Les traits fonctionnels foliaires sont des paramètres clés des processus écologiques dans les forêts. Malgré les progrès réalisés dans la cartographie de traits foliaires à partir d’imagerie hyperspectrale, il subsiste un besoin de construire des modèles spectres-traits spécifiques à un environnement donné, entraînés à partir de mesures à l’échelle de l’arbre, afin d'améliorer la précision des cartes locales de traits. Nous avons cartographié 12 traits fonctionnels foliaires dans une forêt tempérée mixte à partir d’imagerie hyperspectrale aéroportée. Des échantillons foliaires de cimes d'arbres (n = 166), couvrant un total de 16 espèces, ont été recueillis à l'aide d'une plateforme de drone pour mesurer des traits foliaires d’arbres individuels, à partir desquels la réflectance spectrale de la couronne a également été mesurée. Des modèles de régression des moindres carrés partiels (PLSR) ont été utilisés pour prédire les traits foliaires à partir des spectres de réflectance à l’échelle de l’arbre (400-2400 nm). Ces modèles ont prédit la masse foliaire par unité de surface (LMA), la surface foliaire spécifique (SLA) et l'épaisseur en eau (EWT) avec une haute précision (R2 > 0.8, %RMSE < 15). Les modèles de concentration de pigments, de l'azote et de la cellulose ont montré une performance modérée (R2 = 0.53–0.68, %RMSE = 17.24–21.31). Les performances les plus faibles ont été observées pour la lignine, le carbone, le contenu en matière sèche (LDMC) et l'hémicellulose (R2 = 0.24–0.44, %RMSE = 20.67–26.13). Des cartes à haute résolution spatiale (1.25 m pixel-1) de traits foliaires ont été produites pour l'ensemble de la zone d'étude de 16 km2. Notre étude s'ajoute aux recherches approfondies visant à utiliser la télédétection pour surveiller la biodiversité des traits fonctionnels à plus grande échelle et fournit des modèles qui saisissent la variation intraspécifique de nombreuses espèces d'arbres d'une forêt tempérée mixte de l'est du Canada. / Foliar functional traits are key drivers of ecological processes in forests. Despite progress in forest foliar trait mapping from imaging spectroscopy, there is a need to build environment-specific, spectra-trait models trained from tree-level measurements to improve the accuracy of local trait maps. We mapped 12 foliar functional traits in a mixed temperate forest using airborne imaging spectroscopy. Top-of-canopy foliar samples from tree crowns (n = 166), representing a total of 16 species, were collected using a drone platform to measure foliar traits for individual trees, from which tree-level crown spectra were also determined. Partial least squares regression (PLSR) models were used to predict foliar traits from tree-level reflectance spectra (400-2400 nm). These models predicted leaf mass per area (LMA), specific leaf area (SLA) and equivalent water thickness (EWT) with high accuracy (R2 > 0.8, %RMSE < 15). Models for pigment, nitrogen and cellulose concentrations showed a moderate performance (R2 = 0.53–0.68, %RMSE = 17.24–21.31). Poorest performance was observed for lignin, carbon, leaf dry mass content (LDMC) and hemicellulose (R2 = 0.24–0.44, %RMSE = 20.67–26.13). High-resolution (1.25 m pixel-1) foliar trait maps were produced for the entire 16-km2 study area. Our study adds to the extensive research aiming to use remote sensing to monitor forest functional trait biodiversity at larger scales and provides models that capture intraspecific variation across many tree species from a mixed temperate forest in eastern Canada.

