This study analyzes data from a Local Seismic Network affected by documented random time synchronization errors, aiming to develop correction strategies to improve earthquake location accuracy, particularly for moderate to small events. The work focuses on optimizing data usage in hypocentral locations and focal mechanisms in a seismically active area where the Italian National Seismic Network (RSN) has uneven station coverage. With this aim, 15 seismic events recorded by the local Rete Sismica Abruzzo (RSA) and RSN between 2007 and 2016 in the central Apennines were selected. First, we assess the effectiveness of the automatic picking procedure by repicking the first arrivals of Pand Swaves in RSN data that we previously picked manually. Then, we applied the automatic picking procedure to RSA data using the SeisBench toolbox and PhaseNet model to reduce potential analystrelated errors. Both datasets are then processed with Hypoellipse for event relocation and FOCMEC for focal mechanism inversion. Five inversion tests were performed to assess the influence of RSA on final solutions in terms of error, varying the codes assigned to Pand Sphase arrivals. We explore ts-tp differential times (where ts and tp denote the arrival times of the S and P phases) as a correction strategy to integrate RSA data into hypocentral solutions. Results indicate that the sole integration of RSA data improves earthquake detection quality regarding azimuthal gap, minimum epicentral distance, horizontal and vertical errors, and RMS values, except for some events. Moreover, using tstp arrival differences helps mitigate clock synchronization errors, enhancing the event’s location and focal mechanism accuracy. However, assigning lower weights to RSA data can sometimes reduce the reliability of the estimates. The findings suggest that careful weighting of these differences is essential to ensure highquality seismic analyses even when using ts-tp intervals as a corrective measure.
Miccolis et al. (2025) studied this question.