Protection of groundwater sources has become an important issue in Canada. Over the last decade many approaches to the protection of groundwater sources have evolved. Some approaches provide qualitative information while others give quantitative values with respect to protection measures. The objective of the thesis is to examine the existing approaches of source water protection (SWP) using a complex geological setting, and introduce new methodologies towards the quantitative measurement of the various steps of SWP. The information obtained from the studies can be used to set up future guidelines for SWP. The first step in SWP is to assess the vulnerability of an aquifer. In this thesis, we compare three approaches for evaluating aquifer vulnerability: the Index Approach (Intrinsic Susceptibility Index, or ISI), the Hydraulic Resistance (HR) Approach (similar to the Aquifer Vulnerability Index, or AVI) and the Travel Time Approach (Surface to Aquifer Advective Time, or SAAT). The ISI approach uses the thickness and vertical hydraulic conductivity of the layers overlying an aquifer, and the vulnerability is expressed as a numerical score which is related to these parameters but is not physically based. The HR approach is physically based, uses the same parameters as ISI with the addition of porosity, and results are in the form of travel time under a unit gradient. SAAT extends the physically based approach by including the unsaturated zone and using the actual downward gradient; results are given in terms of advective travel time from surface to aquifer. These three approaches are compared, using two different aquifer systems. The second step in SWP is the delineation of wellhead protection areas (WHPAs). The WHPA delineates the area within which a source of contamination could have an impact on the well. The actual impact on the well depends not only on the source, but also on the characteristics of the groundwater system. Important considerations include the dimensionality of the system, the uncertainty in the system characteristics, and the physical processes that could affect the impact. The conventional approach is to define different time of travel (TOT) zones based on backward advective particle tracking. An alternative approach is to apply backward advective-dispersive solute transport modelling, in which dispersion can be taken as representing the uncertainty in defining the hydrogeologic characteristics (e.g. hydraulic conductivity) of the aquifer. The outlines of the TOT zones in the backward advective particle tracking approach is obtained by drawing an envelope around the respective tracks, which may require considerable guesswork. In the backward-in-time transport modelling, the outline of the TOT zones are developed using mass balance principles. The third step is the assessment of well vulnerability. Well vulnerability is based on the source-pathway-receptor concept which analyses the transport and fate of the contaminants along its path from the source to the receptor, and the interaction of the well itself with the flow system, and thus determines the actual impact on the well. The impact can be expressed in terms of the contaminant concentration in the well water. The mapping of the impact can be carried out by using a standard advective-dispersive transport model in either a forward-in-time mode (for a known contaminant source) or in a backward-in-time mode (for unknown sources). Thus, the well vulnerability concept goes beyond the conventional approach of WHPA, which is based solely on advective transport, neglecting dispersion and chemical processes. For any known point or non-point time-varying contaminant sources located arbitrarily within the well capture zone, the expected concentration at the well can simply be evaluated by convoluting the source mass with the results of the well vulnerability without further use of the model. Convolution is a well-known and effective superposition method to deal with arbitrary inputs in time and space for linear systems. The information of the contaminant concentration in the well water can be used to quantify the risk of a well becoming contaminated. Risk can be expressed in terms of the exposure value of the contaminant concentration exceeding the allowable limit and the time frame within which the well becomes contaminated. The exposure value can be integrated with the time element to set up a ranking of priorities, or to calculate the investment that must be made today in order to have the required funds available for remediation at the time it becomes necessary. The concept is applied to a well using hypothetical contaminant sources located arbitrarily within the capture zone. Well vulnerability maps can be used as a powerful tool to identify the optimal locations for Beneficial Management Practices (BMPs). A case study addressing the problem of elevated nitrate levels in a drinking water supply well is used to demonstrate the principle. The reduction of nitrate input concentration within the most vulnerable areas shows the largest impact at the well.

