Main Objectives

Description of interactions during PM-CAES and an approach for coupled PM-CAES simulations in a future energy system

New Aspects

Coupled Power-Plant & Geostorage simulations for compressed air energy storage in porous formations, Display of worked example of developed simulator using homogeneous reservoir models

Summary

Compressed air energy storage in porous formations (PM-CAES) can provide storage capacity in future energy systems largely relying on renewable power generation. The limiting conditions for a storage plant design are given by the geological setting and the load profile the storage plant has to support. Neither of these conditions is known exactly given that the load profiles depend on interactions within the future energy system and the properties of the geologic subsurface being inherently uncertain. Consequently, multiple scenario simulations must be evaluated to estimate the storage behaviour for a given geological setting.

The power a PM-CAES system can provide or take up is a function of the available mass flow and the pressure. Such systems show a positive feedback mechanism during discharging and a nega-tive during charging, i.e. the required mass flow decreases with increasing pressure for a given power rating. To accurately represent these feedback mechanisms a coupling interface was de-veloped, combining the component based power-plant model TESPy and the reservoir simulator package ECLIPSE. In combination with using homogeneous replacement models derived from a heterogeneous ensemble model the developed simulator coupling can be used to accurately as-sess the behaviour of a realistic PM-CAES under various load conditions.

Main Objectives

Flow Diagnostics

New Aspects

Dual Porosity Flow Diagnostics

Summary

Flow diagnostics can rapidly approximate the reservoir dynamics and can be used for screening ensembles of models prior to production forecasting.

We have applied our recently extended dual porosity flow diagnostics to the Teapot Dome, a naturally fractured reservoir in the US. We have demonstrated using flow diagnostics for ensemble screening and fast optimisation studies.

Main Objectives

Geothermal Storage

New Aspects

Impact of subsurface heterogeneity

Summary

In this paper we evaluate the influence of reservoir and thermal rock property heterogeneity on the efficiency of an aquifer thermal energy storage (ATES) located in a carbonate dominated formation. We designed a series of thermal and fluid flow modelling scenarios to assess the impact of the individual parameters. Simulation results for the recovery efficiency of the ATES system in the carbonate aquifer show that reservoir heterogeneity and thermal rock property variations need to be modelled in detail to accurately forecast subsurface fluid flow, thermal interference and reservoir performance. Optimal storage and well planning depend significantly on the incorporation of comprehensive geological and sedimentological characteristics of the aquifer in the geological and dynamic models. The parametric study of thermal conductivity variation showed that (i) thermal conductivity heterogeneity is an import factor to be considered in ATES modelling and (ii) the entire system needs to be modelled in detail. The latter with respect to the accurate characterisation of flow and non-flow intervals in the reservoir as thermal energy is not only transmitted through advection but also conduction through low-permeable intervals. Furthermore, characterisation and explicit modelling of the confining strata is important to provide a holistic impact assessment of the storage system

Summary

No summary available

Main Objectives

Presentation and illustration of strategy for consistent use of CSEM for prospect evaluation

New Aspects

CSEM strategy and case study; CSEM decision input

Summary

Lundin’s CSEM strategy focuses on using CSEM in well-known areas to assess the volume high-case for a prospect. In line with this strategy, a shallow Prospect in the Barents Sea was evaluated with CSEM. Very robust results were obtained, indicating only a very weak CSEM response across part of the Prospect and contradicting the expectation scenario. This was interpreted as an absence or very limited development of the targeted oil-zone, leading to the Prospect being abandoned due to the reduced volume potential.

Main Objectives

we present an efficient approximation method, with which we simulate the effects of the seabed pipeline via line-segment (volume-less) finite element (FE).

New Aspects

line-segment (volume-less) finite element (FE) for EM wave

Summary

In this abstract, we present an efficient approximation method, with which we simulate the effects of the seabed pipeline via line-segment (volume-less) finite element (FE), and validate the method in comparison to a full 3D FE numerical solution. It is seen that we can determine a certain region in a computational model where the approximation method provides inaccurate results (only very close to pipeline). The response calculated in such regions should be excluded e.g. from CSEM data inversion.

Main Objectives

Demonstrate a cost-effective multi-physics workflow for salt exit velocity retrievalflow

New Aspects

New workflow to enhance salt exit velocity and obtain a clearer subsalt image using multi-physics

Summary

Subsalt velocity retrieval is, in many cases, a difficult task to perform because of complex salt overburden. Anomalous high base of salt amplitudes can often be found, indicating relatively strong impedance contrast at the salt-subsalt boundary. Determining the salt exit velocity is not only important to the quality of subsalt images, but essential for drilling hazard prediction using pore pressure estimation. We present a broadband CSEM-Resistivity-to-Velocity transform workflow, as a cost-effective alternative to expensive wide and full azimuth streamer or OBN acquisition and computer intensive FWI, to further enhance the quality of existing streamer data, improve complex subsalt imaging and accurate retrieval of salt exit velocity spatial variation. We demonstrate this workflow to the Espirito Santo ultra-deep-water area offshore Brazil where a greatly enhanced salt exit velocity and a clearer subsalt image are achieved.

Main Objectives

time-lapse, inversion

New Aspects

double-difference, same inversion code for baseline and time-lapse

Summary

Time-lapse resistivity tomography bring valuable information on the physical changes occurring inside a geological reservoir. In this study, resistivity monitoring from CSEM data is investigated through synthetic and real data. We present three different schemes currently used to perform time-lapse inversions and compare these three methods: parallel, sequential and double difference. We demonstrate on synthetic tests that double difference scheme is the best way to perform time-lapse inversion when the survey parameters are fixed between the different time-lapse acquisitions. We show that double difference inversion allows to remove the imprint of correlated noise distortions, static shifts, and most of the non-linearity of the inversion process including numerical noise and acquisition footprint. It also appears that this approach is robust against the baseline resistivity model quality, and even a rough starting resistivity model built from borehole logs or basic geological knowledge can be sufficient to map the time-lapse changes at their right positions. We perform these comparisons with real land time-lapse CSEM data acquired one year apart over the Reykjanes geothermal field.

Main Objectives

Demonstrate the potential of GPU for 3D CSEM modelling

New Aspects

GPU-based 3D CSEM modelling, 40 times speedup over CPU

Summary

Marine controlled source electromagnetism (mCSEM) has been recognized as a valuable tool for de-risking purpose of reservoir exploration. The kernel of mCSEM inversion is the numerical simulation of 3D electromagnetic wave propagation, which is very cost demanding and inefficient. We have parallelized the finite difference time domain (FDTD) modelling using Nvidia graphics processing unit (GPU), to boost the modelling efficiency. We report 40x speedup over the sequential implementation using single threaded CPU.

Main Objectives

Demonstrate improved quality of subsurface resistivity images and increased confidence in their interpretation from Gauss-Newton EM inversion as compared to quasi-Newton inversion

New Aspects

Gauss-Newton EM inversion implementation; new application case with much-improved results from Gauss-Newton inversion where previous quasi-Newton inversion gave ambiguous images

Summary

We have applied a new Gauss-Newton (GN) inversion code to re-image data that previously could not be interpreted with confidence. The data were acquired over an area where seismic data indicate a basement high about 3 km below an exploration target. Previous extensive field data imaging and scenario tests using a quasi-Newton (BFGS) inversion algorithm could not rule out the possibility that a slightly resistive feature at the depth of interest was a misplaced expression of the deeper basement high. GN inversion provides much-improved, more stable images that show a laterally extended, moderately resistive layer in the target depth range. When explicitly including the basement in the inversion starting model, that resistor becomes only marginally weaker, and remains clearly separated from the basement. Unlike BFGS inversion, the GN inversion results leave no room for misinterpreting the basement high as a hydrocarbon reservoir.

Main Objectives

Improving the interpretation accuracy of airborne EM data with topography

New Aspects

A time-domain airborne EM inversion algorithm for topographic earth has been developed

Summary

The topography has a big influence on the time-domain airborne electromagnetic (ATEM) data. Especially the mixture of electromagnetic (EM) responses from the topography and the abnormal bodies makes the survey data very complex, bringing great difficulties to the traditional data interpretation technology based on flat ground model. In this paper, we develop a 3D inversion algorithm with topography for ATEM data to improve the interpretation accuracy. The time-domain finite-element algorithm based on unstructured mesh is used to modeling the ATEM responses. By adopting the direct Gauss-Newton method, our inversion codes obtain reasonably rapid convergence. To speed up the calculation of forward modeling and Jacobian matrix, we decouple the forward modeling and inversion meshes and construct the unstructured local meshes to improve the calculation efficiency. Finally, we apply our 3D inversion codes to the synthetic data to verify its reliability.

Main Objectives

Improving the accuracy of 3D forward modeling for time-domain airborne EM over an anisotropic earth

New Aspects

Developing an adaptive finite-element method for time-domain airborne EM over an anisotropic earth

Summary

The quality of the forward modeling mesh has a huge influence on the accuracy of the 3D electromagnetic numerical simulation. How to generate a reasonable mesh for an anisotropic model with complex electrical structure is the key to obtaining high-precision modeling results. In this paper, we develop an adaptive finite-element forward modeling algorithm for time-domain airborne EM over an anisotropic earth by combining the unstructured time-domain finite-element algorithm and adaptive mesh refinement technology. The reliability of our algorithm is verified by comparing our results with 1D analytic solution.

Main Objectives

To accurately model the Ground penetrating radar signals by the full spectral method

New Aspects

Full spectral modeling

Summary

Accurate differentiation matrices operators by the full spectral method have been applied to the temporal and spatial domains of the transverse electric (TE) and transverse magnetic (TM) modes of the GPR full waveform. By Fourier transform, these operators are simply complex frequency and wavenumber which must be sampled appropriately to produce accurate results. This full spectral method contrasts with the pseudo-spectral method that transforms only a component (either the time or space) to the Fourier domain and leaves the second component in its regular domain. The finite difference approach supplies its approximate differentiation matrices operator in time and space domains thus leading to inherent truncation error. Convolution of the differentiation matrices operator with the medium parameter is performed; and the inversion of the system of linear equations produce the accurate GPR full waveform modeling results for different permittivity models.

Main Objectives

Unique resistivity imaging by joint inversion constrained by Gramian

New Aspects

Linear correlation enforced by the Gramian constraints

Summary

Joint inversion of multi-physics is used to minimize the non-uniqueness associated with under-determined geophysical problem by some constraints. The Gramian stabilizing constraint has been used to enforce the linear correlation between the resistivity models from the frequency and time-domain airborne electromagnetic (AEM) data. The Gramian is the dot product of the two resistivity models which constrains the nonlinear least square optimization towards more reliable interpretation even in the presence of noise. Both synthetic and field data demonstrations give satisfactory results.

Main Objectives

3D MT anisotropic inversion

New Aspects

3D MT anisotropic inversion

Summary

For a long time,geologists believed that the frequent occurrence of earthquakes in the west coast of the United States is closely related to the existence of a north-south Cascadia subduction zone in the deep part of the region. the National Science Foundation EarthScope/USArray program has collected a large amount of long-period magnetotelluric (MT) data in the area. Many geophysicists have studied these data.From the published results, the approximate distribution of the Cascadia subduction zone in the area can be inferred, but they also pointed out that there is significant electrical anisotropy in the deep underground of this area To solve the problem with electrical anisotropy in 3D MT inversions, we propose a 3D MT anisotropic inversion method that considers simultaneously all principal conductivities and rotation angles. After testing on synthetic data (not shown in the abstract), we applied the method to the Long-period MT data collected from the west coast of the United States.We have successfully developed a 3D MT inversion algorithm for anisotropic earth models. The numerical experiments have proved the effectiveness of our inversion method. The anisotropic characteristics in the Cascadia subduction zone are identified from our inversions. Hope that this will help further analyze the earthquake-prone zone.

Main Objectives

The main objective is to enhance the computational properties and interpret or determine the subsurface parameters by joint inversion and Gibb’s sampler.

New Aspects

Joint Inversion; Gibb’s sampler; Probability distribution function; Hybrid PSOGWO

Summary

Employing the hybrid optimization technique with Gibb’s Sampler in association with joint inversion of MT and DC methods over 1D layered earth structures. The hybrid optimization algorithms have ability to balance the exploration and exploitation characteristics required for obtaining global solution. This hybrid technique uses the exploitation characteristic of PSO algorithm and the exploration characteristic of GWO algorithm, and the arrangement are generated from model parameters according to the Gibb’s sampling. The inherent problem of suppression is also studied due to conductive layer above a resistive layer. The results of hybrid algorithm with Gibb’s Sampler converges the solution faster than standard hybrid algorithm and it depicts large number of good fitting solutions lies in narrow region within the search space. Therefore, it is better to analyze histogram and calculate global mean model based on probability distribution function (PDF) with 68.27% confidence interval (CI) for all accepted models instead of selecting global model based on least error. In the present study, two different subsurface structures are optimized with noise free and noisy synthetic data. The efficiency of the algorithm is demonstrated by optimization in the paper.

Main Objectives

Description of the electromagnetic properties of subsurface material

New Aspects

A new method is developed to estimate the loss tangent of subsurface materials

Summary

Amplitude of electromagnetic (EM) waves is attenuated when propagating in a lossy medium. The attenuation which is commonly characterized by the loss tangent depends on the electromagnetic properties of the composing materials, medium structure and the nominal operating frequency of the transmitted signal. Therefore, evaluating the EM waves attenuation can give insights to the material’s constitutive parameters. Assuming a Ricker wavelet as the source wavelet, a nonlinear equation relating the frequency to two-way travel time, loss tangent and nominal frequency is derived that is then solved by probabilistic approach inversion to recover the sought model parameters. The proposed approach is applied into a real dataset acquired on Mount Etna volcano, Italy.

Main Objectives

Report results from a research project on the developpement of EM methods for geothermal exploration

New Aspects

Novel application of active and passive electromagnetic surveying for deep geothermal exploration

Summary

In the present paper, we present a novel combination of 3D Controlled-Source Electromagnetics (CSEM) and Magnetotellurics (MT) adapted to image electrical resistivity of deep granitic fractured reservoirs. This method will be applied in the Upper Rhine Graben (URG) to locate and describe accurately the geothermal resource in order to reduce the risk of future Enhanced Geothermal System (EGS) projects planned in this area. Electrical resistivity logs data from Soultz-sous-Forêts geothermal wells (Alsace, France) have been used to design an optimal survey, since knowledge of resistivity variation in the sedimentary cover is required to properly image deep fluid circulation with high accuracy. As a first validation step, a 2D field trial run in 2019 near the geothermal power plants from Soultz-sous-Forêts and Rittershoffen successfully characterized the subsurface targets. A subsequent large-scale 3D CSEM/MT acquisition campaign was carried out in 2020 to cover 150 km2 in Northern Alsace, in a context of an actual exploration program.

Main Objectives

Application of geophysical tools as an alternative method to know in advance the characteristics of the soils for aquaculture ventures.

New Aspects

Application of geophysical tools for aquaculture ventures.

Summary

In this study, Ground Penetrating Radar (GPR) and electromagnetic (EM34-3) geophysical tools were used, with the support of Unmanned Aerial Vehicles (UAVs) to characterize the subsoil of lands intended for the excavation of fish ponds. The study site was in the Montenegro aquaculture zone, in the municipality of Bragança, in the state of Pará (Brazil). The geophysical data showed that the subsoil of the aquaculture venture has appropriate characteristics for the excavation of the aquaculture ponds for due to the good quality of the land. The results of the geophysical prospecting, together with the analysis of the existing excavations, confirmed the adequacy of the terrain for the implantation of new ponds. The applied non-destructive geophysical tools are recommended for aquaculture ventures because they allow to know in advance the characteristics of the soils, as important step for the success and sustainability of any aquaculture operation.

Main Objectives

Mineral exploitation, application of EM and Induced Polarisation, 3D inversion anomaly mapping, reserves, hires remote sensing

New Aspects

EM/IP anomalies for mineral reserve mapping, new 3D simultaneous inversion algorithm, field study with shallow borehole calibration, exceptional prediction power, cost effective complementary technique

Summary

Geoelectric techniques are applied to identify geobodies in the shallow subsurface (<1km) that correspond to commercial ore deposits (copper-molybdene) in Kazakhstan. A combined CSEM and Induced Polarisation method is chosen to delineate anomalies in the underground. Resistivity and polarisation effects prove diagnostic. The workflow comprises steps like: EM acquisition, quality control and data preconditioning, inversion, interpretation and Principle Component Analysis. Inversion processing is done via a finite elements method solving the Cole-Cole formula simulating Maxwell’s equations. 1D inversion results serve as input for the 3D inversion. Principle Component Analysis (n-dimensional clustering and distance weighting) and computation of composite geoelectric parameters enhance the discrimination power. EM anomalies are circular (hydrothermal injection feature) and/or elongate in shape. Fracture zones and faults provide conduits/barriers and govern hydrothermal processes. Faulting in part controls the outline of the segmented IP anomalies. Three shallow well locations were proposed based on the EMS-IP data. Two of these boreholes demonstrate elevated polarisation phenomena: copper-molybdene metal ore in MN17 and pyrite enrichment in MN16. The mapped geobodies based on EM anomalies give complementary information on volume and distribution of the mineral resources. EMS-IP is a cost-effective investigation tool that deserves more attention in geoscience projects.

Main Objectives

Characterisation of karst

New Aspects

Multiscale and seismic attributes analysis

Summary

This study follows an integrated approach attempting to characterize karst morphologies in Middle Miocene carbonate build-ups of Central Luconia. Karst is a common phenomenon in carbonate build-ups world-wide. It has significant economic implications for exploration, drilling, field development and secondary recovery mechanism. In Malaysia over 250 carbonate buildups occur offshore in the Central Luconia province of Sarawak. Some 60 platforms have been drilled and almost every field has encountered indications for high permeability zones likely associated with karst such as mud losses and drill bit drops during drilling activities. Some fields were left abandoned due the mud losses that could not be controlled. Geometries, distribution and dimension of karst in Central Luconia fields remain unknown. They have not been studied in detail. To improve the geological understanding of karst morphology, a workflow is proposed by integrating drilling parameters data, cores, well logs data and seismic data. MX platform located in the central region of Central Luconia has been studied for this research. Three seismic attributes have been deployed on MX platform seismic to enhance the seismic for the characterisation of karst. Result shows that RMS amplitudes, spectral decomposition, and Acoustic impedance attributes are used to highlight karst features on seismic.

Main Objectives

To demonstrate the influence of various environmental factors on the development of carbonate platforms through stratigraphic forward modelling

New Aspects

A 3D reconstruction of the evolution of the Llucmajor with geologic time, and the application of the results to platform development in general

Summary

Stratigraphic forward modelling affords us the luxury of a digital laboratory to investigate the evolution of sedimentary systems. This technique was employed to understand the processes at play during the approximately 2 My of the evolution of the Llucmajor Platform. The results were subjected to comparative analysis with observed geometries and facies distribution as documented in the extensive work of Pomar and others over the years.

This presentation will focus on the environmental factors responsible for the resultant geometry and facies distribution of the platform and alternative cases that would have resulted if the conditions were different will also be presented.

Main Objectives

Seimsic interpretation of a mixed siliciclastic−carbonate−evaporite system

New Aspects

Quantitative prediction of lithology content in a mix system

Summary

High-quality three-dimensional seismic data acquired in Sichuan Basin, southwestern China, offer an opportunity to map complex lithologies in a mixed siliciclastic–carbonate–evaporite system in the Lower Triassic Jialingjiang (T1j) Formation. The formation consists of siliciclastics, limestone, dolostone, anhydrite, and salt, which consist several source-reservoir-cap assemblages in the area.

Lithologies in the T1j Formation change rapidly in the vertical direction, forming different interbed patterns. In the meantime, the lateral extend of lithology is complex. This vertical and lateral distribution makes it difficult to predict lithology by single seismic attributes. Therefore, principle component analysis (PCA) was applied to tens of seismic attributes to extract useful information. The first three components contain most lithology information preserved in seismic attributes, which were used to correlate with lithology content calculated by core-calibrated wireline logs. Correlation coefficients of the three seismic components with lithologies are higher than those of individual seismic attributes. Different assemblies of end-member lithologies were selected from anhydrite, siliciclastics, tight dolostone, limestone, and salt to perform PCA. Lithologic content distribution of individual end members was shown by color-blending method to map the lithology mixture.

Main Objectives

A regional sequence stratigraphic review to challenge stratigraphic misconceptions on the Arabian Plate.

New Aspects

Dominant role of eustacy and response of the carbonate factory for the creations and infill of the Gotnia Basin.

Summary

The Jurassic stratigraphy of the Middle East contains some of the world’s most economically significant petroleum systems comprising world-class source rock, reservoir and seal packages. Yet these depositional systems are still not fully understood in their regional context, leading to inconsistencies in the use of lithostratigraphic nomenclature across international boundaries. This, in turn, results in misconceptions of stratigraphic architecture and evolution, with implications for the distribution and quality of petroleum systems elements, as well as exploration and production strategies. This revised interpretation utilizing the latest public domain datasets challenges some of these long-standing misconceptions. One of these stratigraphic inconsistencies includes the creation and infill of the Late Jurassic Gotnia Basin. The development of the Gotnia Basin is cited by many authors to be tectonically controlled. However, an alternative model based on the concepts of carbonate sequence stratigraphy and its relationship to eustatic sea-level change can be proposed. This new insight has impacted the prediction and distribution of source, reservoir and seal facies, and the presence of stratigraphic traps.

Main Objectives

Understand the process involved in the deposition of the carbonate minerals and the organic matter diagenesis.

New Aspects

Organic matter diagenesis associated with the carbonate deposition using lipids biomarkers.

Summary

The coastal region of the Rio de Janeiro state (Brazil) is characterized by a semi-arid microclimate associated with the upwelling coastal system of nearby Cabo Frio, which affects the hydrological and biogeochemical cycles in the region. Lakes and lagoons are natural laboratories to study biogeochemical signals that occur over geological times because they generally have higher sedimentation rates than the oceans. Lagoa Vermelha (LV) and Brejo do Espinho (LBE) beside amplify signals can register fluctuation in the local climate which is related to ocean circulation and biomineralization. These lagoons represent one of the few places in the world where modern precipitation of dolomite occurs. This study uses a multiproxy approach to characterize the deposition of carbonate sediments at these evaporitic environments. Sedimentary cores from LBE and LV, 1.6 and 6.1 cal kyr BP respectively, demonstrated mixed organic matter source reaching the lagoons with large input of terrestrial components. The dolomite-rich layers deposited ~2.1 cal kyr BP displayed enriched 18O and depleted 13C suggesting intense microbial activity and dryness. Using an approach combining organic and inorganic geochemical proxies has led to the recognition of dryness as an important regional climatic characteristic on the carbonate sedimentation in these hypersaline coastal lagoons.

Main Objectives

petrophysical property cutoffs of low permeability carbonate reservoirs

New Aspects

multiple centrifugal tests under nuclear magnetic resonance (NMR)

Summary

Focusing on the Paleogene tight lacustrine carbonate rocks in the Yingxi area of Qaidam Basin, based on determining the distribution range of nano-scale dolomite intercrystalline pore throat by high pressure mercury injection test and scanning electron microscope(SEM) analysis, this study proposes an experimental method to determine the petrophysical property cutoffs of reservoirs. The core of the method is to separate the movable fluid from the irreducible fluid through multiple centrifugal tests under nuclear magnetic resonance (NMR), Pore throat radius corresponding to the separation point of movable fluid is taken as the pore throat radius cutoffs for the movable fluid distribution. The petrophysical property cutoffs of reservoirs are determined by the fitting relationship between Pore throat radius and physical properties. Results show that the distribution range of dolomite intercrystalline pore throat determined by mercury injection test and SEM is 40 to 300 nm, and that of pore size of the sample determined by NMR test after saturated distilled water is 50 to 300 nm. The separation point of movable fluid corresponds to the pore throat radius cutoffs of 47 nm for the movable fluid distribution, porosity cutoffs determined by NMR-based centrifugal test is 3.29%, and permeability cutoff is 0. 02 mD．

Main Objectives

To show that siderite precipitation due hydrocarbon migration in petroleum systems is likely

New Aspects

Relating the presence of siderite in hydrocarbon environment directly to hydrocarbon migration

Summary

Migration of hydrocarbons in the subsurface has been shown to create an environment that promotes the precipitation and/or alteration of magnetic minerals. For example, iron oxides and iron sulphides have been shown to precipitate due to the reducing conditions created by hydrocarbon migration. Siderite, a paramagnetic mineral with Neel temperature of 37K has been variously identified in hydrocarbon environment and has also been suggested to be an authigenic product of hydrocarbon migration. However, it is commonly found in sedimentary settings. Here we show via experimental studies that siderite is precipitated due to hydrocarbon migrations and suggested the mechanism responsible for this process. Magnetic minerals precipitation along migration pathways suggests the creation of a magnetic fingerprint that if thoroughly understood can be applied to oil and gas exploration.

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Main Objectives

To reduce costs in geophysical data generation/acquisition and to improve the performance of acoustic FWI in the presence of strong elastic effects.

New Aspects

Novel applications of deep neural networks in geophysics.

Summary

Image-to-image translation using GANs have successfully been applied to a wide variety of problems, from mundane implementations to turn horses into zebras, to stunning synthetic media deepfakes. We explore its application in geophysics as a cost-reduction tool, and demonstrate its potential in 3D field data. We show it can be used to learn the mapping between different data flavours of interest in modern data processing workflows: acoustic/elastic for full-waveform inversion and geophone vertical-component/hydrophone-pressure for up- and down-going wavefield separation.

Main Objectives

Showing how Machine Learning can solve ordinary problems

New Aspects

Application of 2D and 3D U-Nets to solve interpolation problems

Summary

We present Deep Learning solutions for seismic data interpolation problems such as: filling in missing traces, replacing bad data zones (undershoot areas), and creating super-sampled 3D volumes. The models in our examples are 2D and 3D Convolutional Neural Networks with a U-Net type architecture. We construct our training sets by randomly blanking certain parts of the input images (2D) or input cubelets (3D) and test them on blind test areas, i.e.: areas from which no training examples were extracted. Next, we demonstrate that a 3D U-Net for infill of missing traces can be utilized as a generic interpolator. We apply a model that is trained on one dataset to solve a similar problem on another dataset located hundreds of kms away. Both datasets exhibit very different seismic characteristics.

Main Objectives

An approach to help adapt neural networks trained on synthetic data to real data.

New Aspects

We develop a novel approach to enhance our neural network model training on synthetic data with real data features (domain adaptation) using a number of linear operations applied on the training and inference input seismic data.

Summary

The requirement for accurate labels in supervised learning often forces us to train our networks using synthetic data. However, synthetic experiments do not reflect the realities of the field experiment, and we end up with poor performance of the trained neural network (NN) models at the inference stage. Thus, we describe a novel approach to enhance our NN model training with real data features (domain adaptation). This is accomplished by applying two operations on the input data to the NN model, whether they are from the synthetic or real data subset class: 1) The crosscorrelation of the input data section (i.e. shot gather or seismic image) with a fixed reference trace from that section. 2) The convolution of the resulting data with a randomly chosen auto correlated section of the other subset class. In the training stage, as expected, the input data are from the synthetic subset class and the auto-corrected sections are from the real subset class, and in the inference/application stage, it is the opposite. An example application on passive seismic data for microseismic event source location determination is used to demonstrate the power of this approach in improving the applicability of our trained models on real data.

Main Objectives

Setting up the problem of seismic denoising (for coherent noise) in deep-neural network framework. Demonstrating the convergence of solution in reasonable processing time. Demonstrating the reduction of noise with increasing number of training iterations.

New Aspects

Demonstrating that networks can be trained with multiple estimates in common-shot and common-receiver domain for multiple attenuation.

Summary

Denoising seismic data is a crucial processing step in imaging process that requires significant compute time and human effort for quality control. There are limitations to existing noise-removal methods due to limited wavefield sampling and frequency content of trace records, and poor knowledge of the source wavelet. Deep learning networks in essence contain various layers between the input and output images, each layer acting as a filter. Filter coefficients are determined during the training process via minimization of the loss function between desired output and forward modeled output. We present our efforts to construct, train and deploy convolutional neural networks with Unet/ResNet architectures to solve problems of denoising in seismic domain. For synthetic trace data we show that without any patching we can train networks effectively and in reasonable time, and noise can be attenuated significantly.

Main Objectives

Through intelligent deblending to provide separated gathers for the following traditional seismic data migration and inversion.

New Aspects

A U-net framework is proposed together with iterative strategy for intelligent deblending efficiently.

Summary

Blended acquisition can help improve the acquisition efficiency or enhance the data density. However, blended seismic data which contains information of multiple sources, should first be separated for traditional seismic data processing steps. Thus, we propose a U-net based intelligent deblending algorithm, combined with the traditional iterative strategy in this abstract. The proposed method can obtain the optimal parameters through self-learning while it should be selected by trial and error in traditional method. We train and valid the U-net by using a set of synthetic data with labels, and then parts of field data with labels are used to finetuned it. Finally, the finetuned U-net is used for intelligent deblending of the left field data. The deblending performance is promising compared with the curvelet transform based thresholding method, which demonstrates the validity of the proposed intelligent deblending algorithm in deblending accuracy, stability and efficiency.

Main Objectives

Improve the generalization ability of CNN based interpolation method.

New Aspects

A self-evolving CNN framework is proposed to improve the generalization of CNN model when applied to new unlabeled data sets.

Summary

Nowadays, deep learning has been widely explored in the field of seismic data interpolation. However, the generalization ability of deep neural network is still a longstanding issue in the research field related to deep learning. It is important to automatically evaluate the adaptability of CNN and adaptively modify the CNN model according to the new data sets. In this abstract, we propose a self-evolving CNN framework for seismic data interpolation. A cyclic structure is designed to automatically evaluate and modify the CNN model when applied to new unlabeled data sets. And an inadaptability index is defined to quantitatively evaluate the generalization ability of CNN. The experimental results on synthetic examples and field examples demonstrate that the proposed framework can improve the generalization ability of CNN model on new data sets.

Main Objectives

To avoid petrotechnical experts spending time on mundane tasks that can easily be modelled my machine learning and instead focus on geologically challenging areas where the characteristics of the near surface are very complex. This technology can easily be incorporated in current Surface wave analysis workflows.

New Aspects

Use of supervised machine learning for picking the mode with the highest energy in frequency-wavenumber semblances. The machine learning model is global, which means one model built on a large diversity of data applied to a new survey location without the requirement of re-training.

Summary

The quality of onshore seismic data is strongly affected by the characteristics of the near surface. To compensate for the distortion of travel times of seismic energy, workflows have been formulated to analyse, model, and invert surface waves. This approach requires human-intensive picking of high energy modes on conditioned semblances that represent each analysis location on the dispersion survey. In this work, we designed a globally trained supervised machine learning model to perform pixelwise binary segmentation on semblances using fully convolutional architecture with residual units to pick the fundamental mode – the mode with the highest surface-wave energy. We validated our approach on all the analysis locations of the Cooper Basin in South Australia, and compared the results with current conventional methods used in the industry.

Main Objectives

improve the accuracy and quantity of picked first breaks

New Aspects

Deep Learning First Break Picking Technology Based on Operation Scene

Summary

With the wide application of high density and high productivity vibroseis technology in the complex oil fields, picking first breaks from massive data with low signal-to-noise is a big challenge. The traditional first break picking method is poor anti-noise and needs a lot of manual modifications, which affects the accuracy and efficiency of first break picking. To overcome the shortcomings of the traditional methods, this paper proposes a deep learning first break picking technology based on operation scene. This technology uses the techniques, for example, design time window and build classification network model to identify abnormal first breaks and use generative adversarial network to correct abnormal first break, etc., to improve the accuracy and quantity of picked first breaks. Test with actual data has verified that the proposed technology can achieve high accuracy of first break picking and has the ability to stably process primary waves with low signal-to-noise.

Main Objectives

In seismic exploration, the seismic data with high SNR is important for processing and interpretation. We propose an adaptive anomalous amplitude attenuation method based on deep neural networks to suppress the noise with anomalous amplitude.

New Aspects

We adopt the deep neural networks to segment the anomalous amplitude noise, then estimate the parameters of the noise, and at last use these parameters to make our AAA method to be adaptive.

Summary

In seismic exploration, anomalous amplitude noise is very common in seismic data, and owing to its high energy, it will distort the result of subsequent processing steps. Anomalous amplitude attenuation (AAA) is a widely used method to deal with this kind of noise, but the performance is heavily depended on the parameters. We propose an improved AAA method based on the deep neural network. Unlike many other denoising methods based on deep neural networks, we do not generate the denoising result from the networks directly. Instead, we adopt our network to estimate parameters to make our AAA method to be adaptive.

Main Objectives

The main objective was to develop a deep learning workflow for salt interpretation that reduces workflow time and retains high accuracy.

New Aspects

The paper covers the comparison between deep learning approaches for salt interpretation: 1) window-based convolutional neural networks, and 2) the U-Net approach. We successfully applied salt interpretation deep learning to the Julia field in the deep-water Gulf of Mexico, which to our knowledge has not been completed before.

Summary

Interpreting salt bodies in the subsurface is a challenging manual task that can take weeks to complete. Obtaining accurate picks of salt is very important, because errors in the placement of salt can result in severe degradation of the seismic image. To meet the challenges of speeding up imaging workflows and retaining accurate salt picks, we evaluate three deep learning approaches: a 2D window-based convolutional neural network, a 3D window-based convolutional neural network, and finally a 2D “U-Net” approach. A 3D seismic volume from the deep-water field Julia in the Gulf of Mexico was used to test these approaches. The Julia field has complex salt structures with overhangs and inclusions, and the thickness of salt can reach up to 5 km. The U-Net architecture proved to be the most accurate of the three methods tested, predicting the placement of salt at 98% accuracy, as compared to the human interpretation. Beyond accuracy, U-Net also proved to be the fastest, requiring only 3.5 hours to predict salt on the 3D seismic volume. The results presented here along with other recent studies of deep learning for salt interpretation represent a clear shift in the seismic interpretation workflow.

Main Objectives

Identification and uncertainty quantification for automatic fault identification using deep neural networks

New Aspects

Use of a Bayesian neural network to identify faults in seismic data including the associated uncertainty measures

Summary

The interpretation of faults within a geological basin or reservoir from seismic data is a time-consuming, and often manual task associated with high uncertainties. Recently, numerous approaches using machine learning, especially various types of convolutional neural networks, have been presented to automate the process of identifying fault planes within seismic images, which have been shown to outperform traditional fault detection techniques. While these proposed methods show good performance, many of these approaches do not allow investigation of the associated uncertainties that arise in the fault identification process. In this study, we present an application of Bayesian deep convolutional neural networks for identifying faults within seismic datasets. Using an approximate Bayesian inference method a Bayesian deep neural network was trained on a large dataset of synthetic faulted seismic images. The model is then applied to a benchmark dataset and a real data case from NW shelf Australia to identify fault planes, and to investigate the associated uncertainty in the predictive distribution.

Main Objectives

An innovative deep learning network to significantly improve the accuracy of fault interpretation

New Aspects

Multi-scale dilated convolutional neural network

Summary

Assisted fault interpretation leveraging machine learning techniques has become a promising way to identify faults in seismic. In geophysical exploration, faults are often considered as a sealing surface which traps hydrocarbons and forms reservoir zones. Thus, correctly identifying fault locations is critical. Fault identification can be treated as a semantic segmentation issue where we classify each seismic pixel into one of a given set of categories, such as fault or non-fault. To be successful we need to combine pixel-level accuracy with global-level feature identification. In this abstract, we propose a novel deep learning network with multi-scale dilated convolution to identify fault locations. It is based on adaptions of a convolutional neural network architecture which has been used for image classification and semantic segmentation. The motivation is that dilated convolution supports exponentially expanding receptive fields without losing resolution or coverage. We implemented multiple dilated convolution layers with variable dilation rates to systematically aggregate multi-scale seismic information. Several tests are shown and demonstrate the improvement of identification accuracy with higher resolution.

Main Objectives

fault damage zone, seismic attributes, artificial neural network (ANN), Junggar Basin

New Aspects

Subsurface damage zones were quantitively studied from 3D seismic data using an artificial neural network approach.

Summary

To further improve the quality and efficiency of subsurface fault zone image and study its geometry. Herein we adopted post-stack seismic data conditioning and a combination of seismic multi-attribute for producing a new hybrid attribute through a supervised multilayer perceptron (MLP) neural network in the Jurassic formation of Cai36 3D prospect located in the eastern part of the Junggar Basin. We first conditioned original seismic data by using the dip-steering cube extracted from the original seismic data. Secondly, we extracted conventional seismic attributes from the conditioned data sensitive to fault zone signatures. Thirdly, we selected a set of “picks” at a time slice representing the presence or absence of fault zones. Then we adopted the supervised MLP neural network to train over the selected seismic attributes extracted at the fault zone and non-fault zone positions. We obtained a new fault probability cube as new attributes. Finally, we analyzed a typical strike-slip fault zone using the new attributes. This study provides an effective way of fault zone imaging from seismic data and adds new insights into its geometry. Therefore, the workflows used here could be widely applied to other 3D surveys.

Main Objectives

2D horizon extraction without seed points

New Aspects

extract seismic horizons on multiple 2D lines without seed points

Summary

Horizon interpretation is one of the most important yet time consuming steps in a traditional seismic interpretation workflow. Traditionally, horizon interpretation is performed interactively. An interactive horizon tracker takes an interpreter’s sparse picks as constraints and fills in horizon segments in between the sparse picks, usually by comparing the similarity between two adjacent traces within a small temporal window. In some legacy oil fields, 2D seismic data are abundantly available. With the advancement of seismic processing and imaging techniques, reprocessing such legacy 2D seismic data may be the key to unlock additional value and bring new insights to the understanding of the subsurface. . In this study, we investigate using supervised deep learning to extract horizons on all the 2D seismic lines after training on a small subset of the available lines from the same region. Using a 2D shallow seismic dataset from offshore the Netherlands as an example, we are able to demonstrate that the proposed method is able to extract horizons consistently across multiple moderate quality 2D seismic lines.

Main Objectives

Method for dense interpretation of seismic stratigraphy sequence and horizons with deep learning

New Aspects

Optical flow field for seismic interpretation, FlowNet

Summary

We developed a FlowNet based deep learning workflow, to produce flow field cube from a seismic survey. Based on the flow field, stratigraphic sequence information such as horizons can be densely extracted given any seed point. Training of the network is performed on image and distorted image pairs generated by applying elastic deformation with random flow field without manual labeling. A multiscale horizon tracking method was developed to incorporate flow fields from different resolutions.

Main Objectives

Seismic facies identification using Artificial Intelligence

New Aspects

Seismic data augmentation and geophysical customization of metrics

Summary

Facies identification is of high interest for oil and gas exploitation or CO2 storage. It is a time-consuming step that requires high-level human expertise. Indeed, the interpretation is based on geological knowledge and understanding of the field. Nowadays deep learning techniques are very efficient on image segmentation and finally may help this challenge. In that context, SEAM proposed an artificial intelligence hackathon on 3D real data thanks to public seismic data of Parihaka (New Zealand) and free-license labels from Chevron.

Our solution is composed of several ideas. First, we applied a seismic data augmentation that deforms slopes and amplitudes. Then, we added the facies interfaces as an extra output of the deep learning network, so as to constraint the facies detection. Finally, we enhanced multi-resolutions predictions combination using a U-Net with Feature Pyramid Network. This led to the best score in both pixel-by-pixel and pixel-weighted-around-interfaces metrics. We observed that the quality of detection is impacted by the presence of main faults which split the training dataset into two unbalanced seismic sets of structures. At last, we are glad that our results have been distinguished by their relative geophysical quality.

Main Objectives

Investigate if GAN can reproduce the complexity of fluvial deposit captured by process-based modelling

New Aspects

PatchGAN reproduce process-based model, one-hot encoder, multi-discriminator method

Summary

This work shows the advantages and disadvantages of modelling complex geological models using generative adversarial networks (GAN). A process-based model, FLUMY, is used to create the training dataset. Compared to previous work in this area, this dataset contains varied geo-body geometry, asymmetrical channel sinuosity and irregular meander morphology. In short, this training dataset is closer to the real complexity of fluvial reservoir. The results indicate our GAN can capture complex multi-facies distribution, their relationships, and facies geometry. However, the GAN generated realizations contain many geologically unrealistic features. In this paper, we list two types of unrealistic features, named ‘mislabelling’ and ‘incorrect channel-levee relationship’. Two proposed methods are proved that they can reduce the amount of the unrealistic features. Embedding one-hot encoder in GAN can cure the ‘mislabelling’ issue. Multi-discriminator strategy is helpful to assist GAN to learn spatial relationships among different facies better.

Main Objectives

improved stratigraphic mapping using machine learning

New Aspects

Hierarchical Deep Learning Networks

Summary

Training neural networks to detect features on seismic requires care from the interpreter to label all the true positives present on a training image. But often, the interpreter is solely interested in detecting features in a single interval. This narrowed focus arises from either a preference for zones associated with commercial hydrocarbon systems, or because the features in the zone of interest are characteristically different than similar features in overlying or underlying zones. In either case, it would be desirable to create a volumetrically consistent subdivision of stratal layers before feature labeling work begins in earnest.

To achieve these ends, the authors introduce a novel way to perform stratal zonation using a Hierarchical Deep Learning (HDL) network. This approach simultaneously segments the entire seismic image into an arbitrary number of stratal zones using a multi-class network. Each stratal zone can then be further subdivided into a hierarchical arrangement of layers. Once trained, the HDL network’s inference can then be iteratively refined using an increasingly rich set of control points. On each successive iteration, the HDL network returns an inference that more closely approaches the geoscientist’s expert opinion.

Main Objectives

To show the effects of acquisition geometry on FWI Imaging using examples from a source-over-spread survey in the Barents Sea, focusing on lateral resolution, fault imaging and footprint.

New Aspects

Show uplift of FWI Imaging over conventional migration methods and introduction of the new concept of dip-coherency images as an interpretation aid.

Summary

The combination of ever-increasing computational power and more robust algorithms have made it possible to run full-waveform inversion (FWI) to higher frequencies and, also, offer more possibilities to take advantage of the reflections in the inversion. Through a process known as FWI Imaging, the detailed velocity models produced can be used to generate a reflectivity normal to the reflector plane. We outline the methodology and advantages of FWI Imaging, and introduce the concept of a dip-coherency image as an additional interpretation tool, using information parallel to the reflector plane. We show examples from the densely sampled source-over-spread Greater Castberg survey in the Barents Sea, demonstrating the uplift in the FWI Image over conventional imaging methods in terms of more balanced illumination and richer low frequencies. We performed decimation tests to assess the acquisition geometry impact on FWI imaging. Although the benefit of FWI imaging can still be observed on less well-sampled data, the best result remains with the original, densely sampled source-over-spread acquisition.

Main Objectives

An accurate background model built with FWI based on Inverse Scattering Imaging Condition and Vector Reflectivity. High resolution model, providing details about the lithology and the structural extension of local velocity variations in the overburden.

New Aspects

FWI based on Inverse Scattering Imaging Condition and Vector Reflectivity applied to a shallow water OBN data set.

Summary

Depth imaging in complex geology requires an accurate background model. In this case study, over the Clair field, we make use of high density node data set and Full Waveform Inversion (FWI) with Inverse Scattering Imaging Condition and Vector Reflectivity to solve the low wavenumbers in the model. The purpose of FWI was to capture the hard and fast seafloor followed by a strong velocity inversion, numerous localized high velocity bodies and accumulations of gas in the overburden, as well as mapping the rotated fault blocks in the target interval below Base Cretaceous Unconformity.

The inversion was done with a wavelet estimated from the data, to avoid possible bias in model updates related to phase errors in the deterministic wavelet. After solving the background model, the inversion continued up to 60 Hz. The final FWI model provided a well focused image with realistic structures and a flatter gas-oil contact than the vintage model. Significant amount of details have been captured in the FWI model, which correlates well to sonic data and the geological knowledge of the area.

Main Objectives

To describe how FWI has been used to invert for overburden changes in time-lapse processing

New Aspects

Application of multi-parameter full-waveform inversion in 4D

Summary

4D time-lapse seismic imaging is typically performed using the same velocity model to migrate the baseline and monitor data. However, in complex cases where one producing reservoir sits above another, significant changes in the properties of the overlying reservoir can result in a 4D signal with associated 4D coda immediately below it, which would mask the 4D signal of reservoirs underneath. Resolving such an issue requires that the baseline and monitor data be migrated with separate models. We demonstrate how we have used a visco-acoustic full-waveform inversion to resolve 4D changes in velocity and absorption within an overburden gas-charged channel. The resulting 4D image shows minimisation of 4D coda below the overburden channel and the unveiling of 4D signals at deeper targets of interest.

Main Objectives

To propose an innovative workflow to estimate anisotropy in order to have flat CIGs after FWI.

New Aspects

The workflow involving first break traveltime computation instead of picking, with the possibility to reposition the sources and receivers.

Summary

An incorrect anisotropy in the Full Wave Inversion (FWI) velocity model leads to imperfect Common Image Gather (CIG) flatness. The main difficulty in the anisotropy estimation through FWI is the strong coupling with velocity. While FWI jointly updating velocity and anisotropy has been proposed, there is evidence that the long wavelength components of the velocity and anisotropic parameters cannot be reasonably decoupled inverting surface data only. The reason is that the long wavelength components of the velocity model inverted by FWI are mainly recovered from the kinematics of diving waves, while decoupling can only be done considering in addition the kinematics of reflected waves. To solve this challenge, we propose an innovative workflow involving joint reflection and diving wave tomography. To overcome the difficulty of first break picking, we propose a robust estimation of first arrival traveltime using the inverted FWI model of the first pass. With the application of this novelty on deep-water data from offshore Africa, we elaborate further with a sequence of first arrival modeling after tuned repositioning of sources and receivers at the sea floor.

Main Objectives

this paper aims to invert Vp and Vs only use single component data in land seismic exploration

New Aspects

we propose a pseudo acoustic wave full waveform inversion method for elastic parameters inversion using P-wave data only. The gradients of the misfit function with respect to updating the perturbations of elastic parameters based on PAE theory are derived.

