Surface water is essential for the eco-environment and provides various purposes and functions in anthropogenic activities such as domestic water supply, agriculture irrigation and drainage, and energy creation. Due to its important role in human health and ecosystem, it is essential to establish effective models to study the water quantity and quality through systematic research. In this dissertation, machine learning (ML) has been investigated as a modelling approach to simulate water quality and quantity of rivers. It includes the analysis of river pollution sources, downstream river flow prediction, long-term river heavy metal prediction and the influence of environmental factors on river heavy metals. This research was conducted to combine ML methods to enhance the accuracy and applicability of models in water environment studies and further assist relevant sectors to strengthen the management of surface water quantity and quality. (1) The primary source contributors of trace metals in surface water were identified based on the trained optimal model. This study trained and evaluated the typical shallow and deep learning approaches to identifying and classifying source contributors based on a database and analyzed the source apportionment of trace metals in the main stream and tributaries of river basin by the proposed approaches. (2) The interpretable ML models were developed to overcome the under-appreciated issue of the model for predicting heavy metals that are both predictive and transparent. This study employed and compared five tree-based machine learning models and then performed global and local feature importance analyses with the optimum models to predict the most important environmental factors for heavy metals management. (3) The performance of conventional or hybrid ML models with time series decomposition technology was developed and compared for long term prediction of heavy metals. This study examined the effect of inputs time-series data of selection and division on the performance of conventional and hybrid models and evaluated the long-term fore-casts by standard metrics to select the optimum approaches for typical metal long term prediction. (4) The downstream river flow was predicted based on the combined ML models. This study established the hybrid CNN-LSTM model to process the 2D rainfall radar information by convolutional neural network (CNN) and the time series information by long short-term memory (LSTM), and explored the capacity of the hybrid model for river flow forecasting.:Table of Content List of Abbreviations V List of Publications on the Ph.D. Topic VII List of Co-authored Publications during the Ph.D. VII 1 General Introduction 1 1.1 Background 1 1.2 Aim and Objective 3 1.3 Innovation and Contribution 3 1.4 Outline 4 1.5 References 5 2 Traceability Study of Metals 9 2.1 Introduction 10 2.2 Materials and Methods 11 2.2.1 Study Area 11 2.2.2 Ecological Risk Assessment 12 2.2.3 Model Development 13 2.2.4 The Classification Supervised ML Models 14 2.2.5 Model Evaluation 15 2.3 Results 16 2.3.1 Metal Characteristics of the Registered Dataset and Applicable Area 16 2.3.2 Spatial Assessment of Ecological Risk 20 2.3.3 Temporal Trend in Ecologic Risk 21 2.3.4 Performance Analysis of Classification Models 22 2.3.5 Resource Analysis Based on the Trained Model 23 2.4 Discussion 24 2.4.1 RF Outperformed the Other Models 24 2.4.2 Potential Sources of Metals in the Given Area 25 2.5 Conclusion 27 2.6 References 27 3 Elucidation of Environmental Factors’ Influence on Metals 33 3.1 Introduction 34 3.2 Materials and Methods 36 3.2.1 Study Sites and Data 36 3.2.2 Indexing Approach 39 3.2.3 Regression Models 40 3.2.4 Model Performance Metrics and Hyperparameter Tuning 41 3.2.5 SHapley Additive exPlanations (SHAP) 42 3.2.6 The Partial Dependence Plot (PDP) 42 3.3 Results 43 3.3.1 HPI and Environmental Variables 43 3.3.2 Assessment of Approaches 44 3.3.3 Global Feature Importance 46 3.3.4 Local Feature Importance 47 3.3.5 Sensitive Factor Analysis 49 3.4 Discussion 49 3.4.1 SHAP Outperformed the Other Importance Method 49 3.4.2 Robust Association of Top Variables with HPI 50 3.5 Conclusion 51 3.6 References 51 4 Long-Term Prediction of Metals Concentration 57 4.1 Introduction 58 4.2 Materials and Methods 60 4.2.1 Study Area and Water Quality Data 60 4.2.2 Input Identification 61 4.2.3 Wavelet Transform 63 4.2.4 Back-Propagation Neural Network (BPNN) Model 63 4.2.5 Nonlinear Autoregressive Exogenous (NARX) Model 65 4.2.6 Wavelet and BPNN (WNN) Hybrid Model 65 4.2.7 Wavelet and NARX (WNARX) Hybrid Model 66 4.2.8 Model Performance Evaluation 66 4.3 Results 67 4.3.1 Model Establishment 67 4.3.2 Performance Analysis of the Optimal Scenarios 70 4.4 Discussion 74 4.5 Conclusion 75 4.6 References 76 5 Prediction of Downstream River Flow 79 5.1 Introduction 80 5.2 Materials and Methods 82 5.2.1 Study Area and Data Acquisition 82 5.2.2 Convolutional Neural Network (CNN) 83 5.2.3 Long Short-Term Memory (LSTM) 83 5.2.4 River Flow Simulation 84 5.2.5 Performance Evaluation 84 5.3 Results 85 5.3.1 Flow Time Series 85 5.3.2 Input Selection 86 5.3.3 Flow Simulation 87 5.4 Discussion 90 5.5 Conclusion 91 5.6 References 92 6 Conclusions and Future Research 97 6.1 Traceability Study of Metals 97 6.2 Elucidation of Environmental Factors’ Influence on Metals 97 6.3 Long-Term Prediction of Metals Concentration 98 6.4 Prediction of River Flow 98 6.5 Discussion and Future Research 98 6.5.1 Discussion 99 6.5.2 Future Research 100 7 Appendices 103 7.1 Supporting Information for Traceability Study of Metals 103 7.1.1 Naive Bayes 103 7.1.2 Support Vector Machine 103 7.1.3 Neural Network 105 7.1.4 Random Forest 106 7.1.5 Long Short-Term Memory 106 7.1.6 Convolutional Neural Network 107 7.2 Supporting Information for Elucidation of Environmental Factors’ Influence on Metals 108 7.3 Supporting Information for Long-Term Prediction of Metals Concentration 109 7.3.1 Figures 109 7.4 References 120