Dans un marché en permanente progression et pour répondre aux besoins des brasseurs en malt de qualité, la maîtrise du procédé de maltage est indispensable. La qualité du malt est fortement dépendante des conditions opératoires, en particulier des conditions de trempe, mais également de la qualité de la matière première : l'orge. Dans cette étude, nous avons établi des modèles polynomiaux qui mettent en relation les conditions opératoires et la qualité du malt. Ces modèles ont été couplés à nos algorithmes génétiques et nous ont permis de déterminer les conditions optimales de maltage, soit pour atteindre une qualité ciblée de malt (friabilité), soit pour permettre un maltage à faible teneur en eau (pour réduire la consommation en eau et maîtriser les coûts environnementaux de production) tout en conservant une qualité acceptable de malt. Cependant, la variabilité de la matière première est un facteur limitant de notre approche. Les modèles établis sont en effet très sensibles à l'espèce d'orge (printemps, hiver) ou encore à la variété d'orge utilisée. Les modèles sont surtout très dépendants de l'année de récolte. Les variations observées sur les propriétés d'une année de récolte à une autre sont mal caractérisées et ne sont donc pas intégrées dans nos modèles. Elles empêchent ainsi de capitaliser l'information expérimentale au cours du temps. Certaines propriétés structurelles de l'orge (porosité, dureté) ont été envisagées comme nouveaux facteurs pour mieux caractériser la matière première mais ils n'ont pas permis d'expliquer les variations observés en malterie.Afin de caractériser la matière première, 394 échantillons d'orge issus de 3 années de récolte différentes 2009-2010-2011 ont été analysés par spectroscopie MIR. Les analyses ACP ont confirmé l'effet notable des années de récolte, des espèces, des variétés voire des lieux de culture sur les propriétés de l'orge. Une régression PLS a permis, pour certaines années et pour certaines espèces, de prédire les teneurs en protéines et en béta-glucanes de l'orge à partir des spectres MIR. Cependant, ces résultats, pourtant prometteurs, se heurtent toujours à la variabilité. Ces nouveaux modèles PLS peuvent toutefois être exploités pour mettre en place des stratégies de pilotage du procédé de maltage à partir de mesures spectroscopiques MIR / In a continuously growing market and in order to meet the needs of Brewers in high quality malt, control of the malting process is a great challenge. Malt quality is highly dependent on the malting process operating conditions, especially on the steeping conditions, but also the quality of the raw material: barley. In this study, we established polynomial models that relate the operating conditions and the malt quality. These models have been coupled with our genetic algorithms to determine the optimal steeping conditions, either to obtain a targeted quality of malt (friability), or to allow a malting at low water content while maintaining acceptable quality of malt (to reduce water consumption and control the environmental costs of malt production). However, the variability of the raw material is a limiting factor for our approach. Established models are very sensitive to the species (spring and winter barley) or to the barley variety. The models are especially highly dependent on the crop year. Variations on the properties of a crop from one to another year are poorly characterized and are not incorporated in our models. They thus prevent us to capitalize experimental information over time. Some structural properties of barley (porosity, hardness) were considered as new factors to better characterize barley but they did not explain the observed variations.To characterize barley, 394 samples from 3 years of different crops 2009-2010-2011 were analysed by MIR spectroscopy. ACP analyses have confirmed the significant effect of the crop-years, species, varieties and sometimes of places of harvest on the properties of barley. A PLS regression allowed, for some years and for some species, to predict content of protein and beta-glucans of barley using MIR spectra. These results thus still face product variability, however, these new PLS models are very promising and could be exploited to implement control strategies in malting process using MIR spectroscopic measurements