Summary

In land seismic exploration, low-velocity zone causes the ray path of reflected wave propagating to the detector perpendicularly. Therefore, single-component data is regarded as P-wave data. In this paper, we first derive a new pseudo acoustic wave equation (PAE) in elastic world based on acoustic approximation. Compared with the acoustic modeling, pseudo acoustic modeling has obvious elastic AVO effect and S-P converted energy. Then we propose a pseudo acoustic wave full waveform inversion method for elastic parameters inversion using P-wave data only. The gradients of the misfit function with respect to updating the perturbations of elastic parameters based on PAE theory are derived. A field data example in eastern china is carried out by our new method using only the p-wave data. The results of pseudo acoustic full waveform inversion shows that S-wave velocity inverted is reliable and the passion ratio profile is well fitted to the natural potential logging curve.

Main Objectives

Full waveform inversion of seismic-while-drilling. Drill bit source signature modeling/estimation. Compensating for the lack of low frequencies by successive inversion of surface seismic and SWD datasets.

New Aspects

A computational framework for drillbit source signature modeling/estimation is introduced. Full waveform inversion of seismic-while-drilling is developed.

Summary

We have developed a full waveform inversion (FWI) algorithm for the seismic-while-drilling (SWD) dataset. Full waveform inversion is a local optimization method. To avoid the local minima, seismic data should have rich low-frequency content. However, in the real-world, seismic data lacks low-frequency content, and FWI struggles to provide accurate subsurface properties. To remedy this shortcoming, we use SWD data to compensate for the lack of low frequencies in the surface seismic dataset. In SWD data, the drillbit acts as a seismic source. The drillbit generates significant elastic energy, which has different raypaths compared to the surface seismic. We show that if we understand the non-impulsive and correlative behavior of the drillbit source signature, the full waveform inversion of SWD data is possible. To estimate the drillbit source signature, we have developed a computational framework based on wave equation drill string dynamics modeling along with top-drive force and velocity measurements. Then, we feed the estimated dillbit source signature to the FWI algorithm to invert the SWD dataset. Our results show that the successive inversion of surface seismic and SWD datasets can compensate for the lack of low frequencies in the surface seismic data and reduce the uncertainties of the subsurface properties.

Main Objectives

We introduce an inversion approach to simultaneously invert for both velocity and reflectivity and demonstrate the benefits of our scheme using two field data examples.

New Aspects

We present a new approach that combines velocity model building (FWI) and imaging (LSRTM) into a single inversion process with minimum data preprocessing from an inaccurate initial model.

Summary

We present an iterative non-linear inversion method to simultaneously estimate both velocity and reflectivity. The core of the inversion workflow is a full acoustic wavefield modeling relation parameterized in terms of velocity and vector reflectivity. A key aspect is the separation of the low- and high-wavenumber components of the gradient based on inverse scattering theory, enabling the sensitivity kernels to update the velocity and the vector reflectivity, respectively. The estimation problem is formulated as a multi-parameter adjoint-state inversion where the trade-off between velocity and reflectivity is minimized through scale separation. Our approach is equivalent to performing Full Waveform Inversion (FWI) and Least-Squares Reverse Time Migration (LSRTM) in a single framework using the full wavefield. The output of the inversion is a detailed velocity model together with an accurate estimate of the earth reflectivity with compensation for incomplete acquisition, poor illumination, and multiple crosstalk. The new approach reduces the turnaround time of imaging projects by combining velocity model building (FWI) and imaging (LSRTM) into a single inversion process with minimal data pre-processing.

Main Objectives

Unlocking unprecedented seismic resolution with FWI Imaging

New Aspects

Full-waveform inversion (FWI) Imaging models and uses the full-wavefield data, including primaries and multiples (ghost included) and reflection and transmission waves, to iteratively invert for the reflectivity together with velocity and thus is an elegant solution to resolve those issues in one (iterative) inversion

Summary

A high-resolution seismic image is of great importance to exploration and production in many ways, such as bypassing drilling hazards and identifying compartmentalized reservoirs. To achieve seismic resolution as high as possible, the conventional seismic imaging process takes more of a linear approach to deal with one or a few specific issues at a time, such as noise and multiple attenuation, source and receiver deghosting, velocity errors, illumination holes, and migration swings. Full-waveform inversion (FWI) Imaging models and uses the full-wavefield data, including primaries and multiples (ghost included) and reflection and transmission waves, to iteratively invert for the reflectivity together with velocity and thus is an elegant solution to resolve those issues in one (iterative) inversion. FWI Imaging has proven to be a superior method for providing seismic images of greatly improved illumination, S/N, focusing, and thus better resolution, over conventional imaging methods. We demonstrate with a towed-streamer data set and an OBN data set that FWI Imaging with a frequency close to the temporal resolution limit of seismic data (100 Hz or higher) can provide seismic images of unprecedented resolution from the recorded seismic data, which has been impossible to achieve with conventional imaging methods.

Main Objectives

Full-waveform inversion for reservoir characterization. The objective is to build and estimate the elastic properties and simultaneously predict the geologic formation. Lithologic information from borehole can be used to constrain FWI, which helps determining the facies distribution between the borehole. This facies distribution can, thus, be used for reservoir characterization.

New Aspects

We use more accurate physics of subsurface to demonstrate that prior information helps increasing the spatial resolution of elastic parameters, and reduce parameter tradeoffs. Constrained FWI then generate the spatial distribution of facies, which can be used further used for reservoir characterization. Furthermore, we only use hydrophone data and realistic initial models to show that even in the absence of frequencies below 2 Hz, we can reconstruct all the anisotropic parameters with substantial accuracy.

Summary

High-resolution velocity models generated by full-waveform inversion (FWI) can be effectively used in seismic reservoir

characterization. However, FWI in elastic anisotropic media is hampered by the nonlinearity of inversion and parameter trade-offs. Here, we propose a robust way to constrain the inversion workflow using per-facies rock-physics relationships derived from borehole information (well logs). The advantages of the facies-based FWI are demonstrated on a 2D elastic TTI (transversely isotropic with a tilted symmetry axis) model with substantial structural complexity. In particular, the tests show that our

algorithm improves the spatial resolution of the inverted medium parameters without using ultra-low-frequency data required by conventional FWI.

Main Objectives

Propose a methodology to prevent numerical dispersion and instabilities when using the finite-difference method for wavefield simulation when performing full-waveform inversion

New Aspects

We propose to incorporating a stability constraint into the objective function of elastic full-waveform inversion through a penalty term based on a probability density function that represents the range of velocities in which numerical dispersion and instabilities do not occur

Summary

Full-waveform inversion (FWI) is a nonlinear inverse problem that refines models by iteratively matching the recorded and modeled data. The success of FWI relies on accurate and efficient wavefield modeling. The finite-difference method (FDM) is one of the most popular techniques used for this purpose because of its easy implementation and low memory requirement. However, depending on the spatial and temporal discretization, FDM may encounter numerical dispersion and instabilities, which degrade the quality of the simulated wavefields. The FWI objective function formulated based on data misfit may lead to situations where the updated models violate necessary conditions to guarantee modeling stability. We propose to address this problem by explicitly incorporating a stability constraint into the inversion through a penalty term based on a probability density functions (PDF) that represents the range of velocities in which numerical dispersion and instabilities do not occur. This penalty term prevents updated models from entering into a configuration that leads to dispersion and instability for specific choices of sampling parameters. Through synthetic examples, we demonstrate the benefits of including this stability constraint into the FWI framework, as it ensures the efficiency of the FDM engine, while increasing the quality of the recovered models.

Main Objectives

Solving cycle-skipping issue in FWI

New Aspects

Partial shift proposal and enhancing the kinematic information within the graph-space optimal transport FWI

Summary

The use of graph-space optimal transport within full-waveform inversion (FWI) has recently been proposed to enhance robustness to cycle-skipping. In this paper, we discuss several features of the graph-space optimal transport FWI, emphasizing in particular its ability to detect time shifts between modelled and observed data and its characteristics in terms of adjoint source (the signal back-propagated from the receiver side during the iterative optimization process). Our analysis is driven by finding an optimal adjoint-source that is robust to cycle-skipping. This leads us to propose an alternative graph-space-inspired scheme called enhanced kinematic transform FWI. Our approach involves a modification of the adjoint-source introducing a partial shift strategy, allowing to highlight the kinematic information. We illustrate the enhanced robustness with respect to cycle-skipping on synthetic and field datasets with comparison to conventional FWI, Kantorovich-Rubinstein optimal transport FWI and graph-space optimal transport FWI.

Main Objectives

Develop a black-box easy-to-implement regularization which admits empirical priors to be used with ADMM-based wavefield reconstruction inversion (WRI) for complex subsurface building.

New Aspects

Adaptive BM3D regularizer is implemented in the ADMM-based wavefield reconstruction inversion (WRI) method and it is assessed against the complicated benchmark velocity models having different statistical properties.

Summary

Regularization is necessary for full-waveform inversion (FWI). The basis of a good regularization is the prior expressed by the regularizer, which can be non-adaptive or adaptive (data-driven). However, tailoring a suitable and easy to implement prior to describe geophysical models is a nontrivial task. In this abstract, we propose a general black-box regularization algorithm for solving FWI in the framework of iteratively-refined wavefield reconstruction inversion (IR-WRI). Unlike classical regularizations based on prior information, the new formulation allows for empirical priors that are determined adaptively by sophisticated denoising algorithms.

In this approach, the solution is gradually fed by adaptive prior built from the previous iterate by the denoiser without asking for any information about its functional form, thus treating the denoiser/regularizer as a black-box.

Numerical results estimated by IR-WRI with block matching 3D filter (BM3D) denoiser show the high performance of this new regularizer for constructing complicated benchmark velocity models of different statistical properties.

Main Objectives

FWI cycle-skipping mitigation

New Aspects

Constrained cross-correlation based measurement of the traveltime misfit

Summary

A traveltime based FWI algorithm can estimate a kinematically accurate macro model in presence of cycle-skipped data. Such method aims at minimizing the traveltime misfit between the observed and modelled data, in a way that also maximizes their temporal cross-correlation function. Obtaining a reliable traveltime misfit estimate is paramount to the success of traveltime based FWI algorithms in the presence of cycle-skipped data. We propose a constrained correlation-based traveltime inversion to improve the quality of the traveltime misfit estimate. We use synthetic and field data examples to validate our method, which show evidence of cycle skipping mitigation and demonstrate the retrieval of kinematically accurate macro models suitable for subsequent iterations of high-resolution L2-norm FWI.

Main Objectives

Mitigating cycle skipping in FWI.

New Aspects

Use of partial matching filters to mitigate cycle skipping, giving more flexibility than alternative, existing approaches based on dynamic time shifts.

Summary

Existing methods for addressing cycle skipping in full-waveform inversion (FWI) typically involve either a modification of one of the data sets used to compute the least-squares objective function, or a reformulation of the objective function itself, often in terms of a traveltime (or equivalent) misfit. Both approaches can be successful, but they are reliant to varying extents on the notion of event similarity – that is, the requirement that the observed and modeled data contain the same, distinct, seismic events, even if the corresponding kinematics are different. We introduce a new technique for mitigating cycle skipping in FWI based on partial matching filters. The method accommodates amplitude differences between observed and modeled data, and does not require any major modification to an existing inversion engine. The proposed approach is validated on synthetic and real data sets, including an example where we observe a reduced reliance on event similarity compared to an established cycle skipping mitigation technique.

Main Objectives

Enhance towed-streamer seismic data with low frequencies by learning a bandwidth extension function from OBN data. The enhancement in low frequencies enables more accurate inversion of subsurface properties through FWI.

New Aspects

This abstract has two main novelties: 1) Sequence-to-sequence (Seq2Seq) learning of bandwidth extension; 2) Learning from OBN to enhance towed-streamer data shot over the same region.

Summary

The higher S/N at low frequencies in seabed seismic data enables the learning of a bandwidth extension function that can be used to enhance data deficient in low frequencies. By training on a very sparse ocean-bottom node (OBN) set, which is considerably cheaper to deploy than a full OBN set, the trained model can be applied in a “narrow generalization” sense to enhance simultaneously-shot towed-streamer data. The frequency-wavenumber representation of seismic data enables a fine-grained encoding of correlations between high-frequency slices, providing more “context” to extend the low-frequency band. We pose low-frequency extrapolation as a sequence prediction problem that we solve using a generic neural-network-based framework known as Sequence-to-Sequence (Seq2Seq). Results from our Seq2Seq bandwidth extension are presented on field OBN data with available “ground truth” low frequencies. We analyze through several experiments the impact of learning from very sparse OBN on full-wavefield inversion (FWI).

Main Objectives

Design a computationally-efficient frequency-domain wavefield reconstruction inversion (WRI) algorithm based on direct solver by reducing the number of LU factorization.

New Aspects

We reconstruct wavefields from inaccurate LU factors and iteratively refine the wavefields within the WRI iteration. We propose an optimal experimental setup of the method to conciliate computational efficiency and accuracy of the final velocity model.

Summary

We propose a double iteration refinement approach for wavefield reconstruction inversion (WRI).

The first is a refinement loop based on the alternating direction method of multipliers (ADMM), which solves the full waveform inversion (FWI), which is formulated as a PDE-constrained biconvex optimization problem, known as iteratively refined (IR)-WRI.

The second loop, which is the main subject of this abstract, enables decreasing the number of LU factorization required in the IR-WRI. Thus we call this new algorithm double iteration refinement .

The main computation burden of IR-WRI is due to the required LU factorization at each iteration to reconstruct the so-called data-assimilated wavefield.

Here, we use the LU factorization of an approximate static operator for this task, which allows us to perform the LU factorization only once prior to the IR-WRI iteration. Then we use a second inner refinement loop to compensate for the errors introduced by this approximation and clean up the wavefield.

Numerical examples show the effectiveness of this double iteration refinement approach for the efficient application of WRI.

Main Objectives

Improve the stability of elastic FWI when low-frequency data is missing and the source wavelet is inaccurate.

New Aspects

N-th power operation can recover some ultra-low-frequency information of seismic data. Convolved wavefield method is used to mitigate the source wavelet dependence of data after high-order power operation. A robust elastic FWI is proposed by using the n-th power operation and the convolved wavefields.

Summary

The elastic full waveform inversion (FWI) can use the recorded multi-component seismic data to construct high-precision multi-parameter models of the subsurface media such as P- and S-wave velocity models. However, due to the reasons such as data quality and algorithm limitations, there are still many problems in the promotion and application of elastic FWI method. Aiming at alleviating the influence of low-frequency data absence on the inversion results, we propose a robust elastic FWI method based on the n-th power operation. The n-th power of the seismic data can compress the time-domain waveform and expand its frequency-band. The FWI objective function constructed using the n-th power wavefields shows better convexity. By successively lowering the power during the inversion, we can realize a new multiscale FWI strategy, which is also a data-domain layer-stripping strategy. Seismic data will be more sensitive to the source wavelet errors after the n-th power operation. To mitigate this problem, we propose a robust objective function for elastic FWI using the n-th power operation and the convolved wavefields. Finally, the validity of the method is verified by numerical examples.

Main Objectives

Geology, Facies analysis, Seismic Attributes, Well Development, Residual oil

New Aspects

Seismic microfacies analysis, new potential development area, identification of favorable zones of remaining oil

Summary

Well operation accounts for potential geological and technological capabilities is one of the important factors affecting the efficient production of hydrocarbon reserves. The main tool for substantiating the technological efficiency of drilling new wells is a geological hydrodynamic network model of the field of development object. However, the process of creating a development plan is time-consuming, and the result, in certain cases, ambiguous. To address potential well development area in terms of residual oil, new integrated analysis workflow summarized based on the results which directly related to a reliable study of the sedimentation medium, in particular microfacies and various reservoir property data and production behavior of wells. The new workflow includes the following steps: 1. Study a well re-completion potential and idle wells conditions 2. Establish favorable phase areas for static analysis 3. Carry out dynamic parameters with an application of seismic inversion 4. Identify potential sites constrained by seismic, geological studies and initial production of the oil field 5. Provide suggestions in a new well development plan. This workflow has been applied successfully for the selection of potential zones for drilling new wells at the preliminary design stage, before creating a production network for the reservoir site model.

Main Objectives

Integrated modelling, gas production, condensate production

New Aspects

Calculating of the production of the field divided by the HC components (using PVT-modeling) without long-lasting compasitional integrated modelling of the large multilayer oil, gas and condensate field.

Summary

The experience and result of creating a full-scale integrated model of gas layers and gas caps of a large oil and gas field in Russia will be highlighted, and, accordingly, the search for optimal solutions to the problem of optimizing the joint development of oil rims, gas caps of productive layers and dry gas layers, taking into account the influence of the restrictions of the collection facility and preparation network system .

Main Objectives

optimal field development and safe operation

New Aspects

forecast of energy consumption of key process units

Summary

To identify the full potential of a brown field, it is crucial to validate its existing field development and production operation plan, to evaluate the feasible development strategies to further enhance oil recovery, to operate the asset safely and to optimise the overall reservoir recovery. These evaluation and assessment require inputs and close interaction across different domains, to reduce any potential gaps and to optimize its final decision making with considerations of various operational as well as economic factors.

By adopting an integrated asset modeling approach, it enables the ability to bring various technologies under one platform and subsequently allows one to accomplish an inclusive development planning study with consideration of optimized well placement of both producers and injectors (water or gas), requirement of new process units to accommodate the increased production.

This paper presents how an integrated asset model for an onshore brown field has been delivered to enable a comprehensive validation of the current field development strategy and its downstream process plant operation plan; as well as assessment of other possible development scenarios for the studied asset. The challenges faced are to identify the optimal development plan with minimal well count and maximum recovery.

Main Objectives

The purpose of this method is to process the multicomponent data and effectively image the anisotropic geological structures.

New Aspects

We proposed an anisotropic elastic dynamically focused beam migration by modifying the propagator of Gaussian beam. Meanwhile, based on the Kirchhoff-Helmholtz integral of two-dimensional anisotropic elastic wave, the elastic imaging weight coefficients are derived and applied to suppress the crosstalk noise.

Summary

In this paper, an elastic wave ray tracing systems is used to calculate travel time, trajectory of central ray and dynamic information. Then we proposed an anisotropic elastic dynamically focused beam migration by modifying the propagator of Gaussian beam. Meanwhile, based on the Kirchhoff-Helmholtz integral of two-dimensional anisotropic elastic wave, the elastic imaging weight coefficients are derived and applied to suppress the crosstalk noise. In addition, the sign function is introduced in this paper to solve the polarity reversal problem of converted wave imaging. Model tests have demonstrated that the proposed method in this paper can process the multicomponent data and effectively image the anisotropic geological structures.

Moreover, compared with the conventional anisotropic converted wave Gaussian beam imaging method, the research method in this paper can improve the deep-layer energy focus and effectively enhance the deep-layer amplitude energy under the premise of ensuring the accuracy of shallow imaging. The result of model test shows the accuracy and validity of the method.

Main Objectives

To reduce migration artifacts including crosstalk artifacts and acquisition footprint

New Aspects

Introducing an efficient and robust multi-parameter imaging tool based on regularized pseudo-inverse Born operator in the presence of density variations

Summary

Among different migration algorithms, least-squares reverse-time migration is the preferred choice for quantitative seismic imaging. The applicability of such scheme depends on the derivation of proper pre-conditioners. In the context of extended domain in a pure acoustic media, the pseudo-inverse Born operator is the recommended pre-conditioner, providing quantitative results within a few iterations, but limited to purely velocity variations. Recently, an efficient weighted least-squares approach has been proposed to extend the applicability of the pseudo-inverse Born operator in the presence of density variations. As expected in the case of multi-parameter imaging, the results using this method suffer from crosstalk artifacts. In order to mitigate this issue, we present variable density pseudo-inverse Born operator constrained with $\ell_1$-norm for each model parameter. The fast iterative shrinkage-thresholding algorithm is used to solve the optimization problem. This iterative scheme is based on soft-thresholding method where no wave-based operators are involved. Numerical tests are used to demonstrate the robustness of the proposed method against crosstalk artifacts and sparse shot acquisition geometry.

Main Objectives

Use time-shift extended least-squares reverse-time migration to produce angle gathers with physically meaningful amplitudes (proportional to angle-dependent reflection coefficient) under complex overburdens.

New Aspects

Perform time-shift extended iterative least-squares reverse-time migration followed by transform to reflection angles while properly handling reflection amplitudes in the presence of velocity variations.

Summary

I present a method for computing reflection angle gathers using time-shift extended least-squares reverse-time migration. The method is aimed at producing image gathers that can be interpreted in terms of angle-dependent reflection coefficients, also under complex overburdens. It is based on a two-step procedure involving an iterative inversion to estimate a time-shift extended representation of the subsurface reflectivity, followed by a transform of this reflectivity to the reflection angle domain. Using a formulation in terms of time-shift extended imaging allows to naturally handle complex wave-field effects like multi-pathing. The main building blocks of the method include an adjoint pair of time-shift extended linearized modelling and imaging operators, an effective time-shift extended-domain preconditioner and a transform from the time-shift domain to the reflection angle domain that properly handles amplitudes. The method is demonstrated on a synthetic data example.

Main Objectives

Estimate inverse Hessian operator for iterative Least-square Migration

New Aspects

Decomposing the complicated mathematical operator (Hessian) to a chain of elementary operators that carry the physical meaning

Summary

We approximate the inverse Hessian operator by a chain of weights in time/space and frequency domains. Tests on synthetic data show that this approach provides an effective approximation while having the minimal cost of forward and inverse FFTs (Fast FourierTransforms). The method can be applied either for compensating migrated images or in the form of a preconditioner inside iterative least-squares reverse-time migration (LSRTM). As demonstrated by experiments with synthetic data, the latter significantly accelerates the convergence of LSRTM and achieves high-quality imaging results in fewer iterations.

Main Objectives

improve RTM imaging by frequency dependent Q compensation

New Aspects

imaging in frequency domain with Q compensation

Summary

In anelastic media different frequency components of seismic waves propagate with different velocities and their energies are attenuated differently. Without properly accounting frequency dependent dispersion and attenuation, conventional RTM images suffer in phase distortion and reduced resolution for deeper horizons. Unlike other time-domain Q compensation approaches that uses an energy compensation ratio for each time and each image point, we believe the energy attenuation ratio is frequency-dependent and therefore we seek to compensate energy loss at each frequency individually. We proposed a frequency domain approach to account for dispersion and to compensate energy attenuation in reverse time migration. Instead of directly compensating energy loss in wave field propagation which is unstable and tends to amplify noises, we simulate wavefields with and without energy attenuation term, and estimate energy loss ratio in the frequency domain from simulated wavefields. The estimated frequency-dependent energy loss ratio is then used in imaging stage to compensate energy loss presented in the data. This approach is numerically stable but it requires a lot more computational times and disk spaces, GPU acceleration makes it feasible and practical. Synthetic and real date examples demonstrate the correct phase and improved resolution in images produced by this Q compensation RTM.

Main Objectives

Accelerating the convergence rate of the variable-density least-squares reverse time migration

New Aspects

Implementing the multi-parameter asymptotic extended pseudo-inverse Born operator as a preconditioner

Summary

Least-squares reverse time migration aims to iteratively fit the synthetic data to the observed data for a quantitative result. This is

a computationally expensive approach. Recently, an asymptotic

wave-equation based inversion under variable-density acoustics

assumption has been proposed. In this study, we use this asymptotic

extended pseudo-inverse operator to accelerate the convergence rate of the variable-density least-squares reverse time migration. This is implemented in a preconditioned Conjugate Gradient algorithm. Numerical examples demonstrate the effectiveness of the proposed method in accelerating the convergence rate while enhancing the image quality.

Main Objectives

Including converted waves in inversion

New Aspects

Converted waves included in forward modeling and inversion scheme

Summary

Full Wavefield Migration (FWM) is a full-wavefield inversion method based on the so-called WRW model in the context of seismic imaging. This WRW model describes seismic data in terms of convolutional propagation and reflection operators in the space-frequency domain. By recursively applying these operators, multi-scattering data can be described and the FWM algorithm can find the underlying reflection properties (i.e. the ‘image’). However, the current FWM algorithm only allows for acoustic imaging. This approach disregards elastic phenomena such as wave conversions. In this report, an extension to the FWM method is given to incorporate these effects. The resulting algorithm is then applied to a simple 1.5D model, to allow for proof-of-concept numerical simulations. The results of these simulations are given and show that the proposed model can model wave conversions and other elastic effects, yielding a good proof-of-concept for further research into this extended algorithm.

Main Objectives

We want to present a new thin reservoir prediction method, which combines sparse-reflectivity obtained by spectral inversion with the low-frequency model associated with logging data. We call this new method iterative spectrum inversion.

New Aspects

In this paper, an iterative spectral inversion method which successfully integrates the low-frequency information of logging data into the traditional spectral inversion is proposed. We can obtain a stable absolute wave impedance by using this method. In addition, we complete the formula derivation of ray theoretical velocity and equivalent theoretical velocity. We found that the inversion result is the bridge between seismic records and logging results. Then we proposed the crossplot analysis method under the constraint of the inversion frame, which plays a good guiding role for us to carry out inversion research.

Summary

High-resolution seismic exploration has always been a significant and concerned topic. Spectral inversion is a popular seismic inversion method. However, this method does not use the more reliable low-frequency information associated with logging.

In this paper, an iterative spectral inversion method which successfully integrates the low-frequency information of logging data into the traditional spectral inversion is proposed. We can obtain a stable absolute wave impedance by using this method. In addition, we complete the formula derivation of ray theoretical velocity and equivalent theoretical velocity. We found that the inversion result is the bridge between seismic records and logging results. Then we proposed the crossplot analysis method under the constraint of the inversion frame, which plays a good guiding role for us to carry out inversion research.

Main Objectives

To systematically describe the beam prestack depth migration theory and combine it with paraxial one-way wave to achieve high precision beam prestack depth migration imaging.

New Aspects

We use the 15° one-way wave equation in the ray-centred coordinate system and overcome the high frequency approximation of the Gaussian beam.

Summary

At present, there are mainly two kinds of seismic wave prestack depth migration imaging methods: the integral method based on the high frequency approximation solution of wave equation and the finite difference numerical solution or mixed domain solution based on the differential wave equation. Gaussian beam method is an intermediate method for describing wave propagation and imaging, and the problem is that too many approximations are introduced in the process of describing the wave propagation in the ray-centred coordinate system. In addition to high frequency approximation, the amplitude of wave field is merely derived from the amplitude Gauss attenuation of points along centre ray path, which is too simple to characterise the complex wave phenomena. Therefore, we use the paraxial one-way wave equation in the ray-centred coordinate system, and the paraxial one-way wave equation is used to propagate the wave field in the ray beam, so as to accurately describe the local wave field. The objective of this paper is to systematically describe the beam prestack depth migration theory and combine it with paraxial one-way wave to achieve high precision beam prestack depth migration imaging. Numerical results show the effectiveness of our method.

Main Objectives

(i) to construct fully numerical and rapid approximate solvers to simulate a radial fracture driven by Herschel–Bulkley fluid in a permeable reservoir; (ii) to explore the problem parameter space and to derive limiting propagation regimes; (iii) to analyze sensitivity of the solution to yield stress for different values of the flow index.

New Aspects

(i) adaptation of the numerical approaches published in literature to current problem; (ii) identification of two new limiting propagation regimes linked to yield stress; (iii) construction of the regime maps showing zones of high influence of yield stress.

Summary

We investigate the influence of fluid yield stress on propagation of a radial hydraulic fracture in a permeable reservoir. The hydraulic fracturing fluid rheology is governed by Herschel-Bulkley model including yield stress and non-linearity of the shear stress. The rock is linear elastic, and the fracture is formed due to fluid injection at a constant volumetric rate. The crack propagation criterion follows the theory of linear elastic fracture mechanics, and Carter’s leak-off law describes the fluid leak-off into formation. We developed two numerical approaches to compute the problem solution: fully numerical (Gauss-Chebyshev quadrature and Barycentric Lagrange interpolation techniques) and approximate (the global fluid balance equation combined with fracture tip asymptote). The presented simulations representing typical field cases demonstrate that the yield stress can lead to a fracture with a shorter radius and larger aperture compared to the radial fracture model with simpler power-law fluid. We derived limiting propagation regimes characterised by dominance of certain physical phenomena and built parametric maps showing their applicability domains. Such analysis enables one to identify whether the yield stress provides a substantial impact for any given problem parameters.

Main Objectives

To provide significant improvement for H-T-M simulation for extreme temperature ranges

New Aspects

Development of a thermal extension for commonly used constitutive models

Summary

Analysis of complex problems in thermal reservoir engineering, geothermal energy extraction, CO2 and waste storage, etc., require integrated (coupled) modeling techniques typically involving solving the geomechanics, thermal and fluid flow. However, the constitutive models are usually defined as isothermal even for thermally-dominant processes.

This approach is not sufficient for extreme conditions such as internal combustion, SAGDOX, or in-situ gasification. This paper describes a thermal extension of constitutive models for such conditions. Model is based primarily on data from oil sands but can be used with any underlying isothermal constitutive model. An example of its use for the caprock integrity analysis in a SAGDOX project highlights the importance of considering and measuring the dependence of mechanical parameters on temperature.

Main Objectives

Integrated subsidence analysis with model calibration via InSar data for an extended case study

New Aspects

The paper represents the first step in detecting a common deformation response, if any, sheared by the numerous UGS systems in the Po Plain basin.

Summary

The scope of the research is to cast light on the deformation characteristics of underground formations involved in gas storage (UGS) activities in the Italian Po Plain area. The reliable prediction of ground movement caused by reservoir compaction/expansion is mandatory for the safety of both storage systems and urban settlements (especially in high urbanized area) and it is dependent on a solid knowledge of the soil/rock deformation behaviour under changing stress.

Two case studies were presented in terms of subsidence analysis via a multi-disciplinary (static/dynamic/mechanical) 3D numerical simulation approach. A coherent and full dataset is available for each reservoir modelling, which includes also deformation/strength parameters from lab and logs and 10+ years of InSar surface movement data adopted for the geomechanical model calibration.

The cases show strong similarities in terms of structural/geological contest, formation lithologies and UGS strategies, among others. Based on the subsidence analysis results and the analysis of InSar data, a common correlation between induced pressure variation in each reservoir and corresponding ground movement was inferred, resulting in an equivalent deformation behaviour. The paper represents the first step in defining a common deformation response, if any, for the numerous UGS systems present in the Po Plain basin.

Main Objectives

Geomechanics of seismicity created by hydraulic fracturing

New Aspects

A new model explaining the occurence of half-moon events during hydraulic fracturing taking into account fracture tip effects

Summary

Half-moon events are a widely observed source mechanism during hydraulic fracturing. However, the stress conditions and driving mechanisms which cause these events are so far poorly understood. We propose a geomechanical model that can explain the occurrence of half-moon events. Due to the fracturing pressure, the natural stress field close to the fracture can be significantly changed. Under specific injection conditions, the principal stresses can become locally rotated, so that vertical or horizontal cracks can become favorably oriented. Such a rotation can either happen close to the fracture tip or close to interfaces between layers of different elastic parameters. For the second case, the fracture needs to cross a layer interface, whereas the first case is possible within a homogeneous layer. Using our model we can additionally constrain which of the two possible fault planes is ruptured. In a naturally normal faulting domain, as assumed for our model, half-moon events occur most likely on the vertical fault plane.

Main Objectives

Upgrading the structural restoration field

New Aspects

Using creeping flow equations for restoration, and the associated structural restoration scheme.

Summary

Restoration of geological structures is commonly used to assess past basin geometry from present-day structures. In geomechanical restoration, numerical methods to date consider the rock properties as fully elastic and the faults as frictionless contact surfaces. However, salt bodies have been proven to behave as Stokes viscous fluids in geomechanics, and faults appear in rocks reaching a plastic limit inside a shear zone.

In order to take these behaviours into account, we introduce a new geomechanical restoration scheme based on Stokes equations. Such a strategy seems reasonable for three main reasons. First, rocks have been found to be mainly ductile in large deformations under long time periods (1e5 to 1e9 years). Second, these equations allow the modelling of other rheologies and boundary conditions closer to natural ones. Third, the reversibility of the Stokes equations can be used to compute the reverse motion of a geological domain.

Our restoration scheme is implemented in a software called FAIStokes. The classic forward modelling part of this software is validated through relevant benchmarks. First tests, including van Keken (1997)’s benchmark, on the more innovative backward modelling part, show the great potential of the proposed scheme for restoration using only mechanical (i.e. no geometrical) conditions.

Main Objectives

Wellbore stability analysis in ductile formations

New Aspects

Borehole failure prediction in plastic regime based on laboratory core experiments modeling

Summary

Traditional methods of wellbore stability analysis use the assumption of elastic rock behavior and failure criterion, which defines the maximum allowable stresses for secure borehole drilling. These methods can lead to overestimating critical mud pressures and even the inexistence of a safe mud pressure window while planning the rocks’ drilling with significant plastic behavior. The integration of nondestructive plastic deformations of the wellbore and the fracture criterion for plastic materials can extend the estimations for safe mud pressures. In this paper, we formulate the algorithm for numerical modeling of spatial stress-strain state of rock masses in the vicinity of an inclined borehole, validated against analytical solutions, and finally validated the model with the lab pseudo-triaxial loading experiment. We give several examples of near-wellbore states for assumed inclined boreholes drilled in the plastic rock in the reverse fault regime.

Main Objectives

Compaction and subsidence assessment.

New Aspects

Calibration with GPS and Bathymetry data, validation with 4D seismic and including the geomechanical model in the loop for static and dynamic model updates.

Summary

A fit for purpose geomechanical model was built to enable an assessment of the compaction and subsidence. The 3D geomechanical earth model was calibrated with the bathymetry, GPS and validated using the 4D seismic and the results used to estimate the maximum subsidence of the two wellhead platforms on the Yadana Field. Using the modelled results, the expected subsidence at the WP1 is 6.13m in 2026 whilst a sensitivity analysis was also carried out using a low case reservoir pressure resulting in a low case maximum subsidence of 6.99m in 2026. This information is vitally important for the future planning of the Yadana field production operations and can be used to assess options for future tie-back opportunities. The recommendations from the geomechanical model were also taken into account for future updates to the static model and the reservoir model through assisted history matching.

Main Objectives

Geomechanics

New Aspects

Uncertainty Quantification

Summary

A novel methodology is developed to design a reliable safe mud window based on most updated geometrical uncertainty distribution. The developed approach allows to estimate the uncertainty ranges for geometrical parameters and their dependent parameters such as collapse pressure and fracture pressure. A trustable mud window design based on posterior probability reduces the risks of wellbore stability problems and less kick.

in this research, the Markov chain Monte Carlo simulation used to quantify the geomechanical uncertainties in order to make it more clear and trustable.

Main Objectives

Extension of traditional UFD concept

New Aspects

New optimum dimensionless conductivity value derivation for naturally fractured reservoirs

Summary

Improving the effectiveness of hydraulic fracturing requires careful selection of design parameters to achieve specific optimum hydraulic fracture dimensions. However, available optimization approaches solely rely on the assumption which implies the existence of the hydraulic fractures only. But this approach may be biased in case of reservoirs with natural fractures. Hence, assessment of the complex fracture network is crucial to examine the potential impacts of it on the production. Most widely accepted unified fracture design (UFD) method does not encounter the effect of natural fractures; therefore optimum dimensionless number, which was derived as 1.6, may change for the reservoirs with natural fractures system. The objective of this study was to extend the conventional UFD method and come up with a new dimensionless number which incorporates the effects of the natural fractures. Obtained results satisfy the intuitive expectations that natural fractures alter the conventional dimensionless number.

Main Objectives

Integrating rock physics and Quantitative seismic interpretation into basin modeling

New Aspects

New workflow for regional assessment of elastic moduli of reservoirs

Summary

This study presents a new integrative rock physics workflow for improving the prediction of elastic moduli in basin modeling. The workflow utilizes core data to define facies suitable for quantitative seismic interpretation to come up with a cube of most likely facies cube classified from inverted seismic attributes. The resulting most likely facies cube is used to populate facies definition in a high resolution basin model. A comprehensive rock physics template is then calibrated to well data and applied to the high resolution basin models to condition the porosity and elastic moduli outputs. When the workflow is applied to an existing regional basin model, the simulated elastic moduli of the basin model, e.g., Poisson’s ratio, show improvements in matching the elastic moduli values derived from the inverted seismic attributes and the well log data. The featured workflow for estimating elastic moduli from basin modeling after conditioning to rock physics can be combined with existing techniques of estimating elastic moduli to increase the confidence in such estimates and reduce the uncertainty.

Main Objectives

To develop an approach that delivers simple and efficient expressions for the vertical (subsidence) and horizontal displacement components of the free surface and can be used as nucleus for inverse problems and reservoir monitoring.

New Aspects

The use of Fracture Mechanics approach. The proposed approach gives the values of subsidence within the limits observed in the Groningen gas field, while the traditional Geertsma’s method gives the estimations which is 50% higher than the observed ones.

Summary

We propose an approach that delivers simple approximate expressions for the vertical (subsidence) and horizontal displacement components of the free surface for all its points. Also, analytical expressions for the point-like nuclei are determined that allow representation of a planar horizontal reservoir of arbitrary shape as a distribution of infinitesimal dislocation loops. For the case of cylindrical reservoir, the method allows recovery of the full distribution of surface displacements. The proposed approach gives the values of subsidence within the limits observed in the Groningen gas field, while the traditional Geertsma’s method gives the estimations which is 50% higher than the observed ones.

Main Objectives

Study of core recovery and formation of irreversible deformation in the core during the drilling process

New Aspects

3D elasto-plastic model for core recovery based on algorithms of simulating the process of a well creating and separating the core

Summary

We develop and study a 3D elasto-plastic model for core recovery based on algorithms of simulating the process of creating a well and separating the core. Estimates of the rock deformation processes are presented for different cases: taking into account localization zones of deformation, the effect of the drill tool, the presence of very plastic interlayers, a non-equiaxial initial stress state in the reservoir. Our model can be used for evaluation of safe core recovery in different natural reservoirs.

Main Objectives

Study of impurities over rheological behaviour of salt rock

New Aspects

Salt Caverns integrity for UGS

Summary

UGeological storage of gas is not only an effective means to mitigate CO2 emissions by injecting CO2 underground, but also an excellent way to accommodate fluctuations in energy peak demands by storage of CH4 and H2. The main aim of this project is to analyse the impurity content in salt rock formations, used for underground gas storage, and its influence on rheological behaviour and deformation to improve the stability of salt caverns.

Main Objectives

Improving the accuracy of pore pressure prediction in organic-rich shales

New Aspects

Quantiying the effect of TOC variation on log response prior to undertaking pore pressure prediction

Summary

Pore pressure prediction in shales undergoing compaction, including mechanical and chemical diagenesis, is customarily related to the mechanism referred to disequilibrium compaction. However, even when this mechanism is established and the normal compaction trend in sonic velocity, as a proxy for shale porosity, is well constrained, the pore pressure prediction may be in error because of the lithological variation in shale composition. Presence of high levels of organic matter in shales that are immature for hydrocarbon generation is an example, causing marked overprediction of pore pressure unless properly accounted for. All published datasets involving TOC and wireline data record a similar relationship between TOC and the bulk density and P-wave velocity log response, in the sense that the measured wireline data shows a decrease (which implies an increase in porosity) as the TOC content increases. In this paper it has been shown that a rock physics model that links TOC and bulk density can be utilised to correct the measured bulk density in immature shales, and, when limited to immature shale, the correction can be extended to velocity data using simple industry-standard models.

Main Objectives

Overpressure mechanism;pressure prediction

New Aspects

It is the first time to demostrated that the overpressure transfer plays an important role in overpressure generation in in continental basins in China.

Summary

In this paper, the log response–vertical effective stress and acoustic velocity-density crossplots are used to identify the characteristics and generation mechanisms of the overpressure in the Linnan Sag. The analyses of the acoustic velocity/density–vertical effective stress and acoustic velocity-density crossplots demonstrate that the overpressured points consistent with the loading curve. So, the disequilibrium compaction of the thick Paleocene mudstones is the fundamental mechanism resulting in overpressures. Hydrocarbon generation and vertical transfer may be the main unloading mechanisms, that correspond to the overpressure points that deviate from the loading curves. Since organic matter cracking may occur in formations at depths deeper than 3800m, the contribution of hydrocarbon generation to overpressuring should be limited. The transfer of overpressure via opening faults is therefore considered as the main cause of higher overpressure in local sandstones. The results of this analysis provide an indication of the magnitude, mechanism and distribution of the overpressure. This understanding will help to guide further exploration activities in the Linnan Sag and similar geological basins.

Main Objectives

Develop a robust workflow for CO2 plume monitoring by integrating various geophysical data.

New Aspects

Integration of gravity and seismic data for CO2 saturation plume monitoring. Here, a stochastic inversion approach is proposed which is able to quantify uncertainty in the predictions and thus increases the confidence in the estimates.

Summary

Carbon capture and storage in subsurface requires frequent monitoring of CO2 plume movement. The correct assessment of the spatial distribution of CO2 saturation front lowers the risk of leakage and thus environmental hazards. In this work, we propose a stochastic inversion framework, the iterative Ensemble Smoother (iES), to predict changes in CO2 saturation plume (ΔSg) using time-lapse seismic and gravity data simultaneously. Here, change in inverted acoustic impedance (ΔIp) is considered as seismic data. The methodology is based on a Bayesian inversion problem, where the prior is provided as an ensemble of changes in CO2 saturation. The realizations of ΔSg in the prior model are computed using geostatistical method and reservoir flow simulator. The updated ΔSg (posterior) are then obtained based on the misfit between simulated and measured geophysical data. The proposed framework is applied and validated on a 3D reservoir model based on the Johansen formation which is a potential large scale offshore CO2 storage site in the North Sea. The numerical results demonstrate that the proposed

framework is capable to monitor CO2 plume movement efficiently.

Main Objectives

Bridge the gap between 1D well-based geomechanical models and 3D/4D computational geomechanical models

New Aspects

1.) Method to tightly integrate 1D models and 3D geomechanical simulations 2.) Use of tectonic strain term in 1D geomechanical model to calculate displacement boundary conditions in 3D/4D geomechanical models

Summary

Numerical geomechanical models require the application of boundary conditions at the outer limits of model. Either stress or displacement boundary conditions can be applied. The magnitude of the applied boundary stresses or displacements markedly affects the horizontal stress state in the model. Hence choice of boundary conditions is an essential step in computational geomechanical simulations. We present a method that uses the tectonic strain term from log-based 1D geomechanical models to calculate displacements for use in 3D and 4D geomechanical computation geomechanical models. Applying the method to a field dataset demonstrates a very good match between stresses in the 1D and 3D geomechanical model. The 3D/4D geomechanical model is automatically calibrated to the 1D geomechanical model. The presented method results in a tight integration between 1D and 3D geomechanical models and makes full use of the knowledge generated from the calibration of 1D geomechanical models to well-log observations. Using displacement boundary conditions also allow for smaller computational domains than using stress boundary conditions, and result in faster compute times.

Main Objectives

Seismic imaging of geothermal reservoir and time lapse

New Aspects

Using DAS , DTS and surface geophone to image geothermal reservoirs

Summary

Following our first seismic study at the Medipolis geothermal field in southwestern Japan in 2018, we conducted a second seismic study at the same geothermal field in 2019. We installed an optical-fiber system for distributed temperature sensor (DTS) and distributed acoustic sensor (DAS) measurements. We deployed the optical-fiber system at a 1,545-m depth in the IK-4 borehole. The temperature was measured to be 272.8 °C at a 920-m depth and 152.8 °C at a 1,530-m depth. We operated a MiniVib seismic source at five locations and performed a frequency sweep of 10–75 Hz 480 times each day, for seven days. We cross-correlated the seismic records and the source signature and stacked the correlated data to enhance the S/N. Stacking for 480 or 960 times considerably improved the arrival waveforms. Based on an analysis of DAS data, we constructed the 2D seismic profile. We estimated three major hydrothermal layers, at depths of 800–1,000 m, 1,300–1,600 m, and 3,600 m. The zone around 3,600 m suggests a high Vp/Vs value and the possible presence of a fluid layer.

Main Objectives

To give the audience an update of the multi-year SCAN program and to present first results of the seismic acquisition and processing.

New Aspects

2D seismic acquisition in NL is nothing particularly new, but we feel it is important to highlight the Dutch effort to accelerate the energy transition by making subsurface data available through government funding.

Summary

To achieve the Dutch policy objective to reduce carbon emissions by 49% in 2030, a shift from fossil towards renewable energy resources is required. Within the Netherlands, Geothermal Energy is a proven renewable energy resource, but currently only with a limited number of operating installations, which are located in areas of good subsurface data coverage.

However, subsurface data coverage is poor in roughly half of the country, including major residential and industrial areas with high heat demand. Improving the data coverage in these areas would increase the benefit-risk ratio of geothermal projects, which would greatly support the development of these projects.

To address these data shortcomings, Energie Beheer Nederland B.V. (EBN) and the TNO Advisory Group for Economic Affairs (TNO-AGE) have embarked on a geothermal exploration campaign, which includes reprocessing of vintage data as well as the acquisition of new, long-offset 2D seismic to improve the subsurface imaging and allow reliable interpretation to a depth of at least 6km. At a later stage deep research wells will be drilled and all subsurface data will be made public.

This paper will give an update regarding the geophysical work program, discuss results achieved by June 2020 and outline further plans.

Main Objectives

A regional geological evaluation as input for a 2D seismic acquisition programme for geothermal energy

New Aspects

Definition of new exploration strategies in poorly explored areas of the Netherlands, including a model for increased permeability below unconformity surfaces and Rotliegend thickness variations due to paleotopography

Summary

In 2018 a geothermal exploration campaign named SCAN was started to decrease the subsurface uncertainties in the less explored areas of the Netherlands. This program includes the reprocessing of vintage seismic and the acquisition of new 2D seismic lines. Finally, the intention is to drill geothermal research wells. In order to focus the SCAN research and exploration activities a regional geological study was embarked upon to define the plays and map geothermal leads. Some new or improved geological concepts can be envisaged. One is based on the observation that below unconformity surfaces improved reservoir quality is often found in the subcropping formations. This difference in reservoir quality can be explained by the leaching of unstable minerals and cements due to infiltration of fresh surface water. Secondly, exploration in the Dutch offshore has shown that the facies and thickness of the Rotliegend sandstone reservoir is highly dependent on paleotopography, which was formed after the Hercynian tectonic phase. In the SCAN areas a conceptual model for the Rotliegend thickness and facies trends can be made. This model could help to target areas where the Rotliegend reservoir may be thicker and thus more attractive to drill.

