Seismic Dataset Research Articles

Subsoil density field reconstruction through 3-D FWI: a systematic comparison between vertical- and horizontal-force seismic sources

SUMMARY Bulk-density (ρ) of soil is an important indicator of soil compaction and type. A knowledge of the spatial variability of in situ soil density is important in geotechnical engineering, hydrology and agriculture. Surface geophysical methods have so far shown limited success in providing an accurate and high-resolution image of 3-D soil-density distribution. In this pursuit, 3-D seismic full-waveform inversion (FWI) is promising, provided the robustness and accuracy of density inversion via this approach can be established in the near-surface scale. However, simultaneous reconstruction of ρ and seismic wave velocities through multiparameter FWI remains a challenging task. Near-surface seismic data are commonly dominated by dispersive surface waves whose velocities are controlled by the value and distribution of shear-wave velocity (VS). One major difficulty in estimating reliably ρ from near-surface seismic data is due to the relatively low sensitivity of the seismic wavefield to ρ compared to seismic velocities. Additionally, the accuracy of the estimated ρ decreases due to error in VS—an issue known as parameter coupling. Parameter coupling makes it difficult to estimate accurately ρ within the framework of conventional gradient-based FWI. More sophisticated optimization approaches (e.g. truncated Newton) can reduce the effect of parameter coupling, but these approaches are commonly not affordable in near-surface applications due to heavy computational burden. In this research, we have investigated how choosing correctly the force direction of the seismic source can contribute to a higher accuracy of ρ estimates through 3-D FWI. Using scattered wavefields, the Hessian, and inversion tests, an in-depth and systematic investigation of data sets corresponding to different force directions has been carried out. A comparison of the scattered wavefields due to a point-localized ρ perturbation for different force directions shows the robustness of the horizontal-force data set to noise compared to the vertical-force data set. Furthermore, for a point-scatterer model, an analysis of the gradients of the misfit function using the Hessian shows that utilizing a horizontal-force source enables one to reconstruct the high-resolution gradient with relatively small parameter coupling. Finally, inversion tests for two different subsoil models demonstrate that 3-D FWI on a horizontal-force-source seismic data set is capable of providing a more accurate 3-D ρ distribution in soil compared to a vertical-force-source data set. Our results show that the use of a horizontal-force source might allow avoiding computationally demanding, costly optimization approaches in 3-D FWI.

Curvature analysis and its correlation with faults and fractures in presalt carbonates, Santos Basin, Brazil

Presalt lacustrine carbonate reservoirs in the Santos Basin, Brazil, can have a high degree of fracturing. Properly characterizing such properties is important due to their influence on the porosity–permeability system. It may play a key role in the so-called excess permeability features and the flow behaviour, especially in the Aptian Alagoas Stage. Therefore, 3D characterization of the fracture system improves the reservoir model. For this purpose, correctly identifying the fault and fracture zones in seismic data is crucial. This task is generally performed manually and guided by amplitude interpretation or geometric seismic attributes. However, due to the seismic resolution of the pre-salt seismic datasets, many faults and fractured zones are not identified in the seismic amplitude domain or by standard geometric attributes such as coherence. Improving the resolution of seismic data and its geometric attributes can positively impact the delineation of fractured zones, structural lineaments, and their connectivity. Curvature (k) attributes have a high potential to reveal vertical to subvertical subseismic-scale structures, which is the main type of strain generated from the tectonic style in the Pre-salt rocks. Among the curvature attributes, the most-positive (k1) and the most-negative (k2) attributes are some of the most commonly used. The present study evaluates the use of seismic inverse Q-filter and three different dip estimation techniques based on Fast Fourier Transform (FFT), Principal Component Analysis (PCA), and Phase Gradient (PG). It also adds two postfiltering techniques (median and recursive Gaussian filters) to obtain a consistent most-positive curvature. We then analyse two analytical models that correlate the volumetric fracture intensity, measured on Borehole Image Logs (BHI), with the curvature. Then, we perform a comprehensive study to understand the role of seismic-scale strain (translated into curvature) and mechanical stratigraphy in the development of fractures in the presalt, where we address that in addition to seismic-scale deformation, the contents of silica, clay, and organic matter are also relevant variables. These results can help to guide future studies of fracture characterization to prospect naturally fractured reservoirs or to estimate fracture intensity for dual-porosity reservoir modelling on the Santos presalt carbonates.

