4th EAGE Workshop on Fiber Optic Sensing for Energy Applications

Acquisition of borehole seismic data using distributed acoustic sensing (DAS) technology is a rapidly developing trend in borehole geophysics. This includes active VSP surveys for the subsurface characterisation and active and passive surveys for reservoir monitoring. While an increasingly popular technique onshore, time-lapse DAS VSP is still relatively rare in offshore environments as, among other challenges, it often requires going through long umbilical cables to the well.

Recently, some solutions to this problem have emerged. For instance, one approach involves adding passive components, like separate cores on tie-back cables, to transmit and receive pulses to and from the sensing fibre.

A field experiment of the DAS seismic acquisition using various fiber optic cables (FOCs) buried at approximately 50 cm depth in the surface ground was conducted at the Aquistore CCS site in November 2023. The primary purpose is to evaluate the seismic responses of the linear cable and the custom-made helically wound cable (HWC). The total length of the cable was almost 5 km, comprising four kinds of FOCs spliced in series to form a single loop: 1) a 30-degree HWC, 2) a linear FOC, 3) a 60-degree HWC, and 4) a lead-in FOC. The active sources consisted of dynamite sources and a permanent rotary seismic source. Inspection of the acquired shot records shows that the acquired data generally have good quality. Further analysis will focus on the variation of seismic response with the incident angle and azimuth.

Distributed Acoustic Sensing (DAS) enables efficient seismic data acquisition for permanent and passive micro-seismic monitoring, especially using permanent downhole installations. However, acquiring large amounts of data pushes against the limits of existing computational systems and algorithms, especially for continuous passive seismic monitoring applications. Thus, more than ever, novel methods to analyse big data are required. In this abstract, we investigate using a supervised deep learning neural network to detect and locate microseismic events due to CO2 injection. The challenges of using synthetic data to train the neural network were identified and addressed to fill the gap in the context of a microseismic application. The methodology was demonstrated on synthetic data and tested on data from the Otway CO2 injection site. More tests were performed to confirm the observed effects of including time shifts in the training data. Those enlightening results pave the way for a more extensive study and potential applications to more field data.

We present a field experiment using a high-frequency sparker source and a fibre optic cable cemented behind a deep well casing as a receiver array. The trial is conducted at the Curtin University GeoLab research facility in Perth, Western Australia. The study provides insights into using such a combination of high-frequency seismic sources and DAS for practical applications.

Joint surface and borehole seismic are a 3D surface-and-borehole seismic exploration method of the simultaneous acquisition of onshore or offshore 3D seismic and VSP (Vertical Seismic Profile) data using the same sources. When acquiring surface 2D or 3D seismic data in the field, simultaneously acquired 2D or 3D data can provide full well high-resolution structural images around the borehole and enhance surface 2D or 3D seismic data processing significantly. This paper describes the imaging processing of the 3D DAS (Distributed Acoustic Sensor) VSP data acquired by a downhole armored optical cable simultaneously with a 3D OBN data acquisition project in the Middle East. Apart from the conventional processing steps of 3D VSP data, the deblending processing of the multi-well blended acquired 3D DAS-VSP data using multiple airgun sources, special ringing noise removal procedure, joint domain full waveform inversion (JDFWI) for velocity model update, and one way wave equation multiple migration (OWEMM) method were used to generate final imaging. The results provide client with good quality imaging in a relatively large subsurface area.

Field experiment using a buried DAS telecommunication cable shows that propeller aircraft are detectable on spectrograms of DAS data by their characteristic Doppler signature. Their parameters and trajectory can be estimated (using multiple DAS channels). The detection range is about 2 to 3 km. Detecting noise form jet aircraft requires a smaller gauge length.

Depth variations of Rayleigh-wave amplitude in multilayered media are controlled by a complex combination of exponential decay of different modes as well as stiffness of individual layers. The widespread deployment of downhole fibre-optic cables provides an important opportunity for investigating the depth dependence of Rayleigh wave amplitude using distributed acoustic sensing (DAS). However, quantitative understanding of these dependencies is lacking. Here we model the Rayleigh-wave vertical strain amplitude versus depth for frequencies from 0.03 to 0.2 Hz theoretically and numerically, and compare the theoretical and numerical results to observed DAS amplitudes. Modelling results show that both modelled and observed DAS amplitudes display a clear anomaly at the depth of a CO2 saturated reservoir. The numerical amplitudes for a source depth of 5 km, 600 km away from the well location, have a reasonable agreement with the field data at about 0.1 to 0.2 Hz. However, at lower frequencies the agreement is not attainable, which requires further investigation.

Distributed acoustic sensing (DAS) with optical fiber cables has actual and potential applications, including energy applications, because it can densely measure the vibrations along the optical fiber compared with conventional receivers. For example, we already utilize DAS measurement for subsurface seismic monitoring by installing an optical fiber cable in a borehole. DAS with optical fibers deployed at the seafloor has also the potential for CO2 sequestration monitoring. However, seafloor or surface-deployed DAS for seismic acquisition still has a technical challenge for the directional sensitivity of fibers. DAS with a straight fiber cable does not effectively detect the vertical component of seismic waves, such as reflected P-waves. A method to address this challenge is helically wound fiber (HWF), which theoretically measures the summation of the vertical and horizontal components of vibration. However, there is no report on deploying HWFs for DAS measurement at the seafloor. Therefore, this study investigates the validity of seafloor-deployed DAS with HWFs for CO2 sequestration monitoring using time-lapse elastic full waveform inversion (FWI) for 4D seismic data. We show numerical examples to evaluate the relation between the accuracy of the time-lapse change estimated by FWI and wrapping angles of HWFs in seafloor-deployed DAS.

Fiber-optic distributed acoustic sensing (DAS) applied to microseismic has gained popularity in various microseismic monitoring applications. Compared to conventional geophone array based microseismic monitoring, DAS microseimsic has unique advantages in minimizing location uncertainties. However, if only one straight fiber were deployed on surface or in downhole, DAS may not be able to uniquely determine a microseismic event location without other constraints due to circular event location ambiguities. Since a standard DAS fiber cable usually can only record one component seismic wave data which is not able to provide source direction to reduce the circular ambiguities in event location, we developed a innovative joint DAS and 3C geophone array for microseismic monitoring and a three-step microseismic event location workflow to obtain a unique event location. First, we calculate source direction using P-wave first arrival on 3C geophone. Second, we calculate 2D semblance image along the source-geophone direction. Third, we extract the location of maximum semblance value in 2D semblance image as the event location. We presented a microseismic event location example which was obtained by applying the three-step location workflow on microseismic event dataset acquired by a field survey with joint DAS and 3C geophone array.

Distributed Acoustic Sensing (DAS) is an important technology for a long-term monitoring for geological carbon storage as optical fiber cables are durable for years. Theoretically, DAS measurement using optical fibers deployed on surface (S-DAS) record P-P reflected and P-S converted wave simultaneously in an active seismic survey. This study explores the possibility of the P-S converted wave imaging with S-DAS 2D land seismic data from a field trial conducted in Japan.

We performed preliminary P-P reflection processing and P-S converted reflection processing on the same data set. Reflection events that is not clear in the P-P reflection image are observed clearly in the P-S converted reflection image in the shallow area. This may be attributed to the difference in impedance contrasts and sensitivity of DAS for a wave incident angle to cable. Imaging of deep area might be challenging in terms of the degree of S wave attenuation.

We confirm that S-DAS measurement can give subsurface image both by P-P reflection and P-S converted wave, and that P-S converted wave gives additional information to leakage monitoring for shallow zone in the geological carbon storage.