Protection of groundwater sources has become an important issue in Canada. Over the last decade many approaches to the protection of groundwater sources have evolved. Some approaches provide qualitative information while others give quantitative values with respect to protection measures. The objective of the thesis is to examine the existing approaches of source water protection (SWP) using a complex geological setting, and introduce new methodologies towards the quantitative measurement of the various steps of SWP. The information obtained from the studies can be used to set up future guidelines for SWP. The first step in SWP is to assess the vulnerability of an aquifer. In this thesis, we compare three approaches for evaluating aquifer vulnerability: the Index Approach (Intrinsic Susceptibility Index, or ISI), the Hydraulic Resistance (HR) Approach (similar to the Aquifer Vulnerability Index, or AVI) and the Travel Time Approach (Surface to Aquifer Advective Time, or SAAT). The ISI approach uses the thickness and vertical hydraulic conductivity of the layers overlying an aquifer, and the vulnerability is expressed as a numerical score which is related to these parameters but is not physically based. The HR approach is physically based, uses the same parameters as ISI with the addition of porosity, and results are in the form of travel time under a unit gradient. SAAT extends the physically based approach by including the unsaturated zone and using the actual downward gradient; results are given in terms of advective travel time from surface to aquifer. These three approaches are compared, using two different aquifer systems. The second step in SWP is the delineation of wellhead protection areas (WHPAs). The WHPA delineates the area within which a source of contamination could have an impact on the well. The actual impact on the well depends not only on the source, but also on the characteristics of the groundwater system. Important considerations include the dimensionality of the system, the uncertainty in the system characteristics, and the physical processes that could affect the impact. The conventional approach is to define different time of travel (TOT) zones based on backward advective particle tracking. An alternative approach is to apply backward advective-dispersive solute transport modelling, in which dispersion can be taken as representing the uncertainty in defining the hydrogeologic characteristics (e.g. hydraulic conductivity) of the aquifer. The outlines of the TOT zones in the backward advective particle tracking approach is obtained by drawing an envelope around the respective tracks, which may require considerable guesswork. In the backward-in-time transport modelling, the outline of the TOT zones are developed using mass balance principles. The third step is the assessment of well vulnerability. Well vulnerability is based on the source-pathway-receptor concept which analyses the transport and fate of the contaminants along its path from the source to the receptor, and the interaction of the well itself with the flow system, and thus determines the actual impact on the well. The impact can be expressed in terms of the contaminant concentration in the well water. The mapping of the impact can be carried out by using a standard advective-dispersive transport model in either a forward-in-time mode (for a known contaminant source) or in a backward-in-time mode (for unknown sources). Thus, the well vulnerability concept goes beyond the conventional approach of WHPA, which is based solely on advective transport, neglecting dispersion and chemical processes. For any known point or non-point time-varying contaminant sources located arbitrarily within the well capture zone, the expected concentration at the well can simply be evaluated by convoluting the source mass with the results of the well vulnerability without further use of the model. Convolution is a well-known and effective superposition method to deal with arbitrary inputs in time and space for linear systems. The information of the contaminant concentration in the well water can be used to quantify the risk of a well becoming contaminated. Risk can be expressed in terms of the exposure value of the contaminant concentration exceeding the allowable limit and the time frame within which the well becomes contaminated. The exposure value can be integrated with the time element to set up a ranking of priorities, or to calculate the investment that must be made today in order to have the required funds available for remediation at the time it becomes necessary. The concept is applied to a well using hypothetical contaminant sources located arbitrarily within the capture zone. Well vulnerability maps can be used as a powerful tool to identify the optimal locations for Beneficial Management Practices (BMPs). A case study addressing the problem of elevated nitrate levels in a drinking water supply well is used to demonstrate the principle. The reduction of nitrate input concentration within the most vulnerable areas shows the largest impact at the well.

