Moving Towards Phase-Sensitive DAS Interrogators for Improved Waveform Correlation and Advanced Microseismic Processing Workflows

Distributed Acoustic Sensing (DAS) is an emerging measurement technology that provides high-resolution seismic data at a relatively low cost. Despite its various benefits and potential however, many questions remain and in particular the capability to record, process, and locate microseismic events (e.g. the fundamental understanding of the acquired optical data, comparatively high-noise floors, directional sensitivity and optimal processing schemes).

DEMODAS is a collaborative project aiming to develop a DAS interrogation unit and processing methods to confront these challenges, for safe carbon storage related microseismic research. The second-generation interrogator is a phase-sensitive unit that provides a more linear response along the fiber relative to the preceding intensity-based system.

Here we present the results from a series of controlled environment, lab and field tests using a piezoelectric transducer and active sources such as hammer blows, weight drops and airgun shots. The latest results from these show improved signal-to-noise ratios (SNR) and waveform coherency from channel-to-channel when using the updated phase-sensitive system. Longer gauge lengths (the distance over which strain fluctuations are averaged along the fiber) provide higher SNRs. These outcomes consequently allow for the implementation of waveform stacking methods.

Waveform comparisons between DAS channels down a 30 m water-filled vertical borehole and complimentary hydrophones show good correlations (up to 0.7) with reducing signal amplitudes at increasing source offsets. When compared to the hydrophone data however, SNRs are 5-10 dB less for the DAS recordings. Despite this, the system is capable of re-creating a consistent acoustic wavefield in the frequency-wavenumber domain and we are able to locate the shots to within a few meters accuracy when using conventional automated phase-picking and event location routines.

We summarize our findings with a suggested processing workflow for our in-house interrogator. Our results are encouraging, in the aim to establish a research-oriented interrogator to improve the understanding of interesting signals recorded via fiber optics.