Quantification of DAS VSP Quality: SNR vs. Log-Based Metrics

Area of Science:

• Geophysics
• Seismic Data Acquisition
• Distributed Acoustic Sensing (DAS)

Background:

• Initial data quality assessment is crucial for seismic data acquisition design, influencing sensing strategy choices.
• Signal-to-noise ratio (SNR) traditionally guides distributed acoustic sensing (DAS) parameter selection in vertical seismic profiling (VSP).
• Established SNR-based methods are compared against metrics derived from well log data for DAS VSP quality assessment.

Purpose of the Study:

• To compare traditional SNR-based data quality assessment with log-based metrics for DAS VSP.
• To analyze the relationship between seismic-derived and well log-derived data quality metrics.
• To propose the use of fit-for-purpose metrics for optimizing DAS VSP acquisition costs.

Main Methods:

• Kinematic and dynamic data products (e.g., interval velocity, P-wave arrival amplitude) were derived from seismic data.
• Statistical metrics were calculated to quantify data product quality by comparing with well log data.
• A dataset of 220 VSP experiments with varying DAS acquisition parameters (gauge length, conveyance type, lead-in length) was analyzed.

Main Results:

• A decoupling was observed between seismic-based and log-based quality metrics.
• The quality of dynamic and kinematic data products for the same record showed distinct trends.
• The study identified that different processing objectives require different quality metrics.

Conclusions:

• Traditional SNR metrics may not fully capture the quality relevant for all processing workflows.
• Fit-for-purpose metrics, such as traveltime-based metrics for tomography and amplitude-based metrics for dynamic analysis, are essential.
• Adopting fit-for-purpose metrics can optimize DAS VSP acquisition design and reduce costs.