Theses defended at ESPACE

List of theses defended at the ESPACE laboratory by year



4 march 2024, Nice

The potentialities of the “built / unbuilt” interfaces in a territorial and environmental resilience approach : Applied Research to Industrial Zones of the right bank of the Var river (Alpes-Maritimes)

Danielle DAGENAIS, Professor, Université de Montréal
Patrizia LAUDATI, Professor, Université Côte d’Azur
Yaser ABUNNASR, Associate Professor, American University of Beirut
Giovanni FUSCO, Research Director CNRS, Université Côte d’Azur
Christine VOIRON-CANICIO, Professor, Université Côte d’Azur, Thesis Director

AbstractThe thesis consists in a research on the general resilience of a peri-urban territory where industrial zones occupy a predominant place. These industrial zones represent a paradox of being both dynamic and aging. Land is increasingly scarce and constrained, and the natural landscape is neglected. The aim of this thesis is to work on the long-term resilience of this industrial space, whose activities are not in difficulty, but which we propose to make more capable of facing the changes to come, and more attractive, in improving its structure and function. The novelty in this research is to approach the question of territorial resilience through a spatio-morphological entry. The entire territory is structured in a nesting of interfaces at different scales. The organization of these built / unbuilt spaces is seen as a latent, untapped potential for environmental and territorial resilience to be developed, in particular by creating “recycled landscapes” and “utility corridors”. A development aiming to create renatured spaces would make it possible to enhance the area not only on the ecological level, but also on the socio-economic level, by improving the quality of life at work and its internal functioning, by promoting soft mobility, for example. We assume that the built / unbuilt interface is a major link in the chain of factors contributing to environmental and territorial resilience. The thesis therefore pursues a threefold objective: – Draw attention to the types of built / unbuilt interfaces, of various nature, size and shape – interstitial spaces, green links, edges, etc. – by highlighting their roles in the development of the studied area. Demonstrate the potential of each type of interface for the dynamic of change that brings resilience. – Evaluate their combined effects on the general resilience of territorial sub-systems, from a multi-scalar and multidimensional perspective, environment, urbanity, quality of life and economic attractiveness, by anticipating the possible futures of the industrial zones through co-constructed scenarios; integrate the assessment of the impacts of modeled green infrastructure on the resilience of industrial zones in the long term (geoprospective and geogovernance methods). – Provide an approach that can be generalized to other types of territories also seeking general resilience: method of interface analysis, prospective resilience diagnostic protocol, methods of co-construction of geoprospective scenarios and methods of assessing their impacts on the dynamics of change through spatially explicit models. This research is part of the larger framework of the POPSU Métropole NCA project (2019-2022) which aims to co-construct a new model of controlled urban development, attractive and resilient in the face of future changes. It constitutes the local variation of the resilient and attractive territorial model that the POPSU project seeks for the entire metropolitan area of NCA, and is intended to serve as a reproducible operating framework for other territories in France.

Key word
Resilience, green infrastructure, interstices, rehabilitation, industrial zones
Consultable en ligne



9 january 2024, Aix-en-Provence

Artificial intelligence and massive multi-sensor image data for the optimization ofearly warning systems in agrosystems with high variability of climatic extremes

Sébastien GADAL, Professor, Aix Marseille Université, Thesis Director
Loubna El MANSOURI, Professor, Institut Agronomique Vétérinaire Hassan II, Thesis co-director
Albin ULLMANN, Associate Professor, HDR, Université de Bourgogne, Thesis rapporteur
Ahmed EL ABOUDI, Professor, Université Mohammed V, Thesis rapporteur
Julien ANDRIEU, Associate Professor, HDR Université Côte d’Azur, Reviewer
Maman GARBA, Senior researcher, INRA-Niger, Reviewer
Vincent MORON, Professor, Aix Marseille Université, President of the jury
Rachid HADRIA, Head of Research, INRA-Maroc, Reviewer

Agricultural vulnerability to drought is a growing concern in many parts of the world. Agricultural drought occurs when rainfall is insufficient or poorly distributed to meet crop water requirements, leading to reduced agricultural production and disruption to food systems. Sahelian agrosystems are considered extremely sensitive due to their low productivity and vulnerability to climatic hazards. Crops and pastures often face severe drought conditions and extreme heat, which reduce their growth and yields. As a result, the challenges in terms of food security, the resilience of rural communities and the sustainability of Sahelian farming systems are constantly increasing. The aim of this thesis is to develop new approaches to spatial modelling of agricultural vulnerability to drought using machine learning and deep learning, with a view to improving drought early warning metrics. To this end, the first part of the study draws on the historical availability of multisensor geospatial data to model and study trends in the historical dynamics of agricultural drought parameters through the development of a new multivariate composite index for agricultural drought monitoring (MDCI). The MDCI includes, for the first time, the principle of rainfall efficiency in addition to the variables commonly used in descriptive modelling of drought conditions. The second dimension of the study is an integrated approach to spatial modelling of agricultural drought conditions at 30 m resolution by comparatively testing the predictive performance of four artificial intelligence models, namely the Cubist model, DeepBoost, DNN and MLP. The last part of the study proposes a new multivariate index of vulnerability to agricultural drought (MADVI) by highlighting the gradients of indicator trends as vulnerability factors for training machine learning models (RF, SVM, KNN, NB, Cubist), deep learning (DNN, MLP) and an ensemble learning algorithm (DeepBoost) rather than the anomalies of the variables or the intensity of their frequencies commonly used. The MDCI demonstrated a high degree of spatio-temporal concordance between the validation indices, with a maximum correlation of 0.91 with the VHI and an R= 0.85 and a p < 5.8e-10 with the hydro-climatic indices (PDSI). Similarly, the validation of the MADVI modelled by automatic and depth learning showed a close match between the spatial variability of the MADVI vulnerability classes and those of the reference model based on the standard vulnerability modelling equation, with an R>0.85 and a very significant p-value <0.0000. The results of the modelling of historical dynamics suggest that, after the long dry decade of 1981-1990, the severity of agricultural drought has undergone a remarkable decline, even though conditions of moderate pockets of drought are relatively more frequent. In general, the interannual variability of agricultural drought conditions as measured by the MDCI is much less contrasted between 2000-2021 than that observed over the 1981-2000 period. For spatial modelling of vulnerability using spatial regression, Cubist and DeepBoost performed slightly better than DNN and MLP respectively, with RMSEs of 0.29 and 0.33. However, the DNN was characterized by a high sensitivity to the number of input variables. Consequently, the integration of new characteristics not taken into account in the model training is an avenue for improving the model’s predictive performance.

