09:45 〜 10:00
[STT36-04] Elastic full waveform inversion for active seismic data acquired with seafloor DAS technology
キーワード：分布型音響センシング、インバージョン
1. Introduction
In recent years, distributed acoustic sensing (DAS) technology using optical fiber cables has attracted attention in various fields, including geophysics. DAS is widely used in subsurface imaging and monitoring, such as vertical seismic profile, because of its advantage of recording seismic waves along fiber-optic cables at high densities compared to geophones and hydrophones. In addition, DAS technology has the potential to be a cost-effective approach to monitoring earthquakes, CO2 storage, and oil and gas production fields. For example, previous research has discussed the use of DAS with existing fiber cables for earthquake monitoring. Furthermore, a study on permanent reservoir monitoring reported that seafloor setting DAS could be a cost-effective approach compared to conventional ocean bottom seismic acquisition [1]. Therefore, DAS application at the surface or seafloor would be beneficial for subsurface monitoring.
However, DAS using optical fibers installed on the surface or seafloor has its limitations. Since DAS with a straight fiber cable dominantly measures the horizontal component of vibrations along the cable, measuring for the vertical components of vibrations, such as vertically propagated P-waves, is challenging. Therefore, the use of surface waves, refracted waves, and P-S converted waves is expected to play an important role in imaging and velocity model estimation with DAS data recorded at the surface and seafloor.
Recent studies have been conducted to estimate velocity models using the algorithm of full waveform inversion (FWI) [2], which is a tomography based on the manner of waveforms, for seismic data measured with DAS at the surface or in horizontal wells [3-5]. However, the problem of deriving a subsurface velocity model at depth levels deeper than near-surface from seismic data measured at the surface or seafloor with DAS remains unresolved. In this study, we apply FWI [6] to seismic data measured at the seafloor to verify its capability for velocity estimation at depth levels. We also examine the effects of refracted and P-S converted waves on the velocity estimation.
2. Methodology
In this study, we conduct numerical experiments with simple synthetic velocity models to verify the effectiveness of elastic FWI [6], which assumes an elastic medium, with only the horizontal component of seismic data. Synthetic seismic data is generated by the finite difference method of seismic forward modeling. We assume that the seismic data does not include surface waves and phase cycle skipping in these numerical experiments.
3. Results
Numerical experiments showed that elastic FWI can accurately estimate velocity models, including P-wave velocity models only with the horizontal component of the seismic data. They also indicated that the influence of P-S converted waves on the inversion results is limited and the length of offset in the seismic data plays a significant role in the velocity estimation.
REFERENCES.
[1] Å.S. Pedersen et al., “A North Sea Case Study: Does DAS Have Potential for Permanent Reservoir Monitoring?,” 83rd EAGE Annual Conference & Exhibition, Extended Abstracts, 1-5 (2022).
[2] A. Tarantola,, “Inversion of seismic reflection data in the acoustic approximation: Geophysics,” 49, 1259-1266 (1984).
[3] W. Pan et al., “Imaging near-surface S-wave velocity and attenuation models by full-waveform inversion with distributed acoustic sensing-recorded surface waves,”, Geophysics, 88, 1, R65-R78 (2023).
[4] M. B. S. Yust et al., “Near-Surface 2D Imaging via FWI of DAS Data: An Examination on the Impacts of FWI Starting Model,” Geosciences, 13, 3 (63) (2023).
[5] M. V. Eaid et al., “Multiparameter seismic elastic full-waveform inversion with combined geophone and shaped fiber-optic cable data,” Geophysics, 85, 6, R537-R552 (2020).
[6] D. Köhn et al., “On the influence of model parametrization in elastic full waveform tomography,” Geophysical Journal International, 191, 1, 325-345 (2012).
In recent years, distributed acoustic sensing (DAS) technology using optical fiber cables has attracted attention in various fields, including geophysics. DAS is widely used in subsurface imaging and monitoring, such as vertical seismic profile, because of its advantage of recording seismic waves along fiber-optic cables at high densities compared to geophones and hydrophones. In addition, DAS technology has the potential to be a cost-effective approach to monitoring earthquakes, CO2 storage, and oil and gas production fields. For example, previous research has discussed the use of DAS with existing fiber cables for earthquake monitoring. Furthermore, a study on permanent reservoir monitoring reported that seafloor setting DAS could be a cost-effective approach compared to conventional ocean bottom seismic acquisition [1]. Therefore, DAS application at the surface or seafloor would be beneficial for subsurface monitoring.
However, DAS using optical fibers installed on the surface or seafloor has its limitations. Since DAS with a straight fiber cable dominantly measures the horizontal component of vibrations along the cable, measuring for the vertical components of vibrations, such as vertically propagated P-waves, is challenging. Therefore, the use of surface waves, refracted waves, and P-S converted waves is expected to play an important role in imaging and velocity model estimation with DAS data recorded at the surface and seafloor.
Recent studies have been conducted to estimate velocity models using the algorithm of full waveform inversion (FWI) [2], which is a tomography based on the manner of waveforms, for seismic data measured with DAS at the surface or in horizontal wells [3-5]. However, the problem of deriving a subsurface velocity model at depth levels deeper than near-surface from seismic data measured at the surface or seafloor with DAS remains unresolved. In this study, we apply FWI [6] to seismic data measured at the seafloor to verify its capability for velocity estimation at depth levels. We also examine the effects of refracted and P-S converted waves on the velocity estimation.
2. Methodology
In this study, we conduct numerical experiments with simple synthetic velocity models to verify the effectiveness of elastic FWI [6], which assumes an elastic medium, with only the horizontal component of seismic data. Synthetic seismic data is generated by the finite difference method of seismic forward modeling. We assume that the seismic data does not include surface waves and phase cycle skipping in these numerical experiments.
3. Results
Numerical experiments showed that elastic FWI can accurately estimate velocity models, including P-wave velocity models only with the horizontal component of the seismic data. They also indicated that the influence of P-S converted waves on the inversion results is limited and the length of offset in the seismic data plays a significant role in the velocity estimation.
REFERENCES.
[1] Å.S. Pedersen et al., “A North Sea Case Study: Does DAS Have Potential for Permanent Reservoir Monitoring?,” 83rd EAGE Annual Conference & Exhibition, Extended Abstracts, 1-5 (2022).
[2] A. Tarantola,, “Inversion of seismic reflection data in the acoustic approximation: Geophysics,” 49, 1259-1266 (1984).
[3] W. Pan et al., “Imaging near-surface S-wave velocity and attenuation models by full-waveform inversion with distributed acoustic sensing-recorded surface waves,”, Geophysics, 88, 1, R65-R78 (2023).
[4] M. B. S. Yust et al., “Near-Surface 2D Imaging via FWI of DAS Data: An Examination on the Impacts of FWI Starting Model,” Geosciences, 13, 3 (63) (2023).
[5] M. V. Eaid et al., “Multiparameter seismic elastic full-waveform inversion with combined geophone and shaped fiber-optic cable data,” Geophysics, 85, 6, R537-R552 (2020).
[6] D. Köhn et al., “On the influence of model parametrization in elastic full waveform tomography,” Geophysical Journal International, 191, 1, 325-345 (2012).