O sensoriamento remoto é uma estratégia que pode ajudar no monitoramento da qualidade das pastagens. Objetivou-se com esse estudo analisar a resposta espectral das folhas de Brachiaria brizantha cv. Marandú, adubada com doses crescentes de ureia, para diferenciar e predizer teores foliares de nitrogênio (TFN). Os tratamentos foram distribuídos em blocos ao acaso (DBC), composto por quatro blocos e quatro tratamentos, totalizando 16 parcelas. Foram utilizadas doses crescentes de adubação com ureia: 0, 25, 50, 75 kg de N/ha/corte. Ao longo do experimento foram realizadas 7 coletas, sendo coletadas 8 folhas por parcela. Essas folhas foram submetidas à análise hiperespectral e posterior análise química do teor de nitrogênio. Ao analisar a resposta espectral das folhas, observou-se diferenças estatísticas entre os tratamentos na região do visível em todas as coletas, com ênfase na região de 550 nm (verde). Por meio de análise discriminante linear (LDA) realizada para cada coleta, os centróides gerados por todos os tratamentos apresentaram diferenças significativas, com exceção do LD1 nas coletas 6 e 7 que não apresentou distinção entre os tratamentos de 50 e 75 kg de N/ha/corte, e LD2 na coleta 5 que não apresentou distinção entre os tratamentos de 0 e 50 kg de N/ha/corte. As equações de regressão multivariada obtidas pelo método de quadrados mínimos parciais (PLSR), geraram valores razoáveis a bons de R2 (0,53 a 0,83) na predição dos TFN, onde os comprimentos de onda com maior peso nessas regressões estão na região do red edge (715 a 720 nm). Por fim, ao testar a performance de alguns Índices de Vegetação da literatura, as coletas 4, 6 e 7 apresentaram bons coeficientes de determinação (R2) que variaram de 0,65 a 0,73; uma característica em comum nos índices que melhor estimaram os TFN é a presença de comprimentos de ondas que fazem parte da região do red edge. / Remote sensing is a set of techniques that can help to monitor pasture quality. The object of this study is to analyze the spectral response from Brachiaria brizantha cv. Marandú leaves, under contrasting nitrogen levels, to differentiate and predict leaf nitrogen content. The treatments were set in a Randomized Block Design, composed of four blocks and four treatments, totaling 16 plots. Increasing doses of urea fertilization were used: 0, 25, 50, 75 kg N/ha/mowing. During the experiment, 7 data collections were performed, and 8 leaves per plot were extracted for each data collection. These leaves were submitted to hyperspectral data extraction and subsequent chemical analysis to quantify the nitrogen content. When analyzing the spectral pattern of the leaves, statistical differences among samples with different nitrogen levels were noticeable in the visible range of the spectrum in all the collections, with emphasis on the 550 nm region (green). Through linear discriminant analysis (LDA), performed for each collection, the generated centroids by the samples of each nitrogen level presented significant differences, except for LD1 in collections 6 and 7, which did not present a distinction between treatments of 50 and 75 kg of N/ha/mowing, and LD2 in collection 5 that did not distinguish between treatments of 0 and 50 kg of N/ha/mowing. The partial least square regression (PLSR) method generated reasonable to good values of R2 (0.53 to 0.83) for the prediction of leaf nitrogen content, where the wavelengths with the highest coefficient in these models are in the red edge region of the spectrum (715 to 720 nm). Finally, when testing the performance of some Vegetation Indexes from literature, collections 4, 6 and 7 presented good determination coefficients (R2) ranging from 0.65 to 0.73; a common feature in the indexes that best estimate the nitrogen content is the presence of wavelengths from the red edge region of the spectrum.

L’objectif de cette thèse était d’appliquer des méthodes d’analyse multivariées au traitement des données provenant de la spectroscopie de plasma induit par laser (LIBS) et de la spectroscopie térahertz (THz) dans le but d’accroître les performances analytiques de ces techniques.Les spectres LIBS provenaient de campagnes de mesures directes sur différents sites géologiques. Une approche univariée n’a pas été envisageable à cause d’importants effets de matrices et c’est pour cela qu’on a analysé les données provenant des spectres LIBS par réseaux de neurones artificiels (ANN). Cela a permis de quantifier plusieurs éléments mineurs et majeurs dans les échantillons de sol avec un écart relatif de prédiction inférieur à 20% par rapport aux valeurs de référence, jugé acceptable pour des analyses sur site. Dans certains cas, il a cependant été nécessaire de prendre en compte plusieurs modèles ANN, d’une part pour classer les échantillons de sol en fonction d’un seuil de concentration et de la nature de leur matrice, et d’autre part pour prédire la concentration d’un analyte. Cette approche globale a été démontrée avec succès dans le cas particulier de l’analyse du plomb pour un échantillon de sol inconnu. Enfin, le développement d’un outil de traitement par ANN a fait l’objet d’un transfert industriel.Dans un second temps, nous avons traité des spectres d’absorbance terahertz. Ce spectres provenaient de mesures d’absorbance sur des mélanges ternaires de Fructose-Lactose-acide citrique liés par du polyéthylène et préparés sous forme de pastilles. Une