DAS VSP Data from Well Tubing Versus Well Casing in CO2 Injector and Monitor Well

Abstract The decision whether to deploy Distributed Acoustic Sensing (DAS) optical fiber on tubing or casing varies across Carbon Capture and Sequestration (CCS) operators and can impact well integrity, data quality, reliability, and cost. The scope of this work is to share results from a project recently commencing injection and evaluate DAS fiber denoising methods. This case study shares processed Vertical Seismic Profile (VSP) shot gathers where we coupled fiber to both tubing and behind casing, in an active CO2 injector well and monitor well. Both wells have nearly identical well configurations allowing for precise comparison on four independent fiber optic cables. To achieve seismic source consistency and effectiveness, two vibroseis sources were operated on frozen roads. Optimization of sweep parameters and DAS gauge length was achieved via on-site QC of early sweeps. As predicted, it was found that there was a significant variation in signal response between fiber coupled to tubing and fiber coupled to casing. While processed P-wave arrival times were consistent, injection noise was found to have more effect on fiber coupled to tubing in the injection well. This is believed to be due to the fiber placement within the wellbore annulus, as it is not cemented and instead surrounded in a brine fluid between the tubing and production casing. This fiber also has cross coupling protector clamps every 30 feet, the approximate joint length. Whereas fiber installed on casing is commonly cemented in place and may have an increased number of clamps (coupling) per joint. We investigated processing different shot offset gathers with different denoising techniques and strategies for fiber coupled to tubing to enhance low seismic signal responses and remove or minimize interfering noise. Our initial approach applied standard denoising filters that were not effective at removing the noise. Our secondary approach was to use the fiber coupled to casing data as a "guide" for denoising the fiber coupled to tubing data. These results were significantly more successful, as true amplitudes were enhanced while interfering noise was diminished or completely removed. Overall, past research has found that DAS VSPs can deliver contradictory findings in terms of signal to noise ratio (SNR), due to tubing design and suitability. But our findings imply that by revising the CCS well design and seismic processing methods, coupling noise from poor coupling of the fiber on tubing could be minimized to produce reliable results. This could greatly benefit engineers and geophysicists within industry as it would impact well integrity, reliability, and cost in their well programs. Our team plans to continue processing these data to gain more insight for future knowledge sharing.