Enhanced speed of sound measurement in Distributed Acoustic Sensing (DAS) using Slepian multitaper frequency-wavenumber analysis

Distributed Acoustic Sensing (DAS) has emerged as a powerful technology for monitoring wells and pipelines, with Speed of Sound (SoS) being a unique fluid property that can be derived from DAS data. However, accurately estimating the local SoS presents challenges primarily due to the non-stationary behavior of the data in the spatial domain, limited spatial resolution, and spectral leakage when applying FFT along either the time or spatial domains. In this study, we evaluate the performance of spatiotemporal multitaper frequency–wavenumber (FK) analysis for measuring SoS compared to conventional 2-D Fast Fourier Transform (FFT). We also introduce a joint methodology that applies multitapering in the time domain while implementing FFT in the spatial domain. This hybrid technique leverages the spectral concentration and noise robustness of Slepian sequences alongside the frequency resolution and lower variance of FFT. The joint methodology is rigorously validated through comprehensive analyses of synthetic and real-world DAS datasets, demonstrating its effectiveness in reducing the uncertainty of DAS-based SoS measurements.