analyse semi-quantitative a été réalisée avec succès par analyse en composantes principales (ACP). Puis les méthodes quantitatives de régression par moindres carrés partiels (PLS) et de réseaux de neurons artificiels (ANN) ont permis de prédire les concentrations de chaque constituant de l’échantillon avec une valeur d’erreur quadratique moyenne inférieure à 0.95 %. Pour chaque méthode de traitement, le choix des données d’entrée et la validation de la méthode ont été discutés en détail. / The aim of this work was the application of multivariate methods to analyze spectral data from laser-induced breakdown spectroscopy (LIBS) and terahertz (THz) spectroscopy to improve the analytical ability of these techniques.In this work, the LIBS data were derived from on-site measurements of soil samples. The common univariate approach was not efficient enough for accurate quantitative analysis and consequently artificial neural networks (ANN) were applied. This allowed quantifying several major and minor elements into soil samples with relative error of prediction lower than 20% compared to reference values. In specific cases, a single ANN model didn’t allow to successfully achieving the quantitative analysis and it was necessary to exploit a series of ANN models, either for classification purpose against a concentration threshold or a matrix type, or for quantification. This complete approach based on a series of ANN models was efficiently applied to the quantitative analysis of unknown soil samples. Based on this work, a module of data treatment by ANN was included into the software Analibs of the IVEA company. The second part of this work was focused on the data treatment of absorbance spectra in the terahertz range. The samples were pressed pellets of mixtures of three products, namely fructose, lactose and citric acid with polyethylene as binder. A very efficient semi-quantitative analysis was conducted by using principal component analysis (PCA). Then, quantitative analyses based on partial least squares regression (PLS) and ANN allowed quantifying the concentrations of each product with a root mean square error (RMSE) lower than 0.95 %. All along this work on data processing, both the selection of input data and the evaluation of each model have been studied in details.

Human activities have increased the concentration of CO2 into the atmosphere, thus it has caused global warming. Construction equipment are semi-stationary machines and spend at least 30% of its life time during idling. The majority of the construction equipment is diesel powered and emits toxic emission into the environment. In this work, the idling will be investigated through adopting several statistical regressions models to quantify the fuel consumption of construction equipment during idling. The regression models which are studied in this work: Multivariate Linear Regression (ML-R), Support Vector Machine Regression (SVM-R), Gaussian Process regression (GP-R), Artificial Neural Network (ANN), Partial Least Square Regression (PLS-R) and Principal Components Regression (PC-R). Findings show that pre-processing has a significant impact on the goodness of the prediction of the explanatory data analysis in this field. Moreover, through mean centering and application of the max-min scaling feature, the accuracy of models increased remarkably. ANN and GP-R had the highest accuracy (99%), PLS-R was the third accurate model (98% accuracy), ML-R was the fourth-best model (97% accuracy), SVM-R was the fifth-best (73% accuracy) and the lowest accuracy was recorded for PC-R (83% accuracy). The second part of this project estimated the CO2 emission based on the fuel used and by adopting the NONROAD2008 model.  Keywords:

Instrumental disparities and soil moisture are two of the key limitations in implementing spectroscopic techniques in the field. This study sought to address these challenges through two objectives. The first objective was to assess Visible-near infrared (VisNIR) and mid-infrared (MIR) spectroscopic approaches and explore the feasibility of transferring calibration models between laboratory and portable spectrometers. The second objective addressed the challenge of soil moisture and its impact on spectra. The portable spectrometers demonstrated comparable performance to their laboratory-based counterparts in both regions. Spiking with extra-weight, was the most effective calibration transfer method eliminating disparities between instruments. The samples were rewetted under nine controlled conditions for the moisture study. Results showed that spiking with extra weights significantly outperformed other techniques and model enhancement was insensitive to the moisture contents. Findings of this study will be helpful for development and deployment of in situ sensors to enable field implementation of spectroscopy.