La présence de Composés Organiques Halogénés Volatils (COHV) dans les eaux souterraines est particulièrement problématique car celle-ci peut les contaminer pendant des dizaines d’années. Le traitement de ces sites, dont les pollutions sont souvent mixtes et complexes, nécessite de développer des techniques fiables. Particulièrement dans le contexte d’un aquifère hétérogène, les techniques classiques souffrent du faible potentiel de balayage de la pollution par les agents remédiant. La génération de mousse in situ est une technique de dépollution innovante permettant de contrôler la mobilité de ces agents. La démarche générale du travail de thèse porte sur le développement de l’injection de mousse et son applicabilité dans le contexte d’un site industriel en activité contaminé par des solvants chlorés. L’originalité de ce travail porte sur l’utilisation de la mousse comme agent confinant (hydrauliquement) d’une zone source au sein même de l’usine. Le premier objectif de cette thèse consiste en la définition de l’origine de la pollution et des processus responsables du transport des composés dissous dans un système aquifère multicouche et hétérogène. Dans ce cadre, un modèle de transport en 3 dimensions a été développé et contraint à partir de données géologiques (structure 3D), hydrogéologiques (piézométrie, mesure de vitesse), et chimiques (solvant chlorés et ions majeurs). Cette combinaison d’approches a eu un impact réel sur la compréhension de la dynamique hydrogéologique du système souterrain présent au droit du site d’étude, et a permis de définir la zone d’injection de mousse. En parallèle des travaux de caractérisation du site, un travail expérimental en laboratoire a permis de définir les mécanismes à l’origine de la réduction de la perméabilité relative à l’eau par l’injection de mousse. Au travers d’une approche multi échelle, ces travaux ont notamment permis de (i) définir une formulation (concentrations en tensio-actifs et composition) et les paramètres d’injection (qualité de mousse, débit d’injection, mode d’injection) pour générer une mousse favorable à la réduction de la saturation en eau (colonne 1D). Cette baisse de saturation ayant conduit à une réduction de la perméabilité à l’eau d’un facteur supérieur à 100. (ii) D’estimer le comportement de la mousse le long d’un profil d’injection et son impact sur la réduction de la saturation en eau à quelques centimètres d’un point d’injection (pilote 2D décimétrique). (iii) De vérifier l’applicabilité en 3D du système d’injection développé et de suivre les évolutions de l’impact de la mousse dans un aquifère (essai sur un piézomètre réel). Enfin, un test d’injection de mousse en continu pendant 96h au droit de la zone source du site industriel contaminé a été réalisé. La réalisation d’un essai de pompage (post injection) au centre de la zone confinée, couplé à des mesures de flux de polluant (pré et post injection) et implémenté dans un modèle 2D, ont permis de mettre en évidence l’impact réel de la mousse sur un rayon supérieur à 2m avec une réduction du flux de polluant en aval hydraulique d’un facteur 4,5. Les différents travaux de laboratoire et de modélisation mettent en évidence les avancées et limites de la technique développée et permettent de proposer des voies d’amélioration. / The presence of Volatile Organic Compounds (VOCs) in groundwater is particularly problematic because it can contaminate them for decades. The treatment of these sites, whose pollution is often mixed and complex, requires the development of reliable techniques. Particularly in the context of a heterogeneous aquifer, conventional techniques suffer from the low potential for pollution sweeping by remedying agents. In situ foam generation is an innovative technique to control the mobility of these agents. The general approach of the thesis concerns the development of foam injection and its applicability. in the context of an industrial site in activity contaminated by chlorinated solvents. The originality of this work concerns the use of foam as a confining agent (hydraulically) of a source zone within the plant itself. The first objective of this thesis is to define the origin of pollution and the processes responsible for the transport of dissolved compounds in a multilayered and heterogeneous aquifer system. In this context, a 3-dimensional transport model has been developed and constrained based on geological data (3D structure), hydrogeological data (piezometry, groundwater velocity measurement), and chemical data (chlorinated solvent and major ions). The combination of these approaches had a real impact on the understanding of the hydrogeological dynamics of the underground system present at the study site, and allowed us to define the foam injection zone. In parallel with site characterization, experimental work in laboratory allows to define the mechanisms responsible for reducing the water-relative permeability by injecting foam. Through a multi-scale approach, we (i) define a formulation (surfactant concentrations and composition) and injection parameters (foam quality, injection rate, injection mode ) to generate a foam favorable to the reduction of water saturation (column 1D). This saturation decrease led to a reduction of the water relative permeability by a factor greater than 100. (ii) Estimate the behavior of the foam along an injection profile and its impact on the reduction saturation in water a few centimeters from an injection point (2D decimetric pilot). (iii) Verify the 3D applicability of the developed injection system and to follow the evolutions of the impact of the foam in an aquifer (test on a real piezometer). Finally, a continuous foam injection test during 96 hours at the source zone of the contaminated industrial site was carried out. Conducting a pumping test (post injection) in the center of the confined zone, coupled with contaminant fluxes measurements (pre and post injection) and implemented in a 2D model, highlighted the real impact of the injected foam over a radius greater than 2m with a reduction of the flow of pollutant downstream by a factor 4.5. The various laboratory and modeling work highlight the advances and limitations of the developed technique and make it possible to propose ways of improvement.