Main Objectives

Environmental and cost efficient geothermal projects

New Aspects

Innovative hydrogeological solutions

Summary

For a project near Oslo, four deep geothermal wells were planned to 1500 m in order to circulate water in a closed geothermal system. However, due to large influx of water, the drilling had to be stopped short of this depth target. Consequently, the closed geothermal system was discarded, to be replaced by an open geothermal system. Ruden AS were tasked with finding a way to extract 24 L/s of water at a temperature of 12°C, thereby producing the planned amount of heat with wells of less than half the planned depth. Based on the combined well logging and pumping procedure a solution was presented having an energy output similar to a much deeper and more costly well field.

Main Objectives

To update the seismic catalog of induced events reported by the BGS, by refining its epicentre location and focal mechanisms of the highest-magnitude events by complementing the seismic data recorded by the UK national seismic network, with local, low-cost, short-period stations. Also, to evaluate a novel magnitude-forecasting method for the same evens (associated with the operation of two deep geothermal wells) without using injection and production data from these wells (not available for this study).

New Aspects

We observed a NE-SW structure from the updated seismic catalog and focal mechanisms (associated with the Porthtowan Fault Zone targeted for this geothermal field) not visible from the original seismic catalog released by the BGS. Also, the evaluated method for forecasting the magnitude of induced events performed reasonably well despite not having any injection or production data from the geothermal wells.

Summary

The UDDGP located near Redruth, Cornwall, UK, consists of two adjacent deviated wells drilled in a hot granitic formation at depths of 5275 and 2329 metres, and intersecting the Porthtowan Fault Zone to allow a natural hydraulic communication between the wells. Shortly after flow-testing operations began in August 2020, a series of induced seismic events, with a maximum magnitude of 1.7 and a maximum intensity level 3 (i.e., weak vibrations) have occurred. Although these magnitude and intensity levels don’t represent significant seismic hazard to local communities or infrastructure, it’s important to have a detailed characterization of the microseismicity associated with operational wells to better understand the geomechanical processes associated with injection, and to assess the likelihood of continued operations leading to the occurrence of higher magnitude induced events. We utilize publicly-available data from stations within 20 km of the UDDGP site, including one BGS national network station and eight Raspberry Shake stations with short-period, vertical geophones, to relocate the epicentres of the induced events reported by the BGS, and to extract the focal mechanism of the highest-magnitude events. We then use the observed event magnitudes to forecast expected event magnitudes should induced seismicity continue to be generated at the site.

Main Objectives

Improving the accuracy of the estimated geothermal doublet performance based on hydrocarbon well data

New Aspects

The use of geothermal production data

Summary

Now that four geothermal projects have successfully been completed in the Slochteren Formation, the

formation has proven as a valid aquifer for geothermal heating purposes. Before drilling a project, the local reservoir parameters need to be determined in order to determine the economic feasibility of a project. However, predicting this is not an easy task: permeabilities are often strongly overestimated, while temperatures and (net) reservoir thicknesses are often underestimated. This has several causes: first, the available resources for geothermal projects are limited. Therefore, geothermal feasibility studies need to be done with limited amount of time using publicly available data, mostly from hydrocarbon wells. Second, these hydrocarbon wells are drilled with a different objective than geothermal wells, which introduces difficulties when translating the datasets into one another. Thirdly, not all planned geothermal project locations have a high density of well data, which can introduce a large uncertainty in interpreted reservoir properties. As a result, comparing data from these indicative studies to the actual results after drilling of the well discloses certain discrepancies. This study helps to understand where these discrepancies come from, and may improve the accuracy of the estimated geothermal doublet performance on basis of hydrocarbon well data.

Main Objectives

DAS-seismic method for super-critical geothermal development

New Aspects

Improved DAS-seismic method for geothermal applicaiton

Summary

To explore unknown geothermal reservoirs, we carried out a geothermal seismic study using a distributed acoustic sensor (DAS) at the Ohnuma geothermal power plant ownd by Mitubishi Material Co. in September 2020. The Ohnuma geothermal power station, the third commercial geothermal power plant in Japan, was completed in 1974. We installed an optical fibre system for the DAS measurements, on the site. We deployed the optical fibre system down to a depth of 1973 m in the O-13R borehole. To enhance the S/N, we stacked the DAS data for 480 times and correlated it with the source signature. By stacking for a long duration, we obtained excellent DAS records down to the bottom of the boreholes. Using 2D migration of observed and synthetic DAS seismic records, we recognised intense seismic reflections from 2.8–3.0 km depth, suggesting the possibility of geothermal reservoirs. The velocity decrease in this zone could be more than 1 km/s, possibly implying that the fracture zone is filled with fluid. The two field studies in the Medipolis and Ohnuma geothermal fields in Japan showed that the DAS-seismic method in the borehole can efficiently image seismically reflective zones, and the findings suggest high possibility of geothermal reservoirs.

Main Objectives

Quantitative interpretation for the geothermal purpose

New Aspects

Quantitative interpretation of legacy seismic and well data

Summary

To increase the probability of success of exploration for geothermal reservoirs while limiting the costs, a methodology based on the quantitative inversion of legacy seismic and well data is proposed. The five seismic lines chosen are from the 80s and cover part of the Dogger of the Paris Basin (France). The processing is standard techniques and aims at preserving amplitude. From the geothermal and oil wells, statistical relationships between impedance and porosity are established and clear trends for the limestone, and shales facies are identified. The results of the inversion and the following translation to porosity allows identifying the high porosity convex oolitic lens characteristics of the producing geothermal bodies of the Dogger. Despite the uncertainties, the workflow allows reducing exploration risks for a fraction of the cost of a new acquisition.

Main Objectives

Improve the Q-compensated FFD WEM in the TTI media

New Aspects

Propose a method for the Q-compensated FFD WEM in the TTI media

Summary

Attenuation has an influence on a seismic wavefield’s amplitude and phase. Because the Q effect strongly relates to the frequency, it is natural to include Q compensation in the one-way Fourier finite-difference wave-equation migration (FFD WEM) that can produce images with high quality and resolution, with a much lower cost than reverse time migration when there are no steeply dipping structures (e.g., above 75°). FFD WEM propagator involves three components: 1) a phase-shift term, 2) a thin-lens term, and 3) an FFD term. The complicated FFD term has a large impact on the accuracy of imaging structures with relatively large dips. We propose an improved Q compensation scheme for FFD WEM in tilted transversely isotropic media. Analogous to frequency-dependent phase velocity, we use a group of coefficients at each frequency, which are calculated by combining linear and nonlinear inversions to increase the accuracy of the least-squares estimation. By approximately including the Q effect in all three components of the FFD WEM, we increase the simulation accuracy for the wave propagation in relatively large dip angles. Both synthetic and field data examples show the validity of the proposed methodology in producing Q-compensated images.

Main Objectives

Avoiding the risks of the distortion using Helmholtz ´ decomposition and the energy leakage using the vector-based elastic-wave equations.

New Aspects

Decoupled propagator in the medium with heterogeneous Lame parameters

Summary

Elastic reverse-time migration (RTM) has shown significant advantages in obtaining depth-domain multi-wave imaging results, but also remains a considerable challenge in its practical application partly owing to the complexities of multi-mode elastic wavefields in the heterogeneous medium. We present a decoupled P- and S-waves propagator to form an efficient elastic RTM framework, without the assumption of homogeneous Lame´ parameters. Also, there is no mode conversion occurs using the proposed propagator even in the case of shear modulus discontinuities, avoiding the imaging artifacts caused by the unphysical wave-mode conversion. In the proposed elastic RTM framework, the source-side forward wavefield is simulated using a P-wave propagator. The receiver-side wavefield is back extrapolated using the proposed propagator, with the recorded multicomponent seismic data as input. Compared to the elastic RTM framework, the proposed framework reduces the computational complexities effectively but nearly preserves the imaging accuracy. We demonstrate its accuracy and efficiency using two synthetic examples, from which the proposed method shows comparative results but superior efficiency.

Main Objectives

Improving P- and S-wave imaging quality in anisotropic medium

New Aspects

Least-squares TTI migration for improve imaging quality

Summary

In traditional anisotropic elastic reverse-time migration (RTM), P/S-wave decomposition is necessary and requires large memory and computational cost. In addition, finite acquisition apertures and band-limited source functions result in unsatisfactory resolutions and amplitudes. To mitigate these problems, we present an elastic least-squares imaging method for tilted transversely isotropic media. Unlike traditional elastic RTM, we use the relative perturbations of stiffness parameters i.e., δlnC33 and δlnC55, as PP and PS reflectivity models, and estimate them by solving a linear inverse problem. Numerical experiments illustrate that least-squares migration helps to improve spatial resolution and image amplitudes compared with the adjoint-based migration, and enhance the contributions of weak S-wave to estimated subsurface reflectivities, producing high-quality δlnC33 and δlnC55 images.

Main Objectives

3D image using GBM

New Aspects

3D dip-angle gather optimization

Summary

Gaussian beam migration (GBM) is effective and robust to overcome the multipathing problem. As a ray-based method, it carries explicit angle-information naturally during the propagation. Different from that in scatter-angle gathers, reflectors with different spatial geometries produce different responses in dip-angle common-image gathers (DCIGs). We develop 3D-GBM in the dip-angle domain with optimizing gathers using different transform-method. The dip-angle-gather computation is based on geometrical optics, and multiple angle conversions are implemented under the rules of solid geometry, which helps to avoid rounding errors and improving accuracy. Additionally, the multi-azimuth joint presentation strategy is proposed to describe the characteristic of omnidirectional dip angles using a finite number of gathers. Numerical results demonstrate that the proposed GBM in the dip-angle domain provides high-quality migration results, and the optimization-method is feasible and adaptable for imaging a land survey.

Main Objectives

Estimating quantitative elastic parameters within a single iteration

New Aspects

Extending the approximate inverse Born operator from acoustic to elastic media

Summary

Several modifications of reverse time migration provide different expressions for the approximate inverse of the extended Born modeling operator. These approaches, often referred to as true-amplitude migration, estimate quantitative results within a single iteration while having roughly the same computational cost as reverse time migration. The main limitation is the acoustic assumption, leading to inaccurate amplitude predictions as well as the ignorance of S-wave effects. Assuming marine towed-streamer data, we extend the approximate inverse Born operator from acoustic to elastic media to estimate density, P- and S-wave impedance perturbations. These parameters are simultaneously inverted from the angle-dependent response of the elastic approximate inverse Born operator using a weighted least-squares method.

Main Objectives

Proposing a novel joint elastic wave imaging of Towed Streamer and OBN/OBS data to improve the problem of footprints caused by the sparse sampling, and obtain accurate elastic images of complex underground structures in deep-sea oil-gas exploration.

New Aspects

Based on acoustic-elastic coupled equations, we propose a novel method of elastic wave joint imaging of dense Towed Streamer (TS) and sparse OBN/OBS seismic data, including joint elastic reverse-time migration and joint elastic least-square reverse-time migration, to improve the imaging problems caused by ocean-bottom sparse acquisition.

Summary

Elastic wave imaging with ocean bottom four-component (4C) seismic data has unique advantages in offshore oil-gas exploration, but for sparse Ocean Bottom Node/Ocean Bottom Seismometer (OBN/OBS) 4C seismic data, the problems of imaging acquisition footprints, poor phase continuity, and low signal-to-noise (S/N) ratio still exist. Based on acoustic-elastic coupled equations, we propose a novel method of elastic wave joint imaging of dense Towed Streamer (TS) and sparse OBN/OBS seismic data, including joint elastic reverse-time migration (J-ERTM) and joint elastic least-square reverse-time migration (J-ELSRTM), to solve the above problems. J-ERTM of TS and OBN/OBS data can improve the problem of imaging acquisition footprints, but with low S/N ratio. J-ELSRTM uses Hessian information and can theoretically suppress the negative impact caused by sparse acquisition, therefore can obtain better images than J-ERTM with higher S/N ratio. The results of synthetic data and field data show that our method is of prime importance for improving the elastic wave imaging problems of sparse OBN/OBS data in deep-sea oil and gas exploration.

Main Objectives

We will present a new objective function for image-domain least-squares reverse-time migration through point spread functions in order to reduce migration artifacts and improve image resolutions and subsalt imaging.

New Aspects

We propose a new objective function for imaging-domain least-squares reverse-time migration through point spread functions. The objective function incudes a regularization of the difference between a migration image and a reflectivity model and with a total variation (TV) regularization of the reflectivity model. The regularization of model change reduces artifacts and makes a reflectivity model to be similar to a migration image, while the TV regularization maintains structural continuities. For complex subsurface structures with salt or dirty salt bodies, a weighting function is necessarily applied to the migration image in order to avoid degrading imaging quality. The objective function is minimized through a nonlinear conjugate gradient method. We applied the method to complex structures with slat and dirty salt bodies. For the first time we show that the method helps improve subsalt imaging.

Summary

Reverse-time migration (RTM) produces image of subsurface structures. However, migration image can be distorted due to spatial aliasing, limit record aperture, illumination effects, and so on. To partially correct for the distortion effects, we present an image-domain least-squares reverse-time migration (LSRTM) by approximating the Hessian through point spread functions (PSFs) with a regularization of the difference between migration image and a reflectivity model and a total variation regularization of the reflectivity model. The optimization problem is iteratively solved by a nonlinear conjugate gradient method for an optimal reflectivity model. In order to improve image quality and subsalt imaging, a weighting function is necessarily applied to migration image for complex structures with salt or dirty salt bodies. Tests demonstrate that the proposed LSRTM is helpful to remove migration artifacts, improve image resolution and subsalt imaging in optimal reflectivity models.

Main Objectives

Computationally efficient RTM and FWI

New Aspects

Computationally efficient and accurate snapshot-free imaging and FWI

Summary

Computing images in reverse time migration and model parameter gradients from adjoint wavefields in full waveform inversion require the correlation of a forward propagated wavefield with another reverse propagated wavefield. Although in theory only two wavefield propagations are required, one forward propagation and one reverse propagation, it requires storing the forward propagated wavefield as a function of time to carry out the correlations which is associated with significant I/O cost. Alternatively, three wavefield propagations can be carried out to reverse propagate the forward propagated wavefield in tandem with the reverse propagated wavefield. We show how highly accurate reverse time migrated images and full waveform inversion model parameter gradients for anisotropic elastic full waveform inversion can be efficiently computed without significant disk I/O using two wavefield propagations by means of the principle of superposition.

Main Objectives

approximate migration to zero offset, application of the NIP wave theorem to PSDM

New Aspects

Fresnel zone residual migration to zero offset, improvement of S/N ratio

Summary

The concept of residual migration to zero offset is introduced for the case that a prestack migration and a subsequent residual moveout analysis have been performed for a seismic survey and a new depth model has not yet been determined. Travel times of reflected events for individual traces are determined for diffraction points situated along horizons of analysis. These events are downward continued into the model used for the migration . The NIP wave theorem is applied to events exhibiting small subsurface offsets in order to determine residual radii of curvature to be used in the updating of seismic velocity model by seismic stripping. Alternatively, it is suggested to construct aplanats and to determine focus points at zero offset of the reflected event without knowledge of the true velocity model. The distance of the estimated point of focus from the observed zero offset response of the migrated reflector serves as an indication of the focusing of the event at the particular subsurface offset and can be used as a measure for the necessity of a subsequent remigration with a new velocity model. Applications to model computations and to a seismic survey over an overthrust structure show promising results.

Main Objectives

Simple forecast of seismicity

New Aspects

Compaction is only driver of gas reservoir seismicity

Summary

Gas reservoirs in different regions have induced seismicity, with magnitude up 4.5. We review cases in Southern France, western Canada and the Netherlands.

The detailed mechanism is not well understood, but two factors appears to be critical: compaction and faulting of the reservoirs. A plausible explanation has been proposed as differential compaction where part of the compaction energy remains stored at reservoir compartment boundaries, which can be released in seismic slippage. Although details, such as the role of pore pressure in the fault zone and fault properties are poorly understood, the differential compaction mechanism is confirmed by geomechanical modeling. Some examples of these models will be used for illustration.

These fields have been produced with varying rate, which correlates with seismicity, but the cumulative seismic moment is shown to be just a function of depletion. This provides a simple prediction of seismicity after about 50% depletion has occurred. Since seismicity only depends on compaction, there is little scope for management of seismicity: only pressure maintenance appears to be a viable solution. This can be accomplished by injection to preserve the mass balance or by shutting in gas fields.

Main Objectives

Study the temporal evolution of stress changes and induced seismicity for a fluid injection experiment at the Basel geothermal site and for a generic scenario of gas production. Investigate the conditions under which TLSs may fail in limiting earthquake magnitude.

New Aspects

Comparison of ‘characteristic earthquake’ patterns and TLS performance in fluid injection experiments and gas production. Investigation of limiting factors in the use of TLS in gas production and fluid injection.

Summary

Traffic light systems (TLSs) for limiting the strength of induced seismicity are used in different energy technologies. We use physics-based numerical models for investigating under which circumstances TLSs may not provide a robust mitigation measure. For seismicity induced by fluid injection, TLS efficiency can be limited by trailing effects caused by post-injection pressure diffusion and stress concentrations at the periphery of previous seismic activity. Seismicity caused by gas production exhibits a ‘characteristic earthquake’ pattern where earthquakes with similar (maximum) magnitude occur in the course of reservoir depletion. The characteristic earthquakes reflect repeated slip of the same reservoir fault patches. The maximum earthquake magnitude of a sequence can occur without precursors. Although trailing effects do not occur in an idealized reservoir with infinite conductivity, the lack of precursory seismicity limits the robustness of a TLS.

Main Objectives

Improved hazard analysis for induced seismicity. This is an invited presentation for the dedicated session on Induced Seismicity

New Aspects

New and improve forecast capability for induced earthquake magnitudes

Summary

Within probabilistic seismic hazard analysis, the exceedance probability of seismic moments, M₀, is treated as a pure power-law distribution, ~M₀^β, where the power-law exponent, β, may vary in time or space or with stress. Insights from statistical mechanics theories of brittle failure within heterogeneous media, statistical seismology, and acoustic emissions experiments all indicate this pure power-law may contain an exponential taper, M₀~M₀^β exp(-ξM₀), where the taper strength, ξ, decreases with increasing stress. The role of this taper is to significantly reduce the probability of earthquakes larger than 1/ξ relative to the pure power-law. This effect may appear as an apparent increase in  values with stress if taper effects are ignored.

Through a combination of Bayesian inference, hindcast evaluations, and forward-model simulations we assessed the forecast performance capabilities of a wide range of models for stress-dependent β- and ξ-models. Our results show that the stress-dependent ξ-model with constant β likely offer higher performance forecasts than the stress-dependent β-models 75 – 85% of the time. This model also lowers the magnitude with a 1% chance of exceedance over the next 5 years under a given gas production scenario from 5.5 to 4.3, corresponding to a 30% reduction in the seismic hazard.

Main Objectives

Link styles of permeability evolution with seismic versus aseismic fault reactivation to determine fugitive emissions from reservoirs and aquifers

New Aspects

Define seismicity-permeability relationships for faults and fractures.

Summary

The presence of pre-existing faults and fractures in the upper crust contribute to induced seismicity as a result of fluid injection, in hydraulic fracturing, deep storage of CO2, and stimulation of EGS reservoirs. In all of these, either maintaining the low permeability and integrity of caprocks or in controlling the growth of permeability in initially very-low-permeability shales and geothermal reservoirs are key desires. We explore styles of permeability evolution using both experimental and computational methods to explore how fracture permeability changes in response to fracture/fault reactivation and investigate the roles of (1) mineralogy and (2) fracture roughness in conditioning response; together with (3) intrinsic controls of healing on the earthquake cycle and permeability evolution.

Main Objectives

study hypocentre-fault relationships in the Groningen field

New Aspects

-Combining a more detailed fault model from ant tracking results with a high-resolution earthquake location dataset, it is now possible to study hypocentre-fault relationships in more detail and with higher confidence -The majority of the hypocentres are located on the deep-seated NW-SE fault systems. -Fault geometries can now be extracted for individual earthquake hypocentres.

Summary

The Groningen gas field is the largest gas field in Europe and has experienced induced seismicity since 1991. Faults at reservoir level play a major role. Combining a more detailed fault model from ant tracking results with a high-resolution earthquake location dataset, it is now possible to study hypocentre-fault relationships in more detail and with higher confidence.

Seismic reflection data and attribute extractions show that most faults in the Groningen area extend from the Rotliegend reservoir into the underlying Upper Carboniferous strata. Many of these faults extend even further into the Lower Carboniferous. The interpretation of these faults in seismic reflection data indicates the presence of large deep-seated NW-SE strike-slip fault systems.

When comparing the updated hypocentre locations with the ant tracking fault map at Top Rotliegend, we observe that almost all events can be linked to a fault. The majority of the hypocentres are located on the deep-seated NW-SE fault systems. Fault geometries can now be extracted for individual earthquake hypocentres. These analyses on faults associated with 16 larger events (ML≥2.0) indicate a dominant NW-SE strike direction, while the average dip of the faults is 68 degrees. These detailed fault characteristics improve the input parameters in geomechanical analyses.

Main Objectives

Understand risk management.

New Aspects

The importance of realistic scenario’s vs. most likely scenario’s

Summary

In the Netherlands there has been a strong focus on induced seismicity related to mining activities, especially gas production. This paper discusses the regulator view on the risk management of induced seismicity. What are the methodologies used? How to handle the results of quantitative and qualitative risk assessments. With a special focus on the importance for risk management to work with all realistic possible scenarios, and not only the most likely scenario.

Main Objectives

In this paper, we would like to discuss the main pain points operators are facing today and how digital solutions can help explorers minimize the gap between pre- and post-drill resource estimations.

New Aspects

Better exploration planning requires easy access to all meaningful data and knowledge, associated with a consistent approach applied to every prospect. New digital experiences bring automation, improving explorers’ agility and adaptability to technical and market changes. Digital experience also means new ways of working, communicating, and collaborating—bridging the gap between technical and decision-making teams.

Summary

Digital technologies and innovations are disrupting our industry by bringing new cloud-native solutions to enable better planning and execution of exploration workflows. Being able to standardize a portfolio prioritization process globally; instantly ranking and updating with crucial data or market changes, is key to increasing planning efficiency and transparency. The examples provided in this paper show that those new digital solutions break down silos and enable closer collaboration between technical teams and decision-makers.

The standardization of the portfolio prioritization process is expected to play a key role in dramatically reducing early and costly exploration spend by focusing on the opportunities that matter. In addition, the standardization associated with new ways of working is expected to minimize human bias and reduce competition amongst exploration teams. This is expected to have a clear impact on reducing the gap between pre- and post-drill resource estimations.

Finally, capturing and sharing experience and knowledge is key for a sustainable future and we hope that these new solutions, powered by the cloud, will help in attracting and retaining key talent in the oil and gas industry.

Main Objectives

Non invasive subsurface temperature measurement

New Aspects

Remote sensing of temperature through radar waves

Summary

Technological advances and depletion of easily extracted oil reserves

have led to the development of enhanced oil recovery (EOR) methods that

allow significantly more oil to be extracted from a reservoir. An increasingly

commonly used technique uses thermal injection, which requires a good

knowledge of the subsurface thermal conditions. Temperature observation

wells (TOW) are used to measure subsurface temperature profiles, an

expensive and invasive process.

We present a noninvasive method for the remote monitoring of

subsurface temperature using low frequency radar pulses.

Radar surveys were performed at 21 locations near TOWs in an oilfield

and returns were correlated, after signal processing to extract the modulation,

with measured down hole temperatures by training a feedforward neural

network. The results were evaluated by excluding one of the 21 data sets

from training and use the remaining 20 data sets to predict the excluded site, resulting in 21

blind tests. We believe results are encouraging, though not yet

fully reliable and we discuss avenues for improvements.

Main Objectives

Distributed Fibre Sensing for geothermal production monitoring

New Aspects

Combination of various Distributed Fibre Sensing solutions

Summary

Distributed Fibre optic Sensing is an innovative technology enabling to turn a fibre optic cable into hundreds to thousands of sensors. This solution can be adapted for temperature, strain or acoustic sensing using a same optical fibre. It then appears as a powerful leverage for the monitoring of Enhanced Geothermal Systems during geothermal production. In the framework of the European funded project MEET, FEBUS OPTICS has been mandated for deployment and exploitation of a fibre optic cable in an observation well located in the geothermal plant of Soultz-sous-Forêts. Distributed Temperature Sensing, Distributed Strain and Temperature Sensing and Distributed Acoustic Sensing monitoring have been achieved at various moments after the fibre optic cable descent. We observe and discuss here the results of those acquisition campaigns.

Main Objectives

Raise awareness regarding a new technology

New Aspects

new generation of instrument going beyond th state of the art in terms of scientific and operational features

Summary

A new generation of absolute gravimeters based on quantum high-precision measurement starts to be operated in the field for reservoir monitoring. We will present the Absolute Quantum Gravimeter developed by Muquans, its principle of operation, its key advantages and measurement performances, and also the first case studies that it is addressing.

Summary

Velocity model estimation continues to be one of the most important steps in seismic velocity estimation. Full Waveform Inversion has become the tool of choice in building velocity models in complex cases; but, is that really the right direction. There are certainly some cases where FWI performs well, and might be the best we can do. However, I argue that we can get better results than we often expect from older tools like reflection tomography when we fully understand the physics of reflection data and how migration interacts with velocity anomalies. For the most complex cases, we clearly need wave equation based tools, but might there be an approach that uses migration and global optimization to avoid having to predict data in the FWI framework? I will show a simple experiment that suggests that there is power and utility in directly optimizing migrated images using a fast migration approach.

Main Objectives

Real-Time Lookahead, Base of Salt identification, Mitigate drilling hazards

New Aspects

First time operator used real-time lookahead technology to mitigate drilling risks in salt to identify changes in formation ahead of the bit.

Summary

Seismic While Drilling is used effectively to identify the base of salt in a complex salt structure in order to minimize drilling risks in a highly uncertain environment.

Main Objectives

Using neural networks to imitate and potentially replace seismic workflows

New Aspects

Synthetically trained neural networks applied to seismic data processing

Summary

Neural networks that learn to map input data to outputs have proven to be efficient tools applicable in a range of domains, including geoscience. We demonstrate the use of conditional generative adversarial networks (cGANs) in order to imitate and potentially replace seismic workflows related to processing and interpretation, such as denoising, multiple removal and seismic attribute generation. With a case study on pre-stack data, we show that a cGAN trained on synthetic gathers (with and without multiples) successfully remove multiples from real gathers even when the multiples interfere with the primary signals. With conventional radon processing, interference between multiples and primary signals can be difficult to address when the dips are similar. The cGAN approach is significantly faster than conventional radon processing and indicates a potential to be better, particularly on the near offset.

Main Objectives

To inform the audience about quantum computing and discuss some possible use cases in geosciences.

New Aspects

We have been able to identify an interesting and relevant geophysical problem that the HHL algorithm could be used to solve.

Summary

In the recent years quantum technologies have matured enough that modern quantum computers outperform their classical counterparts at some very special computational tasks. It is however not entirely clear what kind of practical problems can be solved using present or future quantum computers. Here we will try to give a first attempt at answering this question with focus on optimization problems in geoscience.

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Main Objectives

Draining Remaining/Bypassed Oil Reserves

New Aspects

Vertical Inter-Reservoir Connectivity

Summary

This paper is a case study of a mature brown field in which a recent surprise in infill drilling resulted in a relook of subsurface understanding, with seismic playing a key role in explaining observations contradicting with the long established reservoir model. More commonly in a field composed of stacked fluvial deposits, an overwhelming amount of data with varying degrees of discrepancy and quality will lead us to conclude a more complex subsurface environment than what the reality is. The reservoir is typically separated into different hydraulic systems, thus treated differently in terms of reservoir management plans. This paper seeks to highlight the complicated connectivity that exists within a fluvial reservoir but from an often overlooked scenario of simplification, using validated production data. This paper also exhibits the importance of alternative scenario thinking, which can potentially open up new leads to monetize overlooked bypassed hydrocarbon reserves in mature brown fields.

Main Objectives

Highlight the importance of data integration to develop fields or divest assets

New Aspects

Maximize efforts and minimize geo/engineers time to effectively develop an asset

Summary

In the past years multiple U.S. operators have sold most of their conventional assets in order to finance their multibillion Oil and Gas shale projects (Erlingsen, 2017) as well as to cover debt obligations. Due to the rapid nature of these transactions, large volumes of data have been transferred from one company to another in different types of formats and structures.

During the acquisition of the subject field, multiple files containing geological, production, reservoir and completion data were delivered to the purchaser in both digital and hard copies. The objective of this project is to show a rather simple workflow to integrate this data to identify and evaluate the unquantified upside of the asset

By creating and combining data bases we were able to identify an upside that can be defined as multiple recompletion opportunities behind pipe and their potential recoverable volumes, undrained regions that may represent future drilling opportunities and new reservoirs that can not be easily mapped due to their complex stratigraphy

Main Objectives

Field development strategy in heavy oil field with high primary recovery factor and facies control on its production

New Aspects

High angle pilot wells and horizontal boreholes placement close to OWC

Summary

Initial Tivacuno development wells found lateral facies changes into M1 and T stacked sandstone channels reservoirs (90’ & 110’ net pay thickness each, respectively), which gradually shift from cleaner sandstone in the western flank to shaly sand facies to the eastern flank of the field. The directional wells started producing 3000 bopd average with 2% water cut, which progressively ramped up to 90% BSW (6000 bwpd) after 5 years, accumulating 2000 Kbls average heavy oil production by well, which represented 11.5% of recovery factor. The dynamic reservoir model showed unflushed oil areas still to be drained. Then, a second development plan drilled high angle pilot wells to monitor the production. The LWD logging found preserved original OWC depth, gamma ray activation and resistivity slump in the middle zone of M1 reservoir, which suggested water formation intrusion by former production. Horizontal wells were placed underneath the shale layer found and close to original OWC. Their initial average oil production was 1200 bopd with high water cut 85% (7200 bwpd). However, they achieved 1300 Kbls average oil production by well. The recovery factor jumped from 13% to 37% in M1 reservoir and 21% to 32% in T reservoir.

Main Objectives

Present an optimal development plan and case study for a foamy extra-heavy oil reservoir

New Aspects

Development optimization of a foamy extra-heavy oil reservoir

Summary

This research provides a more profitable development plan for improving cold recovery of foamy extra-heavy oil reservoirs in the Eastern Orinoco Belt.

Main Objectives

Smart water flooding in small scale setup

New Aspects

Dynamic aging- small core slices

Summary

In this study, the effect of brine salinity in different sequences of core flooding scenarios has been studied. Salinity of injection brines at secondary and tertiary stages of flooding has been changed to investigate low salinity effect. Core plugs from Tor chalk formation in the Danish North Sea were used and the recovery response measured by core flooding experiments. The experiments have been conducted in small scale core flooding setup, employing shorter core plugs of 1-2 cm length. As small slices with length of less than 2 cm and standard diameter are used in these experiments, the pore volume is more than 2.5 times less than the standard core plugs. Hence the amount of materials (oil and brine) and also consumption time is lower than the long core flooding experiments. Ion analysis has been conducted for effluents to see engaged mechanisms as well. This study demonstrates using small slices in core flooding experiments provides comparable results to experiments on standard cores.

Main Objectives

Gain novel insights into controlled salinity water-flooding and study the underlying mechanisms leading to increased oil recovery.

New Aspects

Assessment of secondary low salinity waterflooding in carbonates rock at reservoir conditions combining coreflooding, pore-scale X-ray imaging and wettability analysis.

Summary

Controlled salinity water-flooding (CSW) is a promising enhanced oil recovery technique, yet the pore-scale mechanisms that control the process remain poorly understood especially in carbonate rocks. The aim of this experimental study is, therefore, to gain novel insights into CSW and characterize oil, water and the pore space in carbonates. X-ray imaging combined with a high-pressure high-temperature flow apparatus was used to image and study in situ CSW in a complex carbonate rock. To establish the conditions found in oil reservoirs, the Estaillades limestone core sample (5.9 mm in diameter and 10 mm in length) was aged for three weeks at 11 MPa and 80°C. This weakly oil-wet sample was then flooded by injecting low salinity brine at a range of increasing flow rates. Tomographic images were acquired at 2.9-micron spatial resolution after each flow rate. A total of 60 pore volumes of low salinity brine were injected recovering 85% of the oil initially in place in macro-pores. Contact angles and brine-oil curvatures were obtained to characterize wettability changes within the rock pore space. Our analysis shows that wettability alteration towards a mixed wet system caused by low salinity brine was the main mechanism for increased oil recovery.

Main Objectives

The main objective of this paper is to investigate the effect of brine composition, injection scenario and temperature on oil recovery during low salinity water flooding in a carbonate (chalk) oil reservoir. In addition, the mechanisms governing oil-brine-rock interactions are investigated by analyzing effluent samples.

New Aspects

Since this study is conducted on reservoir materials from a Danish chalk reservoir, for which there is a lack of data in the literature, it supplies a significant body of data for low salinity recovery in chalk as well as investigation of the recovery mechanism. In addition, complete analysis of core effluents – brine and oil – elucidates rock-brine-oil interactions during coreflooding.

Summary

This research aims to investigate the effect of brine composition, injection scenario and temperature on oil recovery during low salinity water flooding in chalk core samples from a Danish North Sea reservoir. In addition, the mechanisms governing oil-brine-rock interactions are investigated by analyzing effluent samples and interfacial tension (IFT) measurements. For this purpose, by means of computed tomography (CT) results, homogenous chalk core samples (without any open fractures) were selected. These cores were saturated with reservoir fluids and aged at reservoir conditions for approximately three weeks. Several synthetic brines (including formation water, seawater and diluted seawater) were introduced through different injection scenarios into the aged cores at reservoir conditions. Brines used in this study were equilibrated with calcite at room temperature to minimize the effect of rock dissolution on oil recovery. Insights into the role of brine chemistry was obtained through effluent analysis performed using ion chromatography and ICP-OES. IFT measurements were conducted using a pendant drop method to quantify the extent of brine-oil interactions at reservoir conditions.

Summary

Reservoir wettability dictates fluid flow in porous media, and adsorption and desorption processes of charged polar crude oil components exert a major impact on carbonate wettability.

Main Objectives

Measure Zeta Potential of Carbonates under reservoir conditions; investigate the link between zeta potential and CSW

New Aspects

Large data set reporting zeta potential of carbonates at reservoir conditions; evidence of variable oil-brine zeta potential polarity; large suite of CSW Corefloods data and links to zeta potential measurements

Summary

This abstract covers measurements of zeta potential in intact carbonates under reservoir conditions and links these measurements to observations of improved oil recovery during controlled salinity water flooding.

Main Objectives

Find best composition for modified salinity waterflood based on existing understanding of oil deposition mechanism

New Aspects

General guidelines for modified salinity brine composition

Summary

This work computationally studies multi-component ion-exchange process and gives guidelines on modifying salinity of seawater used in waterflooding carbonate reservoirs. The study is restricted to \ce{Ca^2+}/\ce{Na^+} cation ion-exchange and cation (\ce{Ca^2+}) bridging mechanisms of oil binding. We find that injected seawater diluted with low salinity brine or de-ionized water shifts ion-exchange equilibrium toward stronger oil binding and is therefore inefficient strategy for improving incremental oil recovery. Instead, keeping salinity high, increasing sodium and decreasing calcium yields weaker oil binding. This should be the preferred strategy for modified salinity waterflooding.

Main Objectives

give an overview of the modeling work carried out in the context of modified-salinity waterflooding in carbonates; systematic study to assess the correlation of proposed wettability interpolators with the remaining oil saturation from spontaneous imbibition tests

New Aspects

Systematic study of comparing and testing different models against experimental data; the outcome of this work is a model that explains the improved oil recovery observed for chalk during modified salinity waterflooding

Summary

The mathematical attempts to model the modified salinity waterflooding (MSW) in carbonates complement the experimental studies and aim at bringing closer the implementation of this improved oil recovery (IOR) method at the field scale. Although multiple causes have been identified for the additional oil recovery during MSW, most modelling work only cover the wettability alteration mechanism. The current mechanistic models describe the shift in the wetting conditions upon injection of a chemically tuned brine composition by defining a parameter to interpolate between the relative permeability curves at water-wet and oil-wet conditions. In this work, we assess the correlation of several interpolants proposed in the literature with the remaining oil saturation from reported spontaneous imbibition tests carried out on chalk. The results show that, among the parameters studied, only the “available adsorption sites” parameter shows a trend and could therefore explain the additional oil recovery observed during MSW in chalk.

Main Objectives

To show how recently devised demultiple methods can be extended to incorporate plane-wave concepts.

New Aspects

We present a new data driven method for synthesis of multiple-free plane-wave responses. We also present corresponding imaging results.

Summary

Seismic images provided by Reverse Time Migration can be contaminated by artefacts associated with the migration of multiples.

Multiples can corrupt seismic images producing both false positives, i.e. by focusing energy at unphysical interfaces, and false negatives,

i.e. by destructively interfering with primaries. Multiple prediction / primary synthesis methods are usually designed to operate on point source gathers, and can therefore be computationally demanding when large problems are considered. Here, a new scheme is presented for fully data-driven retrieval of primary responses to plane-wave sources. The proposed scheme, based on convolutions and cross-correlations of the reflection response with itself, extends a recently devised point-sources primary retrieval method for to plane-wave source data. As a result, the presented algorithm allows fully data-driven synthesis of primary reflections associated with plane-wave source data. Once primary plane-wave responses are estimated, they are used for multiple-free imaging via standard reverse time migration.

Main Objectives

Remove residual multiple pre-migration by using the coherency of multiples post-migration

New Aspects

Uses a matched subtraction technique prevent the need for direct adaption

Summary

In the North Sea, water bottom related surface multiples offer one of the largest challenges to a seismic processing project with a careful balance between multiple attenuation and primary preservation required when using a model and subtract approach. For Ocean Bottom Node (OBN) data Up/down deconvolution can be used to output the reflectivity where multiple and source signature will be removed, eliminating the need for an adaptive subtraction as part of the principle de-multiple process. Residual multiple can however still be present where the assumptions within the up/down deconvolution method do not adequately describe the reality of the data. Often, these residuals are only noticeable once migrated, particularly where a diffracted component is prevalent. In this paper, a method using the coherency of residual multiple evaluated on migrated and stacked data is used to drive pre-migration residual multiple attenuation through the use of a matched-subtraction.

Main Objectives

Removing the effect of long-period sea surface reflection

New Aspects

Reflection-removing approach based on wavefield-extrapolation processing

Summary

We developed a processing method to separate and utilize long-period sea surface reflected waves by applying wavefield extrapolation processing. It would be possible to separate the sea surface reflection waves from the survey data with high accuracy using the pseudo sea surface reflection generated by downward wavefield extrapolation. In addition, it would also be possible to use the separated multiple to enhance the signal-to-noise ratio of the survey result by applying the upward wavefield extrapolation. We applied our method to the survey data obtained by the deep-towable marine seismic vibrator towed at 225 m depth to verify the effectiveness of our method. Our processing method could remove the effect of a long-period sea surface reflection, which has been a problem to realize a seismic survey with a deep-towed seismic source. In addition, our proposed method could use the decoupled reflection waves to enhance the signal strength by considering it as survey data from a mirrored source.

Main Objectives

Processing of the downgoing wavefield of OBN data and the attenuation of downgoing multiples

New Aspects

Prediction of the downgoing multiple wavefield on OBN data

Summary

Efficient acquisition of deep-water ocean bottom seismic data relies on using a grid of sparse nodes, something that is also becoming popular in shallower water. In these cases, the shallow illumination for the primaries are poor and good imaging of the downgoing data (first-order receiver-side free-surface multiple) is needed. An accurate and efficient free-surface demultiple of the downgoing data is therefore required. This paper develops an approximate and fit-for-purpose solution for attenuating free-surface downgoing multiples (with two or more bounces off the free surface) from the downgoing data. The method is run in a “1D fashion”, offers robustness against the effect of a mildly dipping seabed and advantages over conventional demultiple methods. The method is closely related to the up/down deconvolution process and requires the knowledge of the downgoing direct arrival wavefield.

Main Objectives

Data-driven Multiple Suppression for Laterally Varying Overburden with Thin Beds

New Aspects

We extend the 1.5D theory to 2D, in combination with the recently proposed internal de-multiple formula

Summary

Marchenko multiple elimination methods retrieve wavefields which can be used to remove all orders of overburden-borne multiples in a data-driven way. Although powerful, the method only accurately handles media with thin beds if the underlying Marchenko equations are constrained by the information that is typically unavailable a priori. The so-called augmented Marchenko scheme aims at addressing this problem, however, it requires a minimum-phase reconstruction step, which is only well-defined for 1D signals. As a result, the method can only be applied to media with laterally invariant (1.5-D) overburdens. Applications to 2-D are restricted by the fact that a generally-applicable minimum-phase reconstruction algorithm does not exist for arbitrarily-complex 2- and 3-D overburdens. Here, we show that accurate solutions to the augmented Marchenko-type equations can still be found in 2D by including additional processing steps in the 1.5-D approach. Furthermore, the approach presented here opens the possibility of calculating band-limited multidimensional minimum phase operators. 

Main Objectives

To raise awareness of the need for alternative constraints of the Marchenko equation(s), the need for improved algorithm development and hence quality control. We offer a tool to carry out the latter.

New Aspects

To date there has been no method proposed which could directly AND accurately determine focusing functions given the 2d medium properties. This is a key enabler to further Marchenko equation based internal multiple suppression algorithm developments.

Summary

Marchenko equation based internal multiple suppression methods, so far had been largely studied in simple enough media, where solutions can be found and quality controlled relatively easily. More realistically complex media might require augmenting the Marchenko equations with additional constraints, calling for more complex algorithms and a need for improved quality control. This is particularly relevant since benchmarking usually can only take place after several additional processing steps, hence hiding the source of potential errors. As more complex media might hide still undiscovered limitations to the Marchenko method, the availability of numerically robust reference solutions is pivotal to further development of the theory as well as implementation. In this work we show that available methods for frequency-domain depth extrapolation can be used to generate accurate focusing functions. Such an approach can be implemented for laterally varying media and even for the visco-elastic wave equation. Our comparison of a modelled focusing function with one derived from the augmented Marchenko method on a model with a horizontally layered, yet very challenging overburden confirms both fidelity of our modeling scheme as well as the capability of the augmented Marchenko method to successfully handle short-period internal multiples.

Main Objectives

Internal multiple elimination

New Aspects

An extended 3D single-sided autofocusing for internal multiple elimination and primary-only imaging

Summary

We propose an extended 3D single-sided autofocusing guided by inverse-scattering theory for internal multiple elimination. Internal multiple modelling and elimination is formulated as an equivalent-scattering source imaging problem with limited-aperture. In our derivation, the focusing source is obtained by a cumulative TRM (time-reversal mirror) autofocusing process. The extension of TRM process is guided by inverse-scattering theory, and is based on causality, energy conservation and source-receiver reciprocity. From this derivation, the application range of internal multiple elimination can be expanded to complex interfaces and irregular-shaped inclusions. Numerical examples of salt-multiples removal and subsalt imaging proved the broad application range of the theory and method.

Main Objectives

Abstract propose a method to achieve 70% reduction in processing time needed to achieve good adaptive subtraction result through utilizing meta-parametrized approach implemented in new subtraction domain

New Aspects

1- new developed adaptive subtraction domain ,2- new approach in computing match filter guided by temporal and spatial wavelength way for deriving 3- how one meta parameter can control all workflow aspect

Summary

Seismic, multiple attenuation, surface multiples, least square, curvelet, FK-decomposition, meta-parameter, deep marine, multiples,

Main Objectives

Find out the propagation phenomenon of multiple waves and ghost waves in seismic physical modeling experiment

New Aspects

Through seismic physical modeling experiment , there are two types of associated ghost waves of model primary reflected wave, unlike ocean exploration, the ghost waves generated between the excitation probe and the water surface have very weak energy. there have 5 kinds of waves are model secondary reflected waves and their associated ghost waves, but have no ghost waves generated between excitation probe and water surface. Three of them are laboratory-specific waveforms, which are related to the special physical structure of the ultrasonic simulation device.

Summary

Laboratory seismic physical modeling experiments are subject to experimental devices and scales when studying the law of seismic wave propagation. So the propagation characteristics and rules of multiple-waves and ghost waves are different from actual earthquakes. In the experiment, the same performance excitation/receiving piezoelectric transducers were used to measure the plexiglass model, and the ultrasonic vibration was excited in water. Keep the same horizontal distance of the excitation/receiving transducer, and change the depth of the excitation transducer and the receiving transducer under the water respectively, and get 4 sets of model acquisition results with different depth relationships.The experimental results show that there are two types of ghost waves reflected from the water surface accompanied by first reflect of the model. As the main energy of the excitation transducer propagates downwards, the virtual reflection energy from source to water surface is very weak. And there are five types of multiple waves reflected by the model, but have no multiple waves that excite from excitation transducer to water surface. There are three kinds of waves that are unique to the laboratory, which are related to the special physical structure of the excitation receiving device.

Main Objectives

Fast efficient processing of OBS data

New Aspects

Additional insight into range of applicability of ‘1D fashion’ up-down deconvolution; efficient workflow; processing results for Snorre PRM data

Summary

The abstract describes a fast and efficient workflow for geophone to hydrophone calibration, up-down decomposition and up-down deconvolution. All steps are applied to the Radon transformed common receiver gathers, where for each receiver fast Radon transform is performed over total offset for shots within pre-defined azimuth sectors. The efficiency of the workflow is illustrated on Snorre PRM data. While Snorre structure is quite complex, the results of fast ‘1D fashion’ processing and 3D migration are very encouraging. We further investigate the limitations of 1D up-down deconvolution and prove, that for complex 3D structures below the sea-floor, the method works well in preserving the primaries and removing the water-layer multiples and peg-legs, if the sea-floor is locally horizontal and the sea-floor reflectivity does not change rapidly when changing the incident angle. The result differs from that presented by Wang et. al (2010), where the receiver-side ghosts of primaries from a complex part of the structure and of receiver-side peg-legs have not been included in the total down-going wavefield. For Snorre the sea-floor is gently dipping and irregular. Because of this, prior to inverse Radon transform of the results of deconvolution, we have applied source-side re-datuming from the sea-floor to the free-surface.