Examining Infrasound Propagation at High Spatial Resolution Using a Nodal Seismic Array

AbstractInfrasound—acoustic waves in the atmosphere below 20 Hz—is a useful monitoring tool. Topography and atmospheric structure strongly control infrasound propagation, and at common source–receiver distances neither of these effects can be ignored when quantitative source constraints are sought. Detailed spatial measurements of the infrasound wavefield would inform propagation models and improve source estimates. However, the “large‐N” deployment strategy now well‐known in seismology has not yet been realized for infrasound studies. Here, we use the 900‐node seismic array from the 2014 Imaging Magma Under St. Helens (iMUSH) experiment as a proxy for a large‐N infrasound network, by leveraging acoustic–seismic coupled arrivals. The active‐source component of iMUSH consisted of 23 shallowly buried explosions around Mount Saint Helens volcano; these explosions produced epicentral infrasound recorded on the nodes. We find that the bulk presence of ground‐coupled infrasound on the nodes is controlled by wind noise and source–receiver distance, with observed arrivals for eight explosions. Explosions with the most extensive coupling produce complex spatial waveform patterns across the array. These patterns are related to both topographic and atmospheric propagation effects, as well as spatially variable site (coupling) effects. We compare our observations to simple topographic diffraction and high‐resolution wind advection models, and full‐wave numerical simulations. We find strong spatial correlations between (a) coupled arrival strength and modeled topographic obstruction and (b) coupled arrival time and along‐path winds. Our seismoacoustic analyses and results are applicable to other existing and future nodal seismic data sets and can expand the utility of such deployments.

Full waveform inversion with combined misfit functions and application in land seismic data

Full waveform inversion reconstructs subsurface structures by matching the synthetic waveform to the observed waveform. Inaccuracy of the source wavelets can, thus, easily lead to an inaccurate model. Simultaneously updating source wavelets and model parameters is a conventionally used strategy. However, when the initial model is very far from the true model, cycle skipping exists, and estimating a reliable source wavelet is very difficult. We propose a combinatory inversion workflow based on seismic events. We apply a Gaussian time window around the first break and gradually increase its width to include more seismic events. The influence of inaccurate source wavelets is alleviated by applying a Gaussian time window around the first break to evaluate the normalized cross-correlation-based objective function. There are inevitable small model artifacts caused by inter-event interactions when calculating cross-correlations. As a result, we switch to the optimal transport function to clean the model and update the source wavelets simultaneously. The combinatory strategy has been applied to models with different types of geological structures. Starting from a crude initial model, we recovered a high-resolution and high-fidelity model and the source wavelets in two synthetic experiments. Finally, we apply our inversion strategy to a real-land seismic dataset in Southeast China and obtain a higher-resolution velocity model. By comparing an inversion velocity profile with well log information and the recorded data with the simulated data, we conclude that our inversion results for the field data are accurate and this new strategy is effective.

Autogenic and allogenic controls on turbidite lobe stacking pattern and architecture: The case of the Turonian to Coniacian deep‐water turbidite lobe complexes (Dønna Terrace, offshore Mid‐Norway)

ABSTRACTThe mechanisms controlling the stacking patterns of deep‐water turbidite lobes are currently open to a wide range of interpretations. A study of Turonian to Coniacian turbidite lobe complexes in the greater Marulk area (Dønna Terrace, Norwegian Sea) was undertaken to examine the balance and respective influences of various controlling factors using a large sediment core, well‐log and seismic dataset. A four‐tiered lobe hierarchy is described based on a detailed sedimentological study of three Cretaceous turbidite lobe systems, involving a variety of sedimentary processes and flow regimes. The inferred depositional stacking patterns were then used to identify autogenic and allogenic forcings on the large‐scale depositional architecture of turbidite lobes. Autogenic processes (best observed in core data) control the self‐regulation of sediment dispersal and the broad evolution of lobe sub‐environments. Conversely, allogenic forcings (best observed in well‐log data) regulate axial migration within the turbidite lobe succession, and control sediment pulses and ultimately the general evolutionary trend of the turbidite lobe complex. Beyond this, an updated approach is proposed here aiming at deciphering autogenic‐dominant and allogenic‐dominant processes at different hierarchical levels in both confined and unconfined turbidite lobe deposits, which may help with assigning appropriate inputs for geomodels for an improved understanding of the internal and external controls on lobe depositional architecture. It is advocated that this approach may eventually be used in other depositional systems, such as in deltaic complexes from coastal settings, both in actual and ancient sediments.