This study recorded Event Related Potentials (ERPs) during completion of a Continuous Performance Task (CPT) in order to identify the contribution of response inhibition, working memory, and response monitoring to the pattern of hyperactive and impulsive and inattentive behaviour observed in patients with Attention Deficit Hyperactive Disorder (ADHD). Four ERP components, Nogo N2, Nogo P3, Go P3, and the ERN were examined and compared using a symptomatic and asymptomatic ADHD sample, and a healthy control group. The Nogo N2 had the expected frontal scalp distribution and was affected by changes to inhibitory demands. It was also suggested that this component was not wholly determined by inhibitory processing and may have been influenced by differing presentation rates of the Go stimulus, a template matching process or an in-depth response strategy. Source localisation analysis suggested a right frontal generator for this component. The Nogo P3 had the expected central distribution and had equal amplitude for those participants that were more efficient at inhibiting behaviours compared to those participants that were less efficient inhibitors. Contrary to expectations, the Nogo P3 was not affected by increasing the inhibitory demands of the task and was suggested as being a less reliable indicator of response inhibition in the present study. The Go P3 had the expected centro-parietal distribution, and appeared to provide a reliable index of working memory. Response inhibition and working memory were not impaired in the sample of symptomatic and asymptomatic ADHD adults used in this study. The symptomatic group elicited a slightly enhanced ERN compared to the asymptomatic and control groups, indicating that deficits in response monitoring may contribute to the pattern of problematic behaviour observed in people with ADHD.

