 0.31°. The second part of the thesis describes two methods for automated measurement of eye misalignment in adults and infants. These methods are based on the user-calibration-free (UCF) technology presented in the first part of the thesis. The first method is based on the clinical Hirschberg test. It is shown that the UCF-REGT technology can improve significantly the accuracy of the Hirschberg test by enabling the estimation of subject-specific parameters (the Hirschberg ratio and angle kappa) in infants. The maximum error in the estimation of the horizontal and vertical components of eye misalignment in five orthotropic infants was shown to be less than 1°, which is significantly better than the accuracy of a standard clinical Hirschberg test. Finally, a novel Eye-Tracker Based Test (ETBT) for the estimation of the maximum (manifest + latent) angle of deviation is presented. ETBT is based on the UCF-REGT system. ETBT allows free head movements and does not require continuous fixation on specific targets. Experiments with 22 adult subjects demonstrated a good agreement of 0.7 ± 1.7° between ETBT and the gold-standard clinical procedure—the altenate prism and cover test. A pilot study with 5 orthotropic infants and one infant with strabismus demonstrated that the ETBT can be used in infants.

Pour répondre aux besoins des nouvelles techniques de caractérisation sur site in-situ et temps réel, l’unité NOVA de l’INERIS en partenariat avec les laboratoires LP3 et GREMI, a entamé des travaux pour étudier deux approches afin d’améliorer les performances de la technique Laser-Induced Breakdown Spectroscopy (LIBS) pour l’analyse des aérosols. LIBS est une technique optique de spectroscopie atomique. Elle consiste à focaliser un faisceau laser impulsionnel sur un échantillon à analyser créant ainsi un plasma. L’émission optique du plasma contient alors la signature des éléments chimiques présents dans l’échantillon. La première approche concerne la détermination de la composition chimique relative (stoechiométrique) d’aérosols sans étalonnage. En effet, l’étalonnage présente des problèmes pratiques. Pour ce faire, les spectres expérimentaux enregistrés lors de l’analyse des particules d’alumine (Al2O3) suspendues dans de l’hélium (He) ont été comparés à des spectres théoriques calculés pour un plasma contenant les mêmes éléments, à l’Équilibre Thermodynamique Local (ETL). L’ajustement des spectres simulés sur les spectres expérimentaux nous a permis de déterminer correctement la composition chimique relative des éléments présents dans le plasma. L’évolution temporelle du plasma a permis de valider l’ETL, et ainsi estimer la meilleure plage temporelle permettant la détermination de la composition relative de l’aérosol. La deuxième approche utilise une cellule radiofréquence (RF) à pression réduite comme piège à particules pour analyser des aérosols contenant des nanoparticules. Un tel piège permet d’améliorer la détection en concentrant spatialement les particules. Les paramètres optimaux d’utilisation de ce système ont été étudiés. Cette étude a permis d’établir que l’émission continue du plasma est fortement réduite dès ses premiers instants de vie. Le volume d’échantillonnage de ce système a été évalué et la limite de détection a été estimée de manière théorique. / New issues related to process control and workplace surveillance accompany the emergence of nanotechnology industry. This involves the development of new real-time and in-situ characterization techniques. In this context, the NOVA unit from the INERIS institute collaborated with LP3 and GREMI laboratories to study two approaches aiming to enhance the LIBS technic performances. The first approach used a flow cell to determine the relative elemental composition of an aerosol with a calibration-free procedure. The recorded spectra were compared to theoretical spectra calculated for a plasma in the Local Thermodynamic Equilibrium LTE. The best agreement between recorded and computed spectra allowed the determination of the relative composition with a good agreement with the reference value, for an alumina aerosol. The study of the temporal evolution of the plasma allowed the estimation of a temporal range within which the LTE hypothesis was verified. The second approach used a low-pressure radiofrequency plasma generated in an inert gas as a particle trap to analyse aerosols and nanoparticles. The use of such a system allowed the enhancement of particles detection by concentrating them spatially. We determined the optimal parameters for the LIBS analysis using this system. Furthermore, we established the plasma continuum was attenuated even at very low time delays. We evaluated the sampling volume of this new system and compared it to case of LIBS analysis on air. Finally, we estimated the detection limits of this system when analysing nanoparticles.

