Noise Analysis of Distributed Acoustic Sensing (DAS) Data in Borehole Installations

Abstract Distributed acoustic sensing (DAS) is a technology that is revolutionizing seismic data acquisition, particularly in borehole installations. Acting as a dense array of receivers, DAS provides high coverage, revealing time-depth patterns that are often hidden in data acquired with traditional seismometers. Its resilience to extreme temperature and pressure conditions, in which standard instrumentation typically fails, makes DAS reliable for microseismic monitoring operations in deep boreholes in geothermal environments. However, DAS faces challenges such as a lower signal-to-noise ratio compared to conventional geophones. DAS requires advanced denoising workflows in environments with high background noise, for example, from anthropogenic activities. A broader understanding and characterization of the noise observed in optical fibers is thus necessary and is still lacking. In this work, we aim to address this gap by analyzing noise data acquired from a fiber-optic cable installed in a monitoring well at the Utah Frontier Observatory for Research in Geothermal Energy Enhanced Geothermal System pilot project site in southcentral Utah, United States. Our proposed workflow combines power spectral density and phase analysis to assess the modulation of noise over time and depth for different frequencies and consequently to differentiate noise originating by anthropogenic sources at the surface from those further away from the industrial site. In addition, our analysis highlights noise components that may be related to instrumental noise from the interrogator, contributing to future noise mitigation strategies. This is further demonstrated through a direct comparison with noise observed by geophones coupled with the optical fiber in the same monitoring well.