Abstract The deployment of Distributed Acoustic Sensing (DAS) for seismic monitoring has significantly increased in recent years due to its numerous advantages over conventional seismic sensors. DAS has the potential to play a crucial complementary role along with classical seismic networks, particularly in logistic challenging areas such as offshore and volcanic environments. However, DAS data are inherently noisier than seismometer data, primarily due to fiber coupling issues and optical noise associated with the instrument. As a result, effective denoising and signal enhancement techniques are essential to fully exploit the advantages of DAS data. Recent efforts in improving the quality of DAS data have primarily focused on denoising algorithms (mostly deep learning-based) aimed at reducing coherent and background noise. However, improvements in terms of signal-to-noise ratio can also be achieved through the application of characteristic functions to raw or pre-processed data. To date, the application of these methods to DAS data have been largely unexplored, with the exception of few standard algorithms. In this study, we investigate the signal enhancement capability on DAS data of a new characteristic function based on the hyperbolic cosine. More specifically, we assess the performance of this function in improving the signal-to-noise ratio and compare the results against a set of more standard characteristic functions. Our analysis follows a two-step approach. First, we quantify the signal enhancement achieved through the application of the different characteristic functions by computing the signal-to-noise ratio of the preprocessed DAS data. In the second step, we evaluate their capability to enhance signal coherence across all fiber channels. This is achieved through the application of a coherence-based detector, which provides an estimate of the coherence as a function of time. Following a standardized denoising procedure, we systematically evaluate the impact of each characteristic function in increasing both the signal-to-noise ratio and coherence of DAS data. We conduct our analysis both on synthetic data and on a real dataset of 947 events recorded at the Frontier Observatory for Research in Geothermal Energy (FORGE) site, in Utah, USA. Graphical Abstract
Gaviano et al. (Fri,) studied this question.