Dip Filtering Research Articles

Protecting the weak signals in distributed acoustic sensing data processing using local orthogonalization: The FORGE data example

The development of the distributed acoustic sensing (DAS) technique enables us to record seismic data at a significantly improved spatial sampling rate at meter scales, which offers new opportunities for high-resolution subsurface imaging. However, DAS recordings are often characterized by a low signal-to-noise ratio (S/N) due to the presence of data noise, significantly degrading the reliability of imaging and interpretation. Current DAS data noise reduction methods remain insufficient in simultaneously preserving weak signals and eliminating various types of noise. Particularly when dealing with DAS data that are contaminated by four types of noise (i.e., high-frequency noise, high-amplitude erratic noise, horizontal noise, and random background noise), it becomes challenging to attenuate the strong noise while maintaining fine-scale features. To address these issues, we develop an integrated local orthogonalization (LO) method that can remove a mixture of different types of noise while protecting the useful signal. Our LO method effectively eliminates the aforementioned noise by concatenating multiple denoising operators including a band-pass filter, a structure-oriented, spatially varying median filter, a dip filter in the frequency-wavenumber domain, and a curvelet filter. Next, the local orthogonalization weighting operator is applied to extract signal energy from the removed noise section. We demonstrate the robustness of our LO method on various challenging DAS data sets from the Frontier Observatory for Research in Geothermal Energy geothermal field. The denoising results demonstrate that our LO method can successfully minimize the levels of different types of noise while preserving the energy of weak signals.

An advanced median filter for improving the signal-to-noise ratio of seismological datasets

Strong noise usually disturbs the recorded seismic waves, resulting in low-quality seismic data sometimes with an extremely low signal-to-noise ratio (S/N), which negatively impacts the subsequent seismological processes, e.g., imaging, inversion, and interpretation. Suppressing undesirable noise is a meaningful procedure to increase the S/N of seismological datasets. This issue can be partially addressed using the recently proposed structure-oriented space-varying median filter (SOSVMF) by filtering out the unwanted noise from noise-corrupted signals. However, the SOSVMF approach has limitations when the raw data is corrupted by various sources of noise which is more common in practice. To conquer the difficulties in improving the S/N in the face of diverse types of strong noise, we introduce MATamf, an open-source MATLAB code package based on a novel advanced median filter (AMF). The proposed AMF workflow takes advantage of multiple denoising operators, i.e., the bandpass (BP) filter, SOSVMF, dip filter in the frequency–wavenumber (FK) domain, Curvelet, and local orthogonalization (LO), all in the same framework to simultaneously attenuate all types of noise and improve S/N. To demonstrate the usefulness of the MATamf package and its advantage over conventional denoising methods, we introduce the principles behind the proposed AMF framework and present results from a variety of seismological problems, including seismic reflection, distributed acoustic sensing (DAS), as well as global seismological tasks of receiver function and SS-precursors imaging. All results show that the proposed AMF workflow can be conveniently utilized to improve the S/N of a wide spectrum of seismological applications.

The Application of Seismic Attributes in Fault Detection and Direct Hydrocarbon Indicator in Tomboy Field, Western-Offshore Niger Delta Basin

Seismic attribute analysis is important in subsurface data interpretation, such as seismic interpretation, which could involve seismic stratigraphic and structural interpretation. This interpretation is often hampered by seismic resolution and, sometimes, human inability to identify a subtle feature on the seismic. These factors have frequently led to the poor seismic interpretation of geologic features. Thus, an integral approach to studying structural patterns and hydrocarbon bearing zones using seismic attributes was carried out on the Tomboy field using 3D seismic data covering approximately 56 km2 of the western belt of the Niger Delta. The seismic volume underwent post-stack processing, which enhanced seismic discontinuities. A deep steering volume was first created, and several dip filters were applied to enhance faults in the study area. After that, curvature and similarity attributes were calculated on the dip-steered and fault-enhanced volume. These calculations show detailed geometry of the faults and zones of subtle lineaments. Six faults (F1, F2, F3, F4, F5 and F6) were identified and mapped. These faults range from antithetic to crest growth faults. Two major growth faults (F5 and F6) were revealed to dip in the northeast to southwest directions. A near-extensive crest fault (F4) appeared beneath the major faults. Although several minor fractures were displayed in the southern and central portions of the crest fault of the dipping seismic data, the southwest (F4) and growth fault, F6, are responsible for holding the hydrocarbon found within the identified closures. Using attributes on the seismic data increased confidence in mapping and interpreting structural features. Furthermore, energy attributes were used as Direct Hydrocarbon Indicators (DHI) to visualize viable areas within the study, which allows a more robust interpretation. Time slices were taken in regions of flat and bright spots. The spectral decomposition attribute was run on these slices to display areas of high amplitude reflection typical of hydrocarbon-bearing regions trapped mainly by regional to sub-regional growth faults. The surface attribute calculated on the generated surface shows that the field is predominantly controlled by faults serving as traps for hydrocarbon.

