Investigation of the Time-Lapse Changes with the DAS Borehole Data at the Brady Geothermal Field Using Deconvolution Interferometry

The distributed acoustic sensing (DAS) has great potential for monitoring natural-resource reservoirs as well as borehole well-beings. However, the DAS wavefields of these applications are complicated and the noise levels are often high due to unknown coupling conditions. Therefore, we seek for an advanced array processing technique that takes advantages of the high spatial receiver density of DAS. In this study, we apply seismic interferometry based on deconvolution to DAS borehole data observed at the Brady geothermal field in Nevada to extract coherent and interpretable waves. The data is from the PoroTomo project at the Brady geothermal field in Nevada. With the deconvolution, we extract strong reverberating signals between 0-165 m depth due to the resonance of the borehole casing. We investigate the propagating velocity of the extracted waves and the velocity variation compared to depth, observation time, temperature, and pressure. With analytical and numerical modeling, we discover that a simple string model with multiple sources can explain the observed data well. The key to explain our observation is the sources coherency, and reflection coefficients at the boundaries. The amplitude spectra show clear normal modes of such reverberations, which are useful for dispersion analysis of the waves. For DASV below 200 m depth, we only obtain signals during the active seismic operation time due to poor coupling. Deconvolution interferometry is a powerful tool for analyzing the large volume of data observed by DAS and monitoring time-lapse changes of the propagation media and external sources.