Key words
Artificial intelligence, Agrosystems, Agricultural drought, Climatic extremes, Geospatial technologies, Sustainable Developments
Consultable en ligne


19 january 2024, Nice

Environmental assessment of local food systems : Contribution of territorial life cycle analysis for a sustainable construction and assessment of agri-food reteritorialization policies and planning in France

Claire DELFOSSE, Professor, Université Lumières Lyon 2
Eléonore LOISEAU, Chief engineer ofPonts, des Eaux et des Forêts, INRAE
Guy ROBINSON, Professor, University of Adelaide
Nicolas VIAUX, Associate Professeur, Université Côte d’Azur
Salma LOUDIYI, Professor, VetAgro Sup
Stéphane BOUISSOU, Professor, Université Côte d’Azur

The current climatic and environmental upheavals deeply impact our society and ecosystems. Food production and consumption play an essential role in these disruptions. The globalized food system, characterized by mass production and consumption and a disconnect between producers and consumers, places significant pressure on territories and their resources. To address this, the reterritorialization of food is emerging as a promising solution for a more sustainable food system. France is at the forefront of this initiative, with the recent development of Territorial Food Projects (Projets Alimentaires Territoriaux, PAT) led by local authorities. These projects aim to promote a local and sustainable food economy while supporting short supply chains. However, despite their potential, the environmental impacts of these initiatives remain understudied. Most current studies compare short and long food supply chains, focusing on transportation-related impacts. Consequently, they do not consider the systemic benefits associated with the development of more sustainable practices within local food systems. To respond to this issue, this thesis adapts the Territorial Life Cycle Assessment (TLCA) methodology to assess local food systems from a systemic and environmental perspective. This adapted methodology is applied to two diverse territories in France and their local food systems: the municipality of Mouans-Sartoux and the Finistere department. Both territories are pioneers in implementing food territorial projects (PAT). The environmental assessment of the two local food systems reveals the extent of both direct and indirect impacts, notably related to imported products and agricultural inputs. The results demonstrate that transportation played a minor role in these impacts. Both local systems heavily depend on imports to feed their inhabitants. The short food supply chains, and consequently their impact, remain limited. The assessment of the impact of the Mouans-Sartoux local food policies demonstrates positive effects since their implementation. An estimated reduction in environmental impact equals approximately 20 % of the entire local food system’s impact, particularly in terms of climate change and land use. In Finistere, the prospective analysis highlights the potential benefits of local strategies, particularly in connection with the promotion of agroecological practices. In conclusion, this interdisciplinary work confirms the relevance of life cycle assessment methodologies for local planning and assessment. It equally reveals the potential of local food policies to contribute to the ecological transition.

Key words
Food Territorial Project, Ecological transition, Local planning, Life Cycle Assessment, Short food supply chain, Sustainable food
Consultable online



22 may 2024, Nice

Restoration of mangrove ecosystems: Evaluation and improvement of practices through a comparative study of projects worldwide

Laurent GODET, Research Director, Université de Nantes
Valéry GOND, Research Director, CIRAD
Claire GOLLÉRY, Associate Professor, CUFR Mayotte
Marie-Christine CORMIER-SALEM, Research Director, IRD
Sylvain PIOCH, Associate Professor, Université Paul Valéry
Julien ANDRIEUAssociate Professor HDR, Université Côte d’Azur


Mangroves represent unique coastal social-ecosystems situated in tropical and subtropical regions worldwide. They endure both anthropogenic and natural degradations, compromising their ecological functionalities and their numerous contributions to human societies. In response to this situation, a multitude of mangrove restoration projects have been initiated. However, in the long term, a significant portion of these initiatives has failed, highlighting a lack of understanding of the ecosystem and its relationship with human societies. Within this context, this thesis primarily aims to address the following question: How can the effectiveness of mangrove ecological restoration projects be improved? Three research objectives have emerged in response: (i) to assess the restoration projects’ effectiveness for nature and society, (ii) to identify gaps leading to poor practices, and finally, (iii) to propose solutions to enhance the projects. To achieve these objectives, a geographic analysis of restoration projects on a global scale was undertaken, examining each of the projects’ major phases: design, implementation of restoration measures, and monitoring-evaluation.
In the design phase, one of the major identified gaps is the lack of standardized pre-restoration site assessment methods accessible to managers. To address this need, the foundations of the Rapid Assessment Method for Mangroves (RAM MANGROVES) tool were developed. A bibliographic analysis, coupled with semi-structured interviews with experts and supported by a field test, led to a method consisting of 16 ecological and socio-economic descriptors, as well as a standardized field protocol.
During the implementation phase, a state of the art of restoration measures and a synthesis of scientific recommendations for their improvement were carried out. This involved a comparative analysis of global restoration projects and an in-depth study of a particular case in the Philippines. These efforts highlighted the crucial importance of a systemic approach integrating ecological, social, and economic aspects to maximize the chances of a restoration project’s success. They also confirmed a key point: rehabilitation of conditions conducive to mangrove development should be considered a priority, with replanting only considered if the environmental conditions favourable to the maintenance and development of mangroves have been rehabilitated or were not degraded. This work also illuminated that most scientific recommendations are not adhered to in many mangrove restoration projects.
Finally, in the monitoring-evaluation phase, a significant lack of project evaluation was identified. To encourage project holders to undertake this crucial step, a high-resolution remote sensing approach is proposed. This approach was used to evaluate the evolution of the vegetation cover of six restoration sites in Costa Rica, Senegal, Benin, and the Philippines. This analysis utilizes Landsat and Sentinel images. An analysis at multiple spatial and temporal scales was conducted to ensure the robustness of the interpretations. The results revealed a high accuracy of the mapping, with an average Kappa index of 0.95, confirming the reliability of this approach.
In summary, the effectiveness of mangrove restoration projects is closely linked to the quality of project design, the relevance of selected restoration measures, the active engagement of local communities, and the use of appropriate monitoring tools.
Key words
Mangrove restoration, rapid assessment method, remote sensing, social-ecological system, ecological monitoring
Consultable online


5 july 2023, Nice

Assessing spatiotemporal forest fire dynamics in southern France : A geospatial and machine learning approach

Etienne COSSART, Professor, Université Jean Moulin Lyon 3, Thesis rapporteur
Lea WITTENBERG, Professor, University of Haifa, Thesis rapporteur
Marielle JAPPIOT, Research engineer, INRAE, Reviewer
Dennis FOX, Professor, Université Côte d’Azur, Thesis Director
Elena DI BERNARDINO, ProfessorUniversité Côte d’Azur, Thesis co-supervisor