Denoising of Distributed Acoustic Sensing Seismic Data Using an Integrated Framework

Abstract Distributed acoustic sensing (DAS) is an emerging technology that offers great potential in the high-resolution multi-scale seismic investigation due to its dense spatial coverage and cost-effectiveness. However, DAS data notoriously suffer from the low signal-to-noise ratio (SNR) due to various types of strong noise, for example, high-frequency noise, high-amplitude erratic noise, vertical or horizontal noise. Here, we propose a novel denoising framework by cascading several individual denoising methods that are designed for suppressing specific types of noise. First, to suppress the high-frequency noise, we apply a bandpass filter, which is implemented by recursive infinite impulse response filtering in the time domain. Second, to suppress the erratic noise, we apply a structure-oriented median filter that arises from the reflection seismology field. Finally, to suppress the vertical or horizontal noise, we apply a carefully designed dip filter in the frequency–wavenumber domain. The overall effect of these cascaded denoising steps is that the DAS data can be dramatically improved in terms of SNR. We introduce in detail the implementations of each step in the proposed denoising framework and analyze their respective contribution toward the final improvement. We demonstrate the effectiveness of the proposed denoising framework through the open-access Frontier Observatory for Research in Geothermal Energy (FORGE) geothermal DAS dataset and provide the reproducible processing workflows for all the DAS subsets containing the catalog earthquake and microseismic events.

Denoising of distributed acoustic sensing data using supervised deep learning

Distributed acoustic sensing (DAS) is an emerging technology for acquiring seismic data due to its high-density and low-cost advantages. Because of the harsh acquisition environment and other unexpected reasons, the seismic signals acquired in DAS are masked by various types of complex noise, which seriously decreases the signal-to-noise ratio of seismic data. We propose a fully convolutional neural network with dense and residual connections to attenuate complex noise in DAS data. The network is designed to learn features of useful reflection signals recorded from a large number of earthquake and microseismic events, aiming at obtaining an unprecedented generalization ability. First, we generate labels using an integrated framework that attenuates specific types of noise in real DAS data, where the integrated framework includes carefully designed band-pass, structure-oriented median, and dip filters. Then, we use the patching technique to segment the training samples into many small-scale patches to reduce computational cost and improve the extraction of essential features from large-scale passive seismic data. Finally, we use the well-trained network to estimate the heavily polluted hidden signals. Compared with two advanced deep-learning methods and a traditional denoising framework, our proposed method can more effectively attenuate strong and complex noise and recover weak hidden signals in synthetic and real DAS data tests.

Reconstruction of borehole radar images by a modified f–k migration

SUMMARY Borehole radar (BHR) is an effective imaging tool. It can be used to detect and map faults, fractures, folds, domes, partings and mine workings. Most BHRs have azimuthally omnidirectional radiation patterns. The echoes sensed by such BHRs may come from any direction. Considering a radar in a straight borehole that passes through a stack of flat reflection planes, V-shaped events or crosses appear on the time section. One of the arms of each cross is a real image while the other is an ambiguity of known origin. Directional ambiguities such as these obstruct efforts to interpret the data. In this paper, we address this difficulty by using a modified f–k migration algorithm to translate crosses into lines on the final section that are consistent with a priori information about for example bedding. Compared with conventional strategies, for example migration + f–k dip filter, this approach integrates the two separated processes into one and is straightforward, computationally effective and simple to implement. The method is demonstrated using a synthetic model and a real BHR field data set. It allows the interpreter to use a priori information about fault swarms or plausible bedding planes at an early stage. The reconstructed BHR image helps the search for geological anomalies such as fractures, partings, domes and rolls that could be a hazard for mining.