Main Objectives

interbed multiples attenuation on land seismic data

New Aspects

We have extended the 3D wavelet transform’s applications to attenuate multiples based on common image gathers from 3D prestack depth/time migration

Summary

We propose an innovative method to attenuate multiples using 3D complex wavelet transform which transform 3D seismic data into orientation and scale domain while the original axis are still kept. The proposed method applies the transform on 3D volumes of individual migrated common offset and partial stack volume of selected offsets respectively under assumption that prestack depth/time migration is carried out using primary velocities and aligns the primary’s events within common image gathers. In the transform domain, the individual offset is filtered according to the partial stack. Therefore, multiple including the interbed multiples are attenuated while signals are enhanced. The proposed method has been tested using synthetic data sets and applied to 3D field data sets. The results demonstrate that the new method attenuate multiples and the image of reserves gets significantly better recovery than traditional methods, especially for near offsets.

Main Objectives

To demonstrate the capabilities and limitations of Marchenko multiple elimination and subsequent imaging

New Aspects

This is the first analysis on a resonant model with combined effects of multiple thin layers on the ability of multiple elimination and its effects on the resulting images.

Summary

The ability to separate primary reflections from multiples is important for making subsurface images. Many existing methods need some form of model information and adaptive subtraction. Marchenko methods have been modified to operate at the acquisition surface. The associated filters can be computed from the reflection response without any model information. They are a function of a freely chosen time instant that defines the time window of the filter. The scheme can be implemented without adaptive subtraction. Applying the filter to the reflection response removes all multiples from the overburden that would arrive within the time window in which the filter is defined. For this reason, the first event in the result is a primary reflection event that can be taken and stored in a new dataset containing only primary reflections. From data of a resonant wedge model with thin layers, the images after MME show that destructive interference effects are removed by MME. Reflectors are imaged that are missing in the images of the reflection response. Thin layer effects cause incomplete prediction and removal of multiples. When MME treats the combined reflections from thin layers as a single complicated event, their combined multiples from other reflectors are properly removed.

Main Objectives

To improve multiple attenuation in shallow water settings.

New Aspects

Sparseness weights are added to a wave-equation deconvolution imaging approach and combined with source-side model based multiple prediction. Real data examples highlight lower levels of residual multiple with the proposed approach.

Summary

The attenuation of surface related multiples is typically one of the most challenging steps in the processing of shallow water marine projects. Least-squares wave-equation deconvolution imaging is a powerful tool to address this challenge, but images derived from a deep target level may produce sub-optimal demultiple results for the shallower section. We introduce image domain sparseness weights to the least-squares problem, derived from a water-bottom depth estimate. This provides a reflectivity with a sharp contrast at the water-bottom, and the corresponding multiple prediction exhibits improved temporal resolution compared to least-squares wave-equation deconvolution imaging. We also illustrate how the multiple prediction from sparse wave-equation deconvolution imaging may be combined with source-side targeted multiple prediction to improve multiple attenuation for complex multiple generators. Data examples from the Central North Sea and the West of Shetland confirm the benefits of the proposed methods in attenuating residual multiples.

Main Objectives

Improve sub-basalt imaging, efficient and effective internal multiple attenuation, comparing two methods respecting dynamics versus kinematics.

New Aspects

Modelling almost all internal multiples simultaneously. Using self adaption to reduce the risk of over-adaption.

Summary

Internal multiples contaminate weak primary reflection signals received from sub-basalt interfaces. We compare two methods of predicting internal multiples in seismic lines acquired in the Norwegian Sea. We compare a wave-equation based method that respects the structure but predicts only a subset of internal multiples corresponding to a downward bounce at the sea bottom, against a fast approximation which relies on a flat-earth assumption but predicts internal multiples generated at many subsurface boundaries all at once. The latter approach provides a better result by making a more realistic estimation of amplitudes, while compromising the accuracy of temporal dynamics.

Main Objectives

We propose a U-Net-based approach to dealiase the estimated surface multiples from limited sources.

New Aspects

The multiple dealiasing problem is highly non-linear, which suits well for deep learning (DL)-based methods.

Summary

The main prediction engine in surface-related multiple elimination (SRME) is the multidimensional convolution process, where data sampling plays an essential role for accurate surface multiple prediction. Therefore, fully sampled sources and receivers are preferred. If especially the source sampling is far from ideal, the estimated multiples will suffer from the severe aliasing effect. Consequently, this can lead to poorly estimated primaries. Interpolation of coarsely sampled sources is not a trivial task and computation intensive. Dealiasing on the estimated multiples from limited sources might provide a potential solution. In theory, this dealiasing problem is highly non-linear, which suits well for deep learning (DL)-based methods. Therefore, we propose a U-Net-based approach to dealiase the estimated surface multiples from limited sources. Applications on two subsets of the field data demonstrate the effective performance of the proposed method.

Main Objectives

Land multiple attenuation, Middle East, Surface multiples, internal multiples

New Aspects

Advanced and integrated workflow, driven by well analysis and full understanding of the multiple contamination problem. First concrete case of a data domain multiple attenuation capable of attenuating strong multiples masking the unconformity in Oman. Previous works are largely based on post-imaging and post-stack solutions.

Summary

Strong surface and internal multiple contamination in the seismic data are known to impair interpretation and reservoir characterization studies in the northwest part of Oman. Major interferences at the target zones are due the complex near surface composed of thin and high-impedance intercalations in the Tertiary sequence. This works presents a full description of the multiple contamination pattern in the area of interest, followed by a tailored data-driven solution designed to attenuate the multiples present in the data while preserving weak primaries over the sequence of interest.

Main Objectives

Development strategy optimization for Naturally fractured-caved carbonate reservoirs

New Aspects

water injection huff-and-puff, then gas injection huff-and-puff

Summary

Naturally fractured-caved carbonate reservoirs in China have some distinctive characteristics: developed multi-scale fractures, vugs and caves, no moveable oil in pores, poor reservoir connectivity and much difficult to develop. How to effectively develop this kind of reservoirs is a major challenge. This paper presents the corresponding development strategy optimization for different reservoir patterns of this kind of reservoirs.

Based on understanding of geological study and dynamic characterization, typical reservoir patterns are identified and established. Corresponding different reservoir simulation models are built for different reservoir patterns. Then reservoir simulation are used for the development strategy optimization for different reservoir patterns. Finally, the optimization results are applied to the enhanced oil recovery of a fractured caved carbonate reservoir in China.

This paper has been successfully applied to a heterogeneous fractured caved reservoirs in China, which provides a reliable foundation for the effective development of this kind of reservoir. Also the method can be used for development strategy optimization study of other kinds of carbonate reservoirs worldwide.

Main Objectives

Understand the complexity of carbonate reservoirs, identify main features controlling permeability.

New Aspects

Description of structural and depositional features at different scales that can influence permeability on carbonate reservoirs.

Summary

Naturally fractured reservoirs present a challenge when determining the permeability associated with different types and sizes of fractures. Permeability in these reservoirs depends on the heterogeneity and connectivity of open fractures; although depositional and diagenetic features also play an important role. In this study, a multi-scale analysis of the Cariatiz Fringing Reef Unit in SE Spain is completed based on outcrop and LiDAR data. Seven different features were found to influence the permeability of the Cariatiz Fringing Reef Unit. Structural features comprise: (i) joints, (ii) veins, (iii) vertical fractures, (iv) fracture swarms, and (v) karsts. Two types of depositional features were also recognised at outcrop:(vi) vertical Porites and (vii) pseudo-bedding surfaces. All contribute to increased permeability, apart from calcite-filled veins that create barriers to fluid flow. The results of this study highlight the complexity of carbonate systems and the need to collect data at different scales of analysis to decrease uncertainties in reservoir models. The approach in this work is valid in hydrocarbon exploration and production, geothermal reservoir characterisation, environmental studies and carbon sequestration projects.

Main Objectives

Show the state of the art muliscale subsurface and surface integrated workflow

New Aspects

Different mathematical properties over various scale ranges of joint and fault patterns

Summary

A fully integrated multidisciplinary workflow has been applied using up to date technologies in order to prove deliverability of already discovered and proven resources, and explore for new (Palaeozoic and basement) reservoir levels and other prospects in the areas of Tinrhert (Ilizzi-Berkine Basin) and M’Sari Akabli (Ahnet Basin) Blocks.

Extensive analyses of vintage 2d and 3d seismic data and previously drilled wells has been performed over the last 4 years. New 2d seismic data were acquired and vintage 3d seismic data have been reprocessed in various phases.

The multiscale workflow presented shows the results of seismic mapping and depth modeling, seismic attribute analyses, well log correlation, image log analyses including stress field analyses, core data and remote sensing data analyses. Completely new approaches are outlined regarding the multiscale treatment of remote sensing data regarding fractures and faults, and regarding data driven DFFN modeling in the 3d domain.

The results constitute a solid backbone for the planning of future appraisal wells in this complex and challenging geological environment.

Main Objectives

3D modeling of a fractured and vuggy carbonate reservoir for dynamic simulation purposes

New Aspects

3D modeling of vugs as discrete objects in carbonate reservoirs

Summary

The Cretaceous reservoir of the KMZ field consists of brecciated, carbonate rocks characterized by the presence of a complex system of natural fractures and solution vugs. As this high-permeability system plays a major role on the fluid circulation in the field, it is crucial to build-up a 3D model in order to quantify its dynamic impact. To do this, we applied the methodology for the analysis of fractured reservoir FracaFlowTM developed by the IFPEN/Beicip-Franlab group, integrating BHI data, structural information (3D seismic) and dynamic data.

Both diffuse and fault-related fractures were recognized in the field, whereas vugs largely affect mainly the brecciated layers of the reservoir. The innovation of the present study is the modeling of the vugs, with the software FracaFlowTM, as independent objects overprinting the matrix properties and distributed independently from fractures. This allows to take into account and calibrate separately the permeability of the vugs and fractures.

Main Objectives

Establish a fracture classificiation in the Valdemar Field, and relate natural fractures to sedimentary facies, and ultimately reservoir recovery

New Aspects

This is a first study that conducts a detailed analysis of natural fractures in the Valdemar Field, and relate them to stratigraphy and facies. Furthermore we attempt to investigate whether the fracture-facies relationship can enhance hydrcarbon recovery.

Summary

The Valdemar Field (Central Danish North Sea) is a heterogeneous reservoir, with a high abundance of natural fractures, producing from chalk units within the Lower Cretaceous Upper Hauterivian–Lower Albian Tuxen and Sola Formations. The Lower Cretaceous stratigraphic interval in the Danish North Sea is characterized by chalk, argillaceous chalk, marlstones and claystones with hydrocarbons trapped in the chalk reservoir units. The reservoir (chalk) sediments are characterized by high porosity and low matrix permeability, but the presence of natural fractures strongly influences the flow properties by enhancing the effective permeability, and ultimately increases recovery. Several types of natural fractures are recognized on core scale, and some of these form a potential hydrocarbon pathway as they can significantly enhance the natural permeability in the reservoir. This is a first study that conducts a detailed analysis of natural fractures in the Valdemar Field, and relate them to stratigraphy and facies. Furthermore it is considered whether certain fracture types occur more frequent in specific stratigraphic intervals. It is investigated whether the fracture-facies relationship can enhance hydrcarbon recovery.

Main Objectives

Naturally Fractured Reservoir; Carbonate Rocks;

New Aspects

Integrated methodology to characterize natural fractures in carbonate rocks

Summary

Low permeability carbonate reservoirs constitutes significant reserves of oil and gas for Petrobras’ E&P sector. Micro-porosities above ~ 10% in these sedimentary sequences allow the accumulation of significant volumes of hydrocarbon. However, the predominance of pore throats smaller than 10 microns produces low permeability in this type of rock, making it difficult for the flow of interstitial fluid during production. This constitutes the main challenge for making production feasible. The shortage of static and dynamic data in most reservoirs of low permeability makes it extremely difficult to understand the spatial distribution of the different scales of heterogeneities and, consequently, influence in obtaining realistic flow scenarios. In this work, an integrated supervised methodology is proposed for the characterization of natural fractures in a low permeability lacustrine carbonate reservoir in Santos basin, Southeastern Brazilian margin. To achieve the objectives of the study, a workflow was developed that involves the actions, briefly described below: (i) descriptive and kinematic analysis at multiple scales of the brittle structures; (ii) understanding of lithological control in the deformation process; (iii) quantification of the deformation in different phases of movement by 2D / 3D structural restoration techniques and (iv) analysis of seismic anisotropy.

Main Objectives

Introduction of a new integrated protocol for 3d geological discrte fault and fracture modeling

New Aspects

Copletely new protocols and implementation plus examples of 3d models.

Summary

The paper discusses and shows examples of 3 aspects in the application of new technologies: Multiscale fault and fracture mapping and parameter extraction, 3d Data Driven Modeling, and the new DMX protocol for 3d mixed deterministic and probablistic modeling. It shows the role and integration in the multidisciplinary workflow of these three elements, their impact on resource management, and provides indications for future developments.

Main Objectives

Natural Fracture Prediction

New Aspects

Natural Fracture Prediction using Geomechanical Forward Modelling Method

Summary

Geomechanical forward modelling will regenerate the present-day structure through geological times thereby providing best estimate of rock deformations and other geomechanical variable. Such attributes will provide considerable margin of error for distribution of the fracture network within the field. The 3D geomechanical forward modelling followed with critical-stress-fracture analysis can estimate productivity behaviour across the field. From simulation result, the most fractures in the South West part of the field are critically stressed fractures and therefore have better hydrocarbon production potential.

Main Objectives

Exploring the influencing factors of sand production pressure difference

New Aspects

Considering both geological and engineering impacts on sanding

Summary

The Keshen block is a fractured tight sandstone gas reservoir in Tarim Oilfield. The porosity is around 6% to 10%. The permeability varies from 0.1 to 1mD. The fractures of sandstone are developed . During development of this block, 49 wells have sand samples collected, which accounts for 53% of existing wells. Due to sanding, 9 wells have had to be shut-in for a long term, which is 34.6% of sand production wells. Moreover, there are more production wells in the Keshen block that face sand production issues every year. Early research has found that excessive production pressure difference is the direct cause of sand production. Based on this, through analysing production data of more than 40 wells in the block, comparing rock mechanical parameters, and numerical simulation, this paper uncovers the main influencing factors of sand production pressure difference: development of natural fractures, reservoir stimulation, and tail pipe material.

Main Objectives

Based on digital rock, the thermal neutron transport has been simulated with MCM to reveal the correlation between the counts difference and porosity, to investigate the effects of anisotropy and saturation.

New Aspects

A new mathematical model has been proposed to relate the counts difference to the porosity and saturation, which can actually be taken as a new method to calculate the saturation with the porosity and the counts difference of neutrons.

Summary

In this research, the digital rock has been constructed with the pileup of the segmented images of X-ray CT data from the carbonate samples. The transport of thermal neutrons has been simulated with the Monte Carlo method to reveal the correlation between the counts difference of neutrons and porosity, to investigate the effects of water saturation. Instead of a detector used in the conventional researches, the 8×8 array detectors are utilized to detect the transmitted neutrons, which can help to investigate and image the anisotropy of pore structure. Based on the simulated results, it is observed that the calculation of porosity with the neutron data can be affected by the anisotropy of pore structure and the saturation. The significance of fluids content in the porosity evaluation can synchronously increase with the increasing of water saturation. Moreover, a new mathematical model has been proposed to relate the counts difference to the porosity and saturation, which can actually be taken as a new method to calculate the saturation with the porosity and the counts difference of neutrons.

Main Objectives

to accurately calculate the irreducible water saturation for tight sandstone

New Aspects

This paper presents a new method of calculating irreducible water saturation by using nuclear magnetic resonance logging. First, the capillary pressure curve is predicted by the variable scale power function. Then, the transformation equation between NMR relaxation time T2 and pore throat radius r is determined. Finally, an integral model for calculating irreducible water saturation based on nuclear magnetic resonance logging is established by using the transformation of integral.

Summary

The complex pore structure of tight sandstone makes it difficult to accurately calculate the irreducible water saturation by conventional logging method. To solve this problem, this paper proposes a model for calculating irreducible water saturation of tight sandstone reservoirs using nuclear magnetic resonance (NMR) logging. First, the samples are classified into high irreducible water samples and low irreducible water samples using support vector machine with a RBF kernel function. Then, the transformation equations between NMR transverse relaxation time (T2) and pore throat radius of high and low irreducible water samples are determined respectively. Finally, the irreducible water saturation of tight sandstone samples is determined by integral equation. The average error between the calculated irreducible water saturation and the measured value is 6.7%. The method is also applied to a tight sandstone reservoir of a well in the study area, and the calculated value of irreducible water saturation is in good agreement with the measured value by the samples experiment, which further reveals the reliability of the method in calculating irreducible water saturation.

Main Objectives

In this study, we apply a new finite difference method to calculate the permeability of digital rock. And an elliptical pore approximation method is applied to divide the irregular cross sections.

New Aspects

1. A new finite difference method proposed in this paper is suitable for elongated pores. 2. The finite difference method can be used to calculate the effective permeability of digital core with complex pore structure and give reasonable prediction. 3. The results calculated by finite difference method in good agreement with those calculated by lattice Boltzmann method after grid coarsening.

Summary

Digital rock images may capture more detailed pore structure than the traditional laboratory methods. No explicit function can correlate permeability accurately for flow within the pore space. This has motivated researchers to predict permeability through the application of numerical techniques, e.g., finite difference method (FDM). However, in order to get better permeability calculation results, the grid refinement was needed for the traditional FDM. And the accuracy of this method decreased in pores with elongated cross sections. The goal of this study is to develop a new FDM to calculated the permeabilities of digital rocks with complex pore space. An elliptical pore approximation method is invoked to describe the complex pore space. The permeabilities of the digital rock images after grid coarsening are calculated by the new FDM in three orthogonal directions. These results are compared with the previously validated lattice-Boltzmann method (LBM), which indicates that the predicted permeabilities calculated by the new FDM usually agree with permeabilities calculated by LBM. We conclude that the presented new FDM is suitable for complex pore space.

Main Objectives

Demonstrate carbonate factory and depositional environment dependency of the acoustic properties of carbonate to mixed carbonate-clastic sediments

New Aspects

Comparison of all varieties of carbonate sedimentary systems

Summary

In this study a comparison is made of acoustic property measurements covering: (i) Continental deposits: Miocene, Spain and Pleistocene travertines, Turkey and Hungary; (ii) Marine mixed carbonate-siliciclastic sediments: Eocene, N-Spain; Jurassic, Spain, and Carboniferous-Permian, Norway; and (iii) Full marine sediments: Miocene, SE–Spain, and Urgonian carbonates, S–France.

The dataset shows that rock fabric and mineralogy control the elastic properties of sediments with varying mineralogy and texture (carbonate, clay and quartz). High carbonate contents result in high velocities and low carbonate contents and/or highly mixed sediments demonstrate low velocities for a given amount of porosity. Deviations from aforementioned trends depend on the quantity, grain-size, and non-carbonates mineralogy.

Porosity variations, related to the age and diagenetic history of the sediments, have a distinct impact on the P-wave velocity.

The P-wave velocity values of marine and lacustrine carbonates overlap suggesting that pore structure and cementation in both systems determines acoustic properties. These P-wave velocities differ from travertines, which probably relates to pore type and pore distribution linked to sedimentary and diagenetic processes, and biology intrinsic of these depositional systems.

In conclusion, the main parameters controlling the acoustic response are: porosity amount and pore type, mineralogical composition (including clay content), rock texture and fabric (frame).

Main Objectives

present a novel, unique method of creating experimental rock analogues

New Aspects

using 3D printing in rock analysis opens a new avenue for experimentation and validation of numerical modelling

Summary

3D printing is becoming a powerful tool to visualize, reproduce, and experiment with porous media. Natural rocks are part of porous media that have always been a focus of studies on how fluids, such as hydrocarbons, greenhouse gases, and/or water, flow through porous systems. Scale and accuracy are among the most challenging factors for current 3D printing techniques when attempting to replicate the pore architecture of natural porous media such as rocks. However, current 3D printing techniques have resolution restraints during fabrication that make feature reproduction at the 1:1 scale almost impossible. A new emerging technology that uses two-photon lithography and ultraviolet-light curable resin allows for microscopic features to be resolved during fabrication. To test this technology, a pore network was obtained from tomographic data of a reservoir rock sample in Mexico (1 mm in diameter and 2 mm in height) and was 3D-printed at the original scale. The 3D-printed sample was subjected to optical and electron imaging to verify the accuracy of pore geometry. Incorporating lithographic printing into novel rock experiments that concern multi-scale, multi-physics models of fluid flow and deformation open unprecedented opportunity for more controlled prediction of reservoir fluid dynamics, carbon capture and storage, and continuum mechanics.

Main Objectives

correct the effect of the internal gradients on the NMR T2 spectrum.

New Aspects

An effective method is proposed to compensate the influences of internal gradients on the T2 spectrum by the time domain correction method and the relationship between internal gradients and T2.

Summary

The laboratory nuclear magnetic resonance (NMR) measurement and NMR logging have been widely used in formation evaluation. However, the measured signal and the inverted transversal relaxation (T2) spectrum is influenced by the existence of internal gradients. To solve this problem, we conducted laboratory analysis and developed an effective method to compensate the influences of internal gradients on the T2 spectrum by the time domain correction method and the relationship between internal gradients and T2. The results show that the influences of internal gradients can be diminished and the diffusion-free T2 spectrum can be obtained. The correction method is validated by core analysis and field NMR logging data. The proposed method provides an effective way to compensate the influences the internal gradients and get the precise T2 spectrum, which may have broad application prospect in unconventional reservoirs.

Main Objectives

Correction of capillary end effects for imbibition in oil-wet media

New Aspects

Extending Huang and Honarpous methodology and presenting the base and extended methods concisely and clearly for the practicing reservoir engineer and petrophysicist

Summary

Oil displacement experiments constitute a major component of special core analyses for both improved and enhanced oil recovery (IOR and EOR) operations. Capillary end effects (CEE) is a well-recognized phenomena that affects the accuracy of unsteady-state (USS) oil-displacement results. Huang and Honarpour presented the procedure of correcting for CEE in 1998. They derived the equations and demonstrated the procedure for correcting water terminal saturation in drainage of water-wet media. However, since their pioneering work, their methodology has not been extended to correction of oil terminal saturation for imbibition in oil-wet media—despite the well-recognized and impact of CEE on oil-displacement studies. Therefore, in this work, we revisit Huang and Honarpour (HH) pioneering method and extend it to the correction of CEE of terminal saturations for imbibition in oil-wet media. The extended method was also applied and validated using special core analysis data for a Middle Eastern carbonate. In the process, we present the base and extended methods in a concise and clear fashion that is of utility to the practicing reservoir engineer and petrophysicist.

Main Objectives

EOR, Foam, Fractures, Gas Diffusion, Coarsening

New Aspects

We have created a novel method to estimate the height of lamellae, by which we discuss whether coarsening stops as there is no driving force or slows to a near stop as there is little lamella area for gas diffusion.

Summary

In this paper, we study foam coarsening in fractures. To this end, we have built two 1-meter-long model fractures, similar in some ways to microfluidic porous media. The two fractures are made of glass plates and have different roughness and hydraulic aperture. We characterize the roughness and represent the aperture distribution of the fracture as a network of pore bodies and throats. We use a fast-speed camera to visualize foam behavior inside the fractures. We show how foam evolves during coarsening in both model fractures. We also estimate the height of lamellae at the end of our coarsening experiments. Based on this information, we discuss whether coarsening stops because bubble pressures are equalized or slows nearly to a stop because bubbles lose contact through lamellae, at the end of the coarsening experiments.

Main Objectives

this paper aims to provide an in-depth understanding of the impact of prolonged successive water-gas injection on RPM’s performance and to gain further information about DPR mechanisms of the RPMs.

New Aspects

In this work we have focused on the impact of injection volume during successive water-gas injection scenario on relative permeability modifiers performance

Summary

Excessive water production is becoming common in many petroleum reservoirs. Relative permeability modifiers (RPM) have been used to disproportionately reduce water permeability (DPR) with minimum effect on the gas/oil phases. This manuscript reports the results of an experimental study where we examined the effect of prolonged successive water-gas injection on RPM’s performance in gas/water system. The results show that the volume of water coming in contact with the polymer-treated porous medium has a direct impact on the extent of polymer swelling and thus the water permeability of the medium. It was also shown that over a large volume of gas injection, treated porous medium presents a better permability for gas at later times compared to the early times of injection. The reason could be the dehydration of the polymer layer adsorbed to the pore surfaces of the medium. It was found that successive water-gas injection could lead to stronger performance of polymer towards reducing water permeability, but it comes with the cost of a further reduction in gas relative permeability.

Main Objectives

To improve heavy crude oil production

New Aspects

A synthesized viscosity reducer bearing polyaromatic hydrocarbons was used for enhanced heavy oil recovery

Summary

Polyaromatic hydrocarbons were known to efficiently disperse the micro-aggregates of asphaltene compound in heavy oil. Based on this principle, a viscosity reducer (VR) containing pyrenyl group was designed and synthesized in laboratory. A cationic surfactant CAS-3 was used along with the VR (SVR) to improve the solubility of the SVR in brine. The yielded binary VR system (SVR/CAS-3) can form stable emulsion with heavy crude oil. Comparing with four commercial VRs, SVR/CAS-3 exhibited much more pronounced effects on viscosity reduction of heavy oil and heavy oil displacement. The high performance of SVR/CAS-3 flooding in improved heavy oil production comes from heavy oil viscosity reduction associated with strong emulsification and increased sweep efficiency.

Main Objectives

EOR, Mineralogy

New Aspects

Evaluation of the impact of reservoir rock composition in mineralogical scale during EOR process

Summary

Achieving the newest technique for EOR process in existing reservoirs is one of significant matter in future petroleum industry. Among of all influential parameters, wettability might be the most challengeable factor to produce oil in all stages of oil recovery. In wettability evaluation many factors such as chemical structure, surface properties of the rock, composition of the oil and etc have to be taken into account. The surface properties of minerals play an important role in elucidating the behavior of reservoir minerals in presence externally added reagents like surfactants.

This paper presents experimental investigation mixed adsorption of nonionic (Triton X-100) and anionic (SDS) surfactants on minerals of reservoir. This alteration was measured based on the contact angle method. The results obtained show the nonionic Triton X-100 surfactant is better than anionic SDS surfactant for wettability alteration of quartz surface; however, for calcite and dolomite surfaces wettability alteration by SDS is more effective than Triton X- 100. Mixture of surfactants is more effective than either of them alone to alter the wettability of all surface pellets.

Keywords: Wettability alteration; Contact angle; Triton X-100; SDS; Calcite; Dolomite; Quartz.

Main Objectives

This paper presents a rapid test to screen nanofluids which can be applied for deployment in limited core oilfields.

New Aspects

The proposed parallel bottom-up approach can be used to rapidly evaluate and screen nanofluid for detailed coreflooding experiments.

Summary

The mechanisms of oil recovery with nanoparticle is complex and being researched worldwide. Nevertheless, the most agreed influencing parameter among researchers is wettability alteration, which changes towards more water wet. General workflow for screening of nanofluid is top-down approach usually takes months due to sequential process and consumes many cores at rock-fluid tests. This paper presents a rapid test to screen nanofluids which can be applied for deployment in limited core oilfields.

Based on parallel bottom-up approach, it was observed that screening process only took several days instead of months to select suitable nanofluid and glass plate could be utilized in screening process to reduce consuming cores for oilfield with limited core. A series of glass plates experiment showed consistent results with core slab. Surfactant grafted nanoparticles has given marginal effect on IFT reduction at certain concentration and achieved steady in less than an hour. These results suggested the most potential rapidly for further analysis on coreflooding experiments.

The proposed parallel bottom-up approach can be used to rapidly evaluate and screen nanofluid for detailed coreflooding experiments. This approach is readily applicable for uncored or limited cores case.

Main Objectives

Examine whether it is possible to transport alkalinity throughout the sandstone core

New Aspects

Surface Reactivity of Sandstone, Samrt Water Flooding, Alkaline Flooding, Alkalinity Transportation

Summary

Both the alkaline water and smart water gives higher oil recovery by creating an alkaline environment in sandstone where alkaline water creates in situ surfactant and lower the interfacial tension (IFT) between oil and water, and smart water changes the wettability of the reservoir by ion exchange and as a result, creates an alkaline environment. Both the wettability alteration and the reduced IFT are dependent on increased pH. Could the observed EOR processes be actually the same, or a combination of both proposed mechanisms? Therefore, the objective of this research is to examine whether it is possible to transport alkalinity throughout the sandstone core. Our research shows that it is not possible to transport alkalinity throughout the core by injecting low saline alkaline water as the minerals present in sandstone play a vital role to consume alkalinity. It also shows that smart water flooding has a higher potential of creating in situ alkalinity than transporting alkalinity by injection of high pH alkaline water.

Main Objectives

Demonstrate the possibility of using a mixture of two types of nanoparticles (Al2O3, SiO2) and two-dimensional material (2D-Smart Nanosheets) as 2D-nanomaterial for EOR, by study its effect on wettability alteration of sandstone rock, interfacial tension of Oil/ Brine, and increasing the oil recovery

New Aspects

This work reveals the new possibilities for economical production of oil using a mixture of nanoparticles

Summary

In this study, a combined nanofluid based on mixing Silicon-oxide, Aluminum oxide nanoparticles with 2D-smart nanosheet (MAS2DSN) for enhanced oil recovery is developed. The current single nanofluid flooding method for tertiary or enhanced oil recovery is inefficient, especially when used with low nanoparticle concentration. In this work, we show the potential of mixing nanoparticles with two-dimensional smart nanosheets in one patch as a C-EOR agent. The combination of a Mixture of Aluminum-Silicon nanoparticles with 2D-Smart Nanosheet (MAS2DSN) is used to reduce the IFT and altering the wettability of the sandstone core samples even at low concentrations. The IFT decreases up to 0.16 %, and the contact angle measurements show that the wettability of sandstone is changed from the oil-wet to the water-wet in the presence of MAS2DSN. The laboratory core-flooding experiments were conducted in the sandstone core samples saturated with module oil (20 cP). Stable MAS2DSN nanofluid is applied in core-flood experiments using Al2O3, SiO2, and 2DSN at concentrations of 0.05, 0.05 and 0.005 % respectively, resulting in a 22.9% increase in the oil recovery. This work exposes new opportunities for oil production by using a mixture of nanoparticles.

Main Objectives

To investigate the potential of surface-active ionic liquid as an alternative to the conventional surfactants in recovering the residual oil from the reservoirs for application in enhanced oil recovery.

New Aspects

Ionic liquids can be considered as a new generation of surfactants which could be utilized in the chemical EOR processes especially for the reservoir with harsh conditions due to its excellent salt tolerance and thermal stability.

Summary

With the ever-growing global energy demand and decreasing reserves, it is believed that enhanced oil recovery (EOR) technologies will play an active role in future to meet the energy demand. Among the various EOR techniques, chemical flooding is one of the most successful techniques used worldwide because of its capability in reducing oil/water IFT and controlling mobility ratio. The objective of the study is to investigate the potential of surface-active ionic liquid as an alternative to the conventional surfactants in recovering the residual oil from the reservoirs for application in enhanced oil recovery. An imidazolium based IL, 1-dodecyl-3-imidazolium bromide was utilized to determine its ability to reduce interfacial tension and alter the wettability of oil-wet rock. The study showed encouraging results in the presence of ions and increasing temperature. The wettability alteration behavior gave an insight into the capability of IL to alter the wettability towards water wet which is required to recover the residual oil saturations. The core flooding experiments showed an additional recovery of 29.11%, after the conventional water flooding with injection of ionic liquid and polymer slugs.

Main Objectives

Improve Oil Recovery and help to Mitigate Greenhouse Gas Emissions

New Aspects

Heterogeneity (layering)

Summary

In this work, a laboratory test to investigate the influence of permeability heterogeneity and ensuing crossflow on the recovery performance of immiscible water alternating CO2 injection was performed. The results reveal the negative impact of heterogeneity in the vertical direction on ultimate oil recovery from layered cores. Contrary to our previous conclusions about continuous immiscible CO2 flooding, it possible to observe that the crossflow to negatively affect the RF of immiscible WAG in layered samples. This contradiction may suggest that WAG flooding would achieve a stable frontal advance in each layer in non-communication layers.

Main Objectives

The main objective of this study is to characterize the impact of direct current on the interfacial tension between condensate droplet and brine.

New Aspects

Experimental set-up designed for this study is novel. The study explains the results obtaind by Zhang et al 2019.

Summary

Several methods have been applied to mitigate condensate banking effect observed in gas condensate reservoirs. Some of these methods have significant impact on the environment (subsurface and surface). Implementing environmentally sustainable solutions is increasingly becoming an important area of research. The introduction of direct current increases oil displacement efficiency in conventional oil reservoirs while retaining beneficial properties to the environment. To successfully apply this technique on gas condensate reservoirs, the behaviour of condensate droplets and brine in the presence of electric current need to be understood. In this study, a laboratory experiment was designed to capture the effect of electrical current on interfacial tension. Pendant drop tensiometry was used to obtain the interfacial tension while electric voltage (0-46.5V) was varied during the entire experiment. Results from the experiment, reveal an increase in IFT as the voltage is increased. This study concludes that the interfacial tension increases when the DC voltage is increased beyond a threshold.

Main Objectives

the dynamic IFT and swelling factor are measured experimentally. this an lead to a more detail mechanistic study in solvent enhanced oil recover processes.

New Aspects

reporting both dynamic sweling factor and dynamic IFT of oil/brine/ther system

Summary

In present study the effects of diethyl ether (DEE) on the interfacial property of the crude oil/brine system was investigated using a pendant drop method. Both IFT and the dynamic swelling factor in the presence of DEE was measured at 30, 50 and 70 °C and 2000 psi pressure. Seawater(SW) was selected as the base fluid and its different concentrations, i.e., 0.1SW and 2SW were prepared to mimic low and high salinity water. Results showed that at the SW, with adding the DEE to the aqueous solution, there is a meaningful reduction in the IFT profile was observed after 200s and it finally reached an equilibrium-IFT of 2.93 mN/m. With the increase of temperature from 50 and 70 °C, the equilibrium-IFT of 2.5 and 2.20 mN/m, were obtained. By reducing salinity to 0.1SW and adding DEE, IFT decreased with a slower rate to 6 mN/m as compared to SW and showed a longer equilibrium time. As to the salinity of 2SW, the trend of dynamic IFT was similar to the case of SW. Moreover, a larger dynamic swelling factor was observed in the presence of salt and DEE in the aqueous phase.

Main Objectives

Study the possibility to apply cationic surfactant with anionic polymer in carbonate reservoir

New Aspects

Provide the insight of the mixture of cationic surfactant and anionic polymer for application in carbonate reservoir

Summary

This work provides insight into the interactions between the anionic polymers with different structures and the cationic surfactants with different self-assembly structures for the potential application in improving oil production from carbonate reservoirs. Cationic surfactants were selected to reduce the adsorption on carbonate rock surfaces. Anionic polymers include partially hydrolyzed polyacrylamides, sulfonated polyacrylamides and hydrophobic associating polymers. Cationic surfactants with different self-assembly behaviors were used. The mixtures of anionic polymers and cationic surfactants precipitated in deionized water due to the charge neutralization. By increasing the solution salinity, the compatibility became better and the solution turned to be transparent in high salinity water. Rheological measurements indicated that cationic surfactants significantly decreased the viscosity of hydrophobic associating polymers, and showed little effect on the viscosity of partially hydrolyzed polyacrylamide solutions. These results suggest that the polymer side chain affects the properties of surfactant-polymer mixtures. When mixing the hydrophobic associating polymers with cationic viscoelastic surfactants, significant viscosity increase was observed even at high temperature. This indicates that self-assembly structures of cationic surfactants play an important role in the performance of polymer-surfactant mixtures. Both polymer structures and surfactant behaviors should be taken into account in the design of optimal formulations.

Main Objectives

These studies highlight the importance of EOS tuning in compositional modeling especially as the fluid composition varies significantly in the reservoir and in cases where the injected fluid is different from the resident fluid and particularly when the injected fluid is CO2.

New Aspects

Validation of EOS predicted data with swelling data

Summary

The gas-condensate fluids behavior especially when CO2 is added is complex. Therefore, it is important to measure appropriate types and amounts of PVT experimental data that are used to tune the equation of state model that appropriately describe their behavior.

In this study, the phase behavior of supercritical carbon dioxide (CO2) with a binary gas condensate mixture at two temperatures (20C and 60C) has been studied. Various amounts of CO2 was added to the original fluid at a pressure of 5500 psi which is above the dew point pressure of the mixture before performing the constant composition expansion and swelling tests. The dew point pressure, liquid dropout, and swelling Factor were determined for each level of CO2 addition.

Some of these data obtained from these experiments were used for tuning of the equation of state (EOS) whilst others were applied to determine the predictive capability of the tuned EOS.

By identifying the binary interaction of light-heavy component as an important tuning parameter, the effective tuning of EOS parameters for such fluid systems was demonstrated.

The results of this work are beneficial for subsequent studies on efficient EOS tuning and designing CO2 injection for gas condensate recovery and CO2 storage purposes.

Main Objectives

We incorporated the experimentally validated surfactant models into a high-resolution reservoir simulation, in order to understand how complex geological structures and multi-scale heterogeneities impact the recovery processes and effectiveness of surfactant-based EOR at the inter-well scale.

New Aspects

We found that the efficiency of surfactant-based EOR in a high-resolution outcrop analogue model was impacted by huge uncertainties inherent in geological heterogeneity as well as the experimentally derived surfactant model.

Summary

Conventional carbonate reservoirs contain most of the world’s oil and gas reserves. However, the recovery factors after primary and secondary oil recovery remain low overall. Surfactant-based enhanced oil recovery can be used to enhanced oil recovery by changing the wettability of the rock and reducing the interfacial tension.

In this study, we incorporated the experimentally validated surfactant models into a high-resolution outcrop analogue model. We built core-scale simulation models based on the experimental data to identify the key physical mechanisms that lead to increased oil recovery. We then parametrise the respective models in order to obtain a better understanding of physico-chemical processes during surfactant injection. Using advanced assisted history-matching methods, we were able to match the laboratory results for both the Spontaneous Imbibition and core flooding.

We then implemented the results from the core-scale at a larger scale i.e., inter-well scale model. We used a realistic analogue outcrop for the Shuaiba and Kharaib formations in the Middle East, in order to understand how complex geological structures and multi-scale heterogeneities impact the recovery processes and effectiveness of surfactant-based enhanced oil recovery.

Main Objectives

Interplay between wettability alteration and emulsion generation during low salinity injection

New Aspects

role of brine composition to generate unwanted water-in-oil emulsions in spite of wettability alteration

Summary

Low salinity water has been proposed as a promising mean to enhance oil recovery mainly through wettability alteration mechanism. However, role of brine composition to generate unwanted water-in-oil emulsions has not been addressed well, which might lead to operational challenges during oil production. To have a promising LSW process, simultaneous study of wettability and emulsion analyses seems to be necessary. Thus, this paper aims to investigate the effect of water chemistry on both wettability alteration and emulsion stability at 80 ˚C using an asphaltenic crude oil and a clayey sandstone sample. Rock wettability was addressed by measuring dynamic contact angle profile. Emulsion analysis was performed via phase separation study aided by mean droplet area of W/O emulsion. As to results, lowering salinity from seawater (SW) to 0.125SW provided a positive impact on wettability alteration performance, while also generated a strong emulsion. This shows that LSW might not always produce promising results. As to ion effect, among potential determining ions (Mg²+, Ca²+ and SO₄²-) sulphate promoted wettability alteration toward more water wetness with the lowest emulsion stability. Therefore, we conclude that presence of SO₄²- in injected water can stimulate LSW process with the least impact on the formation of stable emulsion.

Main Objectives

Lateral and vertical variation in the Tertiary reservoir succession and the techtonic influence

New Aspects

Injectite indentification and submarine fans/lobes

Summary

The Tertiary succession is a promising reservoir in the western margin of the Barents Sea. Previous studies have mapped injectites and submarine fans in a single 3D seismic dataset in the Sørvestnaget Basin. Wells have confirmed the reservoir potential, but the lateral variation of these reservoir rocks and the role of active tectonics in their distribution is not yet properly understood. Even though both the tectonics and the sedimentation are studied, they are only studied separately and there are no references displaying the role of tectonics on the sedimentation for the Tertiary succession in this area.

The technostratigraphical evolution during Tertiary in the study area can be related to the regional geological mechanism such as, rifting, uplift, subsidence, and glaciation. All of these which have an important factor for controlling the deposition and remobilization of sediments in the subsurface. Salt halokinesis have together with the regional tectonic events also affected the Tertiary succession to a large degree as a major halokinetic event occurred during Eocene, reworking the sediments by truncation, erosion, and the formation of local deep basins.

Main Objectives

Reservoir quality

New Aspects

Using a centralised database to reduce the time currently taken to conduct multi-technique source-to-sink studies so that they are used more frequently when assesing reservoir quality.

Summary

We outline a novel source-to-sink workflow for utilising hinterland datasets to predict reservoir quality and distribution in frontier exploration regions, and apply this methodology to a case study in the Nile Delta.

This source-to-sink example uses datasets otherwise overlooked in the exploration process and reduces the reliance on more speculative inputs such as paleogeographic reconstructions and paleoclimate modelling. The workflow provides quantitative predictions with percentage certainties, allowing explorers to understand the degree to which results can be relied upon.

Main Objectives

Lithological prediction of contourites and bottom-current reworked sediments

New Aspects

Pragmatic classification of contourites with qualification of reservoir & seal potential

Summary

The role of bottom-currents in the reworking and redistribution of sediments has been clearly underestimated. The diagnostic criteria to discriminate contouritic deposits from downstream sediment gravity flow deposits need to be improved. Transport by vigorous bottom currents or in-situ winnowing of sediment can result in reworking and deposition of potential clean sands. This has renewed the interest from both the academia and the O&G industry in search of new exploration opportunities.

Our objective is to provide exploration and reservoir geologists with calibrated analogues for the different sedimentary objects associated to bottom currents, with keys for lithological interpretation and prediction of reservoir and seal potential. The analysis of representative study cases allows us to propose a classification in relation with the main drivers identified, and to define types with specific issues (e.g. reservoir continuity, diagenesis).

The main outcomes of the study are that:

• Contourites have the potential to provide reservoirs and seals

• Their lithology varies depending on type and nature of sediment input and hydrology

• The keys for prediction are based on a review of 22 calibrated analogues

• A pragmatic classification is proposed that links environment, processes and facies, with an estimation of reservoir and seal risks.

Main Objectives

To evaluate possible provenance, depositional setting, petroleum perspectives and economic significance of the early Paleocene successions.

New Aspects

Paleosol analysis

Summary

The lithofacies analysis of the Hangu Formation represents deposition from subaqueous to marginal

marine environment. It is divided into ten distinct lithofacies in five different geological sections. It also represents the development of paleosol. Economic deposits like coal, laterite, bauxite, iron, silica and porous sands for

hydrocarbon reservoir are found in this formation. The diagenesis of the Hangu Formation is not much

complex as it is revealed by petrographic studies. The calcitic, silicic and clayey cements have been

observed which results in porosity loss. The sequence stratigraphic studies of the formation indicates

two depositional sequences i.e., the lower one consists of falling stage system tract and the upper one

consists of lowstand and transgressive system tract.

Main Objectives

to clarify the impacts of three major diagenesis on pore structure

New Aspects

we believe that Feldspar dissolution and calcite cementation occur preferentially in the connected pores with a throat greater than 1um while clay growth have an impact on throat size less than 1um.

Summary

Precipitation and dissolution are two mechanisms that alter porosity and pore size, which have vital implications for a host of geological settings, such as hydrocarbon extraction from a subsurface reservoir, carbon sequestration in geological formations and diagenesis in a sedimentary basin. In this paper, we construct parameters to quantitatively illustrate the effects of three major diagenesis processes (feldspar dissolution, calcite cementation, and clay growth) on the pore structure in geological sample though microscope observation and reservoir characterization experiments. Results show that feldspar dissolution and calcite cementation occur preferentially in the connected pores with a throat greater than 1um while clay growth have an impact on throat size less than 1um. At the same, the parameters have an excellent relationship with pore structure, porosity, and permeability, which is of great importance for reservoir physical prediction in study area.

Main Objectives

Controlling factors of deep buried high quality reservoirs

New Aspects

Study on burial history and microfractures of grains

Summary

The Junggar basin is a large petroliferous, superimposed basin in Northwest China. The burial depth of Cretaceous Qingshuihe Formation sandstone reservoirs in the southern margin is 5500-7000m.Although the burial depth is large, the sandstone reservoirs are of good physical properties. The Qingshuihe reservoirs have mineralogical immaturity and moderate textural maturity. The pore system is mainly primary intergranular pore and a few secondary dissolution pore. Microfractures are developed in some grains. More initial porosity lost by compaction than cementation. Eodiagenesis in Qingshuihe sandstones is dominated by mechanical compaction, cementation by calcite and analcime. Mesodiagenesis mainly consist of further mechanical compaction, dissolution of feldspars and cements, formation of illite and mixed-layered I/S, Fe-calcite cementation. The formation of deeply buried high-quality reservoirs in the Qingshuihe Formation is controlled by four key factors. Firstly, grain size and rock composition played a significant role on initial porosity. Secondly, early long⁃term shallow burial and late rapid deep burial is conducive to the preservation of primary pores in the reservoir. Another factor is intragranular micro-fractures, which can increase the permeability of the reservoir and the overpressure also is an important factor for the formation of high-quality reservoirs.

