Exploration Geophysics - Volume 54, Issue 1, 2023

The magnitude versus offset (MVO) curve, a type of frequency domain marine controlled-source electromagnetic data, is the most common way to identify electromagnetic anomalies in oil and gas reservoirs. However, in actual exploration, it can be difficult to identify the boundary of the high resistance anomaly when there are response signals of multiple emission frequencies. Also, the noise would reduce the accuracy of manually detecting electromagnetic anomalies. The robustness of the bidirectional long short-term memory (LSTM) network is relatively strong, and the LSTM neural network would get the most out of the sequence information of the data for feature extraction purposes and to achieve automatic classification and identification. Therefore, this paper proposes a method of using bidirectional LSTM to solve the problem of anomaly identification in marine controlled-source electromagnetic data. The LSTM unit was applied to establish anomaly identification models of single-layer LSTM, two-layer LSTM, and bidirectional LSTM, respectively. In this paper, theoretical data were calculated by a one-dimensional uniform layered medium model, and the synthetic noise data were constructed by adding random noise with different signal-to-noise ratios. The three types of models were trained, verified, and tested, respectively, to compare the accuracy of electromagnetic anomaly identification. According to the comparison, a conclusion can be drawn that the bidirectional LSTM model suggests the best manifestation of learning the characteristics of the sample. Its electromagnetic anomaly identification accuracy reached 100% in the theoretical dataset and 79.58% in the synthetic noise dataset.

The results of the geophysical surveys carried out in the city of Ugento, specifically in the St. Antonio area and into the Medieval Castle, are presented in this paper. The aim of this study is to integrate the results of Ground-Penetrating Radar and Electrical Resistivity Tomography surveys in order to support archaeological investigations aimed at the elaboration of a detailed archaeological map of the Messapian, Roman and Medieval settlement. Indeed, the obtained data were jointly analysed with archaeological data already known and were georeferenced on the general plans of the site in order to obtain an overall view of the anomalies detected by geophysical instrumentation and related to buried ancient structures. In particular, the geophysical surveys carried out in the St. Antonio area made it possible to reconstruct a stretch of the Messapian city walls and nearby necropolis, while the investigations inside the Castle allowed for the acquisition of very interesting data about previous phases dated to the Middle Ages (and maybe also the Messapian or Roman ages) and about the geo-morphological characteristics of the site.

Karst cave is a sort of special and buried geological structure that was widely developed in the Permo-Carboniferous coal accumulation area of North China. It brings karst collapse and safety hazard in the mining industry. In this study, we propose a seismic structure-constrained inversion of controlled source audio-frequency magnetotelluric (CSAMT) data on a detailed survey and detection of karst caves. Instead of constrained by seismic impedance, the method in this study directly takes the seismic imaging results as structural constraints, which is different from the cross-gradient technique used by conventional structural constraints. First, the seismic migration section is divided according to the CSAMT inversion grid and applied pixel extraction for each grid. Clustering is carried out according to the structural information interpreted by the seismic migration section and the average pixel value of each cluster is calculated. Then the clustered results were used in the seismic structure-constrained inversion of CSAMT data based on cross-gradient technique. After that, as a karst cave model developed in limestone was established, the study compares the structure-constrained inversions with different clustering strategies shows a much more precision of karst cave detection than the method only applies CSAMT data. Moreover, experimental verification is provided in this study, which is for the detection of a suspected karst collapse column from Shandong Province, China. The comparison results further show that the structure-constrained inversion method proposed in this paper is applicable and may effectively improve the locating accuracy of karst caves.

This paper presents a new algorithm to generate edge detection images using a migrated image. The key idea is that discontinuity response are comprised of all dip angle ranges and thus after F-K filtering, they can exist in both images with negative-only and positive-only dips. However, reflection responses can only exist in either image with negative-only dips or image with positive-only dips. Applying multiplication imaging condition (MIC) to two images with opposite dips can generate an edge detection image. This method can be easily extended to 3D cases. More specifically, two strategies are proposed. The first one involves applying a 3D F-K filter to calculate four components with different dip and azimuth angles, then multiplying the decomposed four components to generate the final edge detection image. The other one consists of splitting the 3D data volume into 2D sections and then applying the 2D methodology to generate the edge detection image. Both synthetic and real data tests will be employed in 2D and 3D to demonstrate the validity of edge detection using MIC.