Analysis of direction-decomposed and vector-based elastic reverse time migration using the Hilbert transform

Straightforward implementations of elastic reverse time migration (ERTM) often produce imaging artifacts associated with incorrectly imaged mode conversions, crosstalk, and back-scattered energies. To address these issues, we introduced three approaches: (1) vector-based normalized crosscorrelation imaging conditions (VBNICs), (2) directional separation of wavefields to remove low-wavenumber noise, and (3) postimaging filtering of the dip-angle gathers to eliminate the artifacts caused by nonphysical wave modes. These approaches are combined to create an effective ERTM workflow that can produce high-quality images. Numerical examples demonstrate that, first, VBNICs can produce correct polarities for PP/PS images and can compute migrated dip-angle gathers efficiently by using P/S decomposed Poynting vectors. Second, they achieve improved signal-to-noise and higher resolution when performing up/down decomposition before applying VBNICs, and left/right decomposition enhances steep dips imaging at the computational cost of adding the Hilbert transform to a spatial direction. Third, dip filtering using slope-consistency analysis attenuates the remaining artifacts effectively. An application of the SEG advanced modeling program (SEAM) model demonstrates that our ERTM workflow reduces noise and improves imaging ability for complex geologic areas.

Dip Filter and Random Noise Suppression for GPR B-Scan Data Based on a Hybrid Method in f - x Domain

Ground-penetrating radar (GPR) is a close-range remote-sensing tool applied in a great many near-surface projects for engineering or environmental purposes. In GPR B-scans, there may exist a variety of reflections and diffractions that corresponds to different structures and targets in the subsurface media, and the noise is always embedded. To assist in the interpretation, GPR B-scans can be generally divided into two parts according to the dip attribute of the reflections, where the sub-horizontal layers and dipping structures are properly separated. In this work, we extend the f - x empirical mode decomposition (f - x EMD) to form a semi-adaptive dip filter for GPR data. In f - x domain, each frequency slice is decomposed by EMD and reconstructed to form a dipping profile and a horizontal profile respectively, where the reflections at different dips are separated adaptively. Then the noises mixed in the dipping profile are further separated by rank-deduction methods in f - x domain. The above two-step scheme constitutes the hybrid scheme, which can separate the dipping structures, sub-horizontal layers, and most of the random noise in GPR B-scans. We briefly review the basics of the f - x EMD, and then introduce the derived hybrid scheme in f - x domain. The proposed method is tested by the synthetic data, the forward simulation data, and the field data, respectively.

Structural Analysis of Lunar Regolith from LPR CH-2 Data Based on Adaptive f-x E MD: LPR Data Processed by Adaptive f-x EMD

The Lunar Penetrating Radar (LPR) is one of the important scientific payloads in China’s Chang’E-3 (CE-3) to image within 100 m below the lunar surface. The acquired LPR data is significant for the research of lunar geological structure. Based on the sedimentary mechanism of lunar regolith, the regolith contains many rocks with different sizes. These local anomalies appear as diffraction in LPR data, which reduces the data quality and limits the structural analysis of lunar regolith. According to the kinematics characteristics of rock caused diffraction, we transform these problems to a problem of steep dip decreasing. To reach this goal, we adopt a data preprocessing workflow to improve the quality of the radar image, firstly. Then, a dip filter based on adaptive f-x empirical mode decomposition (EMD) is proposed to extract the rocks in the regolith and the corresponding removed IMF map indicates the degree of rock enrichment and highlights regolith-basement interface. Both simulation and LPR CH-2 data present a great performance. Finally, according to the processed result, we locate the position of each rock and highlight the contact interface of regolith and the basement rock.