AbstractForest fires pose a significant threat to ecosystems and human welfare globally, especially in the Mediterranean region, where hot dry summers facilitate fire ignition and propagation. The fire season of 2022 was the second most severe in terms of burned area and number of fires in Europe since 2006. Recurrent droughts and higher temperatures in a context of climate change are expected to increase burned area by 200% in the Mediterranean region by 2090. Geospatial analysis is a powerful approach that enables a deeper understanding of the complex interactions, relationships, patterns, and trends between spatial phenomena, ultimately informing better decision-making and problem-solving. In addition, Machine Learning (ML) has emerged as a valuable tool in wildfire science, but its effectiveness depends on the availability of extensive, high-quality datasets. Spatial relationships between fire activity and environmental factors change over time due to climate change and fire management practices. Mapping spatial patterns and the primary driving forces impacting fire distribution provides valuable insights for civil protection agencies. Although fire suppression is often prioritized, prevention, mitigation, and community preparedness are crucial components of wildfire management. In addition, targeted wildfire prevention necessitates understanding and documenting forest fire causes, however, there is a substantial percentage of unknown-caused fires both in Europe and in France. This Ph.D. thesis examines the spatiotemporal evolution of burned areas in southern France over a 50-year period (1970-2019) ; it focuses on the interactions with topography and vegetation types using fire geodatabases. Results are divided into two 25-year periods due to the implementation of a new fire suppression policy in 1994. In the last 25 years, the burned area significantly decreased, and the geographic distribution of fires changed, particularly in regions with large fires. South-facing slopes became more fire-prone in the second period, whereas low slope inclinations were increasingly avoided after 1994. The majority of burned areas were strongly associated with the location of sclerophyllous vegetation clusters (shrublands), which are highly fire-prone and expand over time.Furthermore, this thesis presents an ML-based model that uses an explainable framework to predict the cause of fire ignitions in southern France based on environmental and anthropogenic features. Results indicate that the source of fires can be predicted with varying accuracy levels: natural fires have the highest accuracy (F1-score 0.87) compared to human-caused fires such as accidental (F1-score 0.74) and arson (F1-score 0.64). Spatiotemporal properties and topographic characteristics are deemed the most important features for determining the classification of unknown-caused fires in the study area.By leveraging large fire geodatabases, advanced geospatial data analysis, and ML techniques, this research demonstrates how geospatial approaches can quantify spatio-temporal trends in fire dynamics in a landscape and how ML techniques can be effectively used to fill gaps in fire ignition causes. It also underscores the need for better and more harmonized collection of data as this is a key component that will enhance the power of ML techniques in providing useful information for land management strategies.

Key words
Forest Fires, Geomatics, Geospatial analysis, Machine Learning, Spatiotemporal patterns




29 march 2023, Aix-en-Provence

Large coastal cities and their beaches. A public policy approach in Marseille (France), Barcelona and Valencia (Spain)

Michel DESSE, Professor, Université de Nantes, thesis rapporteur
Catherine MEUR-FEREC, Professor, Université de Bretagne Occidentale, Thesis rapporteur
Lise BOURDEAU-LEPAGE, Professor, Université Lyon III Jean-Moulin, Presidente of the jury
Elisabet ROCA BOSCH, Professor, Universitat Politècnica de Catalunya, Reviewer
Associate Professor HDR, La Rochelle Université, Reviewer
Samuel ROBERT, Research Director, CNRS, Thesis Director

Located at the interface of densely populated and complex urban areas, the beaches of large coastal cities are today subject to many pressures. In these cities, defined here as municipalities with more than 500,000 inhabitants, beaches are the first area of contact with the sea. As a result, they are particularly exposed to the risks associated with climate change, such as erosion or marine submersion. Due to their insertion in a dense, complex and sometimes very old urban environment, these beaches are frequently confronted with a degradation of their environmental quality, whether it is the quality of the bathing water or the sand. However, beaches play a key social role in these cities where the number of green and public spaces is sometimes limited. They are thus confronted with particularly large influxes of users, who frequent these beaches all year round and develop specific recreational and social practices. In large cities, these beaches have a dual function: they are both a popular point of contact with nature and an extension of the urban public space. The task of the public authorities responsible for the preservation of these spaces is therefore particularly heavy and difficult. Then, this thesis proposes an analysis of public policies for the management of beaches in three large cities in the north-western Mediterranean: Marseille, Barcelona and Valencia. Faced with the major social and environmental challenges arising on the beaches and coastlines of large coastal cities, the public management policies implemented in these three cities are compared and analysed using integrated coastal management as a conceptual reference. To this end, this thesis reports on the research activities carried out in these three study areas. These include an analysis of documents related to public beach policies; semi-structured interviews with stakeholders and associations; and a study of the answers given by public authorities to the impact of covid-19 on beaches. On the basis of the results obtained from these three methodological protocols, this thesis shows unequal progress in the field of integrated beach management. This work thus highlights the importance of using a systemic approach to beach management and of strengthening the cooperation and dialogue mechanisms between the numerous institutions responsible for beaches. As a valued, developed and endangered area, our analyses also highlight that beaches are a central issue in large cities that should receive the full attention of public authorities.

Key words
Beach, large cities, public policies, management, territory, socio-ecosystems


16 march 2023, Nice

Assessment tools for the health status of restored braided hydrosystems

Dennis FOX, Professor, Université Côte d’Azur, Thesis Director
Margot CHAPUIS, Maître de conférences, Université Côte d’Azur, Thesis co-director
Emanuèle GAUTIER, Professor, Université Paris 1, Thesis rapporteur
Jean-Nicolas BEISEL, Professor, Université de Strasbourg, Thesis rapporteur
Professor (associate), Université de Barcelone, Reviewer
Jérémie RIQUIER, Associate Professor, Université Jean Monet, Reviewer
Barbara BELLETTI, Post-doctoral researcher, Université de Lyon, Guest

Braided rivers have been harnessed for their natural resources and modified profoundly through decades of human activity, sometimes to the point where major ongoing management issues and risks to society have arisen. Restoration is then a process that aims to return functionality to these hydrosystems, especially in order to reach good ecological status according to the Water Framework Directive. This thesis, based on four hydromorphological restoration operations on Alpine braided rivers, aims to adapt and develop analysis tools to evaluate and quantify the success of restoration operations in order to produce restoration feedback. Key considerations for the effective management and understanding of these hydrosystems are also presented and discussed within the evolutive trajectories of the studied sites along with a systemic and transdiciplinary perspective.

Key words
Anthropised braided rivers, Long term management, Restoration feedback, Indicators, Systemic approach, Trandisciplinarity

Consultable online




10 december 2023, Nice

Spatio-temporal dynamics of the Senegalese mangrove: Characterization of the resilience of plant species through spatial modeling of biological processes

Julien ANDRIEU, Associate Professor HDR, Université Côte d’Azur, Thesis Director
Aude NUSCIA TAÏBI, Professor, Université d’Angers, Thesis rapporteur
Marc ROBIN, Professor, Université de Nantes, Thesis Rapporteur
Véronique HELFER, Research Director, Leibniz Centre for Tropical Marine Research, Reviewer
Cheikh MBOW, Research Director, Centre de Suivi Écologique du Sénégal, Reviewer