Currently, the transmitting channel wave technique mainly uses the attenuation coefficient of the channel wave total energy to explore the geological structure of the coal seam face. In the case of weak geophone coupling and intense geological anomaly, the channel wave energy will be attenuated severely, which significantly affects the stability and accuracy of the result. The Q value of the coal channel is a critical parameter to evaluate the energy attenuation characteristics of the channel wave. The Q value is typically estimated by using the attenuation coefficient of the body wave, but the special coal channel model hinders the estimation of the Q value of the coal seam. According to the linear attenuation characteristics of the centroid frequency of the transmitting channel wave, a new method was proposed to assess the quality factor (Q) of the coal channel by using the centroid frequency change of the channel wave signal. The expected frequency was calculated as its centroid frequency according to the energy ratio of each frequency point through the spectral analysis of the channel wave signal. Combined with the transmission tomography technology, the imaging of the coal seam face based on the transmitting channel wave Q value was established. According to the sudden change of the Q value of the coal channel near the geological structure of the coal seam face, a geological interpretation method based on the abnormal Q value was proposed. The two-dimensional numerical simulation demonstrated that the centroid frequency of the transmitting channel wave signal decayed linearly with the propagation distance and the geological structure increased the frequency shift. Furthermore, three-dimensional numerical simulation validated the feasibility and effectiveness of the Q value inversion method. Field Experimental results showed that the algorithm exhibited improved stability and accuracy. This work proposed a novel frequency-domain inversion method of the transmitting channel wave that directly uses the frequency shift characteristics of the channel wave to estimate the Q value of the coal channel, which offers a new strategy in the data processing of channel wave.

Empirical mode decomposition (EMD) is an effective tool for signal analysis that splits the data into individual modes, called the intrinsic mode functions, which are associated with symmetric and narrow-band waveform ensuring that the instantaneous frequencies are smooth and positive. However, some negative features encumber its direct application namely the mode mixing and splitting, aliasing and endpoint artefacts. Two variants, ensemble EMD (EEMD) and complete ensemble EMD (CEEMD) have been recently introduced to overcome these problems. We intend to show the application of the EMD for demarcating the zones of gas-hydrates and free-gas bearing sediments. Gas-hydrates are ice-like crystalline substances that occur in shallow sediments along the outer continental margins and in the permafrost regions, and are considered as viable major future energy resources of the world. Gas-hydrates in marine environment are generally identified by an anomalous reflector, known as the bottom simulating reflector, on seismic section. The present study demonstrates that the EMD can be effectively utilised in demarcating the zones of gas-hydrates and free-gas bearing sediments with a field example in the Mahanadi basin of the eastern Indian margin.

Mafic igneous units within sedimentary basins can be widespread and severely attenuate seismic reflection data. Attenuation obscures imaging of sub-igneous sedimentary units, impeding exploration in prospective and frontier basins. This study compared historical 2D seismic surveys and found two seismic acquisition parameters that have the greatest influence when imaging beneath mafic igneous rocks in offshore and onshore basins from Australia’s Northwest Shelf. These parameters were found by using a 3D model developed with integrated 2D seismic and well data in the Browse, North Carnarvon, Onshore and Offshore Canning basins. Simultaneously comparing the 2D seismic lines in 3D space revealed that the surveys with the longest, streamer length and the most receivers are the most effective at imaging beneath igneous units. Additionally, we identified potential depocenters obscured by igneous horizons from a regional basement map. These depocenters correlate with older basins that are infilled by pre-rift, Paleozoic sediment and capped by mafic igneous rocks formed during late Permian-Mesozoic rifting events. Much of the Northwest Shelf maintains a frontier status, but exploration outcomes can be improved. Therefore, maximising streamer length and number of receivers to future seismic surveys can result in more effective exploration opportunities in basins with known igneous occurrences.