Exploration Geophysics - Volume 51, Issue 6, 2020

Due to the severe interference of background noise, the signal-to-noise ratio of desert seismic data is extremely low. In addition, due to low-frequency characteristics of sand in the Tarim desert region, the background noise in desert seismic data is mainly distributed in low-frequency band, so that the frequency spectrum aliasing of effective signals and background noise is more serious than the general land seismic data. Thus, conventional filtering methods cannot effectively suppress background noise in desert seismic data and recover effective signals. In order to overcome the problem that low-frequency background noise in desert seismic data is hard to suppress, a new method called R-TNRD based on robust principal component analysis (RPCA) algorithm and trainable nonlinear reaction diffusion (TNRD) network is proposed in this paper. By using the good sparsity of RPCA, the input noisy desert seismic data are decomposed into a low-rank matrix and a sparse matrix, and these two matrices contain background noise and effective signals. Due to the serious spectrum aliasing of desert seismic data, conventional thresholds have been unable to extract effective signals from the two matrices obtained by RPCA effectively. Therefore, we introduce TNRD network into desert seismic data denoising. By network training with a low-frequency noise set, the optimisation of TNRD network can be achieved, so as to accurately extract the effective signals from the low-rank matrix and the sparse matrix. In the experimental part, we test the performance of R-TNRD on both synthetic and real seismic data. The results demonstrate that the proposed method can suppress background noise more effectively than conventional methods.

Seismic exploration employing wide-azimuth, broadband, high-density data (i.e. double-width single-height; “double-width” means wide-azimuth and broadband; “single-height” means high-density data, generally referring to small-bin-size data (less than 10 × 10 m); and double-width single-height is an abbreviation) enables more complete wave field information to be recorded, reduces aliasing, and produces abundant low-frequency information, which is conducive to broadband processing and the anisotropic study of seismic data. Based on the acquired double-width single-height seismic data, in this paper, we analyse the wave field characteristics, signal-to-noise ratio and frequency of seismic data. We also design a procedure for processing double-width single-height seismic data. The key techniques in the proposed procedure focus on the high-resolution amplitude-preserved characteristics of double-width single-height seismic data obtained with a low-frequency vibroseis sweep. Faults and sand bodies are characterised and described by using the final imaging results, which embody the advantages of double-width single-height seismic exploration.

Many abandoned water-filled mine tunnels at unknown locations in major coal fields in China pose a hazard to the safe production in modern mines. Geophysical methods with rapid and accurate determination of the location and size of the tunnels are needed. The downhole transient electromagnetic method (TEM) applied in a borehole has significant potential for detection. A 3D finite-difference time–domain (FDTD) modelling method was used to simulate the anomalous response and parameter influence in conjunction with a model of a water-filled tunnel. The results show that the method gives a clear response and has distinct benefits for the detection of water-filled tunnels. Three components of the response signal were collected in an exploration hole in the vicinity of an abandoned mine in Shaanxi Province, China. The three components of the signal, especially the horizontal components, indicated obvious anomalies. We compared three different inversion procedures: (1) using three components of the total (primary + anomalous) field; (2) three components of the anomalous field and (3) using the horizontal components of the anomalous field. The results of the three inversions are almost the same and accord with the exploration result from densely distributed exploration boreholes nearby test borehole. The study showed that the downhole TEM effectively solves the problem of detecting and locating water-filled tunnels in the vicinity of a borehole. This information is a valuable complement to the geophysical detection of abandoned coal mine tunnels.

Seismic data interpolation is a meaningful research topic in the field of seismic data processing. In this paper, we propose deep internal learning for interpolating regularly sampled aliased seismic data, to improve the upsampling accuracy of regularly sampled aliased seismic data. The proposed algorithm, contrary to previous deep external learning-based seismic interpolation relying on prior training for vast external seismic data, exploits the characteristics of the field data itself, based on the feature similarity between the regularly missing and remaining samples. Internal learning generates training samples solely from the currently remaining regularly undersampled seismic data, and then trains a simple convolutional neural network using the training set. Finally, the trained model is used to upsample the current seismic traces regularly with high accuracy, and can adapt itself intelligently to different field data for the upsampling requirement. This enables seismic data antialiasing interpolation on regularly sampled seismic data with a small sample in the case of insufficient data. The performance of the proposed deep internal learning is assessed using synthetic and field data, respectively. Moreover, the comparison of the proposed deep internal learning with a classic prediction-based interpolation method and deep external learning-based seismic interpolation validates the effectiveness of the proposed algorithm.

To improve the computational accuracy of seismic traveltime for complex structures in inhomogeneous media, a mesh generation method that is adaptive to a rugged seafloor or interface is presented. A hybrid mesh discrete velocity model consisting of conventional rectangular mesh and irregular quadrilateral mesh is also proposed. The local traveltime equations for the hybrid mesh are derived from the linear traveltime interpolation (LTI) method and are shown to be stable. The relationship between the group velocity and group angle is converted into that between group velocity and interpolated point coordinates, extending the LTI ray tracing algorithm to multiwave simulation in vertical transversely isotropic (VTI) medium by multi-stage partitioning. Computations of the first arrival, reflection, multiple reflection, multiple transmission conversion, and multiple reflection conversion seismic waves in the VTI medium with rugged seafloor and complex structural interfaces are realised. The results show that the proposed method can adapt to the rugged seafloor and velocity interfaces of complex structures, resulting in higher computational accuracy of the traveltime and ray paths for ocean bottom node (OBN) seismic wave simulation. The accuracy of the proposed method is verified by comparison with the finite-difference method. This method will be useful for modelling of seismic waves, wavefield identification, and study of seismic wave propagation in OBN seismic observations.