Non-Linear Seismic Velocity Variations Observed During a Seismic Swarm in the Alto Tiberina Low Angle Normal Fault From Ambient Noise Correlation Measurements

From seismic interferometry, we investigate the strain sensitivity to seismic velocity variations related to a seismic swarm activity that occurred in 2013 along the Alto Tiberina low angle normal fault. We compute daily auto-correlation functions of ambient noise recorded at seismic stations located in the vicinity of the fault over the course of 10 years. Using the stretching technique, we compute daily velocity variations smoothed over a period of 100 days with a time lapse approach. Through the application of an optimization procedure based on synthetic modeling, we separate the non-tectonic, thermoelastic and rain induced velocity variations, from the tectonic components. Consequently, we unravel a significant velocity drop of 0.033% coinciding with the swarm occurring at seismogenic depth (3-5 km). Additionally, the time evolution of the velocity changes shows a direct relationship with the strain rate rather than the strain indicating a non-linear behavior of the crust induced by aseismic slip. The deduced strain sensitivity, exhibiting an order of magnitude comparable to that observed within volcanic settings, confirms this non-linear behavior and suggests the presence of pressurized fluids at depth. The driving mechanism of seismic swarms is still unclear. Seismic swarms are bursts of small earthquakes without a clear triggering mainshock. Understanding the physical processes driving these sequences is crucial for earthquake hazard assessment. To that scope, we focus in this study on a small seismic swarm that occurred in 2013 in northern Apennines, Italy. We monitor the properties of crustal rocks using continuous recordings of ambient seismic noise recorded at seismic stations located in the vicinity of the Alto Tiberina low angle normal fault over the course of 10 years. With this approach, we demonstrate the ability to retrieve quantitative information about the fault's state and to indirectly infer that the triggering of the swarm is mainly related to the presence of fluids in the crust. Velocity variations from seismic coda wave interferometry Separation of thermoelastic and hydrological stress induced velocity variation from tectonic changes Seismic swarm triggered by aseismic slip favored by the presence of fluids