Copper remains at elevated levels in the aquatic environment of Stockholm due to diffuse urban sources. Management of these diffuse sources requires their quantification but they cannot be measured directly by field observations. The working hypothesis of this thesis was that Copper levels in the sediments of urban lakes would reflect diffuse emissions within their catchment areas. In order to test this hypothesis, a source – transport – storage conceptual model was developed for tracing the urban diffuse sources of Copper to the sediment in the urbanised catchment. A substance flow analysis (SFA) approach was taken in the source module and a fate, mass-balance model was applied in the lake module. Five separate urban lakes (Judarn, Laduviken, Långsjön, Råcksta Träsk and Trekanten) within the Stockholm area and a main water flow pathway from Lake Mälaren to the inner archipelago of the Baltic Sea, through Stockholm, were selected as case studies.  In comparison to actual source strength data in the literature for the five case study lakes, the SFA approach gave similar results to previous models, but with reduced uncertainty. The SFA approach was also able to indicate the actual sources of urban copper, which was not accomplished by the other approaches and which is a great advantage in managing the sources. For the five lakes in Stockholm, traffic and copper roofs were found to be major contributors of Copper. For the three more polluted lakes, good agreement was obtained between simulated sediment copper contents and independent field observations, thereby supporting the applicability of the model in such cases. Furthermore, simulation results showed sediment copper content to be linearly dependent on the urban load. While this suggests that the urban copper sediment level reflects the urban load, considerable integration of this load over time (decade(s)) was suggested by the simulation results, so time must be allowed in order to detect a change in the urban load by field monitoring of the sediments.  Published data on the main water flow pathway from Lake Mälaren to the archipelago showed a peak in sediment copper content close to the city centre, confirming a considerable urban influence. An approach to quantitatively follow Cu from its urban source through such a complex, aquatic system was developed and applied to Stockholm. The compliance of future quantitative model results with monitoring data may help test the choices made in this conceptual model and the applicability of the model. Data availability proved to be a major obstacle to achieving a quantitative model, particularly as several municipalities with different levels of data availability surround the main water flow pathway studied.  Finally, the applicability of the quantitative, coupled source – transport – storage was demonstrated in a simplified scenario analysis. The ability of the model to estimate the copper load to air and soil and to the urban aquatic environment was also demonstrated. / QC 20110324

Configurable software systems adapt to changes in hardware and execution environments, and often exhibit a variety of complex maintenance issues. Many tools exist to aid developers in analysing and maintaining large configurable software systems. Some are standalone applications, while a growing number are becoming part of Integrated Development Environments (IDE) like Eclipse. Reusable tooling frameworks can reduce development time for tools that concentrate on software configuration analysis. This thesis presents C-CLEAR, a common, reusable, and extensible tooling framework for software configuration analysis, where clear separation of concern exists between tooling functionality and definitions that characterise a software system. Special emphasis will be placed on common mechanisms for data abstraction and automatic IDE integration independent of the software system that is being analysed. / Graduate

Diploma thesis creates the project design for changes in the purchase process in the selected company together using the tools and methodologies of project management. Based on an analysis of the business environment and key theoretical background of project management, it brings in the draft plan project, usable for its future company‘s implementation.

La voix est un stimulus auditif omniprésent dans notre environnement sonore. Elle permet non seulement la parole, mais serait aussi l’équivalent d’un visage auditif transmettant notamment des informations identitaires et affectives importantes. Notre capacité à discriminer et reconnaître des voix est socialement et biologiquement importante et elle figure parmi les fonctions les plus importantes du système auditif humain. La présente thèse s’intéressait à l’ontogénèse et à la spécificité de la réponse corticale à la voix humaine et avait pour but trois objectifs : (1) mettre sur pied un protocole électrophysiologique permettant de mesurer objectivement le traitement de la familiarité de la voix chez le sujet adulte; (2) déterminer si ce même protocole pouvait aussi objectiver chez le nouveau-né de 24 heures un traitement préférentiel d’une voix familière, notamment la voix de la mère; et (3) mettre à l’épreuve la robustesse d’une mesure électrophysiologique, notamment la Fronto-Temporal Positivity to Voices, s’intéressant à la discrimination pré-attentionnelle entre des stimuli vocaux et non-vocaux. Les résultats découlant des trois études expérimentales qui composent cette thèse ont permis (1) d’identifier des composantes électrophysiologiques (Mismatch Negativity et P3a) sensibles au traitement de la familiarité d’une voix; (2) de mettre en lumière un patron d’activation corticale singulier à la voix de la mère chez le nouveau-né, fournissant le premier indice neurophysiologique de l’acquisition du langage, processus particulièrement lié à l’interaction mère-enfant; et (3) de confirmer l’aspect pré-attentionnel de la distinction entre une voix et un stimulus non-vocal tout en accentuant la sélectivité et la sensibilité de la réponse corticale réservée au traitement de la voix. / Voice is a very prominent auditory stimulus in our acoustic environment. It is not only the carrier of speech, but would also be an auditory face that conveys important affective and identity information. Our ability to discriminate and recognize voices is socially and biologically important as it is amongst the most important functions of the human auditory system. This thesis was interested in the ontogenesis and specificity of the cortical response to human voice and had three objectives: (1) to develop an electrophysiological protocol to objectively measure the processing of voice familiarity in adult subjects; (2) to assess whether the same electrophysiological protocol could also objectify preferential processing of a familiar voice in 24-hour-old newborns, in particular the mother’s voice; and (3) to test the robustness of an electrophysiological measure, more specifically the Fronto-Temporal Positivity to Voices, interested in pre-attentional discrimination between vocal and non-vocal stimuli. Results from these three experimental designs have enabled (1) to identify electrophysiological components (Mismatch Negativity and P3a) sensitive to the processing of voice familiarity; (2) to highlight a singular pattern of cortical activation to the mother’s voice in newborns, providing the first neurophysiological evidence of language acquisition, a process especially related to the mother-child interaction; and (3) to confirm that vocal/non-vocal discrimination is a pre-attentional process, while enhancing the selectivity and the specificity of voice processing cortical response.

