GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 1.1 GNGTS 2024 Studying the Viability of Kinematc Rupture Models and Source Time Functons with Dynamic Constraints M.E. Locchi 1 , F. Mosconi 1 , M. Supino 2 , E. Casarot 2 , E. Tint 1,2 1 Sapienza Università di Roma, Rome, Italy; 2 Isttuto Nazionale di Geofsica e Vulcanologia, Rome, Italy; Earthquakes are one of the greatest natural hazards and a beter understanding of the physical processes causing earthquake ruptures is required for appropriate seismic hazard assessment. Progresses on the knowledge of the seismic source have been made in both modelling the increasingly dense geophysical data and through laboratory experiments. Kinematc modelling is a standard tool to provide important informaton on the complexity of the earthquake rupture process and for making inferences on earthquake mechanics. Despite recent advances, kinematc models are characterized by uncertaintes and trade-ofs among parameters (inherent non-uniqueness of the problem). It has been documented that, for the same earthquake, source models obtained with diferent methodologies can exhibit signifcant discrepancies in terms of slip distributon, fault planes geometry and rupture tme evoluton. Prescribing the slip velocity on causatve faults (source tme functon) is one of the crucial components in the models because it contains key informaton about the dynamics and prescribes how fast each point on the fault reaches its fnal slip (Tint et al, 2005). Such functon is nonetheless one of the most poorly observatonally constrained characteristcs of faultng. Recently, slip velocity tme histories have been studied with laboratory earthquakes and a systematc change of mechanical propertes and tracton evoluton has been observed to correspond with a change in the shape of slip velocity (Scuderi et al., 2020). However, this functon is assumed a-priori following two diferent approaches: in the single window approach an analytc functon is assumed; while in the mult window approach the complexity and heterogeneity of the process is simulated through the linear sum of cumulatve slip at diferent tme windows. The fnal efect is a heterogeneous functon over tme that accumulates the slip untl it reaches the fnal slip. Multple-window inversions impose the smoothness of the slip distributon, introduce a sort of causality but at the same tme heterogeneity of the rupture front prescribed by the diferent tme windows, solve for the evoluton of rupture, allow the slip at each point to be determined by multple tme window functons with no predetermined shape, and observe the slip functons a posterior (Konca et al., 2013). Nevertheless, the limited resoluton of the kinematc models prevents reliable constraints on the slip velocity tme history. To investgate the efect of the slip velocity functon on the ground moton and on the inverted slip history on the fault we run a series of forward and inverse models. We generate spontaneous