The mangroves of Senegal are characterized by a partial coverage of intertidal zones. Proportionally to the salinity of water and soil, the upper part of the intertidal zone is too salty and acidic for mangrove life to thrive. These areas are referred to as ‘tannes’. The interface between the mangrove and the tannes fluctuates over time, in response to variations in the hydrosystem, following climatic fluctuations. Thus, the fluctuations in the mangrove-tanne boundary define the range of metastability and resilience of the mangrove ecosystem in the estuaries of Senegal. A period of drought between the late 1960s and the mid-1990s led to the conversion of mangroves into tannes due to increased salinity. However, since the late 1990s, precipitation levels have returned to near the centennial average, thus promoting mangrove regeneration in the estuaries of Saloum and Casamance. In the Sahelian-Sudanian part of Senegal, the Saloum delta, in a semi-arid environment, exhibits characteristics of a distribution range margin due to extreme salinities. The mangrove in this delta primarily responds spontaneously to environmental fluctuations, providing an opportunity to study its resilience in the face of hydro-climatic disturbances. Furthermore, knowledge gaps hinder the understanding of the reasons behind this dynamic. Weighing the importance of factors influencing the spatial dynamics of the mangrove is therefore a scientific challenge. Consequently, a methodology based on the synergy of remote sensing, in-situ data, as well as statistical and spatial modeling has enabled us to grasp the mangrove’s reaction to changes in the hydrosystem. This approach first involves the observation and description of the dynamics of species populations at the distribution range margin in a current reference state. Then, the identification of regime changes in environmental dynamics and their impact on the distribution of species populations over the past decades was considered. Finally, experiments were conducted to evaluate the response of reproduction processes to identified environmental transition thresholds. A spatio-temporal modeling first allowed the reconstruction of these regime changes. Then, the germination opportunity window of species based on hydro-climatic temporal variability was characterized in order to identify the potential instability of the system in the face of disturbances. The results of this thesis indicate that climatic fluctuations in recent decades have affected the spatial structure of species distribution. The mangrove of Saloum has shown overall resilience to rainfall changes, oscillating between 500 mm and 700 mm of annual precipitation, particularly in parts of the delta with the highest salt concentrations. Furthermore, current hydro-climatic conditions seem to favor a trajectory of the mangrove from tannes towards a partial coverage of the intertidal zone, where Avicennia germinans rarely finds conditions conducive to its establishment and colonization of the mangrove-tannes interface. Its establishment capacity appears to require higher cumulative rainfall and an earlier time in the rainy season. Thus, the results of this thesis highlight, more precisely, that a rainfall threshold of 600 mm can likely induce modifications in the spatial organization of mangrove species, depending on whether it is reached in August or at the end of the year.

Key words
Mangrove, Socio-Ecological System, Remote Sensing, Spatial Analysis, Multiscale analysis, spatio-temporal modeling


10 may 2023, Naples

Urban coasts in socio-ecological transition. A framework to analyse the city-sea interface

Concetta FALLANCA, Professor of Urban planning, Université de Reggio Calabria “Mediterranea”, Thesis rapporteur
Massimo CLEMENTE, Research Diretor, CNR IRISS, Reviewer
Alexandra SCHLEYER-LINDENMANN, Associate Professor in Psychologie, Aix-Marseille Université, Reviewer
Sylvie DAVIET, Professor of Geography, Aix-Marseille Université, Reviewer
Romano FISTOLA, Professor of Urban planning, Université de Naples “Federico II”, President of the jury
Giovanni FUSCO, Research Director, CNRS ESPACE, Université Côte d’Azur, Reviewer
Antonio ACIERNO, Associate Professor of Urban planning, Université de Naples “Federico II”, Thesis co-director
Samuel ROBERT, Research Director, CNRS ESPACE, Aix-Marseille Université, Thesis co-director

Pascale FROMENT, Professor of Géography, Université Paris 8 Vincennes – Saint-Denis, (Thesis rapporteur,  not taking part in the examination)

The research aims to define the urban coastal society as a community intimately connected to the coast and to the sea, in the context of the socio-spatial and climatic-environmental critical issues connected to the development of coastal cities. Urban coasts are places of both rupture and connection between land and sea, whose integrity appears to be fragmented since it is composed of variegated elements that can also be considered as pieces of a potential green-blue infrastructure. In this sense, the concept of city-sea interface can be mobilised as the physical, ecological, social and functional contact area between the edge of the city and the edge of the water. Its flexibility to social, environmental and management issue are comparable to an urban amphibious. This notion is explored in the Euro-Mediterranean context, in relation with large coastal cities through a methodological framework. A legal exploration describes the main MSP guidelines and how they are incorporated in national littoral governance systems; the geographical approach defines a spatial data model for the socio-environmental features of the interface; finally, the psycho-social approach aims at studying the social representation in relation to accessibility, uses and planning adaptation. This framework is proposed as a benchmarking tool: specifically, it has been applied to Marseille and Naples, two similar Mediterranean cities with complex city-sea interfaces. The framework can also support the implementation of coastal-maritime planning at a local scale.

Key words
Urban coast, City-Sea Interface, Urban coastal society, Transition areas, Ecological planning, Perception of citizens
GODOYE Quentin

Quentin GODOYE

13 december 2022, Avignon

Austerity and territorial planning Impacts of allocation methods of public services on generation/reduction of territorial inequities

Sophie BAUDET-MICHEL, Associate Professor, Université Paris Cité, Reviewer
Guilhem BOULAY, Associate Professor, Université d’Avignon, Thesis Co-supervisor
Laurent CHAPELON, Professor, Université Paul Valéry Montpellier III, Thesis rapporteur
Sophie DE RUFFRAY, Professor, Université de Rouen Normandie, Thesis rapporteur
Associate Professor, Avignon Université, Reviewer
Cyrille GENRE-GRANDPIERRE, Professor, ESPACE, Université d’Avignon, Thesis Director
François TAULELLE, Professor, Institut National Universitaire Champollion, Reviewer

The economic crisis of 2008 has considerably accentuated the budgetary austerity imposed on public services in France, profoundly modifying their functioning as well as their location on the territory. The aim of this thesis is to analyse the reasons for the closure of public services in metropolitan France and their consequences on accessibility to the population between 2007 and 2017. The Permanent Equipment Database was used to obtain the precise location of public services. Spatial analysis methods were used to study their dynamics on the territory. The implementation of optimisation models made it possible to simulate the closure of additional services, as well as the relocation of some of them. Since 2007, the loss of accessibility to public services has been fairly limited. The simulation of the reduction of public services shows that accessibility does not decrease sharply, with the exception of maternity wards. The relocation simulations show a gain in accessibility in rural areas, to the detriment of urban areas.