Main Objectives

Provide sequence stratigraphic model for Triassic and Jurassic in Central and South Viking Grabens, and compare relationships and primary control of depositional facies on porosity and permeability

New Aspects

Volume of porosity and permeability data integrated with sequence stratigraphic model

Summary

This study presents a cross-border sequence stratigraphic framework for the Triassic and Jurassic in the South Viking and Central Grabens in the North Sea. The Triassic remains underexplored in both discoveries and availability of data, including biostratigraphic control and core. By comparison the Jurassic is the dominant targeted play in the Central Graben and is data rich, offering reasonable recovery of biostratigraphy and core data to constrain depositional systems. Porosity and permeability data from conventional and sidewall core analyses from over 1000 wells are integrated into the sequence stratigraphic model to confirm the control of depositional facies on reservoir characteristics. This study aims to present an integrated approach confirming the primary control of depositional facies on reservoir in the Triassic Skarerrak and Smith Bank Formations and the Jurassic Hugin and Sleipner Formations. Local disparities are observed from this trend and are likely due to a combination of factors, which include mechanical compaction, pressure-temperature regimes and diagenetic overprint.

Main Objectives

Prediction of diagenetic facies using well logs

New Aspects

Application of BP neural network in diagenetic facies logging identification

Summary

In this study,four diagenetic facies are identified according to the thin section petrography and core experiments, and they are: (1) tightly compacted facies; (2) dissolution of unstable components facies; (3) carbonate cement facies; (4) microfracture facies. The dissolution of unstable components facies and microfracture facies show the best reservoir quality. BP neural network can be used to build a model for diagenetic facies prediction using well log, and the blind testing of this model in well DB102 suggest it could be applied for uncored wells in the study area.

Main Objectives

to demonstrate the potential of non-replicated 4D seismic in combination with survey design for this particular case

New Aspects

non-replicated, blended survey design for 4D seismic in combination with simultaneous inversion of all available vintages

Summary

Time-lapse, or 4D seismic is capable of satisfying the continuously increasing demand for high-quality subsurface images to reveal both static and dynamic elements during the field development. However, in practice, challenges of pursuing this strategy lie in different perspectives related to budgetary, operational and regulatory constraints. Seismic surveys, performed in a compressed manner in time and/or space, can provide high-quality seismic datasets in a cost-effective and productive manner. The processing of data acquired in this way usually requires decompression, e.g., deblending and data reconstruction. The decompression performance is of fundamental importance in determining the success of compressed measurements. Our decompression approach deals jointly with deblending and data reconstruction via a sparse inversion, coupled with constraints on causality and coherency. Additionally, we carry out the inversion simultaneously for all available vintages, sharing static information between them while extracting the dynamic changes. We use this inversion as the kernel of a survey-design scheme. We use artificial intelligence (convolutional neural network) to speed up the computations. In our experiment, using time-lapse data from the Troll field, the improvement of designing the acquisition geometry combined with the simultaneous inversion of all available vintages was 6 dB.

Main Objectives

Emit low frequency energy in marine surveys to improve application of Full Waveform Inversion

New Aspects

New survey designs for sparse nodal surveys

Summary

The recent success of long offset nodal surveys in the Gulf of Mexico (GoM) for FWI based velocity estimation and subsequent RTM based imaging has rekindled interest in OBN survey designs. A major cost component of such surveys is the size of the source halo surrounding the nodal patch. For “regular” OBN surveys in the GoM, focusing on reflection imaging, the size of this halo can vary between 4 and 10 km depending on imaging objectives. Recent long offset nodal surveys in the GoM typically acquire much larger source halos in the order of 20-30 km to enable FWI based velocity estimation utilizing refractions off basement. In this abstract we will discuss options offered by a new generation of Low Frequency (LF) enhanced marine seismic sources for the design of more efficient long offset nodal surveys. The designs will retain much of the higher low frequency output of these sources, which is crucial for the performance of FWI in salt dominated environments.

Main Objectives

This ultra-deep and sparse source technology intends to dramatically reduce cost and turnaround time for smaller more targeted 4D monitoring projects. Development, design, implementation and full-scale testing of new Ultra Deep Small Seismic Source technology for 4D monitoring of reservoirs.

New Aspects

New revolutionary ultra-deep small seismic source combined with multi-domain deconvolution and full wavefield imaging to obtain large high resolution 4D images from very sparse and simple source distribution.

Summary

With a desire to reduce the time and cost to obtain a 4D image across the Edvard Grieg field, we set out on a journey to re-think and develop a new source method for much faster 4D turnaround time. We envision a new 4D image every two weeks! In order to achieve this, we have designed, built and tested a new ultra-deep small seismic stationary source. The source emits its wavefield close to the seafloor and the upgoing ghost wavefield is used as the primary image point in addition to all orders of multiples. An advanced imaging flow is currently under development capable of handling the very sparse source geometry of 800 x 800m shot spacing. Key steps include multidomain deconvolution and full wavefield imaging technology in order to create virtual source locations from the ghost wavefield at each seafloor receiver location. The new deep source will be further tested for small-scale 4D repeat surveys on a weekly or bi-weekly schedule at the next opportunity.

Main Objectives

Illustrate the feasibility of characterizing the acoustic wavefield generated by the seismic vessel itself from towed streamer data to image the subsurface.

New Aspects

Using “passive” towed streamer data to image the subsurface.

Summary

The acoustic wavefield originating from the seismic vessel itself is generally not treated as acoustic signals in the imaging of marine seismic data, and is typically categorized as ambient noise. If this acoustic wavefield could be characterized, the noise could be attenuated. Also, this ambient noise may be used to image the subsurface, alternatively as a complement to images based on active sources. During a field test of the continuous wavefields method acquired offshore Malaysia late 2019, also more than one hour of “passive” seismic data were acquired without triggering any airguns. In this paper, estimation of the acoustic wavefield generated by the seismic vessel itself using these data is discussed. The continuous wavefields method has been used to create a seismic image of the subsurface from these data. This image is compared with an image produced from data acquired by triggering individual airguns with short random time intervals.

Main Objectives

Determination of Explosive source signature

New Aspects

Needs two shots of different size

Summary

I present a new method to estimate the source time functions of explosion source seismic data, using a modified acquisition method. Two shots of different size are fired into the same geophone spread with source-receiver geometry arranged such that the Green’s functions are essentially the same. The spectral ratio of corresponding seismic traces is then the spectral ratio of the source time functions. A corresponding synthetic ratio filter is calculated from Blake’s explosion source model for the sources within each pair. Each modelled source is defined by five parameters: the minimum radius of the elastic zone, the internal pressure at this radius, the density of the rock, and two elastic constants for an isotropic rock, the P-wave velocity and Poisson’s ratio. A grid search finds the source parameters that minimise the difference between the measured and synthetic filters for each source pair, thus yielding the two source time functions, but not their absolute amplitudes. Deconvolution of the shot records within each pair for the minimum-phase source time functions recovers the estimated earth impulse response gathers, convolved only with the response of the recording system.

Main Objectives

Demonstration using extensive field trial data of how GNSS use can be optimised for low-power land seismic nodes to provide adequate time control and self-survey capability whilst maximising battery life

New Aspects

Large database of measured clock drift behaviours from multiple field trials, correction using recorded temperatures to infill GNSS hiatuses

Summary

Although a GNSS chip remains an essential component of the next generation of small land seismic recording nodes, their power consumption is high compared to other aspects of the electronics. To maximize the lifetime of the battery and minimize node size and weight, GNSS use should be optimized. We use two field trial examples from the development of the latest generation of land seismic receivers to show how GNSS usage can be optimized without compromising on timing accuracy or the ability of the nodes to self-survey to an accuracy sufficient for seismic processing. Without temperature calibration, we find that as few as four time-checks per day are sufficient to ensure that sample time errors are less than ±1ms. With temperature calibration, GNSS hiatuses of up to several days can be corrected. For short surveys, the lower limit on the number of fixes may be set by positional requirements, with at least 60 measurements being required to measure X, Y and Z to within ±2m.

Main Objectives

Full-azimuth, full-offset, broadband marine seismic data

New Aspects

Autonomous marine seismic surveys

Summary

FreeCable™, a new marine seismic acquisition method was tested in the Mediterranean Sea. For this sea trial, the system used two Midwater Stationary Cables (MSCs), four Recording Autonomous Vessels (RAVs) and a control room installed onboard a master vessel. Each MSC is fitted with four-component stations (hydrophone and tri-axial geophone), spaced at 25m interval. The source was a 150in3 airgun and its depth was 7m. The work presented here is two-fold: i) analyse the frequency contents recorded with one MSC of 1.5km length (#6003 line, ‘Pourquoi Pas’ survey) ii) examine the water bottom reflection recorded on both the hydrophone and the vertical geophone in terms of amplitude and acoustic impedance. The data analysis demonstrated that the hydrophones record useful signals in the 2-4Hz frequency band. This is a remarkable result all the more that the airgun emits very few LFs. Despite the small source and the geophone low-cut filter, water bottom reflection is observed in the 6-12Hz band. The amplitude and acoustic impedance maps are very stable for each sensor and each cmp. The sensor coupling is accurate, high-fidelity (in a 4C sense) and repeatable. The FreeCable method avoids the repeatability issue which is critical in 4D seismic surveys.

Main Objectives

Geophysics acquisition design

New Aspects

Zipper design

Summary

The West Kuwait 3D project is massive channels project, which firstly applies INOVA’s G3iHD acquisition system and over 200,000 SL11 digital detectors. The project is adjacent to the border between Iraq and Saudi Arabia. The border area controlled by the military and surrounded by barbed wire. The operation time can only be 12 hours daytime. In addition, many uncertain factors are there. If the operation organization is not good, the production efficiency will seriously affect. There are two air fire ranges inside the survey, inside which are full of debris and many unexploded munitions. The permission for air fire range between zipper 4 and zipper3 only have 5 months. How to complete this short period without causing downtime is big issue. This article describes how to rapidly pass through the military restrict area by technical method.

Main Objectives

The main objective of this work is to analyze the AVO based seismic reflectivity at various oil saturation and pore pressure changes during production or field development by using AVO inversion schemes.

New Aspects

In this study, P converted P and P converted S inversion schemes are used to map the saturation-pressure changes in the reservoir rock and their comparison is made to find the best inversion scheme at low and higher incident angles.

Summary

The fundamental goal of geophysical tools in the 4D seismic data analysis and field development phases is to make a reliable assessment of fluids saturation and pressure changes. In the present study, the sensitivity of variation of seismic reflection amplitudes by various P converted P and P converted S AVO inversion schemes to the saturation-pressure changes during oil production is analyzed and a comparison is made for the better assessment to examine the reflectivity response during production. It is found that the reflectivity response computed by the lowest order expression and integrated differences of the P and S wave velocities and densities can provide more prominent reflection amplitudes on pre-stack seismic data when oil saturation is 20% or less. These methods also give better results at higher incident angles (>30) than the lower ones.

Main Objectives

Proposing a pragmatic approach to account for the noise model in probabilistic seismic inversion.

New Aspects

Assessment of the effect of the noise model (bandwidth and magnitude) on the results of direct probabilistic inversion of seismic data to porosity. The role of the wavelet in the noise covariance matrix and its effect on the inversion results were evaluated.

Summary

Accounting for the noise is a fundamental step in probabilistic inversion approaches as well as a challenge for the practitioners. To investigate a pragmatic approach for noise description in probabilistic seismic inversion, we used a probabilistic sampling-based inversion method to invert seismic data associated with a hard carbonate reservoir in southwest Iran to porosity. We assumed eight different scenarios for the bandwidth and the magnitude of the noise. The posterior statistics assessment shows that ignoring the correlation of the noise samples in the noise covariance matrix generates unrealistic features in porosity realisations. Furthermore, underestimating the noise magnitude leads to overfitting the data and generates a biased model with too little uncertainty. These issues are resolved considerably when the noise bandwidth is considered in the inversion setup. This indicates the tangible effect of the noise bandwidth on the statistics of the posterior realisations. Our analyses showed that constructing the noise covariance matrix using extracted seismic wavelet is a practical solution. Our tests also show that the error propagated to the posterior realisations by using imperfect wavelet frequency spectrum in construction of the noise covariance matrix is insignificant.

Main Objectives

Show case successful application of seismic inversion for reservior characterisation and highlight the importnace of a well defined prior model

New Aspects

Data quality assessment and Prior model impact

Summary

Deep water fields with turbidite depositional systems have been successfully characterised based on deterministic inversion. We describe a case study on seismic reservoir characterization using seismic inversion outputs on a deep-water gas field, offshore Mozambique and highlight the importance of a well-defined prior model to achieve good inversion results. The field is composed of deep-water turbidite reservoirs with thick (up to 100m) homogeneous sand intervals characterised by structurally complex toe-thrust and associated faulting to the west, and more stratigraphic and lithology complexities in the east. Therefore, a successful inversion process is one that can capture the structural and stratigraphic complexities in the results. The prior model is essential to aid seismic inversion give useful output. Therefore, the prior model was carefully built to capture the identified complexities and guide the inversion process. This was pivotal to constraining the inversion especially, in the toe-thrust affected region. Inversion results obtained reflected the field complexities and had good correlation with elastic logs from well data. Pseudo volume of clay and porosity datasets were computed based on the inversion results to help reservoir characterization and used for detailed fairway interpretation and geomodel construction.

Main Objectives

uncertainties evaluation for AVO

New Aspects

uncertainties evaluation for AVO

Summary

AVO (Amplitude Versus Offset) technique is widely used for de-risking fluids inside prospects.In this extended abstract, the objective was to develop a new family of attributes “confidence attributes” allowing a better understanding of AVO uncertainties. So, with the help of confidence attributes geophysicist could better map parts of the anomalies that are highly robust to the choice of the process parameters. All results are here illustrated on a real deep offshore turbiditic field.

Main Objectives

Improve accuracy of reservoir properties derived by post-stack seismic inversion

New Aspects

Accounting for interbed multiples and simulation of stacking as part of the inversion procedure

Summary

Seismic inversion is a well established technique to derive elastic properties of the subsurface. Ideally, the inversion is performed on pre-stack gathers in order to allow recovery of two or sometimes even three independent parameters. During the exploration stage or in case only vintage data is available, this may not be feasible though. Post-stack inversion for a single property, e.g. acoustic impedance, is often the only approach towards project de-risking by seismic inversion. In this case, the post-stack data is frequently approximated as normal-incidence data while multiple scattering is neglected. This means that the underlying inversion engine produces only an approximation of the actual seismic data. In this paper we explain and demonstrate the concept of deploying the wave equation for post-stack seismic inversion. The wave equation is solved iteratively, meaning that not only primaries are accounted for but also interbed multiples are properly modelled. Here the proposed workflow is tested on a dataset from the Brasse field which is located in the Norwegian sector of the North Sea. Comparison of the results with available well log data validates that accounting for interbed multiples in post-stack inversion improves the accuracy of the derived elastic property.

Main Objectives

Create AVO feasibility maps away from well control in areas with complex geology

New Aspects

Innovative integration of basin modeling, seismic stratigraphy, rock physics and AVO.

Summary

A new integrated workflow for generation of AVO feasibility maps to be used in prospect de-risking is presented. We demonstrate the workflow on data from the Barents Sea. The methodology enables rapid extrapolation of expected rock physics properties away from well control, along selected horizon, constrained by seismic velocity information, geological inputs (basin modelling, seismic stratigraphy and facies maps) and rock physics depth trend analysis. The workflow should allow for more rapid, seamless and geologically consistent DHI de-risking of prospects in areas with complex geology and tectonic influence. The AVO feasibility maps can furthermore be utilized to generate non-stationary training data for AVO classification.

Main Objectives

This paper discusses the effect for strong anisotropy on AVO according to the quasi Zoeppritz equation for incident qP wave in VTI medium.

New Aspects

strong anisotropy, AVO, VTI medium

Summary

The rock physical experiments show that shale is often moderately even strong anisotropy. This paper discusses the effect for strong anisotropy on AVO according to the quasi Zoeppritz equation for incident qP wave in TI medium with vertical symmetry axis (VTI medium). Theoretical analysis and numerical examples are indicated that the strong anisotropy will cause observable changes to the reflection coefficient. When the anisotropy is strong, δ has a great effect on Rpp at both small and large angles. In the case of strong anisotropy, the effect of ε on Rpp at large angle is much greater than that of δ.

Main Objectives

AVO, Heavy Oil Reservoirs, Petrophysical Property

New Aspects

Study shows that petrophysical property (porosity) is the main influencing factor of AVO responses for heavy oil reservoirs, while the influence of fluid can be neglected. A new method is therefore proposed in this paper to predict petrophysical property of heavy oil reservoirs based on the variation of AVO responses, which is applied in Bohai Oilfield and proved to be effectice.

Summary

Although Amplitude Versus Offset (AVO) technology has become an effective tool for hydrocarbon detection, few researches have been reported regarding AVO applications on heavy oil reservoirs. Based on Biot-Gassmann theory, fluid substitution, porosity substitution and AVO forward modeling are integrated to analyze the sensitivity factors of AVO characteristics for heavy oil reservoirs. And petrophysical property (porosity in this paper) is proved to be the main influencing factor. A new method is therefore proposed in this paper to predict petrophysical property of heavy oil reservoirs based on the variation of AVO responses. A case study is subsequently illustrated in JZ heavy oil field within Bohai Bay Basin, northeast of China. AVO attribute analysis is conducted based on suitability study of well logs and seismic data. Result shows that the lateral variation of porosity reflected by AVO attribute is consistent with the drilling outcomes, which confirms the feasibility of proposed method for petrophysical property prediction in heavy oil fields, and provides references for well deployment in subsequent exploration and development.

Main Objectives

reservoir characterization, rock physics, pre-stack inversion

New Aspects

A new elastic parameter ρ/Vs is proposed and the lithological identification factor F is further established through crossplot analysis, which can differentiate sand from shale very well. Pre-stack inversion is then conducted and reservoirs associated with different amplitudes are all clearly characterized.

Summary

Reservoirs associated with strong, medium and weak seismic amplitudes are drilled at the same time within Paleogene Dongying Formation in Bohai Oilfield, which indicates P-impedance based post-stack inversion can hardly be effective for reservoir characterization. Rock physics guided pre-stack studies are therefore necessary for further hydrocarbon exploration. However, study shows that it’s difficult to distinguish sand from shale effectively using conventional elastic parameters. In this paper, we have attempted to reveal the essential rock physics regularity and then a new lithological identification factor is constructed. The research starts by rock physics analysis of the original well-logs, which shows that Vp of sand and shale overlaps with each other, making it hard to distinguish between the two using parameters such as Ip/Is, Poisson’s ratio and λρ/μρ. However, density is found to be relatively more sensitive to lithological variations but still can’t differentiate sand from shale due to compaction effect. A new elastic parameter ρ/Vs is therefore proposed and the lithological identification factor is established through crossplot analysis, which can differentiate sand from shale very well. Pre-stack inversion is then conducted within this block, reservoirs with different amplitudes are clearly characterized and the result is successfully applied in exploratory well deployment.

Main Objectives

Bulk density and S-wave velocity of subsurface medium are crucial for reservoir characterization and fluid identification. However, it is difficult to obtain reliable estimate of both parameters, especially density, from the limited-offset reflected PP-wave. In this paper, we present a simultaneous inversion of the reflected SH waves based on a linear approximation of the SH-wave reflection coefficient.

New Aspects

We present a simultaneous inversion of the reflected SH waves based on a linear approximation of the SH-wave reflection coefficient. Both coefficients corresponding to Vs and  in the linear approximation are “model-parameter independent”, there is no need to estimate Vp/Vs ratio, which is necessary in PP-wave or PS-wave AVO/AVA inversions. The field data example demonstrates that our proposed method can recover stable and high-accuracy density and S-wave velocity information.

Summary

Bulk density and S-wave velocity of subsurface medium are crucial for reservoir characterization and fluid identification. However, it is difficult to obtain reliable estimate of both parameters, especially density, from the limited-offset reflected PP-wave. In this paper, we present a simultaneous inversion of the reflected SH waves based on a linear approximation of the SH-wave reflection coefficient. The approximation has high accuracy, and it includes only two parameters (natural logarithm of Vs and ) to be inverted. Both coefficients corresponding to Vs and  in the linear approximation are “model-parameter independent”, there is no need to estimate Vp/Vs ratio, which is necessary in PP-wave or PS-wave AVO/AVA inversions. We show that the inversion of SH-wave is well-posed when using data of moderate incident angle (30°-40°). The new simultaneous inversion method is applied to an SH-wave prestack dataset from a 2D nine-component survey. The field data example demonstrates that our proposed method can recover stable and high-accuracy density and S-wave velocity information.

Main Objectives

To share learnings about suitable inversion methodologies in geological settings with thin beds and coals

New Aspects

Application of a recent probabilistic inversion to a challenging field

Summary

Seismic inversion of thin-bedded multi-lithology stratigraphic systems suffers from problems of non-uniqueness despite significant advances in acquisition and data processing. The Cambo Discovery (West of Shetlands, UKCS) provides a good case study comprising stacked sub-seismic sand reservoirs interbedded with thin coals and shales. Appraisal wells drilled on 1990’s vintage conventional streamer data have proved areas of thicker (but still sub-seismic) sand bodies away from the crestal discovery well. These data were considered too poor quality to confidently define and map sand bodies using quantitative interpretation techniques. The elastic rock physical properties of brine sands and shale overlap at log-scale on rock physics templates and at seismic scale the hydrocarbon-bearing sands can also overlap with variable coals.

Nevertheless, over the past 5-6 years our confidence in quantitative interpretation techniques has significantly improved. Broadband dual azimuth streamer data acquired in 2011 and 2014 provide high quality data. Developments in seismic inversion techniques, specifically direct probabilistic inversion, have allowed more confident lithology prediction. We share results of both deterministic simultaneous AVO inversions and a recent direct probabilistic inversion to lithology. The results of the latter show a significant improvement in prediction of sand distribution.

Main Objectives

To develop a workflow for quantitative shale characterisation at a regional scale.

New Aspects

Predictive generalised linear model for smectite content versus elastic attributes. Modular workflow for quantitative seismic interpretation for shale composition and hydraulic permeability.

Summary

Predicting the mineralogical composition and hydraulic permeability of shales is of utmost importance for drilling operations related to hydrocarbon exploration/production and for the assessment of their sealing capacity as hydrocarbon or CO2 barriers, respectively. Despite the importance of inferring these two parameters for industrial applications few methods have been developed. Here we introduce a workflow to infer shale composition at a regional scale and asses its background hydraulic permeability that combines seismic data, well logs, and laboratory measurements. The workflow is applied to the Duyfken 3D seismic survey acquired in the central part of the Northern Carnarvon Basin where a regional smectite-rich seal with a thickness of more than 1 km is hindering hydrocarbon exploration. Results of the quantitative seismic interpretation for shale mineralogical composition are verified against laboratory XRD measurements from a test well that was not used for interpretation. The results have a good match to test well data within the determined uncertainty bounds. Using these results background permeability of shales is estimated using empirical relationship derived from the experiments on artificial shales.

Main Objectives

Utilize both PP and PS converted wave data from OBC acquisition into joint 4D wave-equation based inversion for 4D pressure- and fluid- prediction.

New Aspects

4D joint inversion of both PP- and PS- data using full wave-equation based inversion

Summary

Over the last years, wave-equation based AVO inversion (WEB-AVO) has become an established geophysical tool to extract elastic subsurface properties from migrated seismic data. In this paper we take full advantage of ocean-bottom 4D seismic by performing a joint WEB-AVO inversion using both PP and PS converted wave data. The 4D data was acquired across the Edvard Grieg field in 2016 and 2018 and both PP and PS migrated gathers have been used as input to the study. As shear wave data is insensitive to fluid changes in the reservoir, inclusion of this datatype allows improved separation between pressure effects and water flooding effects. The WEB-AVO technique is particularly suitable to handle time-lapse seismic data, because the non-linearity caused by reflectivity changes together with travel-time changes is automatically handled by the wave-equation. Inclusion of PS data for the 4D time-laps inversion has provided improved detection and separation of pressure and fluid effects and the work will soon be extended to include the 3rd 4D survey acquired in 2020.

Main Objectives

Unconventional Reservoir Characterization

New Aspects

PP/PS Joint Inversion, Multicomponent Seismic, Horizon Mapping

Summary

The focus of this study was to perform an enhanced characterization of the unconventional play of the Vaca Muerta formation. In 2017 a multicomponent test survey was acquired over an area of about 13km² with the aim of reliable fraction and rock elasticity information to reduce uncertainty of this shale oil development project. To test this concept, we employed a joint inversion method developed by Hampson et al. (2005) and Russell et al. (2005) that makes use of the compressional wave (PP) and the shear wave (PS) components. PP/PS Post-Stack joint inversion delivered straightforward and fast results for P- and S- impedance and gave enhanced insights into the elastic properties of the formation. Comparing horizons picked in PP and PS time provided fast insight and helped in identifying a subtle fault that can have a significant impact on later development stages. On the Vp/Vs ratio dataset, a result from the joint inversion, low values in upper part of lower Vaca Muerta formation were observed. These values could reflect zones of overpressure which can be caused by leak-off of pressure through the underlying Tordillo formation. Another possible explanation might be that this low Vp/Vs values reflect zones of higher organic content.

Main Objectives

To stop S-wave artifacts from propagating in acoustic orthorhombic media.

New Aspects

The new acoustic assumption and the new eikonal equation

Summary

Application of acoustic orthorhombic media can be considered as a modern standard in industrial 3D seismic data processing and modeling stages. However, the presence of significant S-wave artifacts that are associated with the conventional acoustic anisotropic media harms any application that need pure P-wave propagation. Accordingly, we propose a new acoustic assumption for orthorhombic media that mitigates the S-waves artifacts by zeroing the S-waves velocities along the symmetry planes. The accuracy of the new approach compares well with the conventional approach, but the new approach further complicates the governing equations. Noticing the algebraic complexity of the eikonal equation obtained in the new acoustic orthorhombic media, we also propose a rational approximation that simplifies the eikonal equation.

Main Objectives

AVA inversion for VTI media

New Aspects

A nonlinear Bayesian joint AVA inversion method based on the exact Zoeppritz equations for VTI media is developed

Summary

For shale reservoirs with VTI anisotropy characteristic, the inversion methods based on linear approximate formulas are commonly used to estimate the elastic parameters and anisotropy parameters. However, the calculation accuracy of linear formulas is low, which limits the estimation accuracy of these parameters. In fact, the linear approximation formulas are made up of isotropic and anisotropic terms. Numerical experiments show that if we use the exact Zoeppritz equations to replace the isotropic term in these formulas, the calculation accuracy of reflection coefficients can be improved. This can reduce the influence of the calculation error introduced by the forward operator on the inversion result. Therefore, in order to improve the estimation accuracy of reservoir parameters, we constructed a Bayesian nonlinear inversion objective function based on the combined equation obtained by substitution. In addition, the differentiable Laplace distribution blockiness constraint term was also added to the Cauchy background prior model to further improve the vertical resolution of inversion results. Synthetic data test shows that the proposed method can not only inverts Thomsen anisotropy parameters stably, but also accurately estimates the vertical P- and S- wave velocities and density, which demonstrates the effectiveness of this method.

Main Objectives

Shear wave velocities in elliptical orthorhombic media

New Aspects

Shear wave characteristic equation and fundamental singularity point.

Summary

We analyze the shear waves characteristics in elliptical orthorhombic media. It is shown that there is a fundamental characteristic equation for S waves phase velocities and the fundamental singularity point located in one of the symmetry planes.

Main Objectives

Accurate Facies Analysis versus Azimuth (FACIVAZ) to improve the prediction of hydrocarbon-saturated permeable fractures in terrigenous carbonate reservoirs.

New Aspects

1. The use of the amplitude-preserved, full-azimuth local angle domain (LAD) common image gathers. 2. The ability to detect the theoretical and laboratory effects of reflected acoustic signals propagating across fluid-saturated cracks: a greater absorption at low frequencies (due to fluid flow) than along the cracks.

Summary

This work presents a novel technology for azimuth-dependent facies analysis (Facies Analysis versus Azimuth – FACIVAZ) to improve the prediction of hydrocarbon-saturated permeable fractures in terrigenous carbonate reservoirs. The analysis is performed in the depth domain, along high-resolution, full-azimuth angle domain common image gathers created by a special Local Angle Domain (LAD) imaging system. The amplitude and phase preservation obtained by this imaging system is crucial to the proposed analysis. Prior to the facies analysis, the general orientation and intensity of the target fracture systems are analyzed and characterized by azimuth-dependent velocity and amplitude analyses (VVAZ and AVAZ) performed along these LAD gathers. The remaining effect of the azimuth-dependent absorption and dispersion is then detected and further related to the rate of the oil-saturated fractured reservoirs. The examples presented in this work show the effectiveness of the proposed FACIVAZ technology in accurately predicting the distribution of seismic facies in target production areas associated with oil-saturated fractured reservoirs, in an active oil field in Western Siberia. The results highly agree with the corresponding facies characteristics measured in the boreholes along the reservoir area, and therefore serve as valuable information for further drilling of new wells.

Main Objectives

1. An accurate and reliable method to converge into stationary rays whenever conventional two-point ray tracing methods are challenging. 2. Obtain the nonlinear kinematic ray tracing ODE for spatially-varying general anisotropic media, in terms of the ray velocity magnitude and its derivatives with respect to location and direction. 3. Solve this equation with an original finite-element formulation. 3. Test the proposed approach with a numerical example.

New Aspects

An original Lagrangian has been proposed and validated for 3D heterogeneous general anisotropic (triclinic) media, where the higher symmetries can be considered particular cases. Explicit expressions for the local (related to a single Hermite-type finite element) traveltime gradients and Hessians have been developed, with respect to the location and direction degrees of freedom. A new method has been suggested to compute the geometric spreading from the global traveltime Hessian.

Summary

Considering 3D heterogeneous media and general anisotropy, the kinematic Eigenray is an original finite-element optimization method for obtaining stationary ray paths (minimum traveltime or saddle point solutions due to caustics) between two fixed endpoints in cases where conventional two-point ray tracing methods are challenging. Based on Fermat’s principle, we introduce an original Lagrangian that we consider the most convenient for our finite-element solver. For the minimum traveltime, the target function includes the traveltime and two essential penalty functions, and it is optimized with the Newton method. For the saddle point, the traveltime in the target function is replaced by the traveltime gradient squared and a gradient optimization method is used. In each iteration, the trajectory is optimally discretized by a number of nodes and segments, with the Hermite interpolation. The degrees of freedom are node locations along the path and the nodal ray (or ray velocity) directions. In either case, both the global gradient and Hessian of the traveltime are needed and computed. The traveltime Hessian of the stationary path is then used to compute the relative geometric spreading between the source and receiver. Finally, we introduce the normalized relative geometric spreading to be used as a propagation complexity criterion.

Main Objectives

1. Obtain accurate dynamic properties along rays characterized by complex wave phenomena. 2. Obtain the linearized dynamic ray tracing ODE for spatially-varying general anisotropic media, in terms of the normal shift vectors of the paraxial ray, and the velocity magnitude of the central ray, and its location and direction derivatives. 3. Solve this equation with the finite-element method, applying the weak formulation and the Galerkin method. 3. Test the suggested approach with an anisotropic numerical example.

New Aspects

A new Jacobi-type equation has been derived for the dynamic ray tracing in 3D heterogeneous anisotropic media. An original, suitable set of ray coordinates (RC) has been suggested to characterize a point-source paraxial ray. A finite element analysis with Hermite interpolation has been applied to solve the Jacobi ODE in order to compute the ray Jacobian. A new relationship has been developed between the Jacobian and the relative geometric spreading for the chosen RC.

Summary

Considering 3D heterogeneous media and general anisotropy, and a parameterized stationary ray path obtained by the kinematic Eigenray, we analyse the second traveltime variation and derive the linear, second-order, vector-form Jacobi dynamic ray tracing (DRT) ODE. It delivers paraxial rays defined by shift vectors normal to the ray direction in its vicinity. The solution is obtained by applying the same finite element scheme (with the Hermite polynomial interpolation) used in the kinematic Eigenray. The resolving matrix of the linear algebraic DRT equation set coincides with the global traveltime Hessian already computed in the kinematic Eigenray stage. Two different solutions of the Jacobi DRT ODE with their corresponding point-source initial conditions, related to the chosen ray coordinates (RC), are needed to compute the ray Jacobian representing to the signed cross-area of the corresponding ray tube. For the chosen RC, we derive an original relationship between the ray Jacobian and the relative geometric spreading. The proposed Eigenray method has been tested using a benchmark numerical example with orthorhombic elliptic factorized inhomogeneous anisotropy. This model has an analytic solution for the ray path configuration, traveltime, arclength, and parameter sigma. The results demonstrate the high accuracy obtained by the proposed method.

Main Objectives

Analysis of seismic anisotropy due to squirt flow

New Aspects

To our knowledge, this is the fist 3D numerical study focused on seismic anisotropy due to squirt flow.

Summary

Activities involving CO2 geological sequestration, exploitation of geothermal energy, and hydrocarbon exploration require a deep understanding of cracked rocks’ behavior under different conditions, so that non-invasive monitoring methods, based on seismic methods, for example, can be developed. Many studies highlight that seismic waves propagation is highly affected by the presence of micro-cracks, grain-scale discontinuities and moving fluids at the pore scale. Strong velocity dispersion and wave attenuation due to fluid flow at the pore scale, known as squirt flow, have been observed in many experimental studies. We perform a three-dimensional numerical study of the fluid-solid deformation at the pore scale based on solving quasistatic equations for solid and fluid phases. We show that seismic anisotropy exhibits frequency-dependent behavior due to squirt flow between interconnected cracks. We demonstrate that the overall anisotropy of the model mainly increases due to squirt flow. We analyze the Thomsen anisotropic parameters and use another scalar parameter to measure the overall anisotropy of the numerical model. Our analysis suggests that seismic anisotropy variations with frequency are nonlinear with frequency, very sensitive to the material properties and the pore space geometry, thus, a general prediction of the seismic anisotropy behavior in different scenarios is not possible.

Main Objectives

To illustrate changes to the processing sequence necessary to utilize wide azimuth data in a shallow water setting.

New Aspects

For a shallow water WAZ scenario, highlighting the importance of 3D deghosting for WAZ data, using NAZ data to predict multiples on WAZ, and the value WAZ data and LSq migration bring to subsurface imaging.

Summary

We discuss the processing and imaging challenges relating to a marine seismic survey acquired northwest of the Shetland Islands. The proximity of the survey to the islands forced the acquisition direction to be strike to the subsurface geology. A shooting vessel provided wide azimuth data to illuminate in the dip direction, and triple sources increased crossline sampling. We show how 3D source designature and source/receiver deghosting were required for this wide azimuth data. For demultiple we illustrate how utilizing narrow azimuth data in water layer multiple modelling was crucial for the wide azimuth multiple prediction. In addition we highlight the benefits of a wave equation deconvolution based approach in this shallow water setting. For velocity model building we show how using time-lag full waveform inversion helped resolve the specific challenges created by the geological setting. Finally we demonstrate how that imaging was aided by the use of least squares migration and the inclusion of the wide azimuth aspect.

Main Objectives

To promote ultra-high density to a wider exploration market and beyond O&G applications.

New Aspects

Processing of ultra-high density nodal survey, with nominal trace density approaching 200 times and more than that of typical conventional cable-based surveys.

Summary

The desire for ultra-high density (UHD) seismic surveys is now becoming more achievable for future exploration and field development with the increasing availability of versatile nodal land systems. Acquisition geometry design using a higher density of sources as well as receivers considerably reduces the effects of spatial aliasing and also provides better subsurface illumination. By sampling the wavefield more densely, there are improved recordings of both signal and noise. This presents new opportunities for processing and imaging. We use a recent UHD nodal survey with nominal trace density approaching 200 times that of typical conventional cable-based surveys to discuss the challenges and rewards.

Main Objectives

Due to severe surface fluctuations and complex near-surface structures, static correction problems has become a major bottleneck restricting the accurate imaging of underground structures in the complex mountainous regions and piedmont zone in western China, which has become the focus of current exploration.

New Aspects

we propose a tomographic static correction method for complex mountainous region with “Dual Variable velocity” technology

Summary

Due to severe surface fluctuations and complex near-surface structures, static correction problems has become a major bottleneck restricting the accurate imaging of underground structures in the complex mountainous regions and piedmont zone in western China, which has become the focus of current exploration. Traditional tomographic static correction method applies constant replacement velocity and constant bottom boundary of the low deceleration zone, with which the static correction problem cannot be solved perfectly. In this abstract, we propose a tomographic static correction method for complex mountainous region with “Dual Variable velocity” technology, that is, according to the surface conditions of the survey, different static correction values are calculated by the bottom boundary of different low-deceleration zones in different areas and the corresponding replacement velocity (“dual-variable velocity”), and then the optimal static correction parameters are confirmed. Secondly, the variational fitting function is used to fit the static correction of the transition region, and the final static correction of the whole transition region is obtained. The method is applied to the field data in the Kulongshan area in piedmont of the Qilian Mountains, and the results show that the method effectively improves the static correction accuracy of complex mountainous areas.

Main Objectives

Study on the static correction method under the complex surface condition in the Loess Plateau of Ordos Basin

New Aspects

The tomographic static correction technology is applied to solve the problem of static correction under the complex surface conditions in the Loess Plateau area of Ordos Basin, and has achieved good results.

Summary

In the south of Ordos Basin, the Loess Plateau is characterized by vertical and horizontal gullies, dramatic changes in surface undulation and large changes in thickness of loess layer, which lead to serious problems in static correction. How to effectively solve the problem of static correction is the primary task in the process of seismic data in Loess Plateau area. This paper mainly studies the application of tomographic inversion static correction technology in the Loess Plateau area of Ordos Basin, and compares the results with refraction static correction technology. The results show that tomographic inversion static correction technology can effectively solve the problem of static correction under the complex surface conditions of the Loess Plateau.

Main Objectives

Seismic data can be considered as the convolution between reflection coefficient and band limited wavelet. The usual wavelet is band limited, whose main lobe has the long time and side lobe energy is strong. Therefore, seismic signal has the serious interference and low resolution. The main lobe of wide-band Ricker wavelet (also called Yu’s wavelet) is narrow and side lobe amplitude is small. Due to the seismic data with wide-band Ricker wavelet’s spectrum characteristics, the interference effect is weak, which is more suitable for reservoir description.

New Aspects

In this paper, the spectral inversion method based on the geological statistics is used, and the spectral inversion technique is improved.

Summary

Seismic data can be considered as the convolution between reflection coefficient and band limited wavelet. The usual wavelet is band limited, whose main lobe has the long time and side lobe energy is strong. Therefore, seismic signal has the serious interference and low resolution. The main lobe of wide-band Ricker wavelet (also called Yu’s wavelet) is narrow and side lobe amplitude is small. Due to the seismic data with wide-band Ricker wavelet’s spectrum characteristics, the interference effect is weak, which is more suitable for reservoir description. In this paper, the spectral inversion method based on the geological statistics is used, and the spectral inversion technique is improved. We can get the broadband data with high frequency and low frequency, and then the broadband data is transformed to the seismic data with Yu’s wavelet spectrum characteristics in order to improve the accuracy of reservoir description. The method proposed in this paper is tested by model data, and practical applied to real oilfield in Bohai sea, the improvement of seismic data is fitter for the reservoir description, which shows the proposed method is feasible and effective.

Main Objectives

Application of data Regularization Based on optimized Matching Pursuit methods in Qaidam basin

New Aspects

first applicaiton to Qaidam basin

Summary

Due to physical and financial constraints, acquisition geometries rarely deliver a perfect spatial sampling, which may cause arc phenomenon in migration and amplitude distortion, and thus affect imaging quality. In this paper, we use the improved Matching Pursuit Fourier Interpolation technique to regularize the irregular seismic data in Qaidam basin, which is in the northwestern of China. The field data results show that MPFI technology can solve the above problems well while preserving AVO, AVAz and the details of the seismic wavefield, which turns out to provide good pre-stack data for subsequent fidelity imaging processing.

Main Objectives

Mixed Sources

New Aspects

Mixed Sources,merging processing

Summary

In some areas with complicated surface conditions, such as mountains, towns, and reservoir nature reserves. Seismic acquisition by dynamite source is prohibited. Although through the shot-point transferring, the overall coverage can be achieved, the lack of shallow or near-offset information has a great impact on near-surface velocity model inversion, velocity analysis and imaging. Therefore, seismic acquisition by a combination of dynamite source and vibroseis is adopted. However, the wavelets, amplitudes, frequencies, and phases acquired by mixed sources are inconsistent, which will cause the illusion of fine interpretation. In order to solve the mismatches of mixed source, this paper proposed a high-precision matching processing method overcoming the shortcomings of conventional matching filtering processing method, which does not consider the influence of noise. This method solves the inconsistency problem of the data collected by mixed sources, eliminates the differences caused by different excitation methods, and improves the S/N ratio and imaging quality of seismic data. Actual processing results show that the high-precision matching processing method of mixed sources can effectively solve the inconsistency problems and can be applied to merging seismic data processing.

Main Objectives

Application of Full-layer Q Compensation Technology has great promotion and application value

New Aspects

Near-surface Q compensation technology , Q tomography , Q migration imaging technology

Summary

Higher seismic imaging resolution is in need as the targets of exploration and development are transformed from structural reservoirs to lithologic reservoirs, which are becoming more and more complex. To meet the demands of lithologic exploration, seismic data results should be broadband and amplitude preserved. A full-layer Q compensation technology which consists of surface absorption compensation, Q tomography and Q pre-stack depth migration imaging technology is raised. It takes into account the near-surface anomalies with strong absorption and the influence of viscoelastic absorption and attenuation factors during the propagation of seismic waves. Different absorption and attenuation problems from shallow to deep are treated separately. The amplitude attenuation and phase change caused by the absorption of stratum in the travel path from shots to detectors are compensated. The resolution and amplitude attributes are effectively improved in the study area. The seismic result supports the prediction of continental shale oil sweet spots and the deployment need of horizontal wells in Songliao Basin.

Main Objectives

Acquisition of well sampled high quality multi-source towed marine data

New Aspects

Worlds first 3D towed marine signal apparition simultaneous source survey

Summary

Simultaneous source acquisition offers the prospect of productivity increases as well as the acquisition of better sampled data. Whereas a productivity increase is straightforward to achieve for OBS, it is sometimes less so for towed marine configurations. On the other hand, towed marine configurations significantly benefits from better sampled data in terms of near offsets, wide azimuth, inline sampling and crossline midpoint distribution. These are areas that simultaneous source techniques can directly address. In this paper we present results from the worlds first 3D towed marine multi-source acquisition based on the principles of signal apparition. We demonstrate that signal apparition results in a much better sampled data set resolving individual diffractors in a complex overburden without compromising on SNR when compared to conventionally acquired flip/flop data.

Main Objectives

Compare image resolution from conventional and apparition data

New Aspects

First 3D apparition streamer survey, insights into processing challenges with apparition data.

Summary

This paper presents acquisition and processing results from a 3D streamer acquisition trial of triple-source apparition data, with an emphasis on processing. We show that, after apparition de-blending, initial migration results did not show resolution improvements over a reference unblended dual-source dataset. However, attentive 3D processing of the same de-blended data gave substantial improvements to image resolution. Time slices from the fully processed apparition data show higher resolution than the corresponding reference data.

Main Objectives

High quality simultaneous source separation enables highly accurate OBS spoke processing including up/down separation and multiple attenuation

New Aspects

Dense shot grids by dense simultaneous source acquisition enables up/down decon multiple elimination and imaging

Summary

Ocean Bottom Seismic (OBS) data that are adequately sampled on the source side allow for up/down wavefield decomposition and removal of all surface-related multiples by means of deconvolution of the downgoing wavefield from the upgoing wavefield. However, conventional sequential acquisition of an adequately sampled shot grid can be costly. Instead, simultaneous source acquisition provides a means to both achieve a productivity increase over conventionally acquired data while also enabling the acquisition of a better sampled shot grid. In this paper we present the results from the first 3D triple-source simultaneous source OBS survey acquired using the principles of signal apparition. We demonstrate that the decoded data allows for a simple bespoke OBS processing workflow to be used resulting in high quality multiple-free images.

Main Objectives

Investigate multiple methods of deblending for reflection and refraction data

New Aspects

Review of deblending in RTM image domain

Summary

Deblending is a process that removes noise and crosstalk from the imaged volume, but significant care must be taken to leave signal content unharmed when determining the distribution of recorded energy onto the output traces. This paper investigates the effects of multiple methods of deblending, including an iterative denoise approach and a hybrid approach utilizing inversion methods and iterative denoise methods, on long offset sparse node data in both time and RTM image domain.

Main Objectives

Improve efficiency with high density sampling, high resolution images, improve data for AVO analysis

New Aspects

ultra-wide-tow sources, very near offset coverage

Summary

This paper presents a case study of a 3D marine streamer seismic acquisition using an ultra-wide-tow penta source set-up to improve trace density and near offset coverage without compromising acquisition efficiency. The survey was shot in 2020 in the Barents Sea where the water depth is approximately between 300 m and 400 m. The total source separation in crossline direction was 315 m. The acquisition and processing grid was 6.25 m x 6.25 m and the processing was carried out at 2 ms temporal sampling. The high number of sources introduces additional blended signals which were simultaneously separated using an iterative deblending method. We then demonstrate the effectiveness of a standard demultiple workflow which benefited from the rich near offset distribution and the early out migrated images show high resolution quality in the very near surface.

Main Objectives

Introduction of the first pilot survey using a new blending method

New Aspects

Productivity enhancement using a new blending method

Summary

Recently, we established a blended-acquisition method: temporally signatured and/or modulated and spatially dispersed source array, namely S-/M-DSA, that jointly uses various signaturing and/or modulation in the time dimension and dispersed source array in the space dimension. We have acquired the first pilot survey with S-/M-DSA onshore Abu Dhabi. In this paper, we introduce this pilot survey and the resulting acquisition productivity enhancement in the time dimension. Furthermore, we discuss how this method could enhance the acquisition productivity in the space dimension as well. These show that S-/M-DSA significantly enhances the acquisition productivity compared to conventional blending methods.

Main Objectives

Propose a new method to separate simultaneous sources. This paper shows how one can use both inversion and denoising techniques to deblend simultaneous sources datasets following an understandable cost function. This paper also introduces regularization by denoising (RED) as a alternative method to solve inverse problems in geophysics.

New Aspects

For the first time, the regularization by denoising methodology is being employed to solve the sources separation problem. Instead of using a gradient descent method, this paper emphasizes the use other solvers, such as fixed-point iterations or ADMM, to deblend datasets.