The overall-subshear and multi-segment rupture of the 2023 Mw7.8 Kahramanmaraş, Turkey earthquake in millennia supercycle

On February 6, 2023, an Mw7.8 earthquake hit the East Anatolian Fault (EAF) and Narlı Fault (NF), followed by an Mw7.5 event on the Sürgü Fault. We combine multiple seismic datasets, global navigation satellite system recordings, and radar satellite images with finite fault inversion and slowness enhanced back-projection to study the rupture kinematics. Our analysis reveals that the rupture originated on the NF, propagating 120 km northeast at 3.05 km/s and 200 km southwest at 3.11 km/s after reaching the EAF junction, exhibiting overall subshear speeds. Further Mach wave analysis confirms the subshear rupture, matching the prediction using close-Rayleigh speeds. The unexpectedly-large slip on some EAF segments suggests a supercycle lasting ≥900 years. The EAF geometry is similar to the San Andreas-San Jacinto Fault system, while the latter has higher slip rates but without large earthquakes on its southern segments since 1857, carrying the potential of an M8 earthquake.

Automated picking of seismic first arrivals using a single- to multidomain self-trained network

Over the past few decades, seismic data volume has increased owing to the evolution of survey acquisition designs that comprise numerous receivers and shotpoints. Consequently, the demand for efficient algorithms to detect seismic first arrivals has been increasing. Machine-learning methods, particularly convolutional neural networks, have shown great potential in achieving this. However, most existing methods lack generalization abilities or are too sensitive to noise. We develop a new picking strategy based on a single- to multidomain self-trained network to achieve automated picking of first arrivals. Our self-trained network learns from single to multiple domains while rejecting the incorrectly predicted picks and assigning pseudo-first-arrival times to the unlabeled data set. The model inherits the advantages of single- and multidomain picking strategies and has high generalizability and noise robustness. In experiments on an open hard-rock seismic multisurvey data set, our picking method outperforms the benchmark for most evaluation metrics. Remarkably, our approach achieves an accuracy improvement of approximately 10% against the data from one mine site (Lalor), with a sampling rate that differed from the training data. In addition, our method indicates robust results considering the limited amount of labeled data. The experimental results demonstrate the high generalizability and robustness of our method for unseen data sets and reveals the effectiveness of single- to multidomain self-trained networks to improve the accuracy and stability of automated first-arrival picking. Our method contributes to reducing human labor and the total seismic processing time.

Bottom simulating reflectors in the Manila Trench forearc and its implications on the occurrence of gas hydrates in the region