Key words
Accessibility, public service, optimisation, austerity, spatial analysis

Consultable online




28 november 2022, Aix-en-Provence

Spatial structure of the permafrost landscapes of yakutia: geoinformational modeling (on the example of middle taiga and mountain permafrost landscapes)

Philippe CADENE, Professor, Doyen, Université de Paris, Reviewer
Professor, Vice-rectrice de l’Académie Nordique-Baltique, Université de Technologie de Kaunas, President of the Jury
Associate Professor HDRUniversité de Toulouse III Paul Sabatier, Thesis rapporteur
Christophe MORHANGE,
Professor, Aix-Marseille Université, EPHE, Reviewer
Associate Professor HDRUniversité de Bretagne Occidentale, Thesis rapporteur
Sébastien GADAL
, Professor, Aix-Marseille Université, Thesis Director

The solution of many issues of rational land use management is based on information about the state of the landscape complexes. The possibilities of obtaining reliable information on the spatial structure of permafrost landscapes become especially relevant for state assessment and evolution trends. Increasing environmental changes in the cryolithozone zone are associated with trends in climate change and vulnerability to anthropogenic impacts. The aim of this dissertation is to study the spatial structure of the permafrost landscapes of Yakutia on the basis of geoinformation modeling, taking as the object of research the middle taiga permafrost landscapes of Central Yakutia and the mountain permafrost landscapes of the Verkhoyansk region. To achieve this aim, the methods of the analysis and the processing of the time series of multi-sensor remote sensing data and digital elevation model has been developed. The methods allow modeling to map the visible morphological features of permafrost landscapes (relief and vegetation) with the implementation of ontological properties with the permafrost and lithogenic base. Time series of Sentinel 2 and Landsat 8 OLI images for the period 2015-2020 are used to map vegetation classes and analyze the state of vegetation cover. The vegetation classes are recognized by the variation in the photosynthetic activity of plant associations on the spectral reflectance characteristics. This allows us to use differences in phenological phases to increase the recognizable classes of plant associations. The accuracy assessments of the classification results are calculated by the comparison with high resolution images and field data to ensure a high level of validation confidence. Based on the ASTER GDEM relief data, we determined the genetic type of Quaternary sediments according to the landform classification calculated by the Topographic Position Index (TPI) and GIS analysis. Thus, modeling these two morphological features allows the development of a geospatial database of test study area. The database includes information on the existence of intensity transformation factors for the identified landscape complexes. The GIS models of permafrost landscapes provide information on the type of soil cover, soil conditions (moisture, temperature) and cryogenic texture. Obtaining such data demonstrates the effectiveness of the modeling for assessing the spatial structure of permafrost landscape in local level.

Key words
Permafrost landscape, geoformation modeling, remote sensing, Satellite image time series, terrain analysis, spatial structure, Yakutia,Yakoutie


30 march 2021, Nice

The odors spatialization

Paule-Annick DAVOINE, Professor, Université de Grenoble, UMR PACTE, Thesis rapporteur
Martine ADRIAN-SCOTTO, Associate Professor, Université Côte d’Azur, ICN, Reviewer
Jean-Paul THIBAUD, Research Director CNRS, UMR Ambiances, Architectures, Urbanités, Thesis rapporteur
Didier JOSSELIN, Research Director CNRS, UMR ESPACE, Reviewer
Joël CANDAU, Professor emeritus, Université Côte d’Azur, LAPCOS, Thesis Director
Sandra PEREZ, Associate Professor, Université Côte d’Azur, UMR ESPACE, Thesis co-director

Smell is a sensation which the individual perceives during the stimulation of the olfactory system by a complex combination of volatile emanations. How this smell is appreciated will vary greatly depending on the individuals, the cultures, and of course the substance in itself. There are many categories of smells. Research is under way in order to distinguish them more clearly, but it is nevertheless possible to classify them as either nauseous smells which constitute a nuisance, or those which are generally more pleasant, relating to for example an atmosphere or the odor-indentity of a place. If geographers take an interest in pollutants such as Sulphur Dioxide (SO2), Nitrogen Dioxide (NO2) Ozone (O3) and the PM10s, etc, there are few to our knowledge who are interested in odours. Now, they can, however, be experienced by residents as a real pollution of the air (ADEME). Smells can in fact invite themselves into people’s homes at whatever moment, and can be felt as a real intrusion into the living space. Smell nuisances keep on growing, in parallel with the boom in certain activities (refining, waste treatment; water purification; animal rearing….). They are more intolerable at certain moments of the year (summer season, when people live outside more), or of the day(morning). The level of nuisance can sometimes go as far as triggering physical symptoms like nausea, headaches or indeed irritations of the throat or the eyes Our thesis work mainly concerns the spatialization of these odors.

Key words
odors, odor nuisance, spatialization, environmental health, petrochemical complex, étang de Berre

Consultable online


BOPP Esteban

Esteban BOPP

11 june 2021, Avignon

The multi-hazard alert in the era of web 2.0 and the internet of things in the PACA region

Eric DAUDE, DR, UMR CNRS 6266 IDEES, Université de Rouen Normandie, Thesis rapporteur
Johnny DOUVINET, Pr, UMR CNRS 7300 ESPACE, Avignon Université, Thesis Director
Frédéric LEONE, Pr, LAGAM, Université Paul Valéry Montpellier III, Reviewer
Damienne PROVITOLO, Dr, UMR CNRS 7329 Géoazur, Université Nice Côte d’Azur, Reviewer
Magali REGHEZZA-ZITT, MCF HDR, UMR CNRS 8591 LGP, École Normale Supérieure, Thesis rapporteur
Lena SANDERS, Dr, UMR CNRS 8504 Géographie-Cités, Université Sorbonne, Reviewer
Emma HAZIZA, Mayane France Director, Guest
Col. Romain MOUTARD, Programme Director FR-Alert, DENUM, Ministère de l’Intérieur, Guest

On September 24th 2020, the Ministry of the Interior announced the introduction in 2022 of two alert methods based on the location of people in real time: Cell Broadcast (CB) and Location-Based SMS (LB-SMS). These two LBAS (Location-Based Alerting System) have the advantage of being more massive, faster, and spatialised ways of alerts (Aloudat, 2010). This research anticipated such a political choice. We estimate the potential of these LBAS, and we analyse how these means can improve the alerting of the population in France. Using a variety of methodological protocols based on spatial analysis, we demonstrate the high potential of CB and LB-SMS at a national scale. The rates of individuals reached by these solutions are very high and these rates are very homogeneous between municipalities. These tools are also well accepted by the population. Thus, these two solutions offer new opportunities to overcome the weaknesses of traditional alerting means. However, it is necessary to fit these solutions to social and territorial contexts. Their integration in the future FR-Alert platform and their use pattern must be thought, avoiding technological fetishism and adapting the organisations to these changes.

Key words
Alert, LBAS, risk, territory, individual

Consultable online



30 march 2021, Aix-en-Provence

Soil quality assessment and urban changes in a Mediterranean city (Marseilles, France)

Franco AJMONE MARSAN, Professor, Università di Torino, Thesis rapporteur
José SERRANO, Professor, Université de Tours, Thesis rapporteur
Béatrice BÉCHET, Research Director, CNRS, Reviewer
Jean-Noël CONSALES, Associate Professor, Aix-Marseille Université, Reviewer
Cécile DELOLME, Research Director, ENTPE, Reviewer
Christian WALTER, Professor, Agrocampus Ouest, Reviewer
Catherine KELLER, Professor, Aix-Marseille Université, Thesis Director
Samuel ROBERT, Research Director, CNRS, Thesis co-director

The aim of urban planning is to promote local development, while preserving local ecosystems and the quality of life of residents. This is about making cities sustainable and, addressing challenges such as climate change, loss of biodiversity, maintaining a good quality of life and mitigating risks. In this context, this research covers soils, seen as local ecosystems that can help to meet these challenges. The study area is the city of Marseilles, France. Our objectives were: (1) to explore the contribution of urban soils to urban sustainability issues through their in-depth characterization, and (2) to study their consideration in urbanism. The results indicate that the potential of urban soils for urban agriculture and their biodiversity are poor, and they present anomalies in trace elements. However, these soils are also a reservoir of local biodiversity, have an ability to mitigate the risk of flooding by runoff, and the bioavailability of inorganic pollutants is low. The consideration of urban soils is generally reduced in urban planning, but indirectly town planning documents do contain provisions for conserving urban soils. Urban practitioners are aware of the need to improve their consideration in urbanism, and they point out that this relative weak integration stems from the lack of available data and information. Finally, this research shows the potential of urban soils and specifies the needs of urban planners.

