Seismic event location with a small aperture DASarray: a case‑study from DIVE ICDP drilling project

Abstract

Distributed Acoustic Sensing (DAS) has emerged as an innovative technology in seismology, sensing seismic waves along fiber optic cables and, thus, increasing ten‑folds the spatial density of seismic measurements. However DAS potential in seismic monitoring is still under investigation. In this study, we assess the differences in seismic event detection and localization when using DAS, conventional seismometers, and combination of both, by analyzing a dataset acquired during a field experiment in Megolo di Mezzo (Northern Italy). A ~1 km buried fiber optic cable was deployed with an almost L‑shape configuration. Seismic data were continuously recorded from November 2023 to February 2024 and compared with simultaneous observations from the local seismic network (DIVEnet). Several local earthquakes were detected, including microseismic events not listed in the official catalog. P‑wave arrival times were extracted from DAS recordings using different picking algorithms and compared to manual picks from seismometers. Event localization was performed through a Bayesian Monte Carlo approach applied separately to DAS and seismometer data, and jointly. Results demonstrate that DAS shows considerable potential in earthquake detection, particularly for low‑magnitude events and those occurring close to the fiber optic cable, as potentially expected during anthropic activities underground. The joint inversion of DAS and seismometer datasets reduced localization uncertainties and produced solutions consistent with the official INGV catalog. However, differences of up to ~2 km between DAS – and seismometer – based epicenters highlight the limitations of simplified velocity models and the impact of network geometry. These findings confirm the complementarity of DAS and traditional networks and underline the potential of hybrid monitoring strategies for advancing earthquake detection and characterization in complex geological environments.