Accurate modeling of contaminant transport and sorption processes in the soil and landfill liners is a prerequisite for realistic model simulations of contaminant fate and transport in the environment. These studies are also important for the remediation of soil and groundwater contamination. Modeling of contaminant transport through soils and landfill liners consists of either solving the direct/forward problem or the inverse problem. In this thesis, an automated time-stepping implicit procedure is developed from the convergence and error studies of explicit and implicit finite-difference solutions for the advection-dispersion transport of contaminants through soil with different sorption mechanisms. This study is further extended for transient through-diffusion (TTD) transport of contaminant in landfills by considering linear sorption mechanism. To validate the numerical solution and also to study the behavior of finite-difference numerical solutions for TTD transport problem, closed-form analytical solution is derived. Further, a new interface condition is proposed based on the finite-volume procedure for stratified soil or landfill liner system. Solvers are developed for the parameter estimation of inverse problem by integrating the developed procedures for the above forward problem with different optimization procedures. Solvers based on Simulated Annealing (SA) and Genetic Algorithm (GA) are developed for TTD transport in the landfill liners and verified with the existing methods of parameter estimation. Novel swarm intelligence based solver is developed for the first time for parameter estimation in contaminant transport inverse problem to overcome some of the limitations of the classical optimization methods and evolutionary methods such as GA. Additionally, the proposed swarm intelligence based algorithms and a new variant is applied to solve ill-posed problem of contaminant source characterization. The presented work in this dissertation can be unswervingly applied for modeling the contaminant transport in laboratory through-diffusion tests and contaminant transport through landfill liners where the transport is usually considered to be one-dimensional and also diffusion-dominated. Similarly, the advection-dispersion transport through laboratory soil columns can also be modeled with the developed, fast, automated, implicit numerical procedure with very good accuracy. The present study can be applied further for contaminant transport through stratified soil/liner system using fast converging numerical algorithms. Finally, the problems of design parameter estimation and source characterization can be handled accurately by the use of developed automated nature-inspired solvers.

La Nouvelle-Calédonie détient d’importantes réserves de nickel latéritique et est devenue, en 2017, le cinquième producteur mondial de Ni. Ces dépôts sont habituellement considérés comme résultant d’altération latéritique intense de la péridotite, qui constitue la principale source de nickel. Ainsi, le principal modèle conceptuel de la formation des minerais de nickel latéritique en Nouvelle-Calédonie est un modèle per descensum où la plupart des éléments (Mg, Ni et Si) ont été lessivés depuis la surface, en particulier lors du développement du sol latéritique. Le nickel est ensuite reprécipité, soit dans la goethite de la latérite fine, soit au niveau de la saprolite, sous forme de goethite et de silicates Mg-Ni, dont des talc-like ou kérolite. Les observations minéralogiques et structurales récentes ainsi que les données minières ont cependant mis en évidence de nombreux types d’hétérogénéités dans les concentrations, et la distribution des porteurs de Ni. Comprendre les facteurs la mobilité de cet élément, ses mécanismes de piégeage ainsi que les paramètres chimiques et hydrodynamiques à l’origine de ce piégeage, est essentiel afin de prévoir la distribution du nickel dans les profils latéritiques en Nouvelle-Calédonie, et constitue l’objectif de cette thèse. Ce travail est basé sur le développement (i) d’un modèle 1D s’intéressant en particulier au comportement géochimique du nickel lors de l’altération de l’ophiolite, sa comparaison avec les observations in situ et une compréhension détaillée de la mobilité des éléments traces pendant le processus, et (ii) d’un modèle 2D hydro-géochimique couplé avec l’hydrodynamique complexe des profils latéritiques, améliorant ainsi la connaissance du contrôle structural sur la redistribution et la minéralisation du nickel. Tandis que les simulations 1D permettent de mieux comprendre les aspects chimiques contrôlant les processus de rétention du nickel au sein d’un profil, le modèle 2D se révèle être un outil puissant pour la compréhension de la formation des dépôts locaux les plus riches en nickel. Les résultats du modèle 2D montrent une remobilisation du nickel depuis les horizons supérieurs puis sa reprécipitation sous forme de silicates dans la saprolite. Le nickel remobilisé provient principalement de la zone saprolitique à cause de la dissolution des silicates de nickel formés précédemment ainsi que de l’olivine