After an earthquake event, major ground effects include landslides. In Italy, the most common type of earthquake-induced landslides are rockfalls. The assessment of the extent and magnitude of an earthquake-induced landslide event may be of importance for both preparedness and response operations. An earlier work devised a modeling chain including an empirical ground shaking scenario and a three-dimensional model for rockfalls, which can be calibrated and possibly applied in real time after an earthquake event. In this study, we explore the combined use of two physics-based methods for both earthquake and rockfall modeling and their possible contribution towards a better understanding of the triggering mechanisms and characterization of seismically induced rockfalls. Specifically, by modeling a set of ground shaking scenarios, which account for increasingly complex details (including multiple seismic events, as well as point and extended sources) and the related rockfall scenarios, we aim to capture the main spatial patterns of observed rockfalls. Application of the proposed modeling chain suggests an advantage of using multiple sources over a single source, and point sources with respect to approximate extended representations when constrained by limited available data. This follows from the comparison of rockfall trajectory simulations for the Friuli 1976 event in Northern Italy with observed rockfalls induced by the seismic sequence. The obtained results evidence the opportunity of including topographic effects in the ground shaking simulations and highlights the possibility of further investigating the cumulative effect of complex seismic sequences and their influence on modulating landslide susceptibility.