La voix est un stimulus auditif omniprésent dans notre environnement sonore. Elle permet non seulement la parole, mais serait aussi l’équivalent d’un visage auditif transmettant notamment des informations identitaires et affectives importantes. Notre capacité à discriminer et reconnaître des voix est socialement et biologiquement importante et elle figure parmi les fonctions les plus importantes du système auditif humain. La présente thèse s’intéressait à l’ontogénèse et à la spécificité de la réponse corticale à la voix humaine et avait pour but trois objectifs : (1) mettre sur pied un protocole électrophysiologique permettant de mesurer objectivement le traitement de la familiarité de la voix chez le sujet adulte; (2) déterminer si ce même protocole pouvait aussi objectiver chez le nouveau-né de 24 heures un traitement préférentiel d’une voix familière, notamment la voix de la mère; et (3) mettre à l’épreuve la robustesse d’une mesure électrophysiologique, notamment la Fronto-Temporal Positivity to Voices, s’intéressant à la discrimination pré-attentionnelle entre des stimuli vocaux et non-vocaux. Les résultats découlant des trois études expérimentales qui composent cette thèse ont permis (1) d’identifier des composantes électrophysiologiques (Mismatch Negativity et P3a) sensibles au traitement de la familiarité d’une voix; (2) de mettre en lumière un patron d’activation corticale singulier à la voix de la mère chez le nouveau-né, fournissant le premier indice neurophysiologique de l’acquisition du langage, processus particulièrement lié à l’interaction mère-enfant; et (3) de confirmer l’aspect pré-attentionnel de la distinction entre une voix et un stimulus non-vocal tout en accentuant la sélectivité et la sensibilité de la réponse corticale réservée au traitement de la voix. / Voice is a very prominent auditory stimulus in our acoustic environment. It is not only the carrier of speech, but would also be an auditory face that conveys important affective and identity information. Our ability to discriminate and recognize voices is socially and biologically important as it is amongst the most important functions of the human auditory system. This thesis was interested in the ontogenesis and specificity of the cortical response to human voice and had three objectives: (1) to develop an electrophysiological protocol to objectively measure the processing of voice familiarity in adult subjects; (2) to assess whether the same electrophysiological protocol could also objectify preferential processing of a familiar voice in 24-hour-old newborns, in particular the mother’s voice; and (3) to test the robustness of an electrophysiological measure, more specifically the Fronto-Temporal Positivity to Voices, interested in pre-attentional discrimination between vocal and non-vocal stimuli. Results from these three experimental designs have enabled (1) to identify electrophysiological components (Mismatch Negativity and P3a) sensitive to the processing of voice familiarity; (2) to highlight a singular pattern of cortical activation to the mother’s voice in newborns, providing the first neurophysiological evidence of language acquisition, a process especially related to the mother-child interaction; and (3) to confirm that vocal/non-vocal discrimination is a pre-attentional process, while enhancing the selectivity and the specificity of voice processing cortical response.