Ce travail de thèse a été consacré à la compréhension du plasma induit par laser sur des matériaux organiqueset à l’application de la technique de laser-induced breakdown spectroscopy (LIBS) à l’analyse quantitative deces matériaux. L’ensemble des travaux contribue donc à approfondir notre connaissance sur les mécanismesphysiques impliqués dans l’interaction laser-matière, dans la génération du plasma et dans l’évolution decelui-ci au cours de son expansion dans le gaz ambiant avec la spécificité de s’adresser à des plasmas induitssur des cibles organiques. Ces travaux visent également à améliorer la performance de la LIBS pour l’analysede matériaux organiques. La spécificité concernant la cible organique répond au besoin actuel de lacommunauté internationale travaillant sur le développement de la LIBS, de mieux maîtriser le plasma induitsur ce genre de matériau qui présente des propriétés optiques bien spécifiques par rapport à des matériauxmieux connus pour l’ablation laser tels que des métaux. Elle répond également au besoin grandissant d’utiliserla technique LIBS aux matériaux organiques pour des applications dans les domaines environnementaux,agroalimentaires, ou encore biomédicaux. Ce mémoire de thèse a été rédigé selon l’organisation suivante. Après l’Introduction Générale qui préciseles contextes scientifiques et technologiques, le Chapitre I rappelle les principes de base nécessaires à lacompréhension du phénomène de génération de plasma par ablation laser, et de l’évolution de celui-ci dans ungaz ambiant. L’accent a été mis sur l’ablation des matériaux organiques. Les procédures et les techniques dediagnostic du plasma induit par laser sont ensuite présentées en insistant sur la nature transitoire etinhomogène d’un plasma en expansion. Le chapitre II s’intéresse à la génération et l’évolution du plasmainduit sur la peau d’une pomme de terre, un échantillon typique de produit agroalimentaire. La caractéristiquedu plasma induit sur une cible organique molle et humide, comme une pomme de terre fraîche était quelquechose d’inconnu au démarrage de la thèse et constituait une base nécessaire à l’analyse quantitative deséléments métalliques en traces et ultra-traces contenus dans un tel échantillon. A l’issue de cettecaractérisation, les données analytiques semi-quantitatives ont été extraites des spectres LIBS correspondant àla peau d’une pomme de terre. Le Chapitre III se situe dans la continuité du Chapitre II pour l’application de laLIBS à des matériaux organiques. Une étude comparative d’analyse de la poudre de lait par la LIBS et parl’ICP-AES permet une évaluation de la performance d’analyse quantitative de la LIBS pour des échantillonsorganiques, et une validation de la procédure CF-LIBS. A la différence des Chapitres II et III où les élémentsde traces métalliques étaient au centre de notre attention, le Chapitre IV étudie le comportement des élémentsmajeurs qui composent la matrice d’un matériau organique, qui sont les 4 éléments organiques, H, C, O, N. Ala décomposition d’un matériau organique par ablation laser, ces éléments peuvent se présenter sous la formede fragments moléculaires, ou encore se recombiner en des espèces moléculaires. Nous étudions alors dans ceChapitre l’évolution de ces espèces en fonction des paramètres du laser utilisé, et notamment la longueurd’onde. Le mémoire se termine par une conclusion générale et des perspectives. / This PhD work was devoted to the understanding of the laser-induced plasma on organic materials and theapplication of laser-induced breakdown spectroscopy (LIBS) to quantitative analysis of these materials. Itcontributes to deepen our knowledge on the physical mechanisms involved in laser-matter interaction, plasmageneration, evolution and expansion of the plasma into the ambient gas, with emphasis on plasmas induced onorganic targets. It also intends to improve the performance of LIBS for the analysis of organic materials. Thespecificity for organic targets fits the current focus of the international community working on LIBS, toimprove the control of the plasma induced on this kind of material which has a distinguished optical prosperitywith respect to that of metals, better known for laser ablation. It addresses also the growing need to apply theLIBS technique to organic materials for different applications in the environmental, food, or biomedicaldomains. The works in this thesis were therefore presented in this thesis document according to the followingorganization.After the General Introduction which introduces the scientific and technological contexts, Chapter Irecalls the basic theoretical elements necessary to understand the phenomenon of plasma generation by laserablation, and its evolution in the background gas. Ablation of organic material is