科研之友
微信
新浪微博
Facebook
分享链接
摘要
Conventional acquisition in marine seismic exploration cannot remove ghost, a special noise, effectively, which has a great impact on waveforms and the bandwidth of primaries. Due to the incidence angle of the upgoing wavefield and the depth of streamer cable, interference between upgoing and downgoing wavefields creates nulls or notches in the recorded spectrum, which will lead to narrow band. So deghosting is one of the most important factors for enhancing resolution and higher fidelity in marine seismic exploration. @@@ To acquire broad-band data, we discuss a deghosting method based on Green function theory. We start from Green's formula and scattering theory, and firstly introduce a perturbation operator. The actual wave velocity can be treated as a perturbation of the background medium. We replace parameters of Green's formula with actual wavefield and Green function, then the upgoing wavefield can be calculated based on measured wavefield and its vertical derivative. However, the vertical derivative cannot be measured directly, so we adopt Huygens-Fresnel principle and integrate the energy of every virtual receiver. By using "Double Dirichlet" boundary conditions, the pressure and vertical velocity fields at any depth can be predicted. @@@ Three models are used to test our method. The first one is a simple three layer model, using 7 m depth streamer wavefield, we accurately estimate pressure and vertical velocity fields of 2 m. After deghosting, the ghost energy is removed, and simultaneously the signal-noise ratio is improved. Another example is actual dual-sensor data. Compared with original records, the resolution is improved with clearer events followed ghost suppressed. We supplement the notch energy and broaden the band effectively. Finally we apply the idea to actual towed streamer data. After wavefield prediction and deghosting, we improve the energy of low-frequency part. @@@ Based on Green's formula, a deghosting method based on Green theory is presented. Without the need of any subsurface information, it can successfully achieve data-driven ghost removal. By using "Double Dirichlet" boundary conditions, we can predict pressure and vertical velocity fields at any depth. In this paper, we establish a ghost suppression processing. The examples of a synthetic data and a marine data show that ghost can be well attenuated, simultaneously the band especially the low-frequency component is broadened, which is favorable for subsequent processing and interpretation.
- 出版日期2016-3
- 单位吉林大学