Pour progresser dans la compréhension des mécanismes liés aux instabilités des cavités souterraines à partir de la réponse microsismique associée, l'effondrement provoqué d'une cavité saline (~ 200 m en diamètre), créée par dissolution, a été instrumentée sur un site d’exploitation de SOLVAY à Cerville-Buissoncourt (Lorraine, France). Pendant l’expérimentation un vaste ensemble des données a été enregistré (~ 50,000 fichiers d'événements) dont la majorité (80%) est constitué d’essaims microsismiques singuliers. Cette thèse présente une analyse et une interprétation détaillée de cette base de données microsismiques grâce à l’adaptation de méthodologies de traitement originales, dont les résultats améliorent notre compréhension sur la nature de la microsismicité liée aux processus de création et d’effondrement des cavités souterraines, ainsi que sur l’évaluation de l’aléa associé. Les résultats principaux obtenus sont les suivants : les événements microsismiques sont comparables à des petits séismes tectoniques ayant des magnitudes de moment variant entre -3 et 1. (ii) L’ensemble des événements microsismiques montre un mécanisme en cisaillement (double-couple) remarquablement stable et est associé à un régime en faille inverse d’orientation NO - SE, plongeant à environ 35°– 55°. Ce phénomène est probablement lié à la présence de fractures préexistantes sur le site. (iii) L'origine des essaims microsismiques est certainement due à l'incapacité du système à créer des fractures de grandes dimensions capables de libérer des contraintes très importantes. Cela est probablement lié aux propriétés mécaniques du toit de la cavité. (iv) Les périodes d’effondrements du toit de la cavité sont associées à une dynamique de forçage systématique et montrent une réponse microsismique particulière, qui peut-être décrite par des lois statistiques. Les travaux de recherche de cette thèse confirment également, que la surveillance microsismique peut être un outil puissant pour étudier les processus d’instabilité des cavités souterraines, même avec un nombre réduit de capteurs si des outils d’analyse adaptés sont utilisés / In order to improve our understanding of hazardous underground cavities and its microseismic response, the development and collapse of a ~ 200 m wide salt solution mining cavity was monitored at Cerville-Buissoncourt in the Lorraine basin in NE France. The majority of the obtained dataset (~80%) was constituted of numerous unusual microseismic swarming events (~50.000 event files). This thesis presents innovative methods able to treat this specific microseismic data set, whose results provide new and fundamental insights into the principal characteristics of caving and collapsing related microseismicity and hazard assessment of excavated underground formations. The principal results are as follows: (i) the individual microseismic events are comparable to small natural tectonic earthquakes with moment magnitudes Mw ranging from around -3 to 1. (ii) Source mechanisms for most microseismic events are remarkable stable and demonstrate a predominant thrust faulting (double-couple) regime with faults similarly oriented NW-SE, dipping 35°-55° , what might be related to the presence of systematically arranged pre-existing fractures. (iii) The origin of microseismic swarming is suggested in the incapacity to sustain larger strains and to release larger stresses, what seems to be related to the mechanical constitution of the rock strata overlying the cavity (i.e. low strength materials). (iv) Caving and collapsing periods at the cavity roof are associated with systematic, self- reinforcing dynamics and have a distinct microseismic response, clearly observable from statistical analysis, which can be precisely described by empirical laws. The performed analysis and interpretation of the microseismicity at Cerville-Buissoncourt has shown that microseismic monitoring is a useful tool to constrain the mechanical and dynamical characteristics of an evolving and collapsing hazardous underground cavity

The Master´s Thesis deals with the posibility of receiving funding from the European Union. It contains a theoretical background about structural funds and an analysis of the present status of a selected company. The practical part focuses on selection of the most suitable endowment title, checking of financial health of a company as a fundamental prerequisite for applying for EU funds and on preparation of the project development application.

The diploma thesis deals with elaboration of project proposal of new product and its introduction into production using project management methodology. The project proposal is designed for a company that produces agricultural technology