Rock Location and Quantitative Analysis of Regolith at the Chang’e 3 Landing Site Based on Local Similarity Constraint

Structural analysis of lunar regolith not only provides important information about lunar geology but also provides a reference for future lunar sample return missions. The Lunar Penetrating Radar (LPR) onboard China’s Chang’E-3 (CE-3) provides a unique opportunity for mapping the subsurface structure and the near-surface stratigraphic structure of the regolith. The problem of rock positioning and regolith-basement interface highlighting is meaningful. In this paper, we propose an adaptive rock extraction method based on local similarity constraints to achieve the rock location and quantitative analysis for regolith. Firstly, a processing pipeline is designed to image the LPR CH-2 A and B data. Secondly, we adopt an f-x EMD (empirical mode decomposition)-based dip filter to extract low-wavenumber components in the two data. Then, we calculate the local similarity spectrum between the filtered CH-2 A and B. After a soft threshold function, we pick the local maximums in the spectrum as the location of each rock. Finally, according to the extracted result, on the one hand, the depth of regolith is obtained, and on the other hand, the distribution information of the rocks in regolith, which changes with the path and the depth, is also revealed.

Imaging the crustal structure along the KCRT-2008 profile in the Southern Korean Peninsula using regularisation

Information regarding crustal structure is important for geological and geophysical research. Several attempts have been made to delineate the crustal structure of the Korean Peninsula, but the resolution in these studies has generally been too low. To acquire detailed images of the regional subsurface structure, we applied seismic data processing tools that are commonly used in oil exploration to crustal refraction data recorded along the Korea Crust Research Team 2008 (KCRT-2008) profile at irregular intervals. To use conventional seismic data processing tools, the data should be sampled regularly in both time and space domains. In this study, we applied the matching pursuit interpolation technique to regularise the KCRT-2008 data in the space domain, followed by dip filtering to attenuate coherent noise. Then, Kirchhoff prestack depth migration was applied using both refracted and reflected waves to image the subsurface structure at higher signal-to-noise ratios and increase spatial resolution using far offset data. The migrated section shows the interfaces of three mid-crustal layers at depths of 14.27 and 24.37–25.34 km and the downwarped Moho at a depth range of 29.3–36.1 km.

Validation of DAS data integrity against standard geophones — DAS field test at Aquistore site

Distributed acoustic sensing (DAS) using fiber-optic cables is a recent addition to seismic acquisition methods. However, a DAS “sensor” differs significantly from conventional, discrete sensing devices such as geophones or accelerometers. For one, DAS measures something akin to strain instead of particle velocity or acceleration. Other properties of the DAS system also aren't obvious at first. What is its instrument response, noise performance, and repeatability? How are DAS channels properly positioned, e.g., in case of a borehole deployment: depth calibrated? To better understand these issues and their impact on the DAS seismic method's application space, a field test was conducted in which three DAS vendors recorded the same survey using a borehole-installed fiber while recording simultaneously with a conventional downhole array. The results show that all DAS systems achieved good, repeatable signal integrity while exhibiting different noise characteristics. DAS noise can be addressed with well-established processing algorithms, but further benefits can be gained from DAS-specific algorithms. Where required, DAS seismic data can be processed to closely match the vector response of conventional geophones. DAS data converted in this way can assist in the up/down separation step without the need for dip filters. DAS VSP data can also be merged with conventional 3D and 4D seismic, adding value in situations such as undershooting of surface facilities in marine settings.

Efficient anisotropic dip filtering via inverse correlation functions

We have developed a computational framework that allows an efficient, spatially variant correlation filter for anisotropic dip filtering. The approach is based on the Laplace correlation function, for which there exists analytical expressions for the correlation kernel and its inverse kernel in the 1D case. An extension to higher dimensions by adding orthogonal 1D inverse functions provides a linear equation whose solution is identical to applying a Bessel filter. We have found that a good approximation of the Laplace filter function is obtained by applying a cascade of Bessel filters. We implement such an inverse operator on regular grids with a finite-difference stencil, giving a sparse matrix for the linear equation, which can be solved efficiently through a conjugate-gradient algorithm. Computing such a correlation operation by solving the linear system involving the inverse operator is significantly faster than applying the correlation function via convolution or windowed convolution: Solving the linear system is as fast as applying often-used tensorized 1D convolutions.