résiduelle de cette zone. Ce modèle a également révélé que l’horizon latéritique (et en particulier les oxy- hydroxydes de nickel) avait un faible impact dans la remobilisation du nickel. L’infiltration latérale de l’eau contenant le nickel dissout issu des formations surincombantes est à l’origine de la formation des zones les plus riches dans les parties inférieures du profil. Cette redistribution est entièrement contrôlée par l’hydrodynamique locale, la topographie ainsi que l’orientation et la position des fractures. Les modèles présentés permettent d’expliquer les processus de formation des minerais de nickel latéritique saprolitique, améliorant ainsi la compréhension des paramètres contrôlant la mobilité des éléments traces dans un environnement ultramafique. Ceci donne une nouvelle clé de distribution du nickel dans les profils actuels, qui peut devenir un outil pour la prospection minière, et la recherche de nouvelles ressources exploitables / New Caledonia hosts significant lateritic nickel reserves, and presently became the fifth largest Ni producer in the world. These deposits are generally thought to be closely as- sociated with the intense chemical and mechanical weathering of peridotite bedrock that is a principal source of nickel. Thus, the main ore genesis model for Ni ores in New Caledonia is based on a single per descensum model where most elements (Mg, Ni, and Si) are leached from the surface, particularly, during lateritic soil development. Nickel is then concentrated either in the fine-grained laterite where goethite is the main Ni bearer, the so-called ’lateritic ore’, or below the laterites, in the saprolite level, where nickel occurs as goethite and several types of Mg-Ni silicates, in particular kerolite. Recent mineralogical and structural observations together with mining data have revealed a lot of different types of heterogeneities associated with the distribution and mineralogy of Ni bearing minerals. Therefore, in depth investigations of Ni mobility, its retardation processes along with its governing chemical and hydrodynamic parameters are of big importance for understanding and subsequent prediction of Ni distribution in profiles of New Caledonia. Such an investigation is an objective of the present work. The concept is based on the development of i) a powerful 1D model with particular emphasis on Ni geochemical behavior during ophiolite weathering, its comparison with in situ observations, and detailed understanding of trace elements mobility, and ii) 2D hydro-geochemical model coupled with complex hydrodynamics, that would additionally provide new insight into the structural control on Ni redistribution and mineralization. While the 1D simulations provide a remarkable result for understanding the chemical features that drive Ni retention processes in a profile, 2D model appears to be a powerful tool for understanding how local Ni-enrichments may form. The results of this model show the reactivation of Ni from upper horizons and its concentration in neo-formed silicates in bottom of the saprolite. The reactivated Ni comes mostly from the saprolite horizon due to the redissolution of previously formed Ni-bearing silicates and still persisting in this olivine zone. Modeling has revealed minor contribution of the laterite horizon (Ni-oxi-hydroxides) into the Ni remobilization. The lateral infiltration of water with remobilized Ni from areas such as topographic highs to downstream slope areas leads to the formation of richest deposits in this lower part of profile. The manner of redistribution is fully governed by the topographic slopes, orientation and position of the fractures. Presented models appear to be of importance in attempt of explanation of Ni mineralization processes, revealing the main keys to understanding the control of trace elements mobility in ultramafic environment. The latter gives new insights into the Ni distribution in present day profiles and, therefore, may greatly help in mineral prospecting and forecasting the distribution of future resources

Eutrophication is one of the key environmental problems of today, both in terms of complexity and magnitude. For the Baltic Sea (BS), eutrophication is an acute problem, leading to hypoxic conditions at the bottom; a situation that is sustained and amplified, when phosphorus is released from hypoxic sediments. Reducing nutrient loading is a top political priority but the present situation is believed to require active measures within the catchments and recipients to reduce both loading and adverse effects. Implementation of effective and cost-efficient abatement methods requires understanding of natural processes in watersheds, streams and recipients as well as technological expertise in order to compare the effects of measures of different kinds and locations. This thesis tries to combine process understanding of catchment transport behaviour, especially in coastal zones, and feasibility of certain technologies for reducing nutrient loading and effects of eutrophication in-situ. The over-arching theme is the fate of the individual contaminant, from