Key words
Soil quality, urban planning, city, climate change, biodiversity, urban planning practitioners

Consultable online


FERRI Mickaël

Mickaël FERRI

22 january 2021, Avignon

Logics of spatial interactions, emergence and dynamics of city systems over time Multi-agent modelling approach

Céline VACCHIANI-MARCUZZO, Associate Professor – HDR, Université de Reims Champagne-Ardenne, Thesis apporteur
Éric DAUDE, Research Director CNRS, Normandie Université, Thesis rapporteur
Alexandra SCHAFFAR, Professor, Université de Toulon, Reviewer
Giovanni FUSCO, Research Director CNRS, Université Côté d’Azur, Reviewer
Cyrille GENRE-GRANDPIERRE, Professor, Avignon Université, Thesis Director
Tania JIMENEZ, Research Engineer, Avignon Université, Thesis Co-director
Mounir REDJIMI, Associate Professor, Avignon Université, Guest

AbstractThe recurrence of a hierarchical structure in the systems of cities around the world, characterized by the fact that the cities are all the more numerous that they are small, and more precisely that it exist a relation between the size of the cities and their ranks (the size of a city of rank “r” is equal to the population of the biggest city divided by “r”) is a phenomenon which has been described for many years.This statistical regularity, known as the rank size law, its local variations and the research for its explanations, have been widely explored by geographers for more than a century. Certainly, the agglomeration economies, favorizing innovation, its broadcast in urban hierarchy, themselves conditioned by transport networks of which interactions between cities depend, play a crucial role to understand emerging and maintenance systems of hierarchized cities.However, if a necessary theory on the evolution of systems of cities, which regroup now more than half the world population, is already pretty well advanced, it is still impossible to consider a system of settlement at time “t” and to predict in detail the evolution, especially the emergence of new population concentration, because systems of cities are complex.The computer science’s gigantic progresses have been able, since about forty years, to develop simulation models, which can test different hypothesis regarding the evolution of the systems of cities and to better understand their dynamic. It is more especially the case of the Multi-Agent model, created by the P.A.R.I.S team, who allowed huge progresses in understanding the dynamic of systems of cities, allowing the theorical hypothesis test, some old, concerning in particular the innovation diffusion and its impacts on the city system. However, those models, very documented and well used, are presenting the disadvantage to be based on numerous parameters and hypothesis, which makes it difficult to calibrate and render their results sometimes difficult to interpret. Also, if these models allow us to analyze the evolution of a systems of cities, the emergence of new cities is not well studied.In this context, the goal of this thesis is to contribute, using the Multi-Agents simulation, to the research of explanations of the emergence and the maintenance of hierarchized systems of cities. Rather than looking to complexify more the existing models so that their results match the best possible to the empiric realities, it has been decided here to put down a central hypothesis, “simple”, without an economic dimension and which counteract with the most frequent approaches, and to test their validity. This hypothesis state that the allometric growth of the systems of cities is first a function of the way that the interactions between the cities happen at the individual scale. More precisely, it is necessary to look first if, depending if a person living in city A decide to move to the city B, draw at random, or of which the attractiveness’ will depend on its historical past in terms of trade, centrality, accessibility, can one succeed, or not, to generate systems of hierarchized cities type rank size. That is to tend to confirm that, way before economic considerations (economic profile of cities per example), it is before everything the spatial interaction modalities which preside on the evolution of systems of cities.Secondly, always placing the spatial interaction between the individuals at the heart of our work, we are looking to better understand the emergence of the cities.

Key words
Emergence, Systems of cities, Game theory, Self–organization

Consultable online


MASSE Florian

Florian MASSE

26 march 2021, Aix-en-Provence

Leisure mobility in two coastal cities : identification of the determinants of travel behaviour in Marseilles and Nice

Sébastien OLIVEAU, Aix-Marseille Université, Thesis Director
Sophie DE RUFFRAY, Université de Rouen Normandie, Thesis rapporteur
Marie-Hélène VANDERSMISSEN, Université Laval, Québec, Thesis rapporteur
Vincent KAUFMANN, Ecole Polytechnique Fédérale de Lausanne, Suisse, Reviewer
Samuel ROBERT, CNRS, UMR 7300 ESPACE, Reviewer
Samuel CARPENTIER-POSTEL, Université de Franche-Comté, Thesis co-director

Although leisure mobility generates negative externalities, little attention is paid to them. The environmental and social issues involved are particularly sensitive in the context of coastal metropolitan areas, where the desirability of coastlines faces the vulnerability of these places. Through the study of Marseilles and Nice, we aim to understand the travel behaviour of populations for their leisure activities. Based on the weekend leisure activities module of household-travel surveys and a study of the spatial organisation of the leisure services, the analysis focuses on the attractiveness of coastlines, the opinions relating to modes of transport and the determinants of mode choice. In contrast to Nice, the results show a coastal attraction in the case of Marseilles, where longer journeys are observed on average for doing leisureactivity on the coast. Secondly, it appears that opinions regarding modes of transport, shaped by weekday mobility habits, constrain the use of alternative modes – instead of the car – for weekend recreational activities. Finally, lifestyles, and in particular the type of leisure, seem to play a determining role in the choice of transport mode. From an operational point of view, this research shows the need for planning weekend transport services towards leisure locations. They also reveal, thanks to the inclusion of attitudes in behavioural models, the fact that young people may be a more receptive audience to modal shift.

Key words
Leisure mobility, behavoir, social representations, coastline, Metropolis, Marseille, Nice, spatial analysis

Consultable online




22 january 2021, Avignon

Recommendation of personalized cultural sightseeing tours : An Interdisciplinary study on automated proposal of visits


Patrice BELLOT – Aix-Marseille Université,  LIS,  Thesis rapporteur
Sébastien MUSTIÈRE – Université Gustave-Eiffel/ENSG, IGN, Thesis rapporteur
Marie-Sylvie POLI – Avignon Université,  LCC, Reviewer
Enrico NATALIZIO – Université de Lorraine, LORIA, Reviewer
Francesco DE PELLEGRINI – Avignon Université, LIA, President
Juan-Manuel TORRES-MORENO – Avignon Université, LIA, Thesis Director
Didier JOSSELIN – Avignon Université,  ESPACE,  Thesis co-director
Fen ZHOU – IMT Lille Douai, Thesis co-supervisor


