This PhD project aims to develop a robust methodological framework for earthquake location and realistic uncertainty quantification in the context of dense seismic networks. Despite recent advances in automatic detection, deep-learning-based phase picking, and earthquake relocation techniques, uncertainties related to velocity models and network geometry remain a major limitation and are often underestimated by conventional approaches. The project will compare and benchmark different detection, picking, and location methodologies in order to assess their respective strengths and limitations. Particular emphasis will be placed on identifying, quantifying, and disentangling the main sources of uncertainty, including phase picking errors, network configuration, and velocity model assumptions. The research will primarily rely on data from Cephalonia Island (Greece). In a second phase, the developed methodologies will be transferred to the Middle Durance region near Cadarache (France), allowing their applicability to lower-seismicity environments to be assessed. The expected outcomes include improved seismic catalogs, a better understanding of active tectonic processes.