Characterizing earthquake sources and their recurrence requires integrating geological, seismological, geodetic, and geophysical observations across a wide range of spatial and temporal scales, often in the face of sparse, incomplete, or uncertain data. This session welcomes contributions on the characterization and modelling of seismic sources, the development and use of earthquake catalogues for recurrence estimation, and their application in seismic-hazard assessment across a range of tectonic environments. Topics may include mapping and characterization of active faults, particularly results from paleoseismological investigations; representation of fault geometry, segmentation, and alternative rupture models, particularly where geological and paleoseismic constraints are limited; time-dependent approaches to fault recurrence; earthquake catalogue development, declustering, magnitude homogenization, and treatment of induced seismicity; estimation and spatial representation of earthquake recurrence in source zones and gridded-seismicity models; and methods for improving moment magnitude estimates for small earthquakes, including coda-based approaches. We also encourage contributions exploring the use of geodetic observations to constrain tectonic deformation, fault-slip rates, interseismic coupling and locking, and earthquake recurrence; and the use of aftershock studies/catalogues to constrain fault geometry, rupture extent, kinematics, segmentation, and other source characteristics. Studies addressing how sparse or conflicting observations are incorporated into seismic-source models, how epistemic uncertainty is represented, and how alternative source interpretations can be evaluated or compared are particularly welcome. The session aims to bring together researchers and practitioners working across the spectrum from understanding fundamental source processes to developing and applying seismic-source models in seismic-hazard assessment.