Summary

Simultaneous source separation, or deblending, techniques can be classified in inversion or denoising-based methods. This paper exploits the advances in both of these classes and proposes to bring them together to formulate a powerful deblending approach based on regularization by denoising. Such an approach poses the inverse problem through a clear and comprehensible cost function where a regularization term employs denoising techniques in its definition. Through numerically blended real data examples, this paper shows that RED can achieve good deblending results.

Main Objectives

Deblending of spatially aliased apparition seismic data

New Aspects

Deblending of spatially aliased apparition seismic data

Summary

We describe a novel multi-dimensional sparse inversion algorithm for apparition de-blending which can handle spatial aliasing and overlapping signal cones. We compose the forward modelling apparition blending operator using radon-like phase shift kernels or atoms. Furthermore, this operator includes the periodic shot-by-shot time delays for each source-line. The source de-blending exploits signal sparsity in the projected domain and relies on a sparse solver to reconstruct each individual source. Windowing in space, time and overlapping frequency bands enhances signal sparsity and isolation from the interfering sources. Synthetic and field data examples show the efficacy of the methodology in presence of spatially aliased apparition seismic data.

Main Objectives

To demonstrate the use of 4D time-shift signal in conjunction with amplitude change to clearly identify gas and water saturation change in geologically complex fields.

New Aspects

Multiple attributes for 4D interpretation; Generation of dV/V attribute in entirely data-driven way.

Summary

The PSVM development is situated in the Lower Congo basin, deep-water Angola, commencing production in 2012. Towed streamer 4D datasets have been invaluable for reservoir surveillance, allowing the reservoir management team a high-resolution time-lapse view of the dynamic subsurface.

Traditionally, 4D interpretation has been based on the full stack 4D difference (Mon – Base) where the monitor is time-shifted to align to the baseline prior to subtraction. After time-shift compensation, the 4D difference is generally well resolved and in simple geological settings it can be quite straightforward to interpret. However, the PSVM fields are often composed of more geologically complex reservoirs with multiple sands stacked inside a single producing interval. Discriminating signal from side-lobe can therefore be nontrivial. Furthermore, many of the fields are close to saturation pressure. Small amounts of depletion can bring gas out of solution and invoke a widespread softening response in the 4D. So understanding the polarity of a 4D event can be critical to describing water movement through a field.

In this study, time-shift signal is treated as an interpretation product in its own right. In conjunction with amplitude change, fields with very complex geology and dynamic history can be interpreted with greater confidence.

Main Objectives

Geomechanically-consistent overburden time shift inversion for 4D seismic data

New Aspects

Novel method to invert 4D seismic differences through an analytical geomechnical model

Summary

We present a new moethod for estimating overburden time-shifts from 4D seismic data. Using an analytical geomechanical model, the seismic differences are inverted to give a volumetric strain map at reservoir level. Time shift solutions modelled from this volumetric strain are therefore inherently consistent with the geomechnical response to production effects in the reservoir. Since this scheme has a signifcantly reduced null-space, compared to existing purely seismic-based techniques, the solutions are better-constrained and less susceptible to noise. We demonstrate the method, and compare it to existing techniques, on a North Sea example.

Main Objectives

independent injected volumes estimation from 4D geobodies

New Aspects

new method for the signal threshold determination for geobody picking

Summary

4D attributes such as time shift and amplitude changes can allow us to monitor a variety of production mechanisms (fluid injection, depletion or aquifer water sweep). In good quality 4D data, geobodies can easily be picked from 4D attributes like relative impedance changes. In this paper, we compare volumes from 4D seismic geobodies with those of injected fluid volumes through several case studies among operational fields (CO2 storage and water injections). This automatic computation of relative volumes between monitor and/or completions from geobodies allows for a better choice of threshold and assessment of uncertainty. Results show fairly good consistency between the global injected volumes and computed volumes from 4D data. Results show the potential of 4D seismic in monitoring the volume in injected fluid for CCS. For large fields with large amount of data (numerous wells, monitors, …), such automated processes to propagate, define threshold and calculate volumes can add value to our understanding of reservoir behaviour and provide independent measurements of relative injected fluid volumes.

Main Objectives

Demonstration of added value from prestack 4D analysis.

New Aspects

Successfull application to a real dataset, separating pressure and saturation.

Summary

4D AVO remains a rarely-used tool. Why is that? Conventional wisdom is that the majority of 4D effects occurring in the reservoir are already detectable and interpretable using acoustic amplitude differences and/or time shift information. We present 2 ways that we can access pre-stack 4D information: one very simple method, that helps us determine if application of a full elastic 4D inversion workflow is a worthwhile exercise; and an elastic extension to joint 4D seismic inversion. A real data example illustrates the potential added value of such an approach, where we separate changes in pressure and saturation effects at a water injection well.

Main Objectives

To show the possibility and advantages of assimilating the decoupled seismic information, by using the results of the Bayesian 4D seismic inversion to pressure and saturation changes. To show the impact of incorporating the Bayesian inversion uncertainty estimations as areal localisation weights to the seismic data maps.

New Aspects

Seismic history matching technique with seismic data assimilation in the pressure and saturation domain using the results of a Bayesian 4D seismic inversion. Use of the Bayesian uncertainties in a localisation weighting technique.

Summary

We present a seismic history matching workflow where 4D seismic data is assimilated in the pressure and saturation domain, using the results of a previously computed Bayesian 4D seismic inversion. The inversion provides an estimation to the changes in pressure and saturation and also their related uncertainties. This uncertainty estimation is assimilated into the history matching workflow in a weighting approach, by penalizing the seismic inversion maps with respect to the Bayesian inversion uncertainties. We show that the proposed penalization method leads to a focused assimilation of the trustworthy seismic information, which may prove crucial when assimilating seismic data in the pressure and saturation domain. The workflow is applied to a real data case to assess and update global reservoir connectivity in a North Sea reservoir.

Main Objectives

Highlight the successful applications of 4D seismic in the reservoir management of carbonate reservoir and quantify 4D sensitivity to dynamic reservoir properties with his impact into 4D monitoring strategy

New Aspects

4D monitoring strategy including production history analysis and 4D forward modelling for a carbonate reservoir

Summary

Following the convincing impact of 3D seismic and encouraging results of the 4D pilot over Al Shaheen carbonate reservoirs, a significant seismic program has been initiated by North Oil Company (NOC) including Ocean Bottom Node (OBN) technology to be set as a new reference for 4D monitoring. To answer the 4D monitoring strategy question, a qualitative assessment of the 4D sensitivity to dynamic reservoir property variations and a quantitative estimate of some production parameters (cumulative oil production, GOR) was undertaken based on field examples.. A few months of oil production in depletion corresponding to 0.6 MMstb produced, GOR~1000 scf/stb depending on reservoir properties) seem sufficient to obtain an interpretable 4D signal associated with Sg increase. Consequently the observations point towards monitoring gas saturation variation for reservoir heterogeneities characterization with dedicated seismic acquisitions based on drilling schedule considering the quick turn-around of acquisition and processing.. This concept has been validated through 4D forward modelling and will be re-assessed with the 2019 survey, updated PEM model and extended considering new full field models (2020). Other aspects such as logistics, contracting will also need to be considered into the 4D monitoring strategy.

Main Objectives

Assessing data error for 4D Seismic History Matching, Uncertainties from processing workflow

New Aspects

Managing uncertainties in 4D seismic and the simulation model for 4D seismic history matching

Summary

Managing uncertainty in 4D-QI is important, especially for 4D Seismic History Matching (SHM), Most descriptions of 4D seismic error do not capture the full uncertainties in the seismic data or do not make the correct assumptions when looking at error. The error may not only be on 4D seismic but can also be in the simulation model so, to match the 4D observed to the predicated from simulation, assessing the error in both may be required. Here we address the data errors for 4D-QI and SHM.

The error on 4D seismic data is assessed by generating multiple realisations of the post-stack volumes with alternative pre-stack processing workflows. The 4D stacks for all the realisations are inverted for saturation. The error in the legacy model is assessed by creating multiple realisations of the simulation model with differing scenarios for the multipliers of horizontal and vertical permeability then ΔSWpred derived. To identify which ΔSWobs optimally matches with ΔSWpred, the difference between ΔSWobs. and ΔSWpred maps from all the simulation models is taken. The results show there is no single 4D observed realisation that we can rely on for 4D-QI and also that no single simulation model can match the observed data.

Main Objectives

Integration of reservoir simulation, rock physics, seismic modelling, and geomechanics analysis to simulate CO2 injection and monitoring in carbonate reef reservoir

New Aspects

Full-scale use of carbonate rock physics method to be integrated from and to reservoir simulation, seismic modelling, and geomechanics

Summary

Carbonate reef reservoir is a very important reservoir in oil and gas industry worldwide. However, the challenge of studying carbonate reservoir lies on the complexity of its facies distribution and pore structures, which makes it having almost unpredictable properties if not thoroughly examined. The application of CO2 Enhanced Gas Recovery (EGR) to increase gas productivity from carbonate reservoir is also currently on the business. In the purpose of dealing with carbonate complexities, an interlinked study of reservoir simulation, rock physics, seismic modelling, and geomechanical analysis is developed. The study is also applied in the decision-making for monitoring of EGR activity to ensure its safety. A case study of CO2 injection pilot project in Gundih depleted gas field in Indonesia, is used for the implementation of this study. The result is pleasing, allowing us to comprehend more about the physical process of EGR and reservoir monitoring in carbonate reef type.

Main Objectives

Applicability of an ultra-light seismic monitoring system every six months for 5 years on Surmont seismic legacy data and SAGD activities. Results were fact-checked with temperature data from observation wells.

New Aspects

Successful ultra-light monitoring technic using only one source and one receiver couple every 6 months to detect steam effects.

Summary

4D seismic surveys have been completed every 6 months over the years to monitor the development of the steam chamber at Surmont, a heavy oil field located in northeast Alberta, Canada.

Using only one source and one receiver optimally placed in the field, a novel light seismic monitoring has been “blind-tested” on non-migrated data from 2010 and 2015 for 3 spot locations chosen on the pad by ConocoPhillips.

The objective was to see if the steam had reached one zone and not another by plotting the evolution of time-shift in these spots without knowing previous history.

The time-shift changes obtained after processing were successfully fact-checked with temperature data from observation wells, confirming the qualitative variations attributed to the effects of the steam chamber evolution in 2 spots out of 3.

This demonstrated the viability of this focused seismic approach to monitor the evolution of the steam chamber in Surmont. This also paves the way for other applications including offshore reservoir monitoring, inspired from this light and agile seismic acquisition system.

Main Objectives

well logging formation evaluation

New Aspects

machine learning applications to well logging

Summary

The existing study single predict neural network were used, that is, for a neural network one petrophysical parameter can be predicted, such as porosity (POR) or water saturation (SW) with a set of logging data. When the tasks are related to each other, the underlying information extracted from the input features of the model has certain commonality. In this case, the multi-task model can extract higher quality of underlying information with the help of multi-dimensional output information, so as to obtain better model performance and produce an effect similar to “information gain” . We propose a “sharing private” multi-task machine learning method for petrophysical parameter prediction with logs, which can improve the efficiency, simplify the process and reduce the mean absolute error compared with single predict neural network. Using logging data from 64 wells in Chinese oilfields, it is found that, compared with the single-task model, the multi-task model can significantly improve the prediction performance of permeability and water saturation.

Main Objectives

Prediction of fracture aperture by logging data

New Aspects

Prediction of fracture aperture using hierarchical expert committee machine model

Summary

Fracture aperture is an important parameter to evaluate the quality of fracture controlled tight clastic reservoir. The well logs were always used to predict the fracture aperture, but some linear regression methods do not match well with complex logging data due to the characteristics of low porosity and low permeability in tight clastic reservoir. The machine learning method can improve prediction accuracy, but it always generates unstable prediction models. A static committee machine (CM) can reduce errors and uncertainties by combining multiple learners, but the weight of integrating learners is difficult to determine. In order to promote the accuracy and efficiency of weight calculation, the CM is improved by using analytic hierarchy process (AHP) and joint neural network (JNN) model. Based on the prediction performance of each expert network, the hierarchical expert committee machine (HECM) model is formed by adding the hierarchical network module adaptively. The experiment shows that HECM model can reduced the relative error of the prediction results, and increased the stability of the prediction model. The HECM model provides a new method for fracture aperture prediction in tight clastic reservoir by mining potential information of input logging data.

Main Objectives

Improving the simulation efficiency of LWD electromagnetic measurements in high-angle and horizontal wells

New Aspects

A 3D finite difference method is applied to model the EM LWD EDDR. Furthermore, the half-space and the quarter-space methods are proposed to reduce the number of unknowns. We can conclude that 1) the new simplification methods are efficient and accurate; 2) the EM LWD EDDR can be affected by the borehole only when the mud is quite conductive.

Summary

The electromagnetic (EM) logging-while-drilling (LWD) measurement plays an important part in high angle and horizontal wells for geosteering and formation evaluation. Among types of EM LWD tool, the new generation tools with low frequency and large T-R transmitter provides tens of meters investigation depths. As a result, the formation structures and rock properties far away from the borehole contributes to the tool responses. Therefore, a 3D algorithm is required to model the problem accurately. In this paper a 3D finite difference method is proposed for the extra-deep directional logging-while-drilling modelling. Furthermore, by using the symmetrical characteristics of the EM fields, a simplified method by converting a full space EM computational problem to a half-space or a quarter-space problem is developed. Numerical results show that the 3D FDM proposed for the EM LWD tools are accurate and the half-space and quarter-space method reduce the degree of freedom to 1/2 and 1/4 with respect to the original 3D problem, respectively.

Main Objectives

We investigate the dispersion curves of borehole sonic waves to obtain new information to understand the bonding conditions of cement to see how the curves change with the arc angle of a fluid channel in the annulus of a cased hole.

New Aspects

The dispersion curves could estimate the formation shear velocity even in the poorly cemented case in fast formation, but only performs well for the widest fluid channel case in slow formation. Squeeze cementing may be needed in slow formation to run sonic logging in cased holes.

Summary

Fluid channels generated in the annulus of incompletely cemented cased holes are problematic in developing oil, gas, or geothermal resources. The dispersion curves of borehole modes in sonic logging bring additional information on the cement bonding conditions to cement bond logs. Our research investigated the effects of the different central arc angles of a fluid channel in the annulus of a cased borehole on the dispersion curves using numerical experiments. Synthetic 3D numerical models are used to simulate wave propagation. We used a modified matrix pencil algorithm to estimate the dispersion curves both in fast and slow formations. Our results indicated that S-wave velocity measurement by flexural dispersion is stable in all cases in fast formation while only possible in the widest arc angle case in slow formation. The monopole can be utilized to detect large fluid channels. Symmetric flexural in monopole response are useful regardless of the source orientation using monopole source. We notice that the slowness of symmetric flexural is sensitive to the small angle of the fluid channel but not for the large angle case. Moreover, symmetric flexural gives a valuable response to determine fluid layer thickness that is not depending on the fluid position.

Main Objectives

To present novel tool for fast forward modeling for far-field sonic imaging

New Aspects

Fast and efficient spectral element modeling of sonic wavefield far from borehole

Summary

Far-field sonic imaging is becoming a well-established tool for high-resolution structural imaging in the vicinity of the borehole. Spectral element modelling (SEM) is proposed and evaluated as a prospective technique for use as a forward modelling engine for sonic imaging. The correctness of the approach is validated by successfully comparing the results against the analytical solution. The tests of SEM capabilities on a realistic synthetic model clearly show that this technique is worthwhile to be advanced further. The memory and computational time estimate demonstrate the necessity of additional work on the method prior to recommending it for practical processing.

Main Objectives

Obtention of a continuous high-resolution velocity model from the free surface up to 200m

New Aspects

Calibration of Near surface seismic data using drilling parameters and acoustic logging.

Summary

On an experimental site, situated in the Cher region (France), two boreholes have been drilled for field experiments. During the drilling, some parameters such as rate of penetration and Torque have been continuously recorded.

Full Waveform Acoustic logging (FWAL) and seismic experiments were conducted. A linear relationship between Torque-to-ROP ratio and acoustic velocity has been computed, in a root mean square sense, to obtain an estimated P-wave velocity log from drilling parameters. A specific procedure based on zoning process applied on acoustic data is used to force the torque to respect the trends of variation of the P-wave velocity. After calibration with acoustic velocity in the 30 – 192 m depth interval, and validation with tomographic velocity in the 0 – 12 m depth interval, drilling parameters allow a prediction of P-wave velocity from the surface up to the terminal depth of the borehole, with a 10% relative uncertainty. The acoustic velocity log from FWAL is by that way extended over the total heigh of the borehole.

The field case shows the benefit of combining hybrid seismic methods (reflection processing, refraction tomography), drilling parameters and acoustic logging to extend the previous velocity model, laterally in the vicinity of the borehole.

Main Objectives

Examination of the capabilities of electromagnetic sounding with an unconventional toroidal source in vertical and deviated oil wells for studying geological environments of different complexity.

New Aspects

Development of high-performance algorithms and multidimensional numerical simulation of electromagnetic signals of two-coil sounding systems with a toroidal source and receiver to investigate the capabilities of their application in the topical issues of petroleum geophysics.

Summary

The research is focused on examining the capabilities of an unconventional electromagnetic source, namely a toroidal coil, with regard to logging vertical and deviated oil wells. With the application of the developed and software-implemented high-performance algorithms, we carried out a large-scale multidimensional finite-difference numerical simulation of the signals of two-coil sounding systems with a toroidal source and receiver in geoelectric models with various degrees of complexity, including thinly laminated macroanisotropic sandstone-shale reservoirs. It was concluded that the measurements of the various electromagnetic field components are independent and complement each other informationally. The numerical simulation enabled selecting the optimal parameters of the sounding systems with toroids – the ranges of the probe lengths and frequencies, which provide a sufficient level of the measured signals, high sensitivity to the reservoir boundaries and deviation angle, and an unambiguous relationship with the resistivity anisotropy coefficient. Moreover, the developed algorithms are a basis for creating fast inversion procedures that enable real-time processing of the electromagnetic responses from a toroidal source in oil wells with different trajectories.

Main Objectives

A novel fusion method is presented to recognize fractures and vugs automatically and improve the efficiency of electric imaging logging data processing.

New Aspects

A fusion technique combining singular spectrum interpolation and mathematical morphology for automatic recognition and characterization of fractures and vugs from electric logging images.

Summary

In order to the fine evaluation on pore structure in fractured-vuggy reservoirs, the mathematical morphology method combined with singular spectrum interpolation is proposed to automatically extract fractures and vugs based on the electric imaging logging data with the high coverage rate and resolution. The singular spectrum interpolation is applied to reconstruct the full borehole electric logging images in spatial domain by calculating the low-rank conductivity matrix. For implementing the edge detection of the conductivity anomalies and constructing the fractured-vuggy pore structure spectrum, the structural elements on different scales and configurations are selected to perform various kinds of morphological filtering operators. The fusion technology combining mathematical morphology and singular spectrum interpolation is utilized to quantitatively characterize fractures and vugs in carbonate reservoirs in the east of the right bank of Amu Darya Basin. The results show that the novel fusion method can not only recognize fractures and vugs automatically, but also help to improve the efficiency of electric imaging logging data processing.

Main Objectives

VSP DAS

New Aspects

DAS Application

Summary

Great advancement in Distributed Acoustic Sensing (DAS) technique has contributed to its wide application in the borehole seismology for some geophysical purposes, such as structural analysis, parameter estimation and reservoir prediction. Tarim Basin, as one of the most petroliferous basin in the northwest China, has been known for the notoriously deep well conditions, including high pressure and high temperature. In this context, VSP applications are greatly limited by the bearing capacity of conventional geophones. Alternatively, DAS technique displays its superiority in good adaption to complex environment. In order to target oil and gas reservoirs with deep burial and complex features, a Walkaway-VSP technique is employed to confront the undesired well conditions, which is based on joint observation by DAS and geophones. This study first compares VSP data respectively recorded by DAS and geophones, then discusses the processing method, and finally obtains a high-precision imaging profile, which lays a solid foundation for the later comprehensive geological research.

Main Objectives

To develop and test robust anisotropy parameter estimation approach applicable for geophone and DAS data at CO2CRC Otway Site.

New Aspects

1)We developed robust approach of travel-time approximation in anisotropic media which is suitable for geophone and DAS data. 2)Tests of developed approach on different types of data show sustainability of the proposed method.

Summary

Stage 2C of the Otway Project involves monitoring of a small-scale (15 kt) CO2-injection using an extensive time-lapse active seismic program. The main components of this seismic monitoring program are 4D surface seismic and 4D VSP surveys acquired before, during and after the injection. Data analysis reveals significant seismic anisotropy of the subsurface, which needs to be estimated and taken into account to improve the quality of imaging with both VSP and surface seismic data.

A wide range of offsets obtained during fifth monitoring survey of the project provides a unique opportunity for anisotropy estimation from 3D VSP data. In this study we compare geophone and Distributed Acoustic Sensor (DAS) VSP data and their applicability for anisotropy analysis. Analysis of DAS data gives anisotropy parameters for the entire depth of the well.

We estimate P-wave anisotropy by analyzing direct-wave VSP arrival times. The study demonstrates significant presence of both polar and azimuthal anisotropy. While vertical-plane anellipticity remains almost constant at 0.1 level for the whole depth range, azimuthal anisotropy changes significantly with depth: from negligibly small in the shallow part with significant increase below the 600 m depth, which most probably indicates the change of stress field at this depth.

Main Objectives

To present a new methodology to image PS VSP data

New Aspects

New NMO formulation for PS VSP Data

Summary

We have developed a new three-term formulation of normal moveout (NMO) correction for PS converted reflection waves in VSP data. The formulation is accurate for multi-layered media with a large source offset (up to a ratio of ~2.5 of offset to reflector depth) in VSP survey. Based on the new NMO correction formulation, we also developed a new methodology to improve signal to noise (S/N) ratio and the final imaging quality of PS converted reflection waves in VSP data. The methodology consists of three major processing steps in the common receiver gather domain: NMO correction for flattening/aligning PS converted wave reflection events, median filtering for attenuating coherent and incoherent noise, and reverse PS NMO correction for migrating PS converted reflection waves. The processing of a field walkaway VSP dataset with our methodology demonstrates the effectiveness of the methodology in improving the imaging quality of PS converted reflection waves.

Main Objectives

Utilizing a seismic-while-drilling based system to predict ahead of the drill bit and reconstruct a reliable checkshot profile

New Aspects

Using a large 3D array of wireless geophones to acquire drill-bit noise data in a complex desert environment

Summary

DrillCAM is a fully integrated real-time system for predicting and imaging ahead of and around the drill bit by utilizing seismic-while-drilling data (SWD) as well as assisting with drilling optimization and automation. We present results and analysis of several SWD applications from the first field trial in an onshore well in a desert environment. Data was recorded with an adaptable grid of wireless geophones and top drive sensor using depth range of 0-10,000 ft. DrillCAM delivered robust checkshots while drilling down to a depth of 7300 ft that included a combination of roller cone and PDC bits. The obtained velocity profile was found in good agreement with three nearby offset wells. VSP reflection data were successfully processed and delivered a more accurate estimate of over-pressured formation about 2000 ft ahead of the bit. Finally, a reliable VSP corridor stack was generated and shown to tie the surface-seismic data.

Main Objectives

DAS VSP data imaging

New Aspects

DAS VSP data imaging method

Summary

Fiber optics distributed acoustic sensing (DAS) allows vertical seismic profile (VSP) data to be acquired across the entire wellbore trajectory without any costly well intervention. The utilisation of fiber optics receivers also allows for better spatial sampling compared to conventional wireline VSP. Aiming at the characteristics of high density and low signal-to-noise ratio of DAS-VSP seismic data, we have developed a Gaussian beam based pre-stack depth migration method suitable for the DAS-VSP observation system. The method does not have the limitation of conventional CDP transform imaging on the horizontal layered assumption of underground media. This paper shows the application of Gaussian beam migration of DAS-VSP seismic data which verifies the effectiveness of this method for DAS-VSP data and the advantages of high-precision imaging. We have also made a trial of well and surface joint migration and demonstrated it in this abstract.

Main Objectives

Quantitatively determine the time-lapse P-wave velocity changes induced by the complex hydrogeological and geomechanical processes that occur during the injection of CO2.

New Aspects

Reflection wave-equation traveltime inversion is applied to time-lapse VSP data and provides time-lapse velocity changes introduced during three years of CO2 injection at the Shenhua CCS demonstration site.

Summary

Vertical seismic profiling (VSP) is an important technique for characterizing the changes in reservoirs during the geological storage of carbon dioxide (CO2). To quantitatively estimate the time-lapse changes due to CO2 injection from walkaway VSP data, it is necessary to reconstruct the subsurface velocity field with high resolution and wide coverage. We present a joint workflow composed of reflection wave-equation traveltime inversion (RWT) and full waveform inversion (FWI) to quantify changes in the reservoirs. The application of the proposed scheme to field data collected from the Shenhua site demonstrates that this joint workflow is able to image the perturbations of velocity models associated with the injection of 300,000 t of CO2 at depths of 1500 – 2500 m. The presented workflow can be an effective tool for detecting changes in the properties of deep reservoirs during the monitoring of CO2 geosequestration.

Main Objectives

Trial of the cross-hole seismic approach for monitoring shallow CO2 release experiments

New Aspects

Cross-hole seismic is an effective approach that could complement the monitoring strategies for shallow-release CO2 experiments

Summary

The project SRD3.3 is focused on the prediction and verification of shallow CO2 migration, associated with a controlled release of carbon dioxide within a near surface fault. The project aims to gain a better understanding of CO2 leakage in shallow faults in sedimentary basins and establish reliable imaging methodology of the CO2 migration behaviour. The project’s program for the near surface characterisation and monitoring strategies is focused around borehole seismic methods. The program is complemented with a trial of the cross-hole seismic approach. The paper presents and describes the results of the cross-hole data analysis in application to monitoring the shallow CO2 injections.

Main Objectives

Report a real field study of a CO2 storage prospect considering long-term CO2 dissolution and temperature effects

New Aspects

Long-term CO2 dissolution and temperature effects on a real field scale reservoir model

Summary

Field scale simulation studies are presented for the Smeaheia storage prospect, to assess storage capacity, CO2 distribution, long-term CO2 dissolution and temperature effects. The reservoir selected for storage has pressure communication with a neighbouring fault block, where hydrocarbon production will be ongoing when storage takes place. A sensitivity analysis is carried out to quantify how this depletion affects the storage capacity. Long term simulations up to 20,000 years are also reported to model the evolution of structural, residual and solubility trapping over time. Finally, temperature effects on the CO2 plume distribution are investigated.

Main Objectives

Calculation of dispersion coefficient values for CO2-brine systems through experimental tests

New Aspects

A scaling relationship is introduced to obtain the dispersion coefficient values for CO2-brine systems which will pay the way for dissolution flux estimation procedure

Summary

Mass transfer rate estimation is an important aspect of the dissolution trapping mechanism analysis during CO2 sequestration in saline aquifers which has been studied extensively in recent years. Based on laboratory experiments or direct numerical simulation tools, scaling relationships have been introduced for the mass transfer rate estimation. However, there are discrepancies between these results. To investigate the discrepancies between experimental and direct numerical simulation results, we carried out a series of experiments to quantify dispersion during CO2 convective mixing, which enabled us to obtain robust scaling relationships. Two brine compositions including sodium chloride (NaCl) and a mixture of NaCl and calcium chloride (CaCl2) are separately considered in two levels with two levels of permeability to encompass the applicable range of the Peclet number (Pe). Furthermore, we conducted a series of high resolution direct numerical simulations to show the applicability of the proposed relationship in simulation. Our analysis of results reveals that a power-law scaling relationship based on the Pe best fits the dispersion values. Furthermore, numerical simulation results show that dispersion has a considerable impact on the pattern and amount of CO2 dissolution in brine.

Main Objectives

CCS process optimisation, reservoir rock dissolution monitoring, reducing uncertainty of the flow measurement

New Aspects

Novel and innovative applications of equations of state for CCS process, combined with the rock dissolution experiments results

Summary

The aim of this study is to illustrate how brine composition could affect the CO2 solubility in the aqueous phase (which could affect the performance of installed flow meters on the pipelines), geochemical trapping mechanisms, and rock-fluid interactions and induced pore structure alteration of the host rock. This will be achieved through a combination of thermodynamic modelling and experimental work on reservoir rock dissolution owing to solubility of CO2 in brine and resulted pH variations. Herein, we have described and evaluated different equations of state (EoS) which utilise robust thermodynamic basis for demonstrating the solubility of gas species in the aqueous phase and integrated with Pitzer’s theory for determination of the activity coefficients of the ionic species involved. This methodology proved to be able to describe the chemical equilibria of the ionic species in the aqueous phase under HPHT conditions and in systems of interest with presence of gas, which can lead to rock dissolution phenomenon through rock-acidic fluid interactions. Generally, good agreements between predictions and experimental results are observed, and the thermodynamic modelling results could explain one of the dominating drivers of the rock dissolution phenomenon, which is pH alteration.

Main Objectives

CO2-EOR monitor

New Aspects

Using Fast neutorn cross section

Summary

we propose a method to evaluate the gas-bearing properties of reservoirs quantitatively by using the fast neutron elasticity parameter. However, the fast neutron elastic scattering cross section (FNXS) cannot be measured directly in the process of fast neutron deceleration. Based on the three-detector detection system, we use inelastic gamma and capture gamma information to characterize FNXS. After that, we proposed a model for quantitative monitoring CO2 saturation. Monte Carlo method is used to simulate the response of FNXS with different porosity and CO2 gas saturation. Particularly, the effect of formation pressure, temperature and borehole fluid is corrected. The effectiveness of this method can be demonstrated by simulation application. This study provides an effective method for monitoring CO2 reserves in CO2 gas drive reservoirs.

Main Objectives

The status of a program to develop a wireline deployable three-axis borehole gravity sensor with a target sensitivity of ~ 5 µ Gal is firstly introduced. The paper has also demonstrated through numerical modelling the advantage of this technology with an application in the Aquistore CO2 sequestration site in Canada.

New Aspects

A three-axis borehole gravity tool with a form factor enabling it to be deployed through cased hole is completely novel and has not been presented previously. A workflow that uses three components gravity for understanding survey feasibility and optimal survey-time intervals is novel. A systematic and comparative study of three-axis borehole gravity responses through modelling of CO2 injection is novel.

Summary

Geophysical monitoring of underground CO2 injection forms an important component of a carbon capture and storage program. Until recently, seismic methods were used to map the spreading of CO2 plume and are the dominant technique used for monitoring. However, there is a disadvantage in use of seismic whose relationship with fluid saturation are interpreted based on the empirical rock physics models using effective medium theory and often only known by statistical methods with large error bars. Gravity measurements have a unique advantage among many geophysical methods that the mass density change being detected by the gravity within a reservoir is directly and uniquely related to the dynamic fluid redistribution. In this paper the status of a program to develop a wireline deployable three-axis borehole gravity sensor with a target sensitivity of ~ 5 µ Gal is firstly introduced, using a resonant MEMS (Microelectromechanical systems) vibrating beam technology innovation. Followed by a feasibility study for monitoring of CO2 in a deep reservoir at the Aquistore storage site in Canada. We model and predict the gravity variation due to density changes during a period of CO2 injection. This study demonstrates the pre-survey feasibility modelling of the emerging field of time-lapse gravity monitoring.

Main Objectives

Introducing a novel machine learning multitasking platform for multidisciplinary applications in geosciences

New Aspects

Systemic approach based on multiple machine learning algorithms chained in effective workflows addressed to multi-scale, multi-disciplinary datasets.

Summary

In this paper, I introduce a comprehensive machine learning framework that combines the benefits of complementary algorithms. The user can design his/her own workflow through easy combination of a large number of Python libraries. This approach is addressed to many different types of applications in geosciences at variable spatial scale and for different purposes. I discuss briefly two applications: the first is a case of litho-facies classification of well log data; the second concerns the construction of probabilistic maps of oil distribution using multidisciplinary geophysical data.

Main Objectives

To evaluate the performance of machine learning algorithms in predicting dewpoint pressure by considering Number of Training Data points vs testing data points. Also to rank machine learning algorithms using Precision vs Accuracy criteria.

New Aspects

Built Stochastic Decision tree based models, Support vector machine models and simple feed forward neural network models.

Summary

Accurate knowledge of the dew point pressure for a gas condensate reservoir is necessary for the design of a field development plan and timing for optimization of mitigation operations for resources management. This study explores the use of machine learning models in predicting the dew point pressure of gas condensate reservoirs. 535 experimental dew point pressure data-points with max temperature and pressure of 304F and 10500psi were used for this analysis. First, multiple linear regression (MLR) was used as a benchmark for comparing the performance of the machine learning models. Neural Networks (NN) [optimized for the number of neurons and hidden layers], Support Vector Machine (SVM) [using radial basis function kernel] and Decision Tree [Gradient boost Method (GBM) and XG Boost (XGB)] algorithms were then used in predicting the dew point pressure using gas composition, specific gravity, the molecular weight of the heavier component and compressibility factor as input parameters. The performances of these algorithms were analyzed using root mean square error (RMSE), absolute average relative deviation percentage (AARD %) and coefficient of determination (R2). This work concludes that for large data sets neural network is preferred but for smaller data sizes, SVM shows better performance

Main Objectives

Rapid Tsunami Estimation for predefined tide gaguge stations

New Aspects

Providing Useful charts for tide gauge stations

Summary

In tsunami modeling, the source parameter of an earthquake or landslide is computed and an initial water displacement precisely similar to the seabed dislocation is obtained. The Initial Water Displacement (IWD) is propagated in a predefined mesh grid using Computational Fluid Dynamics (CFD). CFD processes depend mostly on the resolution of the mesh grid in the modeling area. In near real scenarios, these procedures consist of high cost (runtime). Here, aside from the source of an earthquake or landslide, we have proposed a formula consisting of three scaling’s and one rotation parameter, for creating a 3-D Initial Water Displacement (IWD). Several CFD processes was performed, and at each time one parameter was changed, and the IWD’s were modeled in a sea like modeling area. The ratio of maximum amplitude and energy of the signal was computed in all of the arbitrary tide gauges. Our results show that if an IWD is extended in the direction of the shoreline, the impact would be much lower than when the IWD is extended perpendicularly to the shoreline. The result of this study is beneficial in obtaining maximum amplitude and energy of IWD’s with variable scaling and rotation parameters in the modeling area.

Main Objectives

The development of method aimed to simplifying the identification of sedimentary and transition to digital sedimentology

New Aspects

New method

Summary

Digital sedimentology as a tool for obtaining numerical parameters of previously poorly formalized petrographic characteristics of rocks provides important advantages in analyzing the structure and properties of a hydrocarbon reservoir. New theoretical and methodological approaches to determining a number of parameters have already been implemented in the software product, which can significantly reduce labor costs, unify data and improve the accuracy of petrographic analysis.

Main Objectives

Present a method that unlocks the true reservoir potential when future production predictions are heavily affected by surface constraints.

New Aspects

Demonstrate a real field example of an operator where the surface constraints heavily impact the future production predictions, how these predictions can be improved and how successful the implementation of the method was.

Summary

Operators of large well count oil fields often experience a production difference between the sum of all individual well test quantities and the total sales quantities. This paper demonstrates a case in which limitations in the surface facilities’ capacity increasingly act as a constraint for production from the reservoir and ultimately impact future production predictions.

A method is presented to improve these predictions using a real field example of a South-American operator. The periodic observed production difference at this operator is far beyond values commonly seen in the industry. It affects the predictions on future reservoir performance and required improvement. The method unlocks the true reservoir potential by leveraging identified surface constraints and plotting valuable rate-time well test data on a set of dimensionless type curves to establish a more representative decline exponent.

The method has been successfully implemented at an operator that is responsible for over 700 wells. It improved the oil rate’s prediction and increased the expected future reservoir performance. It simultaneously firmed-up the estimated reserves, extended the reservoir’s lifetime and effectively upgraded its value without additional costs. Accordingly, it enabled the operator to report more than 200% for the Reserves-Replacement Ratio in its latest audited reserves reporting.

Main Objectives

an efficient upscaling method based on homogenization theory is developed for the HM process in shale matrix

New Aspects

an efficient upscaling method based on homogenization theory is developed for the HM process in shale matrix

Summary

Shale matrix is the main gas storage space, so the development of its Hydro-Mechanical coupling (HM) model is important to macroscopic HM simulation in shale gas reservoir. At microscopic scale, shale matrix is composed of organic and inorganic matter, while the mechanical properties of these two media are quite different, and both gas storage type and transport mechanism are also different in these two media, thus we need to develope different microscale models to describe the HM process in shale matrix. However, microscale models cannot be straightly applied to macro simulation due to their huge calculation cost. In this paper, an efficient upscaling method based on homogenization theory is developed for the HM process in shale matrix, which can accurately represent the microscale characteristics of organic and inorganic matter in macroscale simulations. Firstly, shale matrix is assumed as a heterogeneous poroelastic medium composed of organic and inorganic matter, and according to different storage type and transport mechanism of real gas in these two media, the microscale HM model is developed. Then, the microscale HM model is homogenized to obtained the equivalent macroscopic HM model for shale matrix. Lastly, the accuracy of the proposed method is proved through a numerical examples.

Main Objectives

Estimate elastic parameters based on geostatistics

New Aspects

build initial models for seismic inversion based on geostatistics

Summary

Geostatistical technologies, which include two-point geostatistics (TPS) and multiple-point geostatistics (MPS), are significant in both geological modelling and geophysical inversion. However, TPS is incapable to characterise complex geological structures, due to the dependency on the variogram. When using MPS in the simulation of the continuous variables such as velocity, the calculation and memory burdens are heavy. We integrate TPS with MPS to release these issues. Based on kriging theory, we present a geostatistical interpolation strategy constrained by the lithofacies probability distribution, which is obtained by MPS. The proposed approach utilizes the advantages of TPS and MPS simultaneously. The model test illustrates the effect of the method on depicting complicated geological structure.

Main Objectives

Modeling asphaltene precipitation using CPA and AEOS, Finding the most sensitive tuning variables of the EOSs using Monte-Carlo algorithm, Proposing a generalized Auto-tune method for asphaltene precipitation modeling by EOS

New Aspects

A new and efficient procedure based on optimization and sensitivity algorithms is presented for auto-tune of equations of state such as Cubic-Plus-Association (CPA) and Association Equation of State (AEOS) in modeling asphaltene precipitation

Summary

Complex structure of the asphaltene causes many challenges for predicting its phase behavior. Recently, Cubic-Plus-Association (CPA) and (Association Equation of State) AEOS models, which consider association term, are widely used. Although these two models predict phase behavior of the asphaltene properly, they contain different adjusting parameters. Since, most of the methods that have been presented for tuning of the EOS models are time consuming, a generalized auto-tune technique based on the genetic and Monte-Carlo algorithms is suggested in this work. The proposed method of tuning was applied to the CPA and AEOS models. To confirm the applicability of this technique, the amount of the precipitated asphaltene during titration data for two oil samples was predicted by CPA and AEOS. Results demonstrated that the outcomes of the EOSs were properly matched with experimental values of asphaltene precipitation, which proves the appropriateness of the proposed tuning procedure.

Main Objectives

Clarification the mechanisms relating Marangoni convection which can be profoundly effective in increasing the phase mixing in the ultra-low IFT conditions.

New Aspects

Using numerical simulation to show a Simultaneous Marangoni-driven Convection in a Chemical Flooding Process

Summary

The characterization of Marangoni-driven flow in low interfacial tension for processes like chemical flooding is not well-established. In this study, direct numerical simulation approach was used to investigate the Marangoni-driven convection and also to compare the results with the mass diffusion phenomenon. In this approach, a moving mesh algorithm along with two-phase level-set system and tracer transport module were used. The results showed high convection velocity when an interfacial gradient was exerted to the interface of fluids by the Marangoni effects. The results also revealed substantial differences between the flow velocity due to Marangoni convection and mass diffusion. Furthermore, the Marangoni convection helped with the stabilization of concentration inequalities and also helped the wettability alteration process due to chemical adsorption.

Keywords: Direct Numerical Simulation, Marangoni Convection, Mass Diffusion

Main Objectives

improving oil recovery by investigating dynamic heterogeneity

New Aspects

investigating the impact of new parameters on dynamic heterogeneity

Summary

During a waterflooding process, the oil type and viscosity difference between oil-water will impact the mobility ratio which influences the performance of oil recovery. Mobility ratio and other impacting parameters such as porosity, permeability, well pattern, and average reservoir pressure that directly impacts changes in fluid velocity causes heterogeneity in a reservoir. If we narrow our observations exclusively on fluid behaviour it will fall in the category of dynamic heterogeneity. In this research we have focused our attention on oil type and average reservoir pressure. It is essential to reduce the effects of other parameters (e.g., porosity, permeability) on reservoir heterogeneity to a minimum, in order to observe oil type and pressure exclusively, therefor we have assigned the same permeability and porosity for all grids and have used simulation to choose the least heterogenous well pattern. Eventually, the results demonstrate that the right choice in flow pattern will greatly control the damaging effects of mobility ratio on dynamic heterogeneity, in addition average reservoir pressure and dynamic heterogeneity follow a similar trend in the beginning and the end of water flooding.

Main Objectives

Enhanced Oil Recovery

New Aspects

low salinity water flooding with considering dynamic fluid-fluid and rock-fluid interactions

Summary

The objective of this work is to understand low salinity effect during tertiary water flooding in carbonate reservoirs. Therefore, a coreflooding tests with sequence of formation water (FW), then sea water (SW) followed by diluted sea water (SW10D) are performed on an oil-wet pure calcite synthetic core at 30 °C. It is shown that tertiary SW flooding recovered more additional oil (2.8% of IOIP) compared to the final stage of diluted sea water (0.4% of IOIP). To understand this observation, dynamic interfacial tension (IFT) and contact angle (CA) are measured for different brines.

Tertiary water flooding was also replicated through dynamic IFT measurements. The results show that replacing FW by SW leads to IFT reduction (~5 mN/m). However, replacement of SW by its diluted counterpart leads to increase in IFT by about 6 mN/m, which is in line with the observed trend in tertiary oil recovery data. Dynamic CA measurement is performed for a pure calcite crystal/brine/crude oil system. It is shown, contact angle change is negligible over the time scale of coreflooding test. Therefore, it is concluded that the fluid-fluid interactions may critically affect the low salinity water flooding into pure calcite media.

Main Objectives

using SVR to predict water saturation

New Aspects

improving machine learning for reservoir parameter prediction

Summary

Estimation of reservoir parameters is one of the most important factors in oil and gas reservoir investigations. One of the most important parameters for modeling reservoir is water saturation, which any errors in evaluating it can cause sever financial damages. By core analysis, calculating water saturation is possible, but there isn’t core in every place. Therefor there is a need to estimate water saturation. In this investigation for estimating water saturation, support vector regression was used, which is one of the applications of machine learning and can solve the curse of dimensionality. The well’s data was used for training in the method and water saturation was considered as labelled data. In the end for obtaining the best estimation two different kernels was used.

Main Objectives

the application of contemporary model building techniques to develop a new and more accurate earth model for a shallow-water, shallow-gas dataset despite the limitations of legacy data

New Aspects

Support field developement planning evergreening of legacy dataset using QFWI

Summary

This paper illustrates application of visco-acoustic full-waveform inversion (Q-FWI) and common image point (CIP) tomography techniques to develop a more accurate earth model for a shallow-water, shallow-gas dataset from Gulf of Thailand, despite the limited bandwidth and short-offset attributes of the dataset available. The subsequent Q-imaging successfully reduced geological uncertainties through the compensation of absorption effects, improved fault definition and structural information for the key subsurface targets.

Main Objectives

Reduce geological uncertainties though reliable subsurface imaging and compensation of absorption effect in a complex marine environment

New Aspects

Design a complementary earth modelling workflow based on simultaneous Vp/Q FWI and Q-tomography to build a consistent visco-acoustic earth model

Summary

We illustrate applying a multistep integrated earth modelling workflow based on full-waveform inversion and Q-tomography to build high-resolution velocity and Q models for seismic data acquired offshore Cameroon. The subsequent Q-imaging successfully resolved complex structures with reduced geological uncertainties through the compensation of absorption effects, improved fault definition, and structural and stratigraphic information for the key subsurface targets.

Main Objectives

To obtain a detailed velocity model and improved seismic image quality at the Northwest area of the Agbami field

New Aspects

A full 4D co-processed pre-processing workflow followed by a two-step top down velocity model build-ing workflow consisting of both refraction FWI and ray-based reflection tomography using the 4D OBN data was implemented: a first for time-lapse processing in West Africa.

Summary

The Agbami field is a major field offshore Nigeria, covered with two towed-streamer and three ocean bottom node (OBN) seismic surveys. The existing processed data and associated velocity model gives satisfactory seismic images at the center of the field. However, near-surface velocity anomalies ad-versely affect the image in the northwest area. To address these issues a full 4D co-processed pre-processing workflow followed by a two-step top down velocity model building workflow consisting of both refraction FWI and ray-based reflection tomography using the 4D OBN data was implemented: a first for time-lapse processing in West Africa.

In this paper, we present the results from the TTI FWI velocity model building and imaging. The work-flow allowed the incorporation of detailed shallow structures into the output model which enhanced the migrated images. Ultimately, this facilitated a better understanding of the reservoir architecture to aid with de-risking of any future drilling.

Main Objectives

To achieve reliability, efficient and robustness in earth model building

New Aspects

New objective function for full-waveform inversion

Summary

We demonstrate the ability of enhanced template-matching full-waveform inversion (ETM-FWI), used in conjunction with long-offset ocean-bottom node (OBN) data, for resolving complex geological features. Two important pillars of this work are the enhanced template-matching objective function for FWI and the ultra-long-offset and full-azimuth data from OBN acquisition. ETM-FWI matches the local pattern between observed and predicted shots, which can handle salt-related wavefield and converge to a kinematically correct and high-resolution model. Sparse-node acquisition greatly improves the acquisition efficiency and benefits FWI with better convergence and penetration depth. We demonstrate that the resulting FWI model is kinematically correct and contains high-resolution details. The kinematics of the FWI-updated model can be validated by geologically plausible image and flatter and better-focussed reverse time migration surface offset gathers. The short-wavelength component can also be extracted as a representation of the earth reflectivity.

Main Objectives

Development of the geologically consistent extension of 2004 BP model for benchmarking velocity model-building from the long-offset OBN acquisition.

