The aim of this PhD project is to develop a global, multi-scale model of the magma systems  feeding volcanoes.  These complex systems, made up of magma reservoirs, fractures and conduits, have a strong influence on the frequency and intensity of volcanic eruptions. influence the frequency and intensity of volcanic eruptions, which follow a power-law distribution.  power-law distribution. These eruptions can have significant local impacts (e.g.: Soufrière Hills, Montserrat) or major global consequences (e.g. Yellowstone).  The aim is to combine detailed physical models (microscale) representing individual components individual components (magma chambers, dykes) with a large-scale representation (macroscale) using
spatialized dynamic graphs.  These graphs model reservoirs as nodes and conduits as links, while also conduits as links, while integrating filling and emptying dynamics.  The model will  will explore how the topology and spatio-temporal structure of the system influence magmatic flows and eruption characteristics.  The work plan includes a literature review, the development of the global model, its validation on case studies case studies, and its application to test current conceptual models. Supervised by an international team in complex sciences and geophysics, this PhD is part of a collaboration between Lyon.