Gas hydrate occurrence and stability in active plate margins have been reported in various localities, with investigations focusing on its potential as an alternative energy resource or the threat it poses due to dissociation and methane release into the atmosphere. The Manila Trench forearc, proximal to active margins and probable methane-rich sediments, can be an analog for gas hydrate formation and geologic preconditions constrained by tectonics and sedimentation processes. The possible occurrence of gas hydrates in offshore western Luzon Island is reported for the first time based on bottom simulating reflectors (BSRs) in multi-channel seismic reflection datasets. Results indicate a prevalence of BSRs on the frontal wedge seaward of the North Luzon Trough (NLT) and in the West Luzon Trough (WLT) basin fill. Continuous, discontinuous, double, and pluming BSRs in the forearc encompass a total area of approximately 15,400 km2 at depths of 223–553 mbsf in the frontal wedge and 486 to 1428 mbsf within the basins. Enhanced amplitude reflectors (EARs) above and below the BSRs indicate possible gas hydrate and free gas accumulations, respectively. Gas chimneys associated with the upward migration of methane-rich fluids appear to be controlled by deep structural features. The accretion of continent-derived sediments along the northern segment of the frontal wedge promotes the formation of fluid migration structures, enhancing fluid migration and gas hydrate formation at shallower depths. Estimated geothermal gradient values for the frontal wedge range from 28 to 92 °C/km, consistent with previously reported in-situ measurements from offshore SW Taiwan. In contrast, arc-derived sediments within the NLT and west-ward fluid migration along the landward-tilted sequences limit gas hydrate accumulation within the basin. To the south, the preferential formation of gas hydrates within the WLT basin points to deep gas sources within the forearc basin sediments, with upward movement of methane-rich fluids along normal faults. Estimated geothermal gradient values in the basin are lower, ranging from 12 to 45 °C/km, reflecting lower thermal conditions. The absence of BSRs seaward of the WLT signifies widespread gas hydrate dissociation in the frontal wedge slope.

Zones of Weakness Within the Juan de Fuca Plate Mapped From the Integration of Multiple Geophysical Data and Their Relation to Observed Seismicity

AbstractThis study aims to explain the nonuniform earthquake pattern along the Cascadia Subduction Zone. In particular, we investigate the relationship between the tectonic features of the subducting oceanic Juan de Fuca slab and the onshore seismicity pattern. We have integrated multiple geophysical data sets toward three general objectives. The first study intends to study variations in physical properties along three 2‐dimensional models through regions of different seismicities that combine public gravity, magnetic, and seismic data sets. These models reveal multiple zones of decreased crustal density that we interpret as regions of weaker oceanic crust. The second objective is to delineate major tectonic features by performing spatial analysis of potential fields. The overall methodology comprises gravity and magnetic data filtering, followed by lineaments mapping and cross‐referencing interpretation with available seismic reflection data. This process allows delineating zones of crustal weakness by extrapolating outside our three 2‐D models. We also map multiple seamounts that appear to cluster along identified zones of weaker crust. Third, we investigate the relationship between the mapped tectonic elements, namely the zones of weak crust with accompanying seamounts, and the observed seismicity trends within the subducted slab. The alignment between those suggests that mapped weak crust zones and associated seamounts may have an influence on the overall subduction process. As more of these structures are heading toward the Washington portion of the margin than to the Oregon portion, more earthquakes are observed in the north than in the south.

Spatio‐Temporal Variations in Sediment Delivery as a Response to Rapid Quaternary Climate Change in the Lake Malawi Rift, East Africa

AbstractThe interplay of rapid climate change and tectonics drives landscape development, sediment routing, and deposition in early‐stage continental rift systems. The Lake Malawi Rift, in the Western Branch of the East African Rift, is an archetype of a juvenile rift and an ideal natural laboratory for evaluating lacustrine source‐to‐sink systems on orbital or shorter timescales. We examine the interplay of these processes over the past 140 kyr using observations from nested seismic reflection data sets tied to scientific drill cores, which calibrate numerical forward models of this closed sedimentary system. Fault slip rates measured from seismic data drive tectonic displacements in the model. Satellite‐derived precipitation maps constrain modern precipitation and are scaled to previous hydrologic balance studies to reconstruct past climates. Our model reproduces known sediment thicknesses across the rift and accounts for 96% of the estimated siliciclastic sediment deposited over the past 140 kyr. The results demonstrate that the onset of arid climate conditions (140–95 kyr BP) causes extreme drainage adjustments downstream and the formation of mega‐catchments that flow axially into a shallow restricted paleo‐lake. Sedimentation rates during this time are twice the present values due to increased sediment focusing via these axial systems into a much smaller, hydrologically closed lake. As the climate became wetter (95–50 kyr BP), the lake rapidly expanded, decreasing both erosion and sedimentation rates across the rift. This closed‐loop approach allows us to evaluate the role of high‐frequency climate change in modulating basin physiography as well as sediment fluxes in juvenile rift systems.