New Aspects

Extension of the BP 2004 model suitable to simulate the long-offset acquisition settings. New variant of the robust benchmark model allowing to test the undershooting of the salt structures.

Summary

We present an extension of the 2004 BP velocity model which is suitable for the assessment of cutting-edge seismic imaging methods as FWI applied to ultra long-offset ocean-bottom node (OBN) acquisitions. The 2004 BP model is routinely utilized to benchmark various velocity-model building approaches – in particular those developed to tackle the challenges encountered in geological settings comprising salt structures. Those challenges are typically related to the correct reconstruction of the subsalt structures or the sharp velocity contrasts between the salt bodies and the surrounding sediments. To make this model suitable for testing the emerging long-offset OBN acquisitions, we embed the original 2004 BP model within a crustal-scale velocity model inspired by the structural interpretation of the tomographic results from the GUMBO experiment (Gulf of Mexico). The resulting model allows for wavefield propagation within the rifted continental crust and the upper mantle and therefore for the undershooting of the salt and subsalt structures. Consequently, there is no need for extrapolation of the original BP model boundaries or resizing/resampling of its spatial dimensions. The GOMCRUST can therefore be seen as a geologically consistent evolution of the 2004 BP model, which allows to benchmark various seismic imaging workflow with long-offset OBN surveys.

Main Objectives

Investigation on an optimal walkaway VSP survey for successful reservoir monitoring with FWI

New Aspects

This work suggests an appropriate acquisition geometry and feasible timing of time-lapse VSP surveys for successful reservoir monitoring with FWI

Summary

Application of FWI to time-lapse borehole seismic data has the potential to be a powerful solution for the successful CO2-EOR reservoir monitoring. A previous study applied acoustic FWI to the field data of time-lapse walkaway VSP acquired in Abu Dhabi. Their FWI study obtained an accurate velocity model of the baseline data with quite high resolution. In this paper, we reviewed the time-lapse FWI results for the field data. After that, we investigated an appropriate acquisition geometry and feasible timing of monitoring surveys through time-lapse velocity model building assuming CO2-EOR and a synthetic FWI study. As a result of the time-lapse FWI for the field data, 4D signals were not captured, even with the high repeatability between the baseline and monitoring data, due to the quite small amount of fluid volume injected between the surveys. As for the synthetic FWI study, we confirmed that extending receivers upward contributed to the extension of imaging area outward. On the other hand, from the monitoring perspective, deploying the receivers both above and below the target reservoir was recommended. Without the receivers below the reservoir, we found that target-oriented time-lapse FWI was effective.

Main Objectives

apply FWI to land data

New Aspects

practical work flow

Summary

full-waveform inversion, land data, cross-correlation, first break

Main Objectives

High-resolution velocity model estimation of fault-karst reservoir

New Aspects

Application of advanced velocity model estimation method to models containing a special and new type of reservoir

Summary

The fault-karst reservoir is a type of hydrocarbon-enriched carbonate reservoirs in the Tarim Basin. It is a combination of small-scale caves, vugs and fractures developing near deep-seated faults, which makes it difficult to delineate using traditional seismic imaging methods. Recent researches on the characteristics of paleokarst reservoirs facilitate hydrocarbon exploration and exploitation in western China. To interpret karstic fault systems and reservoirs, the seismic image is an effective tool. A high-quality velocity model is essential for high-resolution imaging. In order to image the complicated and irregular fractured-cavity reservoirs accurately, we applied multi-scale FWI method accelerated on graphics processing unit (GPU) devices to models containing fault-controlled paleokarst reservoirs. According to numerical test results, FWI shows high performance of delineating boundaries of fault-controlled karstic reservoirs. The proposed method also has the potential to characterize inner structures. Based on the high-precision velocity model provided by FWI, the seismic image obtained by reverse time migration (RTM) is also improved. These seismic results will show the location, width and shape of the fault-karst structure, which gives detailed information for reservoir interpretation.

Main Objectives

Understand the hydrocarbon migration and fluid PVT phase behavior evolution during fold and thrust belts deformation.

New Aspects

The proposed integrated workflow allows to assess fault-related petroleum systems in complex structural domains, at field level or basin scale, for a dynamic fault seal analysis fluid interactions and PVT fluid phase behaviour evaluation.

Summary

The Cantrell complex, located in the continental shelf in the Southern Gulf of Mexico, is a naturally fractured carbonate field discovered in the late 70’s by PEMEX. The main field known as Akal is the largest with 32 Bboe OOIP. Underlying the Akal filed, the namely Sihil field (1.2 Bboe OOIP) constitutes the thrusted footwall block of a complex fault propagation fold and thrust belt structure.

Even if the Cantarell complex has been studied for several decades and its exploitation reaches a mature declination stage, several questions remain related to the physical and geochemical processes occurring during the trap fold thrust evolution and the synchronous thermo-fluid dynamics: How was the thermal-pressure evolution between the overlying Akal and the Sihil footwall during fold thrust formation? What was the timing of HC expulsion and the fluid migration interactions between both blocks? How was the evolution of the fluid composition and PVT phase behavior during HC charge? What are the implications for near field exploration (NFE) and field development?

The applied methodology is a 2D structural kinematic restoration meshing tool (KronosFlow), coupled with a fault-related petroleum system modelling (TemisFlow) and detailed 1D reservoir fluid PVT phase behavior modeling.

Main Objectives

Reveal the hydrocarbon carrier systems in tight formation.

New Aspects

The effects of micro-coal line an micro-fractures are revealed in the hydrocarbon carrier systems.

Summary

The study of natural gas migration and accumulation of tight gas has become the focus of the energy industry in recent years. One of the priorities is the effects of micro-fracture and micro-coal lines. In this study, based on the real core, three models are established to describe the micro-fracture/micro-coal line characters. According to the three typical models, the Voronoi method is used to rebuild the network models. The migration and accumulation process in tight formation is simulated with the Multiple Relaxation Time lattice Boltzmann method in numerical simulation and with displacement apparatus in physical simulation. The numerical and physical simulation results show that the micro-fracture/micro-coal line will significantly improve the percolation capacity of sandstone. When the micro-fracture/micro-coal line direction is along with the natural gas migration direction, the sandstone will have the highest gas saturation. The micro-fractures/micro-coal lines are favorable for natural gas accumulation in tight formation—combined with the field statistics of gas saturation from more than 1,200 wells in the Xujiahe Formation. Unlike the conventional formation, in tight formation, micro-fractures/micro-coal lines tend to be the main migration channel during the gas accumulation process.

Main Objectives

Analyse the impact of fault tip extension and connectivity uncertainty on compartmentalization, to quantify uncertainty and reduce risk in economic forecasting

New Aspects

probabilistic linear regression analysis

Summary

We present a new approach to address the issue of fault tip extension and connectivity uncertainty and demonstrate its impact on compartmentalization, drawing on an example from the undeveloped AMN area of the greater Ameland field. The Ameland field is a producing gas field, offshore The Netherlands, where a discrepancy between estimated in place volume and dynamic estimations from producing parts of the field leads to significant risk in field development planning and economic forecasting.

We demonstrate how uncertainty cases based on stochastic linear regression analysis of fault throw data calculated directly from seismic interpretation data, can be rapidly generated and incorporated into structural modelling workflows in a way that allows compartmentalization uncertainty to be routinely incorporated into the reservoir modelling workflow.

Main Objectives

Evaluation of sealing performance of the multi-stage extensional abutting faults in petroleum exploration

New Aspects

Comprehensive analysis about the evolution level，time interval and transpressional stress

Summary

Bohai Sea area has experienced complex tectonic evolution including the multi-stage extension which has led to the development of abutting faults system. And the formation of abutting faults has been demonstrated to significantly contribute to the hydrocarbon accumulation in the fault block traps, while strong hydrocarbon productivity heterogeneity has appeared in lots of drilling practices, and hydrocarbon leakage was the main reason for some drilling failure. Comprehensive analysis and comparison combining the 3D seismic and well data reveal that the evolution level of abutting faults determining the degree of leakage in fault intersection, time interval between charge and leak determining the effective charging timing, difference of stress among faults resulted from oblique extension of pre-existing faults determining the magnitude of transpressional stress are all related to the sealing ability of abutting faults. The height of trapped hydrocarbon column derived from drilling data in the structures with abutting faults of Bohai Oilfield has excellent correlation with the sealing ability analysis based on the above three factors, which proved that this method has good application prospects in the petroleum exploration of similar structures.

Main Objectives

Displacement transfer along oblique extension faults

New Aspects

The displacement transfer along Chengbei fault is achieved by rotating the fault orientation and the individual components of slip rate but keeping the net-slip vector fixed. The strike-slip deformation reaches relatively strongest and the extension deformation is the weakest in the S2 fault segment.

Summary

The Chengbei fault with curved fault trace is the boundary fault of Chengbei low uplift. It can be divided into three fault segments (S1, S2 and S3) based on its strikes. Each fault segment is different from another in geometry characteristics.

The change in fault kinematics could be achieved by rigid-block translations with displacements confined mainly to the principal fault surfaces. Oblique extension results where the two displacement fields (strike-slip and normal) are superimposed on one another. The displacement transfer along oblique-extension faults may be achieved by rotating the fault orientation and the individual components of slip rate but keeping the net-slip vector fixed.

Main Objectives

subsurface imaging of a mineral exploration site using passive surface waves

New Aspects

semi-automatic processing workflow

Summary

We propose a workflow to obtain the path-average dispersion curves, which are the input of SW tomography, from 3D ambient noise records. The workflow starts with a pre-processing step, which separates the time windows at which significant surface-wave energy has been recorded and sorts them, based on the azimuthal direction of the SW origin. For each direction, aligned pairs of receivers are found and the path-average dispersion curves are extracted for each pair. The availability of long records allows stacking, improving the data quality. We applied the proposed workflow on a dataset recorded in 2018 in the Siilinjärvi mining site in Finland, with the purpose of increasing the knowledge of the extension of the phosphate mineralization. Comparison with active dispersion curves proves the reliability of our results and indicates that the passive data allow deeper investigation, increasing the possibility of mapping deeper mineralization targets.

Main Objectives

Define additional prospecting criteria to increase the chance to find a new gold deposit. Finding the instrument for separation of LS-epithermal gold veins deposit alterations by geophysics in Chucotka region.

New Aspects

The instrument for separation of LS-epithermal gold veins deposit alterations by geophysics in Chucotka region. New prospecting geophysical criteria (the chargeability anomalies at 0.31 Hz) was suggested.

Summary

A number of epithermal gold-silver deposits are located within Okhotsk-Chukotka volcanic belt (Far East of Russia). LS-epithermal veins with the thickness of several meters can contain up to several hundred tons of gold. The search of covered veins is a complicated task for exploration and it is difficult to solve it without geophysics. The example of successful application of audiomagnitotelluric method in Chukotka region was presented by author. As the result of all data analysis the geophysical prospecting criteria were suggested. Unfortunately, not every found vein is commercial. To increase the chance of finding a new deposit the author set the task to define additional prospecting criteria. The induced polarization (IP) method was used to achieve this goal. The main result of the investigation is that the instrument for separation of alterations types was suggested. The chargeability anomalies at 0.31 Hz and 15 mV/V amplitude are connected with illite chlorite alterations of a deposit core. Illite chlorite alteration of the core deposit is also characterized by high resistivity (up to 2000 Ohm-m) and low magnetic susceptibility. The zones of illite and kaolinite alterations surrounding the deposit core are characterized by reduced electrical resistivity (from 50 to 300 Ohm-m).

Main Objectives

The main objectives are 1)Determination the mnimum equipment requirements for local microseismic monitoring for ensuring safety in mining 2) suggestion and discussion methodology of selection the low-cost seismic equipment for microseismic monitoring task

New Aspects

The complex of methods described in this paper offers a useful approach to evaluate the suitability of recording equipment for specific problems of microseismic monitoring of mineral deposits and to significantly reduce the microseismic monitoring costs

Summary

This work is devoted to improving technology and conducting microseismic monitoring of mineral deposits by choosing the optimal low-cost seismic equipment. During the last few years much progress has been made in the development and production of seismic exploration geophones. There are modifications with natural frequency values 4.5-5 Hz and an increased sensitivity of about 100 V/m/s. The problems that are solved by monitoring of weak local seismicity (where it is sufficient to ensure the registration of frequencies above 0.5–1 Hz), become important. However, these values are still below the operating frequency band of a standard geophone. In this paper, we consider low-frequency deconvolution method to expand operating frequency band of geophones. Then, authors discussed the need for careful choice of low-noise digitizer. It is proposed a method for estimation of digitizer’s self noises – calculation of Power Spectral Dencity. The using of cheaper equipment (compared with seismological devices) will help significantly reduce the cost of microseismic monitoring and increase the number of observation points.

Main Objectives

We report a test result of deep Seafloor Massive Sulfide (SMS) deposits exploration at the Longqi hydrothermal field on the ultraslow-spreading Southwest Indian Ridge (SWIR) by using transient electromagnetic method (TEM)

New Aspects

The Longqi-1 hydrothermal field discovered was the first active field to be found in this region. Here we show a 2D conductivity profile of a Seafloor Massive Sulfide deposit

Summary

In this study, we report a test result of deep Seafloor Massive Sulfide (SMS) deposits exploration at the Longqi hydrothermal field on the ultraslow-spreading Southwest Indian Ridge (SWIR) by using transient electromagnetic method (TEM). The TEM dataset was acquired during the 30th cruise of the China Ocean Mineral Resources R & D Association (COMRA). Conductivity profile has been obtained through 2D inversion of TEM data, and high conductivity anomalies under the seafloor covered by the profile was consistent with the hydrothermal venting sites. Combined with the geochemical and hydrothermal anomalies, it is suggested that the conductive anomalies are attributed to SMS deposits and that the TEM is helpful for exploration and characterization of SMS deposits at hydrothermal fields.

Main Objectives

Near surface

New Aspects

Post-coding Mini-Sosie

Summary

Due to the vibration rammer impacting a regular frequency, there is high side-lobe and correlation noise in the seismic record, leading to the dispersion energy scatter. The traditional method design the corresponding control system based on pseudorandom coding technology, but it is challenging to achieve without complicated (and expensive) electronics. Therefore we describe a technique, the post-coding technology, that can result in a high SNR seismic record comparable to that of Mini-Sosie while without complicated electronics. Through synthetic and real data examples, we designed the linear impact sequence for post-coding. The results show that the decoding record has a high signal-to-noise ratio, the dispersion curve with the frequency range of 8-40 Hz can be extracted effectively.

Main Objectives

Identification of shallow geohazards

New Aspects

New and advanced use of technology to aid interpretation

Summary

Understanding the conditions of the seabed and the shallow subsurface are critical to avoidance of many types of geohazards which could potentially impact on the safety and cost of engineering operations, as well as the longevity of the structures being installed. A dedicated site investigation survey includes numerous shallow geophysical imaging techniques, often deployed simultaneously, focusing on very high vertical resolution with a shallow depth of penetration. These datasets can be supplemented by any 3D exploration seismic data available in the area, which can offer additional perspectives. Geohazard assessment can be greatly aided by high-resolution 3D seismic attributes, machine learning based techniques and artificial intelligence, combined with enhanced visualisation. Analysis of the different aspects of the seismic signal, such as frequency, amplitude and phase and their joint interpretation contribute to a comprehensive understanding of the shallow subsurface. A complete geological risk assessment can often also require understanding of structures which extend from shallow to depth, for example faults, this can only be achieved by integration of low frequency data exploration data. The seismic analysis techniques discussed in this presentation can be easily applied to help identify hazards which may compromise locations for the seabed and sub-seabed infrastructure.

Main Objectives

Integration of Machine Learning in Distributed Acoustic Sensing solutions

New Aspects

Distributed Acoustic Sensing for pieline integrity monitoring

Summary

Distributed Acoustic Sensing (DAS) is a new and innovative technique that allows to convert fibre optic cables into hundreds to thousands of acoustic sensors. Among its many different applications, DAS can be used for pipeline monitoring when intrusion detection is necessary. However, DAS, because of its ability to acquire data with dense temporal and spatial samplings, is creating large datasets often difficult to handle. Thus, Machine Learning appears as a major leverage for potential threats identification along the structure. In this work, we present the results of the integration of the Random Forests algorithm for classification of events recorded by the FEBUS A1-R DAS system. The study is focussed on tests lead along a gas pipeline instrumented with fibre-optics cable. We work on the discrimination of six classes of acoustic sources: ambient noise, manual compactor, mechanical excavation, drill, jackhammer, sheet piling and circular saw. We demonstrate the ability of the algorithm for detection and identification of these classes with an estimated classification accuracy of around 97%.

Main Objectives

To generate the experimental dispersion curve and to delineate the 1D shear wave velocity profile using Nature inspired optimization algorithms namely the Particle Swarm Optimization and Grey Wolf Optimization.

New Aspects

Comparative study of 1D Shear wave velocity estimation using global search Meta heuristic approaches namely the Particle Swarm Optimization (PSO), standard Grey wolf optimizer (GWO) and Improved Grey wolf optimizer (GWO).

Summary

This study uses the Ground Rolls (Rayleigh waves) contained in the synthetic dataset to generate the experimental fundamental mode dispersion curve using the phase-shift method. These experimental fundamental mode dispersion curve were later on inverted to obtain the 1 D shear wave velocity profiles using two meta-heuristic approaches, focused on the most widely used stochastic-based Nature-inspired optimization algorithms, namely the Particle Swarm optimization and Grey wolf Optimization algorithms. We choose to use the global search optimization algorithms for the inversion because they are independent of the initial model and provide the cost function’s global optimal solutions iteratively from a range of possible solutions. Comparing inversion results from PSO and GWO (standard and improved) variants yield errors of ~4.3 % for layer 1, ~4.5 % for layer 2, and ~2.5% for the half-space.

Main Objectives

Use of satellites observations for derisking seismic survey preparation and operations

New Aspects

Derived bathymetry technics

Summary

For Oil & Gas exploration projects close to shore or in shallow waters, bathymetry is needed for :

-De-risking seismic survey operations,

-Assessing how to perform seismic acquisition : towed streamers or Ocean Bottom Surveys (OBS),

-Defining survey shape and suitable line directions.

However, some of these potential exploration areas might have only sparse bathymetry data either no bathymetry at all and planning a standard bathymetry survey or airborne bathymetric LiDAR survey is not always possible at the seismic survey feasibility stage.

To mitigate this, Total E&P has started using Satellite Derived Bathymetry (SDB) technics as an alternate solution for providing, within the time frame and when feasible, derived depth data as an intermediate product/deliverable for a first assessment of survey areas.

Main Objectives

Demonstrate associated carbon storage in EOR projects, highlight time-lapse model development

New Aspects

Novel uses of Winland R35 approach for assigning facies in hydraulic flow models

Summary

The evolution of five versions of a geologic model at Farnsworth unit have highlighted the ability to integrate new interpretations, newly generated data, and feedback from monitoring and simulation efforts into increasingly comprehensive models over extended periods of time, while allowing for useful models at each stage of development. Technical advances in model building and design of comprehensive workflows should aid future projects with the goal of commercial carbon storage at EOR sites.

Main Objectives

Safe Seismic Operations in any infrastructures and ensure the quality of final seismic imaging

New Aspects

Determine the safe distance from different infrastructure

Summary

The present paper will focus on a mega single sensor 3D seismic survey, where different environments

that needed to be covered. These includes producing oil fields, oil installations, refinery’s, cities, farms

and construction sites. Due to these changing as well as challenging environments, a safety standard

had to be adopted to support vibrator and explosive sources for safe and successful completion of the

project. Peak Particle Velocity (PPV) surveys were conducted using the DIN 4150 German Standard as

a reference whenever the seismic operation approached to any oil field infrastructure, populated area or

any possible vulnerable hazards.

A total of 666 PPV measurements were conducted in multiple environments within the seismic survey

boundary. Based on these tests a set of optimum possible safety distances were determined to oil

installations, residential structures, construction sites and other infrastructures to ensure that the quality

of the final seismic imaging is optimized and survey objective is achieved.

Main Objectives

To acquire quality 3D Seismic Data Volume without zero incidents and any environmental footprints

New Aspects

A two stage de-risking method was establised to acquire high resolution 3D Seismic data volume in former battle filed area also 82 UXO was identified and destroyed.

Summary

The Gulf War conflict of 1991 turned Partitioned Zone (PZ) of the Kingdom of Saudi Arabia and the State of Kuwait into a high-risk area with regards to Unexploded Ordnance(UXO) and Explosive Remnants of War (ERW). A large national and international effort has previously cleared UXO/ERW from most of the country, but to date UXO/ERW is still found because of the dynamic nature of the environment. During 2014-16, a high resolution, multi-azimuth 3D seismic Survey was completed utilizing 168,000 UniQ channels, with 20 DX-80 vibrators, shooting flip flop with four fleets 24/7, covering approximately 4,612 km2 (5,346 km2 including the 2 zippers). The survey was planned to satisfy the geological objectives, safety and security threats, and restricted access due to oil production facilities. An innovative two-phase risk-based approach was implemented to ensure personnel and operational safety while maximizing seismic production. A total of 551 days was spent for UXO/ERW identification, with a total of 82 UXO found and destroyed. The seismic acquisition project of more than 100 vehicles and 600 personnel was completed without any incidents or accidents and a high standard of verification was maintained throughout the project execution.

Main Objectives

COVID-19, 2-D land seismic

New Aspects

Original COVID-19 infection prevention measures in land seismic survey

Summary

Under the severe COVID-19 infection circumstance as rapidly spreading in JAPAN since March 2020, we conducted 2-D land seismic survey in residential area, Niigata, JAPAN from November to December 2020 with necessary infection prevention measures. We set the criteria for the survey start propriety in advance and discussed about the survey implementation with related parties. We also obtained the understanding about this survey from local governments and residents before the survey start. During seismic survey, we took two types of original infection prevention measures as “Enclosure” and “Separation”, in addition to basic measures indicated by guidelines of the Japanese government. “Enclosure” measures aimed to prevent coronavirus from entering inside of the survey site. “Separation” measures were risk diversification of spreading coronavirus in the survey site. By planning and thoroughly adhering original countermeasures that matched the survey area COVID-19 status, we were able to complete the survey in the planned duration without interrupted the survey. In this paper, we introduce our original on-site infection prevention measures “Enclosure” and “Separation” and discuss how to manage the risk of COVID-19 infection at the seismic survey site.

Main Objectives

Seafloor Surface Characterization

New Aspects

Innovative data collection and processing techniques

Summary

The overall objective of this work has been to develop a consistent and semi-automated seafloor survey method for generating high-resolution, benthic maps for environmental assessments and monitoring of offshore energy sites. While we show examples from sites in North America, the approach described herein has been used globally to assess hydrocarbon exploration prospects, environmental clearance of candidate wellsites, seep detection studies, and pre- and post-drilling surveys in frontier and established exploration basins. We present these results as an example of the benefits realized to any explorer looking to efficiently derisk their exploration targets and lease areas through the creation of photographic evidence of the upper seabed and seafloor surface.

Sediment profile and plan view imaging (SPI-PV) technology were combined with multibeam bathymetry and acoustic backscatter methods to demonstrate a rapid, cost-effective benthic mapping protocol. Benthic habitat maps generated for areas offshore of Oregon, USA and are presented to illustrate the effectiveness of this approach across a range of environmental conditions (i.e., low to high energy settings). A key technical innovation of this project was the development of a computer-automated image processing platform that automatically measures key features in the images.

Main Objectives

Thorough geoscientific site characterization of the aging salt cavern facility Etzel; including a critical analysis of fault reactivation risks within the salt overburden.

New Aspects

Taking critically stressed faults within the salt overburden into account while analyzing an aging salt cavern facility for potential subsurface risks.

Summary

Any reutilization of preexisting salt cavern gas storages, as well as the construction of new cavern facilities for future storage of regenerative energy, will require a thorough geoscientific site (re)characterization in advance. The publication describes a rarely published case, where legacy information has been exploited to qualitatively reassess the short-, mid- and long-term structural integrity of the aging salt cavern gas storage facility Etzel, also considered for future storage of hydrogen. Assessment results reveal a seriously critical stress state of the salt overburden of gas caverns directly underneath the critically stressed main fault. Qualitative reassessment conclusions are evidenced by big gas cavern convergence rates (tendency to shrink; up to 3 vol% of the initially leached volume per year) which in turn caused highly elevated subsidence rates up to 84 mm/yr; spatially coinciding with the main fault.

Main Objectives

Speed-up Deep Learning models training to accurately predict oil slicks in satellite images.

New Aspects

Multiple new HPC and ML techniques to speed-up Deep Learning models training

Summary

Oil slicks from natural seeps and man-made spills of hydrocarbons need to be monitored in real-time to prevent environmental hazards. Remote sensing techniques are suitable for the task. Unfortunately, human intervention prevents real-time monitoring due to the limited area that can be processed by remote sensing specialists in GIS applications at a time. In this study, the focus is on how Machine Learning -in particular Deep Learning (DL)- can help automating the above mentioned task. Thus, a DL architecture is trained with a Total’s proprietary dataset for segmenting oil slick regions from satellite images. Further, the training stage is scaled and enhanced with high-performance computing techniques. These techniques allowed the original task to be solved up to 2.5× faster than the baseline for single GPU and pseudo-linear scalability for the distributed case. The later takes this application closer to real-time use case level.

Main Objectives

magnetometry, electromagnetic studies, near surface geoscience, magnetometry, dumps, landfills

New Aspects

new screening approach for geoenvironmental studies

Summary

Dumps and landfills are the end place of unwanted material and disposed products. Burried resources may be landfill mined and environmental pollution diminished. The problem is lack of information on unknown dump sites of former times – there remote sensing and traditional geodesy, proximal sensing techniques could be used. Near surface geophysical methods are valuable for screening of areas where drilling is limited due to technological limitations and anthropogenic unhomogenousity of material. The aim of this study was to determine whether screening of magnetometry and geoelectrical methods may be useful for old burried dumps recognition. Protonmagnetometer was used in Eastern Latvia to detect burried dump in forest, already covered by soil and vegetation. Induced polarisation and electric resistivity research was done in Southern Sweden for the macro-content analysis of dump hills composed of glass industry residuals and construction waste mixture. Surveying helped to determine macroproperties such as geomorphology and physical type of material underneath the surface. Results allowed spatially characterize dumpsite masses (location and dimensions) and identify the internal structure of a these sites. This is valuable information in order to estimate the material recovery potential of landfills. This study was supported by project No.1.1.1.2/VIAA/3/19/531.

Main Objectives

To offer solutions on the sustainable management of waste in Ghana

New Aspects

towards renewable energy

Summary

The United Nations 17 sustainable development goals is a plan that seeks to achieve sustainable development of the planet earth and its people. Achieving the SDGs by 2030 will require more and better financing, a renewed focus on implementation to improve the lives of those hardest to reach, significant improvements in data collection and analysis.

In order to achieve sustainability, strategies must be developed so that resources are only used at a rate which allows them to be replenished so that they will continue to be available, while at the same time emissions of waste are confined to levels which do not exceed the capacity of the environment to absorb them.

However, the world’s population is growing at an alarming rate. This has raised a number of environmental concerns.

As the population grows, the generation and management of solid and liquid waste becomes crucial. Municipalities and Metropolitan areas need to put policies in place that will deal with waste collection in a sustainable way.

This paper presents a proposal to improve the planning, collection and treatment of solid and liquid waste using sophisticated methods of waste management in the form of regulation, education, investment, and motivation (REIM).

Main Objectives

Development of a new sustainable and renewable materials for heavy metal removal from drilling mud

New Aspects

Removal of lead and cadmium from synthetic-based fluids

Summary

Due to their persistence, removal of heavy metals from synthetic metal-bearing wastes is of particular important nowadays. In this study, sorption potential of a renewable resource, viz. activated palm kernel husk (APKH) for two heavy metals, vis. cadmium (Cd) and lead (Pb), was assessed in a synthetic drilling mud. The utilized agricultural wastes used were ground, sieved to a defined size range, and carbonized. Results revealed that the utilization of this renewable agricultural waste is effective for removal of Cd and Pb from synthetic drilling mud. The percent removal of both the metals was found to increase with the presence of APKH and certain contact time. The efficiency of APKH for the removal of Cd and Pb were 5.3% and 30.9% for synthetic drill cuttings (SDC), and 73.5% and 76.8% for SDC and conventional additives, respectively. The mean of metals concentrations from drill cuttings decreased with increasing of time and reached equilibrium at 120 minutes for all metals. The present data confirms that APKH may be used as efficient sorbent for the removal of cadmium and lead ions from synthetic drilling mud.

Main Objectives

To constrain depth errors in 2D seismic data in complex land settings.

New Aspects

A synthetic case study exploring the 3D errors in 2D seismic data in a fold and thrust belt and derived equations to predict the positioning errors of a target horizon beneath a thrust fault.

Summary

Seismic data collected in complex land settings is often two-dimensional (2D). The data is then processed in 2D, and these 2D images are ideally representative of the subsurface beneath the profile. This may not be the case, however, if the subsurface has complicated three-dimensional (3D) structure and lateral heterogeneity, like in fold-and-thrust belt settings. In order to quantify these 3D errors in 2D seismic images, we carried out a synthetic case study using a 3D model based on the Caipipendi block in Bolivia, where a target horizon lies beneath a thrust fault. We compare results from illumination studies and 2D migrated images with predicted errors due to a single thrust fault. Illumination studies reveal that seismic energy can reflect off subsurface boundaries kilometers outside of the crosslines. The target in the crosslines consequently has migrated depth errors of tens to hundreds of meters. The thrust fault explains the majority of the errors, both in the lateral direction perpendicular to the 2D plane, and in depth. Our thrust fault error prediction equations have the potential to correct for errors in a seismic strikeline due to a cross-dipping thrust fault and can be incorporated into uncertainty analysis and risk assessment.

Main Objectives

Prestack least-squares Kirchhoff migration providing increased resolution and more accurate seismic amplitudes than conventional imaging.

New Aspects

Local-calibrated least-squares Kirchhoff migration to optimize the inversion to changes in the background model and account for missing physics in the modelling engine.

Summary

Reliable seismic amplitudes are crucial for the estimation of rock properties. In conventional depth imaging, amplitudes and resolution will be influence by propagation effects in the imaging model. These limitations origin from the formulation of the migration operator, implemented as the adjoint rather than the inverse of modeling. Least-squares migration (LSM) tries to eliminate these effects and resolve the real reflectivity model.

In this study, we make use of a newly developed local calibrated image-domain Kirchhoff least-squares migration to deconvolve the system response from the depth migrated gathers. We demonstrate how the inversion de-blurs the image and adjusts the prestack amplitude response, following better the expected response from well synthetic. The method is demonstrated on a North Sea dataset from the Viking Graben area, covering the Verdandi/Lille Prinsen discovery.

Main Objectives

A cross-discipline integrated approach of multimeasurement acquisition, broadband processing, imaging and interpretation the setting to accelerate prospect generation through data-driven prospect ranking, reducing uncertainty and unleashing the regional potential for exploration for a deep-water offshore Sabah, Malaysia.

New Aspects

A cross-discipline integrated approach of multimeasurement acquisition, broadband processing, imaging and interpretation the setting to accelerate prospect generation through data-driven prospect ranking, reducing uncertainty and unleashing the regional potential for exploration for a deep-water offshore Sabah, Malaysia.

Summary

Deep-water (1000 m+) Sabah, Malaysia, is a frontier exploration province proximal to successful exploration trends within the Sabah thrusts zone and Luconia carbonate shelf. The basin is on trend with numerous proven petroleum systems; however, it remains underexplored with minimal drilling activities targeting the deep-water hydrocarbon plays. A large broadband towed-streamer seismic survey acquired in 2017 and 2018 provides an opportunity to build on the regional petroleum systems, create exploration opportunities, and unlock the hydrocarbon potential in this under-explored frontier area. This work illustrates how to mitigate exploration risks through contemporary broadband processing, interpretation-guided, high-resolution, anisotropic earth model building, multiphysics input to interpretation, and integrated reconnaissance AVO inversion.

Main Objectives

Sub-salt exploration using heritage seismic

New Aspects

Demultiple and model building

Summary

Hydrocarbon exploration and production in the Southern North Sea remains a key part of the energy mix in the United Kingdom. At the same time, the area is also of interest for carbon storage as part of the ongoing transition to renewable energy and a low-carbon economy. While the energy transition is important, there is still a requirement to continue production from existing fields with an efficient and economic workplan. In this area of the Southern North Sea, hydrocarbon prospectivity is primarily in the pre-salt section, and mapping these often-subtle traps is challenged by a complex overburden and heritage seismic data with limited offsets. Here, we present a case study that demonstrates that broadband reprocessing and high-resolution depth imaging can be used on heritage data to reduce uncertainties in imaging and reservoir characterization. Key limitations of heritage data sets in the area include limited bandwidth, residual multiple contamination, and distortions associated with unresolved overburden heterogeneity. Here, we show that modern deghosting and shallow-water demultiple techniques, coupled with high-resolution model building to capture tertiary and intra-Zechstein complexities, ultimately reduce interpretation uncertainty and ensure continuing, efficient hydrocarbon exploration and production in the area.

Main Objectives

Mitigation of the fault shadow zone effect on the geological target

New Aspects

Fault constraint tomography; TTI velocity model building

Summary

Llanos Basin is well known for its hydrocarbon potential and proven reserves. One of the main Imaging challenges is the fault shadow zone impact on the the geological targets and therefore its reserves. This presentation highlights the positive contribution of PSDM Imaging to the mitigation of the fault shadow zone impact applied on Entrerrios 3D survey. the mapped hydrocarbon reserves increased by a significant factor.

Main Objectives

delineate fault compartmentalised oil field

New Aspects

broadband high resolution dual azimuth imaging

Summary

We present the results of a reprocessing and imaging project designed to better understand the reservoir architecture in deep-water offshore Nigeria. The overall goal was to enhance imaging of the faulting seen at reservoir level and to identify the level of compartmentalization present within the structure, in order to better understand the challenges that could be faced in bringing this field into production.

The first depth velocity model over this field was created in 2011, using data acquired in 2008, as part of a high definition (HD) survey collected on a 6.25 x 12.5 binning grid with relatively shallow streamer and gun depths (5 m and 6 m respectively) aimed at maximizing the signal bandwidth and spatial resolution of the heavily faulted sandstone reservoir.

Here we present a dual azimuth velocity model building project, updating the 2011 model by combining the 2008 HD survey with an earlier 1998 regional survey, as these surveys were shot in orthogonal directions. The stated objectives were achieved, in that successful reprocessing. and merging of disparate input data volumes combined with high resolution dual azimuth TTI tomographic model building facilitated a better understanding of the overall reservoir architecture, and most importantly, the fault compartmentalization.

Main Objectives

Geohazard risk evaluation, 2D high-resolution seismic data processing, Pseudo 3D imaging

New Aspects

Accelerating seismic data interpretation procedure & 3D attribute calculation

Summary

Detecting shallow geohazard risk is one of the critical challenging issues during well drilling for gas fields development. 3D conventional seismic surveys typically contain information from deep parts, since they are designed for deep reservoir targets. In current study, several 2D high resolution seismic data with dense line spacing were acquired for shallow geohazard evaluation over a shallow marine gas field, offshore Iran. Such data is beneficial in terms of available near offset information which are necessary for shallow marine hazards evaluations. On the other hand, 2D imaging cannot be accurate to distinguish correct subsurface location of the drilling hazardous features. In this situation, pseudo- 3D migration is an appropriate method to achieve the correct subsurface image with more details relative to 3D conventional seismic. Pseudo-3D cube generation also has some other advantages that can be profitable in seismic data interpretation including: 3D attribute calculation, surface (horizon) attribute estimation, 3D geo-body extraction and accelerating seismic data interpretation procedure.

Main Objectives

High Resolution and High Precision Imaging of Deep Thin Reservoir

New Aspects

Viscoelastic Q pre-stack depth migration and an improved method for obtaining high precision layer Q

Summary

Since the underground medium is not completely elastic, the seismic waves will undergo viscous absorption and attenuation during the downward propagation process. The deep volcanic rock and glutenite oil and gas resources in Songliao Basin are abundant, but the burial depth is large and the reservoir is thin. The conventional pre-stack depth migration data have low dominant frequency and narrow frequency band, which is difficult to meet the needs of deep thin reservoirs.

Based on the acquisition and processing of seismic data of ” Wide band wide azimuth high density”, this paper develops viscoelastic Q pre-stack depth migration, which effectively compensates the absorption and attenuation of high-frequency seismic waves caused by the viscosity of the earth’s medium and thin-layer scattering, and restores the attenuated high-frequency components, which further obtained high-resolution migration imaging results. The application of actual data shows that this method can obtain higher-resolution migration imaging results than conventional pre-stack depth migration and provide high-quality results for fine structure interpretation and reservoir prediction of deep thin reservoirs, which effectively supports well location deployment and reserve submission in this area.

Main Objectives

Create method to provide forecast of turbidity reservoir

New Aspects

Provide seismic prediction in new geological region

Summary

In paper shown prediction of turbidities reservoir in West Siberia (Achimov reservoir) based on complexity geological and seismic data. Defined the main method to provide this forecast, its simultaneous inversion. Its allowed to perform quantitative forecast of pay thicknesses and their distribution over the area, probabilistic evaluation of reserves and to create an exploration program. It helps to decrease uncertainties and make the decision on the construction of surface facility and early development drilling.

Main Objectives

electrofacies analysis

New Aspects

using electrofacies analysis” methods for channel determination

Summary

In this study, 2 clastic sandstone reservoirs have been selected for identification of sandstone layers (as pay zones) by “using electrofacies analysis” methods in two different Iranian oil fields (Northeast and Southwest of Iran). Therefore, electrofacies analysis has been applied for sandy channel determination in both fields in order to be used in static reservoir modeling. Two key wells, have been used as reference wells for electrofacies analysis in both reservoirs. Some petrophysical logs such as PHIE, NPHI, shale and quartz volumes have been used to define electrofacies according to relationship results between imported logs obtained from checking relative cross plot. This study in both clastic reservoirs showed that sandy channels can be determined by using electrofacies analysis” as pay zones. Then MRGC and SOM methods have been applied for electrofacies analysis among other different approaches after checking the results of all algorithms. In first field, two clusters (EFAC-3 and EFAC-4) are consist of clean sandstone with moderate and relatively high porosity using both MRGC and SOM methods. In the second field only the first cluster (EFAC-1) is constructed of sandstone with porosity of 6-11%.

Main Objectives

to identify heterogeneous thin sands via machine learning (artificial neural network) and evaluate the impact of tuning thickness on the recognition by ANN.

New Aspects

(1) The prediction of hetergeneous thin sand by machine learning (ANN); (2) The role of seismic limits (tuning thickness) in ANN is evaluated quantitatively.

Summary

The objectives of this work are to identify heterogeneous thin sands via machine learning (artificial neural network) and evaluate the impact of tuning thickness on the recognition. The thin sands within the study interval mainly developed in a complex fluvial to shallow marine environment. Multiattribute classification using supervised Artificial Neural Networks (ANNs) is employed to predict the distribution of these thin sands within six subintervals and the role of tuning thickness in the prediction is evaluated quantitatively.

Main Objectives

To assess the feasibility of distinguishing different lithologies and fluid types in the Volve Field

New Aspects

Reservoir quality investigation

Summary

Summary

Before any seismic inversion work is carried out, a conventional project starts with a rock physics feasibility study during which we evaluate whether facies classification is feasible.

In this study, an attempt was made at identifying the reservoirs present at well locations in the Volve field located in the Norwegian North Sea. Rock physics modeling and AVO analysis were applied in an integrated approach to study the seismic response of the reservoirs and assess the feasibility of distinguishing different lithologies and fluid types.

Perturbational Modeling (porosity, lithology, and fluid) was also carried out on the Hugin sands. The result shows the effect of changing porosity, volume clay, and fluid on the elastic properties. Due to the good porosities and thickness of the sands, good Lithofacies and fluid discrimination were observed especially in the upscaled and AVA domain.

Main Objectives

electrofacies analysis

New Aspects

Dynamis and static electrofacies analysis versus HFU

Summary

In this study, new static and dynamic aspects of electrofacies analysis have been performed in order to check relationship between static and dynamic electrofacies with HFU and petrophysical parameters with different electrofacies in Ilam reservoir in one Iranian oil field. 6 static and dynamic electrofacies models have been generated using MRGC, SOM and AHC methods in order to check relationship between static and dynamic electrofacies with HFU in studied field. Some petrophysical logs such as porosity, predicted permeability, mineral volumes, CGR, DT, NPHI and RHOB have been used as imput data for electrofacies analysis. Based on this study, both static and dynamic electrofacies (without and with applying permeability log as an input data) can detect pay zones from, others considering the HFU. However, electrofacies can identify shaly zones (poor reservoir or seal) from other payzones.

Main Objectives

EOR screening

New Aspects

High resolution ‘truth’ modelling

Summary

EOR screening studies require integrated reservoir characterisation and static-dynamic modelling capable of capturing both subtle reservoir heterogeneities and the complexities of multi-phase flow. This often leads to interwell modelling where some compromises are required on reservoir detail, or detailed modelling at a small-scale which then has to be upscaled. Taking advantage of current high-end computing power we adopt an approach (‘truth modelling’) in which the models are resolved at the scale of the data (cell size=core plug size) but scaled at the scale of the development question (up to typical offshore well spacing). The multi-million cell models which result have no cut-offs applied, use full physics and are not upscaled. The models reveal the fluid flow process in very high resolution not previously possible for integrated study work. The approach is applied here for EOR screening on mature oil reservoirs in the Niger Delta, focussing on WAG and ASP EOR mechanisms, revealing and quantifying the impact of fine-scale heterogeneities, often missed in ‘standard’ resolution models on the efficiency of the EOR process.

Main Objectives

Apply multiple-parameter versions of the simple descent and the conjugate-gradient method to elastic iterative reverse-time migration and compare their efficiency to the single-parameter methods.

New Aspects

A matrix-free data-domain reformulation of the multiple-parameter conjugate-gradient method is presented. Its performance is compared to the simpler multiple-parameter descent method as well as to the single-parameter versions on a small isotropic elastic reverse-time migration problem.

Summary

Multi-parameter inversion of linear systems appears in many problems. The focus here is on isotropic elastic iterative reverse-time migration for three position-dependent subsurface model parameters, which amounts to data fitting of processed seismic data with synthetics from the Born approximation of the elastic wave equation. In that case, the matrix of the linear system is the hessian. As it is impractical to form, a matrix-free formulation is needed, which is readily derived for the gradient descent method. For single-parameter inversion, the conjugate-gradient (CG) method is generally more efficient than simple descent. However, the multiple-parameter CG method has a significantly higher cost than the descent method. Here, first a matrix-free data-domain reformulation is derived. Then, its performance is compared to the simple descent method to see of its faster convergence justifies the higher cost. A comparison on a marine 2-D toy problem with a salt body and sea-bottom receivers shows that the multiple-parameter descent method wins in terms of efficiency if the number of iterations is limited and that the single-parameter CG method is even faster.

Main Objectives

Enhance the amplitude and improve the imaging quality in the middle and deep layer without reducing the imaging quality in the shallow layer

New Aspects

we extend Nowack (2011)’s method to anisotropic media and present an anisotropic dynamically focused beam migration by modifying the propagator of Gaussian beam.

Summary

As an improved ray method, Gaussian beam migration gives not only ideal calculation efficiency, but also high imaging accuracy. However, the accuracy of the Gaussian beam migration is controlled by the initial beam width at the surface. In this paper, we extend Nowack (2011)’s method to anisotropic media and present an anisotropic dynamically focused beam migration by modifying the propagator of Gaussian beam. We use anisotropic kinematic and dynamic ray tracing to obtain travel time, trajectory and dynamic information. Then we use dynamically focused beam of anisotropic media to calculate the Green’s function and the Claerbout’s imaging condition to obtain images. This strategy enables us to enhance the amplitude and improve the imaging quality in the middle and deep layer without reducing the imaging quality in the shallow layer, which will help to overcome the limitation of the initial beam width in Gaussian beam migration. The results of Shengli complex structure model for VTI media show the accuracy and validity of the research method in this paper which can solve the amplitude preserving imaging problem of deep complex anisotropic subsurface structure.

Main Objectives

To compensate the seismic wave attenuation caused by the subsurface viscoelasticity, improve the imaging quality with iterations and produces better imaging results.

New Aspects

A topography-dependent Q least-squares reverse time migration (Q-LSRTM) based on the first-order viscoacoustic quasi-differential equations is proposed by deriving Q-compensated forward-propagated operators, Q-compensated adjoint operators and Q-attenuated born modeling operators.

Summary

Seismic wave attenuation caused by the subsurface viscoelasicity reduces the quality of migration and the reliability of interpretation. A variety of Q-compensated migration methods are developed based on the second-order viscoacoustic quasi-differential equations. However, these second-order-equation based methods are difficult to handle with density perturbation and surface topography. In addition, the staggered grid scheme, which has an advantage over the collocated grid scheme because of its reduced numerical dispersion and enhanced stability, is unavaiable to implement in these methods. A topography-dependent Q least-squares reverse time migration (Q-LSRTM) based on the first-order viscoacoustic quasi-differential equations is proposed by deriving Q-compensated forward-propagated operators, Q-compensated adjoint operators and Q-attenuated born modeling operators. In addition, the proposed method using curvilinear grids is available even when the attenuating medium has severe surface topography and can conduct Q-compensated migration with density perturbation. Numerical example on a typical model shows that our method improves the imaging quality with iterations and produces better imaging results with clearer structures, higher signal-noise ratio, higher resolution and more balanced amplitude by correcting the energy loss and the phase dispersion caused by the Q attenuation, and suppressing the scattering and diffracted noise caused by the surface topography.

Main Objectives

Investigating the influences of free-surface ghosts to imaging resolution with novel proposed method

New Aspects

Using point spread function to quantify the missing wavenumber caused by ghosts and assist real acquisition

Summary

In reverse time migration, when sources or receivers are buried at certain depth, the interference between the primary waves and the free-surface reflections can cause missing frequency contents in seismic data. If the free-surface effects are not properly removed, they will be further mapped to the subsurface and deteriorate the depth image. In this study, we propose a method to investigate how the free-surface ghosts can affect the seismic data and depth image. We numerically calculate and illustrate the free-surface reflections on the seismic data, and then on the target wavenumber contents. Their effects on depth images are investigated by analyzing the point spread function in both space and wavenumber domains. The proposed approach provides a useful tool for evaluating the effect of free-surface ghosts on the depth image. The understandings generated from this method may help interpreting the depth images achieved under the existing of free surface reflections, or helping designing the acquisition system to minimize the free-surface effects.

