Characterizing Seismic Swarm Activity in the Aftershock Zone of the 2016 Amatrice-Norcia Seismic Sequence

Abstract

The 2016 Central Italy seismic sequence unfolded in an area dominated by normal faulting along the Apennines. Our study models this complex sequence to explore the role of fluids and aseismic slip in triggering small seismic swarms. Leveraging a high-resolution earthquake catalog, we apply a three-dimensional density-based clustering approach to group earthquakes into dense clusters. We further define a four-dimensional seismicity model based on the Epidemic-Type Aftershock Sequence (ETAS) model, introducing an earthquake detection probability to accommodate rapid fluctuations in earthquake detection. By computing the ratio between observed and ETAS-modeled rates of earthquakes in high-density clusters, we identify 40 seismic swarms. These swarms occurred relatively uniform throughout the sequence, with significant time delays of weeks to months following the mainshocks. We investigated earthquake migration patterns during these swarms and found that 29 of them exhibited significant migration, particularly in the eastern parts of Visso and Norcia, as well as the southern part of the seismic sequence. The computation of migration velocity for each swarm yielded a mean migration rate of 3.7 m/hr, indicating a connection to fluid migration. The pressure changes exerted by the three largest shocks of the sequence are found to be mostly negative at the swarm locations, suggesting that the time delays between the mainshocks and the swarms are controlled by fluid migration from high- to low-pressure zones co-seismically. We propose that the poroelastic rebound, involving stress redistribution due to fluid movement, plays a crucial role in triggering seismic swarms in the geological context of Central Italy.