Description of the thesis topic

Modelling Urban Territorial Transformations of the Baltic Region: A Spatio-Temporal Computational Approach. (What is an Integrating Mathematics, Remote Sensing, and Geo-AI Simulation?)

In the context of rapid and unprecedented global urbanisation affecting the Baltic region of Europe, conventional approaches to spatial analysis for urban planning and governance are no longer adequate for monitoring territorial and development changes. Baltic cities are transforming at scales and rates that challenge conventional methods of spatial analysis. Today, urban areas are no longer just the setting for social and economic activity, but are more complex and evolving systems, shaped by dynamic interactions between population growth, land use planning, infrastructure development, and the integration of environmental adaptation processes. These interactions unfold simultaneously in space and time and exhibit non-linear patterns of change that require new ways of thinking about and modelling the urban condition.
Nordic Baltic cities such as Helsinki, Tallinn and Kaunas are ideal places to study urban territorial transformations for the following reasons: firstly, they represent a rich history of urbanisation and urban innovation in different political contexts (Soviet cities, liberal cities, “mathematical” cities, smart cities, green cities, etc.). Secondly, they are extremely sensitive to the environment and to the context of global climate change and sustainable urban development, which particularly affects the Baltic region of Europe. Finally, they also offer reliable geographical data that can be combined with advanced urban planning policies. Their strategic locations in Northern Europe make them an ideal case study for modelling urban dynamics using advanced remote sensing, AI and new mathematical approaches.

The PhD. project is part of the GeoMath PEPR Math-Vive, developed and designed to refresh the relationship between urban geography and mathematics. It aims to address urban challenges and dynamic interactions to model and find answers on how new applied mathematical approaches could help in the fields of modelling and analysing urban territorial transformations based on three guiding principles: 1) Geography and mathematics, new areas for the co-construction of urban models. 2) Use new reproducible mathematical methods to define and model geographical objects related to the challenge of urban territorial transformation for the benefit of other geographical areas. 3) Directly bringing together the most advanced concepts from both fields, so as not to reduce mathematics to mere “modelling tools” for geographers, but rather to consider the interpretative correspondences between the concepts of the two “disciplines”.

Regarding the three principles mentioned in the context of this doctoral research project, an interdisciplinary framework for analysing and geo-simulating urban territorial transformations in the Baltic regions of northern Europe will be proposed. This framework integrates tools and concepts from mathematics, geography, spatial remote sensing, and artificial intelligence. In this project, urban space is considered as an interconnected system with multi-scalar spatial and temporal dimensions that should be studied holistically. Exploring and understanding the evolution of urban areas over time involves simultaneously examining many aspects, such as spatial patterns and socio-economic and environmental processes, which requires mathematical methods capable of handling complexity, scale, and uncertainty.

Urban transformations are often the result of interdependent changes. For example, increased population density can lead to vertical growth or horizontal sprawl, while transport infrastructure reshapes accessibility and land values. Other factors, such as environmental and geopolitical risks, influence zoning and construction. These changes are not evenly distributed, creating diverse impacts across neighbourhoods, regions, and Baltic cities due to specific development patterns. In fact, descriptive and predictive spatio-temporal modelling is central to the analysis of urban territorial transformations to capture new geographical knowledge.

Mathematics can provide powerful modelling and representation solutions based on spatial remote sensing and geographic information systems (GIS). The new mathematical models developed as part of the Math-Vive PEPR will serve as a basis for building predictive analysis models and recognising hidden (unknown) spatial structures.

The predictive geo-simulation platform created is intended for researchers, urban planners, and policy makers in real-world situations.
This doctoral project aims to enrich theoretical and practical knowledge related to the challenge of territorial transformation in the Baltic region. On a theoretical level, it aims to improve the modelling of cities as multi-scalar spatio-temporal systems while proposing new frameworks for understanding urban dynamics through the combination of mathematical formalisms and geoinformatics. Methodologically, it aims to demonstrate the advantages of merging satellite data, artificial intelligence, and spatial analysis. This research project is characterised by the convergence of geography and mathematics.

Work Context

The research will be conducted at the ESPACE laboratory (AMU site) in collaboration with colleagues from the Orsay Mathematics Laboratory and the PEPR Math-Vive GEOMATH. Research will also be conducted in collaboration with Baltic research teams in Finland, Lithuania, Latvia, etc. This research project is part of the ESPACE UMR's 'Urban Transformation and Futures' theme and cross-disciplinary areas (methodology, future of territories, etc.).

The position is located in a sector under the protection of scientific and technical potential (PPST), and therefore requires, in accordance with the regulations, that your arrival is authorized by the competent authority of the MESR.