injection to removal. Transport and dispersion in catchments are investigated, combining physically-based, distributed, numerical groundwater models with Lagrangian stochastic advective reactive solute (LaSAR) transport modelling. The approach is powerful in the sense that it incorporates catchment structural, geomorphological dispersion in the numerical model with hydrodynamic and sub-scale dispersion as well as uncertainty in the LaSAR framework. The study exemplifies the complex nature of transport time distributions in catchments in general and when varying source size and location, importance of dispersion parameters and retention due to molecular diffusion. It is shown that geomorphological control on dispersion is present even for relatively heterogeneous systems and that neither the mean residence time nor a statistical distribution may provide accurate representations of hydrological systems. To combat internal loading of P from sediments in-situ, large-scale aeration of deep waters, halocline ventilation, has been suggested. This study further investigates the feasibility of wave-powered devices to meet the energy demands for such an operation. It is shown that the required amount of oxygen needed to keep the sediments at oxic conditions could be provided, cheaply and efficiently, through the use of wave power. / QC 20120511

Travel demand management policies such as congestion charges encourage car-users to change among other things route, mode and departure time. Departure time may be especially affected by time-varying charges, since car-users can avoid high peak hour charges by travelling earlier or later, so called peak spreading effects. Conventional transport models do not include departure time choice as a response. For evaluation of time-varying congestion charges departure time choice is essential. In this thesis a transport model called SILVESTER is implemented for Stockholm. It includes departure time, mode and route choice. Morning trips, commuting as well as other trips, are modelled and time is discretized into fifteen-minute time periods. This way peak spreading effects can be analysed. The implementation is made around an existing route choice model called CONTRAM, for which a Stockholm network already exists. The CONTRAM network has been in use for a long time in Stockholm and an origin-destination matrix calibrated against local traffic counts and travel times guarantee local credibility. On the demand side, an earlier developed departure time and mode choice model of mixed logit type is used. It was estimated on CONTRAM travel times to be consistent with the route choice model. The behavioural response under time-varying congestion charges was estimated from a hypothetical study conducted in Stockholm. Paper I describes the implementation of SILVESTER. The paper shows model structure, how model run time was reduced and tests of convergence. As regards run time, a 75% cut down was achieved by reducing the number of origin-destination pairs while not changing travel time and distance distributions too much. In Paper II car-users underlying preferred departure times are derived using a method called reverse engineering. This method derives preferred departure times that reproduce as well as possible the observed travel pattern of the base year. Reverse engineering has previously only been used on small example road networks. Paper II shows that application of reverse engineering to a real-life road network is possible and gives reasonable results. / <p>QC 20170222</p> / Silvester

Predicting the passenger flow inside a city is a vital component of the intelligent transportation management system. The proposal for a new residential area, an office space, post­pandemic policy implications for work from home, behavioral changes for revised traffic patterns, infrastructural improvements, require a visual and analytical backing which can be provided through a macro simulation model. This research explores the performance of the Machine learning (ML) based transport model against the predictions provided by the traditional Spatial Interaction Models (SIM) for the city of Oslo. The transport models and their parameters are analyzed for sensitivity analysis and scenario analysis to derive city character. Furthermore, the derived model is deployed over an interactive dashboard for analytical and their practical visualizations through infographics. The results show that the ML model outperforms the SIM. Although the traditional SIM has a clear advantage of being interpreted by design and requiring a few parameters, it suffers from its inability to accurately capture the structure of real flows and greater variability as compared to the ML model. Extensive statistical analyses are conducted to obtain significant results and realize the pros and cons of both the models which question the validity of results for the ML model over SIM. With this thesis, we discuss the potential of ML model detected trends of passenger flows, andtheir capacity to simulate city development­related scenarios for the traffic flows within the city.