This thesis focuses on the recommendation of cultural visits through an interdisciplinary approach. This work mixes techniques from Operational Research and Natural Language Processing and is based on concepts from the sociology and geography. We propose new methods around the evaluation of cultural points of interest as well as the automatic creation of sightseeing tours, taking into account the preferences of the visitor. These principles are applied to two different scales and contexts, museum visits and cultural tours in a city.In the first part, we focus on visits in art museums, according to the preferences expressed by the visitor and the prestige of the artworks. This double approach allows us to classify the exhibits, both according to the cultural affinities of the visitor, but also according to their importance within the museum. The latter is calculated by applying automatic text summarization algorithms to the museum’s official texts that describe the artworks and provides a profile reflecting the discovery of a museum based on its masterpieces. This profile can then be modified according to the visitor’s tastes to obtain a visit corresponding to him or her while preserving the museum’s point of view.Subsequently, we liken the construction of a tour to a routing problem, aiming to find a path among the different rooms and exhibits whose purpose is to maximize visitor satisfaction while respecting time constraints. Two methods are proposed, an integer linear programming model and an heuristic that can be used to propose a route in real time, for example upon arrival in the museum.In a second part, we emphasize on the tourist recommendation in the city, by establishing metrics to be considered for route computation. Based on an interdisciplinary study, we highlight the importance of customizing the routes and identify an essential factor in their construction in addition to cultural tastes, the pace of the visit. A new method of measuring the quality of experience of a tour that combines these two criteria is used. The latter combines methods from the literature to evaluate cultural interest and uses actograms as a geographical representation of a route and thus defines a measure of the pace of a visit.Subsequently, we develop a system to recommend sightseeing tours, in the form of an integer linear programming model based on an extensible formalism that allows for a wide variety of constraints and that integrates three criteria for the evaluation of the route. On the one hand, the cultural interest and the rate of visit, which depend on tourist preferences, are measured at different scales, making it possible to introduce coherence into the construction of the itinerary. On the other hand, we propose to integrate into the objective function, the peak-end rule, a famous psychological heuristic that has already been applied in many other fields.Based on concrete case studies, we show that the combination of techniques from various disciplines allows us to obtain good results, both in the estimation of the attractiveness of points of interest and in the construction of tourist routes.

key words
Recommendation system, sightseeing tour, natural language processing, routing optimization, mixed integer linear programming

Consultable online


MICHEL Camille

Camille MICHEL

2 february 2021, Aix-en-Provence

The old districts in Lyon, an example to study the dynamics of socio-economic inequalities of territories

Jean-Yves AUTHIER, Professor, Université Lumière Lyon 2, Reviewer
Brigitte BACCAÏNI, Inspectrice Générale de l’Administration du Développement Durable, Reviewer
Sébastien OLIVEAU, Associate Professor – HDR, Aix-Marseille Université, Thesis Director
Catherine RHEIN, Research Director emeritus CNRS, Thesis rapporteur
Stéphanie VERMEERSCH, Research Director CNRS, Thesis rapporteur

The “old districts ” of the French city of Lyon have long been territories characterized by a strong public action, particularly in the context of what is called “la politique de la ville” (literally, the urban policy). These districts have undergone many transformations over time and since 2014 some of them are no longer in the priority geography of the “politique de la ville”. The many actions carried out within these districts have favoured gentrification processes and have contributed to change the social profile of their inhabitants. The strong downtown attractiveness and property pressure are also contributing to their transformationOld districts are therefore an interesting field for understanding the dynamics of change of territories. In this study we focus on the social transformations of the space and the socio-economic inequalities at the scale of the city of Lyon. Our analyses allow us to investigate how these districts are affected by these dynamics and to question the role of public policies in these evolutions.

[This doctorate is the result of a CIFRE agreement supported by the Pluricité research consultancy, in partnership with the City of Lyon]

Key words
Inequalities, quantitative analysis, public policies, “politique de la ville”, social diversity, gentrification,old districts, Lyon

Consultable online




14 june 2021, Nice

Sustainability Transitions: Multi-scale modelling of renewable energy technologies diffusion and urban resilience under a network approach in the Swiss Alps and in the South Region of France

Dominique BADARIOTTI, Professor, Université de Strasbourg, Thesis rapporteur
Ola HALL, Professor, Lund University, Thesis rapporteur
Didier JOSSELIN, Research Director CNRS, Université d’Avignon, Reviewer
Serge LHOMME, Associate Professor, Université Paris-Est Créteil, Reviewer
Jean-Christophe LOUBIER, Professor, University of Applied Sciences and Arts Western Switzerland, Thesis co-director
Christine VOIRON-CANICIO, Professor, Université Côte d’Azur, Thesis director

This doctoral dissertation is an interdisciplinary research work at the intersection of geography and innovation studies through the prism of two subjects: diffusion of sustainable innovations and urban resilience to energy transition. The research work was developed in the Swiss Alps and in the South Region of France in order to deploy comparative analyses. This Ph.D dissertation investigates the regions receptivity to sustainable innovations, specifically renewable energy technologies (RET) such as solar photovoltaics, solar thermal collectors in the Swiss Alps and electric and hybrid vehicles in Switzerland at the national level. As well as the research was developed in the South Region of France were six innovation indicators were analysed: solar photovoltaics, solar thermal collectors, wind power, small and big hydroelectric power plants, biogas and biomass. The research aimed at improving our understanding on regions’ ability to integrate these innovations into their dynamics and the embedded urban resilience to energy transition, to adapt to change, accommodating disruptions in the diffusion process and develop new spatial diffusion paths. The research questions aimed at underpinning our understanding on the network effects of the RET diffusion and on the potential insights that spatial information might provide regarding such diffusion processes. The underlying assumptions were that RET diffuses across scales in a non-random fashion and describe a preferential attachment mechanism. The implication of the latter assumption is that the urban renewable energy systems exhibit fractality, which is the signature of self-organized systems. By definition, resilient systems are self-organized, so in this context, the innovation systems are analyzed in the framework of urban resilience to the energy transition. Thus, a further assumption is made and proposes that more innovative places and locations are more resilient than less innovative locations. The methodological approaches to address these research questions and verify the assumptions are described as follows. In the Swiss region a model called ‘Spatial Preferential Attachment’ (SPA) was created based on spatial interaction theory, relying on a gravity model that was built through agent-based modelling and systems dynamics approaches. The integration of the gravity model with the spatial information of the RET allowed to build a spatial network, which simulated the urban energy system of the region. The results allowed to accept the assumptions in which a preferential mechanism in the diffusion process take place, since the spatial diffusion distribution follow power laws. The model was also applied in Switzerland and in the South Region of France, obtaining similar results, following multi-level and hierarchical mechanisms. These results are in line with the path development theory proposed by economic geographers, where the specific-place legacy has at least a partial impact in the future intensity of diffusion processesThese results are important within the sustainability paradigm from a research perspective and challenging for the current innovation framework, the so-called Transformative Change, which aims at establishing a fairer view on socio-economic and environmental issues. The preferential attachment mechanisms in RET diffusion imply that there are hubs of innovation ruled by urban scaling laws that put in disadvantage other locations. The SPA was also used to simulate the urban resilience to energy transition in the Swiss canton of Valais. The energy urban network was ‘attacked’ by removing the hubs of the structure and the simulations showed that the system could reorganized itself at global level, showing strong sings of resilience however not at local level. Resilient systems are self-organized however it does not imply that resilient itself is fractal as differences were found from a multiscale spatial perspective.