emphasized. Procedures andtechniques of diagnostics of laser-induced plasma were then presented with a focus on the transient andinhomogeneous nature of the expanding plasma. Chapter II focuses on the generation and the evolution of theplasma induced on the skin of a potato, a typical sample of agricultural product. The characteristics of plasmainduced on a soft and wet organic target, such as a fresh potato, was something unknown when the thesis workstarted. These characteristics provide the necessary basis for the quantitative analysis of the trace andultra-trace metallic elements in these samples. Following this characterization, semi-quantitative analyticalresults were extracted from LIBS spectra corresponding to potato skin. Chapter III is presented in thecontinuity of Chapter II for the application of LIBS to the quantitative analysis of organic materials. Acomparative study on the analytical results with LIBS and ICP-AES for milk powders allows an assessment ofthe performances of quantitative analysis by LIBS for organic materials, and a validation of the CF-LIBSprocedure that we have developed. Different from Chapters II and III where attention was paid to trace metalelements, Chapter IV studies the behavior of the major elements that make up the matrix of organic material,which are 4 known organic elements: H, C, O, N. During the decomposition of organic material by laserablation, these elements can be found in the form of molecular fragments, or recombined into molecularspecies. We then study in this Chapter the evolution of these species as a function of the laser ablationparameters, the laser wavelength in particular. The thesis document ends with a general conclusion andoutlooks.

This PhD work was devoted to the understanding of the laser-induced plasma on organic materials and theapplication of laser-induced breakdown spectroscopy (LIBS) to quantitative analysis of these materials. Itcontributes to deepen our knowledge on the physical mechanisms involved in laser-matter interaction, plasmageneration, evolution and expansion of the plasma into the ambient gas, with emphasis on plasmas induced onorganic targets. It also intends to improve the performance of LIBS for the analysis of organic materials. Thespecificity for organic targets fits the current focus of the international community working on LIBS, toimprove the control of the plasma induced on this kind of material which has a distinguished optical prosperitywith respect to that of metals, better known for laser ablation. It addresses also the growing need to apply theLIBS technique to organic materials for different applications in the environmental, food, or biomedicaldomains. The works in this thesis were therefore presented in this thesis document according to the followingorganization.After the General Introduction which introduces the scientific and technological contexts, Chapter Irecalls the basic theoretical elements necessary to understand the phenomenon of plasma generation by laserablation, and its evolution in the background gas. Ablation of organic material is emphasized. Procedures andtechniques of diagnostics of laser-induced plasma were then presented with a focus on the transient andinhomogeneous nature of the expanding plasma. Chapter II focuses on the generation and the evolution of theplasma induced on the skin of a potato, a typical sample of agricultural product. The characteristics of plasmainduced on a soft and wet organic target, such as a fresh potato, was something unknown when the thesis workstarted. These characteristics provide the necessary basis for the quantitative analysis of the trace andultra-trace metallic elements in these samples. Following this characterization, semi-quantitative analyticalresults were extracted from LIBS spectra corresponding to potato skin. Chapter III is presented in thecontinuity of Chapter II for the application of LIBS to the quantitative analysis of organic materials. Acomparative study on the analytical results with LIBS and ICP-AES for milk powders allows an assessment ofthe performances of quantitative analysis by LIBS for organic materials, and a validation of the CF-LIBSprocedure that we have developed. Different from Chapters II and III where attention was paid to trace metalelements, Chapter IV studies the behavior of the major elements that make up the matrix of organic material,which are 4 known organic elements: H, C, O, N. During the decomposition of organic material by laserablation, these elements can be found in the form of molecular fragments, or recombined into molecularspecies. We then study in this Chapter the evolution of these species as a function of the laser ablationparameters, the laser wavelength in particular. The thesis document ends with a general conclusion andoutlooks.