Autonomous vehicles are predicted to disrupt the current landscape of urban mobility.Many studies have investigated how autonomous vehicles, either operated as a serviceor as private cars, could compete against public transport and even replace it. Fewerstudies have investigated how autonomous vehicles could actually be an opportunityfor the public transport sector, as a new type of offer that would cover specific needsalong traditional modes such as buses or metros.The aim of this project is to quantify the effect of replacing part of the public transportnetwork of Uppsala by demand-responsive autonomous fleets. This is achieved bybuilding a transport model based on the traditional four-step transport model andcalculating the total cost of the network both from the passenger and the operator’sperspective.The study shows that autonomous vehicles can slightly improve the performance of thenetwork and work best when combined with traditional bus lines. However, they alsoincrease the traffic and have a risk to cause congestion.

La finalité de notre thèse est d'interroger la relation entre la planification urbaine et le développement durable en regard des politiques publiques urbaines de développement et plus particulièrement du management territorial. Dans notre conception de la démarche, nous mettons l'accent sur les différentes dynamiques (déplacements et mobilité) et mutations morphologiques et structurelles (occupations du sol) à l'échelle de l'agglomération afin de singulariser la décision publique en matière du développement urbain durable. Nos choix se sont inscrits volontairement dans une logique transdisciplinaire qui s'est révélée particulièrement adaptée à nos différents recours théoriques, méthodologiques et empiriques.Traitant le contexte tunisien en pleine transition, ce travail propose un modèle de décision publique hybride permettant de déterminer les traits d'une planification urbaine adaptée aux différents contextes actuels et aux exigences de la durabilité. Le recours à la fois aux techniques de la prospective, aux modèles intégrés de déplacements et d'occupation du sol, à la simulation et aux techniques d'analyse multicritère nous a permis une construction intégrée et itérative de plusieurs niveaux d'évaluation partielle et d'un niveau d'évaluation globale. Le modèle conçu et testé pour la ville de Sousse permettra aux décideurs publics de disposer d'une grille synthétique d'informations issues d'une prise en compte aussi complète que possible de la réalité urbaine. Il offre différents niveaux d'évaluation thématique et un niveau global intégrant l'ensemble. La démarche pourra servir de référent à d'autres villes tunisiennes et aura par conséquent un impact réel sur la qualité de leur développement. / The purpose of this thesis is to investigate the relationship between urban planning and sustainable development in relation to urban development policy and in particular the territorial management. In our design approach, we focus on the different dynamics (travel and mobility) moreover, morphological and structural changes (land uses) on the scale of the urban area in order to single public decision in urban development long lasting. Our choices were enrolled voluntarily in a trans-disciplinary logic has proved to be particularly suited to our different theoretical, methodological and empirical appeal.Treat the Tunisian context in transition; this research proposes a hybrid public decision model to determine the features of an adapted urban planning to different contexts and current requirements of sustainability. Using both foresight technics, integrated land use-transportation models, simulation and multi-criteria analysis technics allowed us an integrated and iterative construction of several levels of partial evaluation and a level of overall evaluation. The model designed and tested for the town of Sousse will allow policy makers to have a synthetic grid information from a decision as complete as possible account of urban reality. It offers different levels of thematic evaluation and a global level integrating all. The approach can be used as reference to other Tunisian cities and will therefore have a real impact on the quality of their development.