Dictionary Learning for Single-Channel Passive Seismic Denoising

Abstract Passive seismic denoising is mostly performed using a simple band-pass filter, which can be problematic when signal and noise share the same frequency band. More advanced passive seismic denoising methods take advantage of fixed-basis transforms, for example, the wavelet, to remove noise. Here, we present an open-source package for data-driven denoising based on adaptively learning sparse transform. Contrary to the fixed-basis transforms, the proposed method belongs to the adaptive-basis transforms. We learn the 1D features embedded in the passive seismic data from all the available waveform data sets without requiring spatial coherency in a data-driven way. Thus, the new method is flexible to apply in any passive seismic monitoring project because of its data-driven and single-channel nature when implemented. Considering the computationally expensive K-singular-value-decomposition (KSVD) in the traditional dictionary learning framework, we suggest applying a fast SVD-free dictionary learning method that can be readily applicable to process massive seismic data during passive seismic monitoring. The proposed method is applied to two synthetic data examples and three real passive seismic data sets to demonstrate its effectiveness in improving the signal-to-noise ratio, and its potential in applications like arrival picking. The open-source reproducible package can be found in the Data and Resources section.

The Middle and Lower Cambrian salt tectonics in the central Tarim Basin, China: A case study based on strike-slip fault characterization

Due to the considerable depth of the salt layers and the lack of calibration by exploratory drilling, the interpretation of the Middle and Lower Cambrian salt formations in the central Tarim Basin poses a challenge. In this paper, we apply the coupling and decoupling deformation theory in salt tectonics to analyze the No.7 fault mapped in the seismic datasets by the response characteristics of the Middle and Lower Cambrian layers. By quantifying the stratigraphic framework of the Middle and Lower Cambrian strata, we define the position of the salt layer with the seismic data. Structural decoupling is observed in the Middle and Lower Cambrian sequences in the Shuntuoguole Low Uplift, while deformation coupling is observed in these two sequences in the Shaya Uplift.

A 2D seismic reflection dataset of the Caosiyao giant porphyry Mo deposit in the shallow coverage area in Jining, Inner Mongolia, China

AbstractProspecting for and exploiting buried mineral deposits is currently challenging. Given their high precision and resolution, reflection seismic methods might be useful in such applications involving deep mineral deposits. However, there are few open seismic datasets available from mineral deposit exploration, especially in hardrock environments. The world‐class Caosiyao porphyry Molybdenum deposit (1.76 Mt) in the Jining area of Inner Mongolia, China, is largely covered by loess layers, which poses challenges to its exploration. Seismic reflection surveys were conducted to help delineate the deep granite porphyry intrusions and associated orebodies. This paper presents the raw seismic reflection dataset from three profiles on the Caosiyao deposit area, which can be used as a standard dataset for reflection seismic processing in shallow coverage and hardrock areas. Situated at the juxtaposition of the Khondalite Belt and the Trans‐North China Orogen in the northern North China Craton, the Jining region hosts considerable known porphyry Mo deposits. As such, this open dataset can assist in research on deep geological structures and hence increase prospecting efficiency in geologically similar areas.

Enhanced seismic data segmentation using an assembled scSE-Res-UNet deep neural network

To enhance the accuracy and robustness of seismic data segmentation, we propose a novel deep-learning network, named scSE-Res-UNet. We select a U-Net as the backbone and augment it with residual blocks to form a Res-UNet, enabling multi-scale feature extraction while reducing edge loss between strata. Furthermore, we introduce Spatial and Channel Squeeze & Excitation (scSE) attention blocks in the decoder to selectively emphasize meaningful features and long-range dependencies in the data. The scSE blocks suppress unimportant features and noise while enhancing spatial context. We apply the proposed scSE-Res-UNet to a public F3 seismic dataset and compare performance against a CNN benchmark model. Experiments demonstrate the proposed architecture achieves superior data segmentation accuracy and robustness. The integration of residual learning and scSE attention mechanisms enhances the network's capacity to extract geologically meaningful representations from the seismic data. This leads to improved delineation of stratigraphic units with similar characters that are challenging to differentiate. The proposed scSE-Res-UNet effectively combines proven deep learning techniques to advance seismic interpretation.

