A Python Code for Detecting True Repeating Earthquakes from Self-Similar Waveforms (FINDRES)

Abstract Seismic data are generally scrutinized for repeating earthquakes (REs) to evaluate slip rates, changes in the mechanical properties of a fault zone, and accelerating nucleation processes in foreshock and aftershock sequences. They are also used to study velocity changes in the medium, earthquake physics and prediction, and for constraining creep rate models at depth. For a robust detection of repeaters, multiple constraints and different parameter configurations related to waveform similarity have been proposed to measure cross-correlation values at a local seismic network and evaluate the location of overlapping sources. In this work, we developed a Python code to identify REs (FINDRES), inspired by previous literature, which combines both seismic waveform similarity and differential S-P travel time measured at each seismic station. A cross-spectral method is applied to evaluate precise differential arrival travel times between earthquake pairs, allowing a subsample precision and increasing the capacity to resolve an overlapping common source radius. FINDRES is versatile and works with and without P- and S-wave phase pickings, and has been validated using synthetic and real data, and provides reliable results. It would contribute to the implementation of open-source Python packages in seismology, supporting the activities of researchers and the reproducibility of scientific results.