Attenuation of noise and simultaneous source interference using wavelet denoising

High-resolution seismic imaging requires noise attenuation to achieve signal-to-noise ratio (S/N) improvements without compromising data bandwidth. Amplitude versus offset analysis requires good amplitude fidelity in premigration processes. Any nonreflected wavefield energy in the data will degrade the seismic image quality. Despite significant progress over the years, preserving low-frequency signals without compromising the S/N and avoiding the smearing of aliased signal are still a challenge for conventional methods. This problem is compounded when additional interference noise is added with simultaneous source acquisition. Because noise characteristics vary from shot to shot and receiver to receiver, we need a method that is robust and effective. In addition, we also want the method to be efficient and easy to use from a practical perspective. We have recently developed an approach using a wavelet transform to deterministically separate the primary signal from the noise, including simultaneous source interference. The goals are (1) improving the S/N without compromising bandwidth, (2) preserving the low-frequency and near-offset primaries without compromising the S/N, and (3) preserving the local primary wavefield while attenuating noise. For distance-separated simultaneous source acquisition, the goal is preserving long-offset primaries while removing interference. This wavelet denoising flow consists of a linear transformation and filtering using the complex wavelet transform (CWT). For reflection signals, normal moveout (NMO) is used. NMO transforms the low-velocity surface waves and the interference noise to where it is easily identified and rejected with a dip filter in the multidimensional CWT domain. Land field data examples have demonstrated significantly improved S/Ns and low-frequency signal preservation in migrated images after wavelet denoising. Since the numerical implementation of the CWT is as fast as a fast Fourier transform, this flow is able to suppress noise and interference simultaneously on the 3D land data much faster than the other inversion methods.

Dip-separated structural filtering using seislet transform and adaptive empirical mode decomposition based dip filter

Dip-separated structural filtering using seislet transform and adaptive empirical mode decomposition based dip filter

Interference Removal for Autofocusing of GPR Data From RC Bridge Decks

Strong interferences can substantially deteriorate the performance of autofocusing techniques in ground penetrating radar (GPR) data analysis for reinforced concrete (RC) bridge deck evaluation. In this paper, we propose a new approach based on F–K dip filtering to remove interferences, including direct waves and cross rebar reflections. Different from previously related work, the surface roughness and uneven cross rebar reflections resulted from GPR signals propagating through rough surface and inhomogeneous medium are carefully modeled as reflectors with small dipping angles, and hence dip relaxations are introduced in the F–K dip filtering to remove strong interferences. The mechanism of the F–K filter without dip relaxation and the correctness of the rebar modeling are first illustrated using simulation data. Then its effectiveness with adequate dip relaxation is demonstrated by both experimental and field data. Additionally, the performance of the F–K dip filter with adequate dip relaxation is also compared with that of the background subtraction (BS) method that is commonly used for direct wave removal in GPR data processing for RC bridge deck evaluation.

Random noise attenuation byf-xempirical-mode decomposition predictive filtering

Random noise attenuation always played an important role in seismic data processing. One of the most widely used methods for suppressing random noise was [Formula: see text] predictive filtering. When the subsurface structure becomes complex, this method suffered from higher prediction errors owing to the large number of different dip components that need to be predicted. We developed a novel denoising method termed [Formula: see text] empirical-mode decomposition (EMD) predictive filtering. This new scheme solved the problem that makes [Formula: see text] EMD ineffective with complex seismic data. Also, by making the prediction more precise, the new scheme removed the limitation of conventional [Formula: see text] predictive filtering when dealing with multidip seismic profiles. In this new method, we first applied EMD to each frequency slice in the [Formula: see text] domain and obtained several intrinsic mode functions (IMFs). Then, an autoregressive model was applied to the sum of the first few IMFs, which contained the high-dip-angle components, to predict the useful steeper events. Finally, the predicted events were added to the sum of the remaining IMFs. This process improved the prediction precision by using an EMD-based dip filter to reduce the dip components before [Formula: see text] predictive filtering. Synthetic and real data sets demonstrated the performance of our proposed method in preserving more useful energy.