The IDA Very Long Period and Very Broadband Networks

Abstract This article describes the rich global digital seismic data set that has been collected from 1975 to the present using a variety of instruments and progressively updated equipment and communication systems. The original IDA very long period (VLP) network operated from 1975 to 1995. The successor IRIS/IDA very broadband network began in 1986 and continues to the present. Collection of these data has supported advances in our understanding of the structure of the Earth’s deep interior and of the long wavelength heterogeneous structure of the mantle; the detailed behavior of large earthquakes; and the discovery of nontraditional seismic sources such as slow earthquakes, glaciers, and landslides. Near-real-time transmission of the data to hazard warning centers has expanded the network’s usefulness as a tool for rapid response to earthquakes, warning to communities affected by tsunamis, and for monitoring compliance with nuclear test ban treaties.

Deep semi-supervised learning using generative adversarial networks for automated seismic facies classification of mass transport complex

Geological and geophysical interpretation is characterised by large and localised datasets that are extremely expensive to acquire. There are clear advantages in applying deep learning techniques to such datasets, but this requires a large amount of suitable data for effective training. Creation of training data can be time-consuming, but novel countermeasures such as Generative Adversarial Networks (GANs) are an enticing alternative with the potential to alleviate this issue by providing synthetic data that is representative of the real data. This paper details an investigation of the potential of GANs for the augmentation of labelled seismic facies from a mass transport complex in training a convolutional neural network (CNN) for facies classification. The study adopts a specific GAN approach, known as the Conditional Style-Based Logo GAN (LoGANv2), because of its capability to generate conditioned output with improved training stability. By using LoGANv2 to synthesise examples that mimic the behaviour of the real data, based on a 3D seismic dataset from the North Carnarvon Basin in Australia, the accuracy of the traditional CNN for facies classification improved from 38% (the benchmark result where no augmentation was applied) to 52%, 61%, 94% and 74% for four trials with different degrees of augmentation. The results show that increasing training size, either through manual annotation (trial 2) or standard manipulations such as image rotation (trial 3 and trial 4), improves the performance of LoGANv2 and hence the CNN classification. The experiments indicate that although standard manipulation can increase the diversity of the training dataset, the balance between diversity and consistency within the datasets is also important, and this should be optimized for LoGANv2 to achieve better performance. In addition, we explored the adaptability of the proposed LoGANv2-CNN approach to unlabelled samples from another survey to demonstrate the robustness and flexibility of the model.

Heterogeneous Strain Distribution in the Malawi (Nyasa) Rift, East Africa: Implications for Rifting in Magma‐Poor, Multi‐Segment Rift Systems

AbstractHalf‐graben basins bounded by border faults typify early‐stage continental rifts. Deciphering the role that intra‐rift faults play in rift basin development is challenging as patterns of early‐stage faulting are commonly overprinted by subsequent deformation; yet the characterization of these faults is crucial to understand the fundamental controls on their evolution, their contribution to rift opening, and to assess their seismic hazard. By integrating multiple offshore seismic reflection data sets with age‐dated drill core, late‐Quaternary and cumulative faulting patterns are characterized in the Central and South Basins of the Malawi (Nyasa) Rift, an active, early‐stage rift system. Almost all intra‐rift faults offset a late‐Quaternary lake lowstand surface, suggesting they are active and should be considered in hazard assessments. Fault throw profiles reveal sawtooth patterns indicating segmented slip histories. Observed extension on intra‐rift faults is approximately twice that predicted from hanging wall flexure of the border fault, suggesting that intra‐rift faults accommodate a proportion of the regional extension. Cumulative and late‐Quaternary throws on intra‐rift faults are correlated with throw measured on the border fault in the Central Basin, whereas an anticorrelation is observed in the South Basin. Viewed in a regional context, these differences do not relate solely to the proposed southward younging of the rift. Instead, it is inferred that the distribution of extension is also influenced by variations in lithospheric structure and crustal heterogeneities that are documented along the rift axis.

