Analysis and modelling of space-time patterns in complex regions

Auteur Carmen Delia Vega Orozco
Directeur /trice Prof. Mikhail Kanevski
Co-directeur(s) /trice(s) Dr. Marj Tonini Dr. Marco Conedera
Résumé de la thèse The project deals with a fundamental development of a contemporary methodology and exploratory tools for detection, analysis, and interpretation of patterns in spatio-temporal data. Important real case studies concern a variety of phenomena and data from natural hazards to socio-economic and demographic patterns. The space-time point pattern characterization represents a powerful tool to understand the distribution and behavior of the underling processes and its correlation with environmental, socio-economic and other relevant factors. Many measures describing the space-time distribution of environmental phenomena have been proposed in a wide number of disciplines; however, most of these measures are of global character and do not consider complex spatial constraints, high variability and multivariate nature of the events. The main objective of the present project is to carry out a basic research with a strong application part to investigate complex phenomena and to overcome unsolved methodological problems in characterization of space-time patterns, in particular, in forest fires. The methodology will benefit from solid background of recent developments in fractal-based point processes, spatial and spatio-temporal statistics, geomatics (GIS, remote sensing), environmental modelling and analysis. Heterogeneity, complexity of the space, multi-scale (spatial and temporal) variability, and multivariate nature of the phenomena will be taken into account. The practical part of the research deals with an application of the methods developed for real and simulated space-time environmental data, particularly, in forest fires, landslides, earthquakes, biodiversity and socio-economic data. For instance, in the study case forest fires, new concepts and definitions of the Wildland-Urban Interface (WUI) – the interaction zone between burnable vegetation and anthropogenic infrastructures – are implemented combining non-classical techniques of spatial data clustering and remote sensing methods. The Alpine region is considered as the most important zone of the study. This is a multidisciplinary research project that implies expertise in geoinformatics, spatial statistics, forest and environmental sciences. It provides a response to the increasing demand for environmental monitoring and management tools (assessment of natural and anthropogenic hazards and risks, sustainable development, retrospective success analysis, etc.).
Délai administratif de soutenance de thèse