Subsalt imaging by target-oriented inversion-based imaging:A 3D field-data example

Reflectivity images obtained by prestack depth migration are often distorted by uneven subsurface illumination, especially in areas with complex geology, such as subsalt regions. We address the problem of uneven illumination in subsalt imaging by posing the reflectivity-imaging problem as a linear inverse problem and solving it in the image domain in a target-oriented fashion. The most computationally intensive part of the image-domain inversion is the explicit computation of the so-called Hessian operator. The Hessian is defined to be the normal operator of the associated modeling/imaging operator, which is a direct measure of the illumination deficiency of the imaging system. We can overcome the cost issue by using the phase-encoding technique in the 3D conical-wave domain for marine streamer acquisitions. We apply the inversion-based imaging methodology to a 3D field data set acquired from the Gulf of Mexico, and we precondition the inversion with nonstationary dip filters, which naturally incorporate interpreted geologic information. Numerical examples demonstrate that imaging by regularized inversion successfully recovers the reflectivities from the effects of uneven illumination, yielding images with more balanced amplitudes and higher spatial resolution.

Seismic Imaging of the Waltham Canyon Fault, California: Comparison of Ray‐Theoretical and Fresnel Volume Prestack Depth Migration

Abstract Near‐vertical faults can be imaged using reflected refractions identified in controlled‐source seismic data. Often theses phases are observed on a few neighboring shot or receiver gathers, resulting in a low‐fold data set. Imaging can be carried out with Kirchhoff prestack depth migration in which migration noise is suppressed by constructive stacking of large amounts of multifold data. Fresnel volume migration can be used for low‐fold data without severe migration noise, as the smearing along isochrones is limited to the first Fresnel zone around the reflection point. We developed a modified Fresnel volume migration technique to enhance imaging of steep faults and to suppress noise and undesired coherent phases. The modifications include target‐oriented filters to separate reflected refractions from steep‐dipping faults and reflections with hyperbolic moveout. Undesired phases like multiple reflections, mode conversions, direct P and S waves, and surface waves are suppressed by these filters. As an alternative approach, we developed a new prestack line‐drawing migration method, which can be considered as a proxy to an infinite frequency approximation of the Fresnel volume migration. The line‐drawing migration is not considering waveform information but requires significantly shorter computational time. Target‐oriented filters were extended by dip filters in the line‐drawing migration method. The migration methods were tested with synthetic data and applied to real data from the Waltham Canyon fault, California. The two techniques are applied best in combination, to design filters and to generate complementary images of steep faults.

3D imaging challenges in steeply dipping mining structures: New lights on acquisition geometry and processing from the Brunswick no. 6 seismic data, Canada

We have analyzed and processed a [Formula: see text] nonorthogonal 3D surface reflection seismic data in the Brunswick no. 6 area to better understand the effect of acquisition geometry on the resultant image and to provide 3D information about the main geologic structures hosting the mineralization. The 3D data were processed using a conventional prestack dip moveout (DMO) and poststack migration algorithm with special focus on refraction static corrections, velocity analysis, and DMO corrections that are important for the data recorded in crystalline environment. However, the nonorthogonal nature of the 3D data combined with its narrow-azimuth, irregular offset distributions, and 2D nature of midpoint distribution in common depth point bins resulted in a lower quality seismic image than those observed on a series of 2D seismic profiles collected in the area prior to the 3D data acquisition. 2D wavenumber spectrum of the data suggests acquisition footprint associated with the data. Most of the noise associated with the acquisition footprint manifested itself as short-length, high-amplitude shallow reflections but largely were attenuated using a dip filter running in the wavenumber domain. Various bin size and geometries were tested, and the best result was obtained using rectangular bins aligned in the orientation of the shot lines. The processing results indicated that the highly prospective and mineralized Brunswick horizon is part of a continuous reflective package that could guide future deep mineral exploration in this mining camp.