Reprocessing of Legacy Seismic Reflection Profile Data and Its Implications for Plate Flexure in the Vicinity of the Hawaiian Islands

AbstractDuring 1975–1988, an academic research ship, R/V Robert D. Conrad, acquired more than 150,000‐line‐km of multichannel seismic reflection profile data from each of the world's main ocean basins and their margins. This extensive legacy seismic data set, which involved both single ship and two‐ship data acquisition, has been widely used by the marine geoscience community. We report on our experience in reprocessing seismic reflection profile data acquired during Conrad cruise RC2308 to the Hawaiian Islands region in August/September 1982. We show that the application of modern, industry standard processing techniques, including filtering, de‐bubble, deconvolution, and migration, can significantly enhance 40+ year old legacy seismic reflection profile data. The reprocessed data reveals more precisely, and with much less scatter, the flexure of Cretaceous Pacific oceanic crust caused by the Pliocene‐Recent volcanic loads that comprise the Hawaiian Islands. A comparison of observed picks of top oceanic crust which has been corrected for the Hawaiian swell and the Molokai Fracture Zone with the calculations of a simple 3‐dimensional elastic plate (flexure) model reveals a best fit elastic plate thickness of the lithosphere, Te, of 26.7 km, an average infill density of 2,701 kg m−3, and a Root Mean Square difference between observations and calculations of 305 m. Tests show these results depend weakly on the load density assumed and that the average infill density is close to what would be predicted from an arithmetic average of the flanking moat infill density and the infill density that immediately underlies the volcanic edifice.

Seismic Image Identification and Detection Based on Tchebichef Moment Invariant

The research focuses on the analysis of seismic data, specifically targeting the detection, edge segmentation, and classification of seismic images. These processes are fundamental in image processing and are crucial in understanding the stratigraphic structure and identifying oil and natural gas resources. However, there is a lack of sufficient resources in the field of seismic image detection, and interpreting 2D seismic image slices based on 3D seismic data sets can be challenging. In this research, image segmentation involves image preprocessing and the use of a U-net network. Preprocessing techniques, such as Gaussian filter and anisotropic diffusion, are employed to reduce blur and noise in seismic images. The U-net network, based on the Canny descriptor is used for segmentation. For image classification, the ResNet-50 and Inception-v3 models are applied to classify different types of seismic images. In image detection, Tchebichef invariants are computed using the Tchebichef polynomials’ recurrence relation. These invariants are then used in an optimized multi-class SVM network for detecting and classifying various types of seismic images. The promising results of the SVM model based on Tchebichef invariants suggest its potential to replace Hu moment invariants (HMIs) and Zernike moment invariants (ZMIs) for seismic image detection. This approach offers a more efficient and dependable solution for seismic image analysis in the future.

Defining the Scale to Build Complex Networks with a 40-Year Norwegian Intraplate Seismicity Dataset.

We present a new complex network-based study focused on intraplate earthquakes recorded in southern Norway during the period 1980-2020. One of the most recognized limitations of spatial complex network procedures and analyses concerns the definition of adequate cell size, which is the focus of this approach. In the present study, we analyze the influence of observational errors of hypocentral and epicentral locations of seismic events in the construction of a complex network, looking for the best cell size to build it and to develop a basis for interpreting the results in terms of the structure of the complex network in this seismic region. We focus the analysis on the degree distribution of the complex networks. We observed a strong result of the cell size for the slope of the degree distribution of the nodes, called the critical exponent γ. Based on the Abe-Suzuki method, the slope (γ) showed a negligible variation between the construction of 3- and 2-dimensional complex networks. The results were also very similar for a complex network built with subsets of seismic events. These results suggest a weak influence of observational errors measured for the coordinates latitude, longitude, and depth in the outcomes obtained with this particular methodology and for this high-quality dataset. These results imply stable behavior of the complex network, which shows a structure of hubs for small values of the cell size and a more homogeneous degree distribution when the cell size increases. In all the analyses, the γ parameter showed smaller values of the error bars for greater values of the cell size. To keep the structure of hubs and small error bars, a better range of the side sizes was determined to be between 8 to 16 km. From now on, these values can be used as the most stable cell sizes to perform any kind of study concerning complex network studies in southern Norway.