Key words
Complex systems, sustainability transition, renewable energy technologies diffusion, network modelling, self-organization

Consultable online

THOMAS Lauriane

Lauriane THOMAS

17 december 2021, Avignon

Is Land dedicated to Economic Activities an adjustment variable of local development models ? : An assessment to the Provence – Alpes – Côte d ‘Azur region

Elodie CASTEX, Professor, TVES, Université de Lille, Thesis rapporteur
Samuel CARPENTIER-POSTEL, Professor, ThéMA, Université Bourgogne-Franche-Comté, Thesis rapporteur
Sonia GUELTON, Professor, Lab’Urba, Université Paris-Est, Reviewer
Cyrille GENRE-GRANDPIERRE, Professor, ESPACE, Université d’Avignon, Thesis director
Claude BERTOLINO, Architecte et Urbaniste de l’État, Directrice de l’EPF PACA, Thesis co-director
Guilhem BOULAY, Associate Professor, ESPACE, Université d’Avignon

Although the issues of employment, technological innovation and reindustrialisation have been at the centre of many debates in France over the last twenty years, these issues are treated in a reterritorialized manner. In particular, the land dimension of economic activity is never addressed. This situation contributes to a lack of knowledge of business land, both in terms of definition and quantification. Based on the case of the PACA region this thesis proposes a typology of economic land, a quantification of its state and dynamics, and a characterisation of its internal evolutions. Through the study of economic land (ZAE) from 1999 to 2017, we estimate and locate the dynamics of creation, regularisation and disappearance of economic land. The poor quantitative balance of economic land is accentuated by internal mechanisms that weaken the ZAEs, resulting in particular from competition between economic and residential uses of the communal land. It is shown that these mechanisms contribute to a development model influenced by pre-presidential jobs and the residential economy, which is likely to lead to a dependence of the territorial development model on past land choices.

Key words
Land, regulation, productive, economic activity zone, planning

Consultable online




mai 2020, à Klaipėda (Lithuanie)

Change in Land Use and Land Cover Coastal Zone: Classification Methods Comparison and Assessment

Thèse réalisée sous la direction de Inga DAILIDIENĖ, Professeure à Klaipėda University et la co-direction de Sébastien Gadal, Professeur à Aix-Marseille Université.

Darijus VETEIKIS, Professeur, Vilnius University, Président du jury
Algimantas ČESNULEVIČIUS, Professeur, Vilnius University
Boris CHUBARENKO, Directeur de recherche, Russian Academy of Sciences, P. P. Shirshov Institute
of Oceanology
Justas KAŽYS, Maître de conférences, Vilnius University
Julius TAMINSKAS, Directeur de recherche, Lithuanian Natural Research Center

This research examines the physical and environmental processes that affect and alter the land use and land cover (LULC) of areas of the Earth’s surface. Some of these changes are anthropogenic, instigated by humans either using the land for new activities or as a result of management policies. Methods are developed so that new technologies can be used to monitor these changes in real-time or for comparison over longer periods of time. The methods are general, but here are applied to two chosen areas formed by, firstly the coastal area of the Curonian Spit and then the continental part of the Lithuanian coastal zone, forming two case studies or test zones. These areas are particularly interesting to study since data is available for the period pre- and post-1990 when the political situation changed, heralding new environmental, forest management, urban planning and management practices in the test zones. This research was carried out utilising remote sensing techniques and comparisons have been made between two methods for retrieving or evaluating LULC, each checked against ground truth data, in order to find the most suitable way to establish geo-graphic details. Specifically, for each patch of land within the test zone, the type of cover was categorised as cropland, arable, sand, forest, artificial, water or grassland as determined using remote sensing Landsat images over the period 1989–2017. An as-sessment of the accuracy of this new analysis was then compared with Google Earth, other images and in situ data, to produce a high level of confidence in the evalua-tions. Results are displayed graphically by extracting the levels of the different types of land cover where some areas remained unchanged over a period of almost thirty years while others were subjected to the significant transformations. The forestation of moving dunes, the human impact on the area and wider changes in the environment are all discussed. The results demonstrate how remote sensing and GIS methods can be used as an important tool for planners, dispensing with the need for costly field trips to establish up-to-date data on land cover change and use.

Key words
Land use, land cover, forest management, urbanisation, remote sensing, Landsat

Consultable online


Paul Gérard GBETKOM

Paul Gérard GBETKOM

24 september 2020, Aix-en-Provence

Studies of spatial dynamics of land use and land cover change in the Cameroonian window of Lake Chad and its hinterland from the Sahelian great droughts of 1970


Laila RHAZI, Université Mohammed V, Rabat, President of the jury
Sébastien GADAL, Professeur, Aix-Marseille Université, Thesis Director
Ahmed EL ABOUDI, Université Mohammed V, Rabat, Thesis co-Director
Julien ANDRIEU, Associate Professor – HDR, Université Côte d’Azur, Thesis rapporteur
Mohammed CHIKHAOUI, Institut Agronomique Vétérinaire Hassan II, Rabat, Thesis rapporteur
Jean-François CRETAUX, Centre National des Études Spatiales, Toulouse, Reviewer

The environmental context in the Sahelian region, from the 1970s is marked by the Sahelian great droughts which followed the wet periods of the 1950-1970 decades, and preceded the return of humidity observed at the beginning of the 2000s. These environmental changes have brought out important modifications in the ecosystems and their uses. The aim of this doctoral thesis is therefore to study, from a multi-scale, multi-sensor and multi-temporal analysis of land use and land cover, the consequences of these environmental changes on the Sahelian ecosystems, through the Cameroonian shores of Lake Chad and its hinterland. The MODIS MOD13Q1 time series images available from 2000 are then used to map the main vegetation classes, analyze the state of the vegetation cover and determine its spatial and temporal evolution trends. The Landsat, Sentinel 2 and SPOT 6 images available since 1972 (for Landsat), are used for a diachronic study of land use and land cover changes of the study area according to Sahelian climatic and environmental changes, with the aim of to identify and highlight the effects of these changes on the use of naturals resources. The results shows that after the aridification of land cover during the great droughts, a dynamic of re-greening of the study area takes hold from the 2000s, through the annual vegetation cover which gradually grows on the bare soils. The re-greening observed increases the availability of the pastures, and improves the accentuation of anthropic pressures on the natural’s resources by the increase of agricultural areas and built-up surfaces. The modeling of the land use and land cover according to the geographical factors of changes and according to the morphological structure and the functional characteristics of the elements of surface, shows then that it is possible to determine large territorial units present in the study area.

Key words

Remote sensing, great droughts, vegetation cover, land use and land cover, spatial dynamics, geographical factors, modelization, Sahel

Consultable online