Main Objectives

we propose an elastic ray tracing systems for multi-component wave and apply it to elastic Gaussian beam migration to achieve an elastic Gaussian beam migration algorithm in anisotropic media for multi-component seismic data

New Aspects

propose a new ray tracing algorithm in anisotropic TTI media and applied to elastic Gaussian beam migration imaging method

Summary

As elastic Gaussian beam migration has the advantages of Kirchhoff migration and wave equation migration that can process the multicomponent seismic data, we propose an elastic ray tracing systems for multi-component wave and apply it to elastic Gaussian beam migration to achieve an elastic Gaussian beam migration algorithm in anisotropic media for multi-component seismic data. According to the previous studies, we derive the anisotropic forward and backward extrapolation and imaging formula of vector wavefield by using Green tensor characterized by elastic dynamics Gaussian beam on the basis of 2D Kirchhoff-Helmholtz integral for elastic wave in anisotropic media. We also migrate the crosstalk between different waves by introducing a weight function. Compared with the traditional method based on scalar wave, it does not need to separate the wave field which indicates its suitability for real seismic data processing. Meanwhile we propose a correction method by sign function to solve the polarity reversion of qp and qsv wave. Numerical examples of subsag model and thrust model in TTI media show that the method proposed in this paper is accurate and effective.

Main Objectives

Highlight capabilites and operation of Devitoboundary for implementing topography in Devito models

New Aspects

Automated construction of a 3D immersed boundary from a point cloud of topography

Summary

Partial differential equation solvers based on the finite-difference method have for many years been a keystone of seismic processing and modelling applications, commonly found in practical migration and full-waveform inversion methods. Sharp density contrasts within the computational domain have potential to introduce numerical error: problematic when introducing topography to a seismic model. Including a simple step change in density to approximate an air layer compromises both stability and numerical accuracy, often requiring smoothing of the contrast, and inducing both first and second order errors in space. Topography can instead be implemented via an immersed boundary conforming to the surface. This is achieved by extrapolating the wavefield across the boundary to find solution values at necessary external nodes. As this process is confined to the pre-processing step, it has negligible effect on the computational cost of the simulation.

Devitoboundary is a tool in its early stages of development, intended to compliment Devito as a user-friendly means of including immersed boundaries in practical applications. 3D immersed boundaries can be constructed from irregularly sampled topography point clouds, via Delaunay triangulation coupled with a 1D extrapolation scheme. The result is a stable, error-free boundary which can be readily integrated with Devito models.

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Summary

Summary not available

Main Objectives

Developing a robust rock physics model for anisotropic unconventional reservoirs

New Aspects

We developed a consistent rock physics model and workflow to estimate elastic properties of both clays and shales. Also, this new model honors the aspect ratio of multiple inhomogeneities inside the medium. The results are constrained and verified by the field data from Eagle Ford shale.

Summary

We present a novel rock physics model for organic-rich shales. The extended Maxwell homogenization scheme is utilized as a rock physics model for transversely isotropic media. Since shales have complex structures, different components of the rock are modeled as multiple inclusions. First, we estimate the anisotropic clay matrix. This is then used as the host matrix, and quartz, calcite, kerogen, and fluid-filled pores are modeled as inclusions with different aspect ratios. Representation of multiple inhomogeneities with different aspect ratios is non-trivial. The Maxwell homogenization scheme honors the aspect ratio of each inclusion embedded in an effective inclusion domain. Combined rock physics models have been used to obtain elastic properties of clays and shales. Notwithstanding, there is no consistent method for modeling both. Our rock physics model and workflow thoroughly handle the estimation of elastic stiffness coefficients of both clays and shales in anisotropic media. The estimated stiffness coefficients using the rock physics model are constrained by dipole sonic logs from the Eagle Ford shale. This study shows that this rock physics model can be readily applied to other unconventional reservoirs.

Main Objectives

Evaluation of water saturation for gas dissolved in water sandstone using rock physics modelling and compare the result with conventional petrophysical well log analysis result. Besides, understanding impact of isolated gas in the sandstone on seismic data.

New Aspects

Demonstrate rock physics modelling result of gas dissolved in water sandstone using an example from the Nakajo oil and gas field, Niigata basin, Japan. The study also revealed water saturation estimated from sonic logs are well matched to gas data of mud gas readings. Application of rock physics modelling results to seismic data is also discussed.

Summary

Natural gas dissolved in water (GDW) is a common form of hydrocarbon occurrence in Japan, and gas production from the GDW accounts for 17% (approximately 290 mmscf/day) of total natural gas production in the country. Nakajo field is one of the fields which has long production history from the GDW reservoir. However, as GDW sandstone itself has not been well understood in aspect of the petrophysics and rock physics behaviour, it was a new finding for us that gas-effect and gas-effect-like sonic responses were observed in the GDW sandstone. In this paper, we show result of rock physics modelling, fluid substitutions, synthetic seismogram calculations, wedge modelling and AVO modelling for GDW sandstone. The rock physics modelling results revealed presence of isolated gas in GDW reservoir. Besides, water saturation estimated from rock physics modelling was regarded as useful tool to detect GDW reservoir with gas isolation. An application to seismic data is also encouraged according to synthetic seismogram calculations and AVO modelling. These outcomes allow a better evaluation approach of the surrounding area where future exploration and development potential may exist.

Main Objectives

The petrophysical properties estimation of dolomite reservoirs in China is very challenging for their low porosity and complex pore structures caused by the burial depth and diagenesis. In order to obtain petrophysical properties more reasonably and accurately, a petrophysical properties estimation method with pore structure constraint is proposed.

New Aspects

We proposed a statistical rock physics model for dolomite reservoirs to highlight the effect of pore structure on the relationship between petrophysical properties and elastic parameters, and then combine it with the pre-stack seismic inversion results and Bayesian inversion framework to quantitatively characterize the Petrophysical properties of deep dolomite reservoirs.

Summary

The dolomite reservoirs in China are generally deep buried and the prediction of petrophysical properties is very challenging for their low porosity and complex pore structures caused by the burial depth and diagenesis. The frame flexibility factor can be used as proxy for pore structure quantification such as quantitative carbonate pore type analysis and classification. In this paper, we propose to build the statistical rock physics relationship between petrophysical properties and elastic parameters under the constraint of frame flexibility factor. We used Bayesian inversion framework to successfully obtain the estimation results of petrophysical properties under the constraint of pore structure. show that the new rock physics relationship is more reasonable and the proposed approach has a higher estimation accuracy of petrophysical properties. Real data tests show that our approach is more reasonable and has a higher estimation accuracy.

Main Objectives

Rock physics modeling

New Aspects

Cement quantitively identification in lab conditions

Summary

When building the rock physics model for clayey sandstones, it is very important to know the contact type and content of cement clay in clayey sandstone for correctly understanding the influence of the cementation on acoustic velocity of clayey sandstone and how to reasonably establish rock physical model. At present, there is no laboratory method to directly quantitatively estimate the content of the cement clay, which leads to large errors in predicting acoustic velocities when using the cemented sands model for cemented sands formation because of the over-estimated of the cement clay. In this paper, a method for distinguishing cemented clay is proposed by observing the contact relationship and relative distribution between clay and particles in thin sections. The cement clay is used as the input parameter of the constant cement model. The comparison shows the velocities error predicted in our method is 20% lower than the original model, and the predicted results are significantly improved. The method proposed in this paper is a suitable rock physics model for weak cemented formation. It can predict acoustic velocity to identify favorable reservoirs and quantitatively evaluate reservoir parameters by combining seismic and logging data.

Main Objectives

Predict the limestone reservoir and dolomite reservoir in the complex carbonate formation quantitatively, and identify gas reservoir in these two types of reservoir.

New Aspects

Optimize Xu-Payne modeling and create two new sensitive elastic factors of lithology and gas-bearing.

Summary

Faced with the complex lithology and gas characteristics of carbonate rocks, the original acoustic wave curve, density curve, and calculated elastic parameter curves have been unable to identify reservoirs and gas layers accurately. In this paper, we use Xu-Payne optimization modeling method to take the measured P-wave logging as the standard curve, and use the least square method to calculate the lithology content iteratively and obtain the corrected velocity curves and density curve. Next, we use the coordinate rotation method to construct lithology-sensitive density projection factor and identify dolomite reservoirs and limestone reservoirs, and finally we construct fluid-sensitive elastic factors under different lithology reservoirs. In this way, the respective gas layer response characteristics of the dolomite reservoir and the limestone reservoir are revealed, and the threshold is set to provide a basis for the quantitative prediction of subsequent seismic inversion.

Main Objectives

Bridge the quantitative relationships between kerogen content and directional elastic and dynamic mechanical properties which take accounts anisotropy in organic-rich shale

New Aspects

Apparent dynamic Poisson’s ratios υ have good agreement with the true dynamic Poisson’s ratio υ31 and υ12 which may simplify real field mechanical problems. At low kerogen content (<10%), VP/VS ratio and kerogen content have linearly negative correlation, and the magnitude of VP/VS ratio which is less than 1.7 may be a significant characterization for high organic−rich (>10%) shale.

Summary

Understanding the quantitative responses of anisotropic behaviours in shale, which are induced by kerogen content, is of great significance in assessing hydrocarbon production potential and hydraulic fracturing design. Controlled experiment is difficult to be conducted on natural shales due to their characterizations of high variations. In this work, hot-pressing technique is applied to create well-controlled artificial organic-rich shale (AORS) for quantitatively researches. 11 AORS anisotropic samples with different kerogen content are constructed and machined into prism-like shape where the opposing faces allow for wave propagation along different directions. Quantitative relationships between kerogen content and directional elastic properties (velocities, anisotropic parameters and VP/VS ratio), dynamic mechanical properties (Poisson’s ratio, Young’s modules and stiffness) to the bedding plane are obtained at ambient conditions with ultrasonic experiments. Bedding perpendicular P-wave velocities are more sensitive to the increasing kerogen content than corresponding bedding paralleling velocities. Apparent dynamic Poisson’s ratios υ have good agreement with the true dynamic Poisson’s ratio υ31 and υ12 which may simplify real field mechanical problems. At low kerogen content (<10%), VP/VS ratio and kerogen content have linearly negative correlation, and the magnitude of VP/VS ratio which is less than 1.7 may be a significant characterization for high organic−rich (>10%) shale.

Main Objectives

Seismic Rock Physical Analysis

New Aspects

Fixed modeling steps

Summary

The deep tight sandstone reservoir has the characteristics of strong heterogeneity, complex lithology and reservoir space. Different pore structures have a great influence on the elastic parameters of the tight sandstone reservoir, which makes the conventional rock physics model unable to get the accurate elastic parameters of reservoir. To solve this problem, this paper improves the construction method of conventional tight sandstone rock physics model based on the linear fitting method and Mori-Tanaka model, obtains more reasonable shear wave velocity, fracture weakness parameters and pore structure parameters. The relationship between pore structure parameters and elastic parameters is obtained through rock physics analysis, and then combine it with the prestack seismic inversion results to quantitatively characterize of pore structure of deep sandstone reservoirs. The effectiveness and accuracy of the rock physics model are verified by the application results from Sichuan Basin in China, which can provide theoretical and practical guidance for further exploration of deep tight sandstone gas reservoirs.

Main Objectives

Rock Physics

New Aspects

Simulation of the effect of rock diagenetic processe on rock properties

Summary

The effects of compaction and cementation on rock permeability are studied by simulating the deposition and diagenetic processes of spherical grains with uniform size. Compaction and cementation are essential parts of the formation of consolidated rock. The simulation of the cementation process uses the morphology algorithm, which ignores many complications and therefore may not represent the real process of rock formation. The single phase fluid flow through the modeled rock is simulated by solving the Navier-Stokes equations. The absolute permeability of the sample is obtained by integrating the flow velocity at the outlet or inlet boundary and combining with Darcy’s law. The simulation shows that compaction and cementation reduce the connectivity and permeability of the deposition model. The results described here are only a preliminary quantitative analysis of the relationship between permeability and diagenetic process, and will be analysed and interpreted in combination with the experimental results in the future.

Main Objectives

Polarization analysis and 3D patterns for the six basic seismic moment tensors, provide the basic for seismic moment tensor inversion using polarization characters.

New Aspects

First time analyze the polarization characters for six basic moment tensor sources over the ground surface. Proposed that it is possible that such patterns could be exploited to invert for hypocenter location and moment tensor of source. Considering the actual borehole microseismic field data and point out that a much wider aperture monitoring system is required to sense rapid and/or appreciable angular changes in the polarization vector for effective moment tensor inversion.

Summary

Polarization is an important property of seismic waves that specifies the direction of particle motion. A full polarization measurement system in 3D space could be exploited for passive seismic event detection, seismic direction finding, and wavefield filtering. Seismic sources with different moment tensors will cause polarization differences which could in principle be used for focal mechanism analysis, even for moment tensor inversion. Here, we simulate the 3D vectoral seismic data set for the six basic moment tensor sources which are then used for polarization analysis. Then, we analyze the polarization patterns (azimuth and inclination angle variations) for P and S waves over the ground surface for a buried source, taking into account polarity changes. The results show that the seismic waves from the different basic moment tensor sources have dissimilar polarization characteristics. Such patterns could be exploited to invert for the hypocenter location and the moment tensor of the source. We next consider an actual 3C borehole microseismic field data set and extract the polarization angle. It is clear that a much wider aperture monitoring system is required to sense rapid and/or appreciable angular changes in the polarization vector for effective moment tensor inversion.

Main Objectives

Research on method of computing the backazimuth of the microseismic events

New Aspects

We introduce a relative azimuth method named maximum projection based method to compute the relative azimuth between the microseismic events which is more robust to the low S/N situation

Summary

In order to reduce calculation error of backazimuth influence on locating microseismic events generated by hydraulic fracturing with low signal to noise ratio (SNR), we introduce a method called maximum projection, which computes the backazimuth difference between the reference event and the target event by searching the optimal relative back azimuth that yields the maximum dot product of the two back azimuthal unit vectors. Furthermore, when array data are available, we can stack the inner product at each sensor to reduce the noise-related error. We test the maximum projection method extensively with a large amount of synthetic data, and the results suggest that measurements with the maximum projection method is more robust than those from the traditional covariance matrix method when the SNR of the microseismic records are low. We also apply the method to a field dataset and the relocated microseismicity tends to be more tightly clustered.

Main Objectives

Fast processing of seismic records for early warning and surveillance projects

New Aspects

Combination of compressive sensing and deep learning for fast data processing of seismic data

Summary

This work investigates the performance of a processing methodology based on Compressive Sensing and Deep Learning. In this application, the objective of the methodology is the estimation of location and moment tensor of seismic events. The method facilitates data processing in compressed form, which reduces computational effort both during neural network training and use for prediction. The compression is applied over individual channels, for which the benefits are more relevant for dense monitoring networks such as, for example, those using Distributed Acoustic Sensing. Performance results in terms of prediction errors are presented with respect to variations in the dimensions of the solution space and level of compression applied. Additionally, the results offer an idea of the size of the training sets required to prepare a prediction system based on the proposed approach.

Main Objectives

Combine Gaussian Process Regression with covariance-based noise modelling for the generation of realistic, geometry-independent noise models.

New Aspects

A noise modelling procedure that can accurately capture the spatio-temporal characteristics of noise and allows for a spatial transformation of the noise into any desired geometry.

Summary

Synthetic datasets are vital for the development and benchmarking of new processing and imaging algorithms as well as in the training of machine learning models. It is therefore important that such datasets are generated with realistic noise conditions making them resemble as much as possible their corresponding field datasets. Building on previously developed covariance-based noise modelling, we propose an extension of such an approach that aims to translate a noise model onto a user-defined geometry by means of Gaussian Process Regression. Starting from a synthetic data, we show that noise models can be generated and transformed into a desired geometry whilst keeping the same underlying statistical properties (i.e., covariance and variogram). The modelling procedure is subsequently applied to the ToC2ME passive noise dataset transforming the actual 69-sensor acquisition geometry into a gridded, 56-sensor array. The ability to generate realistic, geometry-independent noise models opens up a host of new opportunities in the area of survey design. We argue that by coupling the noise generation and monitoring algorithms, the placement of sensors could be further optimised based on the expected microseismic signatures as well as the surrounding noise behaviour.

Main Objectives

Implement alternative methods for epicentre location and magnitude estimation of fluid-induced seismic events.

New Aspects

The observed correlation between the measured coda lengths and magnitudes from the studied seismic events can be implemented to calibrate a coda-magnitude scale for the Groningen area.

Summary

Microseismic monitoring is an essential tool required to better understand the induced-seismicity processes observed in multiple oil and gas, and geothermal reservoirs. The data obtained from a seismic monitoring program, as the location and magnitude of the observed induced events, can be integrated with the near-surface conditions to assess the ground motions generated by these events at surface, and its impact to local infrastructure. However, its reliability strongly depends on the implemented methods for hypocentre location, where most of the conventional methods rely on a pre-defined velocity model. As an alternative, an epicenter location method based on the linearized normal moveout equation is proposed to locate events without the need of a velocity model. A magnitude estimation method based on the duration of the coda wave is also analyzed, which can be measured in seismic data acquired with conventional geophones. These methods are applied to locate and estimate the magnitude of some induced seismic events reported in the Groningen gas field in the Netherlands, detected with a dense shallow-borehole monitoring array that also allows a reliable measurement of seismic velocities of the near surface that can be integrated with a subsequent ground motion analysis of the same events.

Main Objectives

Seismic tomography and Joint inversion

New Aspects

We have presented an optimized workflow to simultaneously determine velocity distribution consistent with the passive seismic data, good source locations and source excitation times. It is largely automatic and independent of a priori information.

Summary

Almost all techniques to locate seismic events require a velocity model which is reasonably close to the true subsurface structure. We propose an optimized workflow for joint source localization and velocity estimation using wavefront attributes of passive seismic data. The optimized workflow comprises two main steps: the first step is to refine the source excitation time estimation for each considered source, and in the second step we simultaneously invert for the excitation times and the velocity model with all sources. In each iteration of the inversion, wavefront tomography and reverse modelling are carried out sequentially. Wavefront attributes extracted by a coherence analysis, serve as input for wavefront tomography. The velocity model used for reverse modelling is obtained via wavefront tomography while the source excitation times used for wavefront tomography is estimated by reverse modelling. Next to the near surface velocity no prior information is required by the method to invert for a velocity model consistent with the passive seismic data, good source locations and source excitation times. We illustrate the optimized workflow with a 2-D synthetic numerical example. The maximum source location errors are within the dominant wavelength and the maximum source excitation time errors are within the prevailing period.

Main Objectives

Investigating the new introduced 3D transdimensional tomography for using in a seismic ambient noise data recorded over Reykjanes Island

New Aspects

Using 3D transdimensional tomography method for the recovery of 3D surface wave velocities and modifying the algorithm to have a better sampling of the posterior

Summary

Ambient-noise surface wave tomography has proven to be an effective tool for 3D crustal imaging. Conventionally, a two-step inversion approach is adopted. That is, a first inversion results in separate, frequency-dependent 2D surface-wave velocity maps, upon which local frequency-to-depth inversions are performed. In our case, phase velocities are used. A one-step 3D non-linear algorithm has recently been proposed in a Bayesian framework using reversible jump Markov chain Monte Carlo which is called transdimensional tomography. This approach has a number of advantages over the two-step approach with the most notable being the fact that the one-step approach preserves spatial correlation information in 3D. The 3D volume is parameterized using a polyhedral Voronoi tessellation. Here, we investigate the feasibility of using this transdimensional algorithm to recover the 3D seismic velocity structure below the Reykjanes peninsula, SW Iceland. To that end, we design synthetic tests using the station configuration of Reykjanes seismic network, which has been recording the field data. We show that the algorithm successfully recovers the 3D velocity structure of the area. This work is a first step towards 3D non-linearized surface-wave tomography using the field data recorded by the stations of the Reykjanes seismic network.

Main Objectives

The main objectives of this work are to compare the possibilities of recording weak local seismic events with surface and shallow seismological networks, to estimate an increase of network sensitivity and verify results with the use of the catalog of the downhole microseismic monitoring system.

New Aspects

The results may be applied to the fields characterized by induced seismicity and high level of industrial noises because such an increase in the number of recorded weak earthquakes can have a significant impact on the results of the seismicity analysis.

Summary

We compare sensitivity of surface and shallow seismological networks in recording weak local seismic events. We analyzed real records from the existing seismological networks containing six monitoring points. Initially surface seismologic network was operated for several years. Then the seismometers were buried at the same locations to depths ranging from 70 to 90 meters. In the same region another downhole microseismic array was operating. This downhole array had the highest sensitivity providing the reference catalog of local weak microseismic events for comparison with the results of the surface and shallow seismological networks.

Analysis of real-data records revealed high variability of the improvement of the station sensitivity after installing it at shallow depth. Background noise level decrease ranged from 2.5 to 40 times depending on the point. Such variability is indicating that the area is characterized by considerably different nature of background noise level. There is also significant variation in noise levels during working and non-working hours. The average noise reduction for the entire network was 11 times. Comparison with the reference catalog (from downhole array) revealed that the network sensitivity increased by 0.92 (ML) and the number of recorded weak earthquakes increased by 10 times after shallow installation.

Main Objectives

We derive the second-order coefficients (principal curvature) of the slowness surface for two S waves in the vicinity of three symmetry axes and define the elliptic form function to examine the existence of the on-axis triplication in ORT model. The existence of the on-axis triplication is found by the sign of the defined curvature coefficients. An ORT model is defined in the numerical examples to analyse the behaviour of the on-axis triplication.

New Aspects

We extend the condition of the on-axis triplication in the ORT model from the vertical axis to all three symmetry axes. This on-axis triplication research could be used as the complement to the one described in the off-axis triplication in the elastic ORT model. The explicit forms of second-order coefficients in the vicinity of three symmetry axes are derived in terms of the anisotropy parameters. The existence of the triplication is possible to analyze by studying the negative value of these curvature coefficients and the defined elliptic function.

Summary

We derive the second-order coefficients (principal curvature) of the slowness surface for two S waves in the vicinity of three symmetry axes and define the elliptic form function to examine the existence of the on-axis triplication in ORT model. The existence of the on-axis triplication is found by the sign of the defined curvature coefficients. An ORT model is defined in the numerical examples to analyze the behavior of the on-axis triplication. The plots of the group velocity surface in the vicinity of three symmetry axes are shown for the ORT model where different shapes: convex or the saddle-shaped (concave along one direction and convex along with another) indicates the existence of the on-axis triplication.

Main Objectives

The goal of our study is to include laws deduced from geological observations on fractures’ distribution in simulations of seismic wave propagation.

New Aspects

Our approach uses the non-periodic homogenization, which hasn’t been used to study fractured media but has proven its efficiency on complex geological model, to consider geological observations.

Summary

The presence of multi-scaled fractures in the crust of the Earth has been largely evidenced by geological observations that support power laws to describe fracture distribution.

Two main approaches have been developed to study fractured media : numerical simulations of waves or effective medium theories that assumes an Elementary Representative Volume or ERV. Each of these methods focuses on a restricted range of fracture size because of computational or theoretical consideration.

In this work, the non-periodic homogenization is exploited to go beyond size restriction and explore the effective properties of multiscale fractured media. We first ensure that this method can be applied to various scales by comparing its solutions to those of the two previous approaches.

Then, we build a 2D medium that accommodates fractures of various lengths. The density of each fracture set is predicted by a power law function of the fracture size, as supported by geological studies. We evidenced that the non-periodic homogenization is efficient to retrieve the effective properties of fractured media and that this method is well adapted to investigate multiscale fractured medium.

Main Objectives

Find an effective method of evaluating the pore structure and the water saturation.

New Aspects

The pore structure model consisted of spherical and cylindrical pores, the renormalization method, the iterative bisect inversion method and the Newton iteration method.

Summary

Fractured-vuggy reservoirs are currently one of the main exploration and development areas in the oil and gas industry. Due to the many types of complex pore structures and strong heterogeneity of pore distribution in the formation, there is no universally effective methods of evaluating the pore structure and the water saturation, therefore, investigation of the effect of pore structure and fluid saturation on resistivity has important theoretical and applicable significance, and it is also helpful for the development of effective saturation inversion method. In this study, the pore structure model consisted of spherical and cylindrical pores is adopted to simulate porous rocks with different electrical conductivity in three orthogonal directions, and the effective resistivity of the set of partitional blocks within the response domain of the logging tool is evaluated through the renormalization method. By adjusting the internal pore structure parameters, we can calculate the resistivity of the sphere-cylinder pore system that is approximate to the actual porosity formation, and predict the average saturation of the complex formation through iterative bisect inversion method and the Newton iteration method. The numerical results show high accuracy of inverted saturation and break a sound path for evaluation of fractured-vuggy reservoirs.

Main Objectives

fractures ” seismic anisotropy” “waveform inversion” “fracture stiffness”

New Aspects

Stochastic waveform inversion approach to characterise a cracked medium

Summary

We numerically assess the effectiveness of the equivalent medium (EM) theory for discrete fracture network (DFN) models by means of seismic waveforms. More specifically, we analyse how the discrete crack parameters, such as crack density, crack stiffness, and crack size, influence the effective properties of a cracked medium by utilising the stochastic waveform inversion method, GA-FWFI. Interestingly, we observe that the equivalent fracture stiffness of a cracked medium and the stiffness of its constituent cracks are not linearly related as we might expect. Furthermore, even though the seismic wavelength is longer than the crack size, we show that crack size has an impact on the variation of the effective parameters. Additionally we find that this waveform inversion approach is robust and accurate in estimating the orientation of the fracture-set, even though the dataset is from a single azimuth.

Main Objectives

The main objective of this method is to increase the geological information of fracture zone to seismic impedance for enhancing the resolving power to distinguish the characteristics of the medium-small scale fractured-vuggy reservoirs controlled by strike-slip fault zone, and solve the contradiction between inter-well production dynamic data and conventional seismic interpretation data, provide a further basis for solving the contradiction between the inter-well production dynamic data and finally provide further basis for determining reservoir unit boundary, evaluating reserves and formulating development plan.

New Aspects

The new aspect of this abstract is to propose a method to improve the accuracy of reservoir prediction. In the low-frequency impedance model (LFIM) of heterogeneous limestone reservoir inversion, the sweet-spot information of fractured-vuggy reservoir controlled by strike-slip fault zone is added, which makes the characteristics of the low-frequency model more in line with the geological characteristics of strike slip fault. Second or more constrained inversions based on the newly LFIM are implemented in the process. In the last stage, by combining the relative impedance with the new LFIM, a satisfactory absolute impedance characteristic is obtained.

Summary

The fractured-vuggy reservoir (FVR) of the Ordovician ultra-deep tight limestone strata in the southern Tabei uplift of the Tarim Basin is the core targets for oilfield capacity building. The seismic response of the FVRs is multi-peak-trough single or group bend-like reflections. With the complexity of the reservoirs, using the constrained spares spike inversion method based on the conventional low-frequency impedance model (LFIM) is difficult to describe the FVRs, as well as contradictions between wells production performance and seismic data. To solve the bottleneck, we propose a seismic iterative inversion method that incorporates a priori knowledge through a preconditioning LFIM of strike-slip fault zone encased in tight limestone to improve reservoir prediction. The process carries out the second or more constrained inversion based on the newly LFIM. The last stage obtains a satisfactory absolute impedance property by merging the relative impedance with the newly LFIM. This method enhances the resolving power to distinguish the characteristics of the medium-small scale FVRs controlled by strike-slip fault zone, and solves the contradiction between the inter-well production dynamic data and finally provides a further basis for formulating a reservoir development plan. These results are corroborated by the application of this method in the HD-YK area.

Main Objectives

Exploring the capabilities and advantages of the full azimuth seismic data in fracture prediction, and comparing azimuth anisotropy characteristic of OVT gather and common reflection angle gather, and analysing the fracture prediction difference and advantage of the two types of gathers.

New Aspects

Based on the full azimuth seismic data, aiming at carbonate fracture-cavity reservoirs, made comparison of azimuth anisotropy characteristic of OVT gather and common reflection angle gather, and finally analysed the accuracy and advantage of fracture prediction about the two types of gathers.

Summary

Compared with traditional common offset gather, the common reflection angle gathers can obtain true azimuth information of the formation, overcome the illusions existing in the conventional common offset gathers, and improve the accuracy of seismic imaging，which is conducive to seismic amplitude attribute analysis and fracture prediction. Based on the full azimuth seismic data of the Sichuan Basin, aiming at the Lower Permian Maokou Formation fracture-cavity reservoir fine prediction a, the OVT processing and the common reflection angle imaging processing are carried out, and the two types of azimuth gathers are compared, and further pre-stack fracture prediction are conducted using Ruger formula. The results show that the AVAZ prediction method based on common reflection angle gathers has achieved good application effects in the prediction of small-scale fracture-caves. The predicted spatial distribution of fracture-caves is significantly better than that of OVT gathers and the prediction result conform to the law of karst reservoir development, which is beneficial to fine study of fracture-cavity reservoirs.

Main Objectives

poster

New Aspects

none

Summary

This paper uses the moment tensor source model to define explosive and shear sources. Based on the three-dimensional anisotropic medium elastic wave equation and staggered-grid finite-difference method(Graves, 1996), we perform the forward modeling of the passive source microseismic wavefield. By comparing theoretical calculation results of waveform records, arrival time and amplitude difference between anisotropic model and isotropic model, the impact of shale anisotropy on microseismic wavefield under different focal mechanisms is quantitatively analyzed.

Main Objectives

The main objectiovers are that there is no need to find reflections that share the same ray paths as the target reflections, and we just pick another adjacent reflection in the overburden for inversion.

New Aspects

We have developed a novel Q estimation method, called logarithmic spectral simultaneous inversion (LSSI). We first calculate the travel time of picked reflections by ray tracing, and then separate the attenuations from different layers according to the travel time.

Summary

Frequency domain Q estimation methods usually pick two reflections at different traveltimes, and estimate Q according to the variation of their amplitude spectra. However, the variation of the amplitude spectra is not only affected by the target layer, but also affected by the overburden, which lead to error of Q estimation results. We have developed a novel Q estimation method, called logarithmic spectral simultaneous inversion (LSSI), to address this problem. This proposed method account for the effect of the overburden by picking another adjacent reflection, which is used as a constraint for the Q value estimation in the overburden, and separating attenuations from different layers according to the traveltime through the process of inversion. Q values of both target layer and overburden are estimated simultaneously. Model test and field data application indicate the validity of the method.

Main Objectives

Oriented inverse Q filtering

New Aspects

Prestack attenuation factor estimation via warpped mapping

Summary

Attenuation is the main factors responsible for degrading the resolution of seismic data. It seriously decreases the energy of signal components especially those with higher frequency. The attenuation-induced energy loss can be partially compensated via inverse Q filtering. However, inverse Q filtering is inherently not stable, which suffers from overcompensation and noise burst. Among the existing compensation methods, poststack data are used more frequently. Nevertheless, poststack compensation can’t provide us with compensated prestack seismic data directly, which are commonly used in prestack impedance inversion and other prestack processing. We have developed an oriented and stable method by introducing plane-wave destruction (PWD), predictive painting and stable division based on Taylor expansion for compensation of attenuating common midpoint (CMP) gathers. We call it warped mapping (WP). The compensation of synthetic data is carried out to illustrate the well performance of the proposed method even though in the presence of ambient noise.

Main Objectives

VSP, P and S wave, shale oil

New Aspects

Research of shale oil reservoirs geophysical characteristics based on P and S wave VSP data

Summary

As the exploration of unconventional oil and gas reservoirs continues to deepen, traditional P wave seismic exploration methods cannot well meet the requirements of exploration and development, and geophysics methods with higher accuracy and more abundant information urgent need to research and development. In this paper, a method of using P and S wave VSP data for shale oil reservoir research with low porosity and low permeability is proposed. Firstly, 9 typical wells were selected, and the 3 component VSP data was acquired by P and S wave vibrators. Then the characteristics of seismic velocities, amplitude attenuation and frequency attenuation in this area were analyzed. It can deeply understand the laws of seismic wave propagation and attenuation in this area, and theoretically calculate the seismic resolution of this area, which could provide support for high-quality surface seismic processing and comprehensive reservoir research.

Main Objectives

This paper achieves attenuation compensation for prestack data.

New Aspects

We propose a novel prestack attenuation compensation method based on inversion considering the influence of ray paths on the absorption attenuation.

Summary

Due to the absorption of subsurface media, seismic waves experience energy attenuation and waveform distortion, which seriously decreases the resolution of seismic data. For prestack seismic data, since the effect of absorption attenuation varies with the propagation path, the amplitude variation with angle (AVA) trend will be distorted. Therefore, we propose a novel prestack attenuation compensation method based on inversion considering the influence of ray paths on the absorption attenuation. We first derive the frequency domain forward formula of the prestack gather in the attenuation media, then reduce the attenuation compensation to an inverse problem, and utilize Tikhonov regularization for stability processing to achieve compensation. Numerical tests, comparative analysis of different compensation methods and noise immunity experiment demonstrate that our method has higher accuracy and can perform attenuation compensation for prestack gather more stably and effectively.

Main Objectives

Advanced edge-aware filtering of seismic images to enhance S/N and retain discontinuities.

New Aspects

Use of a Siamese convnet to learn a similarity function between seismic image patches.

Summary

Siamese neural networks have proved powerful at learning discriminative features for tasks such as face verification. We present a novel application, called Edge-aware Filtering, which employs a deep Siamese network ranking similarity between seismic image patches. Once the network is trained, we capitalize on the learned discriminative features to achieve within-image stacking power endowed with edge awareness. We show on field data examples that the learned representations lead to superior filtering performance at various discontinuity scales compared to 3D anisotropic diffusion and structure-oriented filtering.

Main Objectives

The proposed denoising method can preserve the structural continuities of signals including low-frequency components, together with edges.

New Aspects

We adopt simultaneous sparsity constraints of the first-order difference of signals along the structural direction and time direction, described by minimizing the Cauchy function, as a combined constraint term imposed on the time-domain data misfit to propose an inversion-based edge-preserving and signal-preserving noise reduction method.

Summary

Attenuating random noise while preserving edges and structures in seismic dataset is of great significance for the following seismic inversion and interpretation. From the viewpoint of inversion, the utilization of more information is an effective way to improve signal-to-noise ratio of seismic data. In this study, we adopt simultaneous sparsity constraints of the first-order difference of signals along the structural direction and time direction, described by minimizing the Cauchy function, as a combined constraint term imposed on the time-domain data misfit to propose an inversion-based edge-preserving and signal-preserving noise reduction method. In this way, the redundancies along both time slices and seismic sections are simultaneously considered, and the edges along the spatial directions can be preserved. The performance of the method is mainly dependent on a trade-off parameter and a scale parameter, which makes it easier to obtain a relatively perfect noise reduction result. The applications on two real poststack datasets and a real prestack dataset demonstrated that the proposed method is an effective edge-preservation and amplitude-preservation denoising tool.

Main Objectives

Using the method proposed in this paper to remove the noise and improve the signal-to-noise ratio of the microseismic data, so that we can recognise the effective microseismic events easily.

New Aspects

In this paper, we propose a single channel independent component analysis algorithm based on phase space reconstruction for noise attenuation. Through PSR, the one-dimensional signal is reconstructed into a high-dimensional phase space without destroying the dynamic characteristics of the original data, while meeting the condition of the ICA algorithm. Combine the dynamic and higher statistical characteristics, the single channel PSR-ICA algorithm can effectively separate the noise and improve S/N ratio of the microseismic data.

Summary

The signal-to-noise (S/N) ratio of the microseismic data is very low and the energy of the effective signal is also weak. In addition, due to complex noise, the effective signal is usually overwhelmed by the random noise. So it is important to improve the S/N ratio of the microseismic data for recognising the effective signal. However, conventional denoising methods may not be fully qualified to remove the noise and improve the S/N ratio. In this paper, we propose a single channel independent component analysis (ICA) method based on phase space reconstruction (PSR) for noise attenuation. Through PSR, the one-dimensional signal is reconstructed into a high-dimensional phase space without destroying the dynamic characteristics of the original data, while meeting the condition of the ICA algorithm. Combine the dynamic and higher statistical characteristics, the single channel PSR-ICA algorithm can effectively separate the noise and improve S/N ratio of the microseismic data, so that the effective signal can be recognised easily.

Main Objectives

In this paper, according to the time-frequency characteristics of seismic signals, a deniosing method based on variational mode decomposition (VMD) and independent component analysis (ICA) algorithm is proposed to suppress the random noise.

New Aspects

A random noise suppression method based on the VMD-ICA algorithm is proposed in this paper. Because VMD method mainly used for high frequency random noise suppression, ICA algorithm uses the weak non-Gaussian nature of random noise to identify random noise by extracting independent sources, so the VMD-ICA method has a better suppression effect on random noise.

Summary

Seismic random noise has a serious impact on the resolution and signal-to-noise (S/N) ratio of seismic data. It can also cause unclear layers and structural artefacts in the seismic profile, which will affect the subsequent processing and interpretation effects. So it’s very important to remove the random noise effectively, especially in high-precision seismic data processing. However, for data with low S/N ratio, it’s difficult to separate the effective signal from the noise. In this paper, according to the time-frequency characteristics of seismic signals, a deniosing method based on variational mode decomposition (VMD) and independent component analysis (ICA) algorithm is proposed to suppress the random noise. Seismic signals can be decomposed into several mode components from high frequency to low frequency with a certain bandwidth by VMD, and combine the advantage of ICA that can extract the independent source signals. The noise and effective signal can be separated effectively, so that achieve the purpose of removing the random noise and improving the S/N ratio of seismic profile.

Main Objectives

suppress linear coherent noise

New Aspects

using mathematical morphological filters in common offset domain to suppress

Summary

The attenuation of linear coherent noise is a persistent problem in seismic processing and imaging. Traditional methods utilize the differences in frequency, wavenumber or amplitude between useful signals and noise to separate them. However, in some cases, the differences are too small to be distinguished, and the traditional method are limited or even invalid. So we introduce mathematical morphological filter to attenuate the linear coherent noise utilize the differences in the shape of seismic waves. In the seismic exploration, we can see that the same linear coherent noises exist in some different common shot gathers. After investigation, we find that the trajectories of linear coherent noises in the common offset gather are horizontal continuous ones. So we implement the MMF attenuation the linear noises in the common offset gathers by using the horizontal consistency. And apply the proposed method on field seismic to show the excellent performance.

Main Objectives

We propose a denoising autoencoder based on unsupervised machine learning to suppress random noise.

New Aspects

we introduce a denoising autoencoder for random noise attenuation, which can extract and compose the robustness features from seismic data without any labels.

Summary

In the field of exploration geophysics, seismic waves received by near-surface geophones are usually corrupted by random noise, which degrades the performance of the following seismic exploration process, such as imaging and inversion. Therefore, random noise attenuation plays an essential step in seismic data processing. In this research, we propose a denoising autoencoder to remove random noise from seismic records. Different from traditional autoencoders that constrain representations, the denoising autoencoder trys to attain appropriate representations by changing the reconstruction criterion, which allows neural network to capture the true seismic wave composition and then attenuate random noise. Compared with the other methods, real data shows that the proposed method achieves better performance in terms of the weak signal preservation and random noise attenuation.

Main Objectives

This paper uses corresponding methods to gradually suppress the denoise: First of all, the method of near-shot abnormal energy attenuation is used to suppress surface waves and strong scattered waves in thick loess areas; Secondly, The coherent noise attenuation technology of up-going and down-going wave field separation is used to suppress shallow multiple refraction; Finally, five-dimensional interpolation is used to further improve the SRN of the data for the irregularities of the observation system caused by the complex surface.

New Aspects

The coherent noise attenuation technology of up-going and down-going wave field separation is used to suppress shallow multiple refraction

Summary

The signal-to-noise ratio (SRN) of the original data in the piedmont zone in western China is extremely poor, with serious noise interference such as surface waves, strong scattering, shallow multiple refraction and random noise. This paper uses corresponding methods to gradually suppress the denoise: First of all, the method of near-shot abnormal energy attenuation is used to suppress surface waves and strong scattered waves in thick loess areas; Secondly, The coherent noise attenuation technology of up-going and down-going wave field separation is used to suppress shallow multiple refraction; Finally, five-dimensional interpolation is used to further improve the SRN of the data for the irregularities of the observation system caused by the complex surface. Through the denoising method in this paper, the SNR in the piedmont zone in western China is significantly improved, which provides a reference for structural imaging under similar geological conditions with low SNR.

Main Objectives

random noise attenuation

New Aspects

We develop entropy sampling to select the effective training point pairs and reduce the training set based on the texture complexity to improve the efficiency of training of CAE denoising.

Summary

We evaluated an improve Convolutional Auto-Encode method for seismic data denoising. The method learn extremely complex functions to effectively attenuate noise by learning and extracting features from a large amount of training data set based on statistical techniques. However, the large quantity of training point pairs may increase the burden of memory and computation during the training. To solve the problem, we develop entropy sampling to select the effective training point pairs and reduce the training set based on the texture complexity. That is, complex texture regions represent the dominant characterization of the seismic data, and these regions are sampled with higher probability as training data set. Numerical illustrations on 2D seismic data show that the proposed method reduces the training data pairs as much as possible to improve the efficiency of training, while ensuring accurate denoising results.

Main Objectives

environment noise attenuation

New Aspects

Multichannel abnormal amplitude preserving attenuation

Summary

In some complex seismic exploration environment, seismic data often contains some continuous multi-channel strong energy amplitudes caused by environment factor (wind, grass, water, etc.), which seriously affects the signal-to-noise ratio and imaging quality of seismic data. This kind of noise has no certain frequency and apparent velocity, existing almost in the whole record length in time, but it happens in some certain areas in survey. Due to the random properties and the continuous existence in a large number traces, it is very hard to remove this kind of noise completely under the amplitude preserved conditions just in shot domain or detection domain, and the residual strong energy noise will still seriously affects the final imaging quality. In this paper, a method of amplitude preserving removal of continuous strong energy amplitude is introduced. First, sorted the prestack seismic data in common offset domain, and order the traces in random sequence, then use the median filter technology with proper parameters in the frequency domain to remove the strong environment noise amplitude. The practice shows that the method is very effective in removing the continuous abnormal amplitude noise of land seismic data in a developed water system area in southwest china.

Main Objectives

seismic deblurring

New Aspects

prestack seismic deblurring using the fast extengding primal-dual hybrid gradient method

Summary

Seismic image, especially for the prestack image, performs a blurred version of the reflectivity image due to spatial aliasing, poor acquisition aperture and nonuniform illumination. The blurring effects can be quantified by the point spread function (PSF). We herein adopt an explicit space-variant PSF formula, which can be defined as a sequential application of the forward and adjoint operators with the asymptotic Green’s function. The deblurred images are restored using the nonstationary deconvolution with total variation regularization in which the blurred images are described by the convolution between the space-variant PSF and the reflectivity image. However, nonstationary deconvolution is computationally challenging. We introduce an extending primal-dual hybrid gradient method (ePDHG) to decompose the complex problem into a sequence of simple subproblems that have closed-form solutions. Numerical results on the synthetic and field data show the promising results in the prestack seismic image deblurring.

Main Objectives

Removal of linear noise such as mud-roll in shallow water marine data using the Curvelet Transform

New Aspects

Usage of Curvelet Transform in seismic data processing

Summary

Mud-roll comprises of dispersive seismic waves that propagate along the unconsolidated sediment layers at the sea floor in shallow water marine environments, where the water depth is normally less than 30 m. Mud-roll’s characteristics are spatially variable, i.e. the dispersion properties change from one shot to another across a seismic survey area. These complex kinematic properties make noise elimination very challenging using conventional seismic processing workflows. Our proposed method is a hybrid, Curvelet transform-based workflow that takes advantage of conventional seismic processing filtering to estimate the noise components, followed by the Curvelet transform that attenuates the residual noise energy that is difficult to remove with a conventional subtraction algorithm. In this paper, we illustrate the proposed Curvelet transform-based workflow using both synthetic and field data and demonstrate its effectiveness.

Main Objectives

noise reduction by unsupervised deep learning

New Aspects

application of an unsupervised denoising deep learning algorithm at an early processing stage

Summary

Noise is a major concern in seismic data and influences the processing and interpretability of seismic data at various steps. However, noise has a certain pattern, which can be exploited by machine learning algorithms, that rose drastically in popularity within the last decade. We aim to remove random noise at an early stage in the processing workflow in the shot-gather domain. We use an unsupervised approach without the time consuming necessity of generating labels.

In our work, we use an autoencoder, that resembles the U-Net structure but uses a ResNext50 encoder variant. The autoencoder aims to reconstruct its input. Due to the design of an autoencoder, such reconstruction is never perfect and omits the least correlating contributions, which is usually noise. We apply our approach to a marine dataset from the Levantine Basin in the Mediterranean Sea. We are able to remove without damaging primary signal, apart from the sea floor reflection, which acts as an outlier during training.

Main Objectives

preprocessing of the blended data

New Aspects

Deghosting on the blended data

Summary

In marine acquisition, the strong reflectivity of the sea surface results in ghost wavefields at the source and the receiver side. The removal of these ghost wavefields is a well-known data preprocessing step to improve the image resolution. Simultaneous source acquisition, or blending, which allows for a temporal overlap between shot records, has been proposed as a method for substantially reducing the acquisition cost while improving data quality. The effect of the blended data on the receiver degosting has not been well investigated. We propose to carry out receiver deghosting on blended seismic data (i.e., prior to deblending) using sparse inversion. Synthetic data example demonstrates the validity of this approach.

Main Objectives

Random noise,Adative non-local means filter,Direction estimation,Time-frequency analysis

New Aspects

An improved non-local means filter with adaptive search window determined by the direction and the instantaneous center frequency of seismic signals

Summary

The random noise in seismic data seriously limits the extraction of information related to underground geological structure and lithology parameters. How to remove such noise sufficiently while keeping the fidelity of original seismic data is an important issue in the seismic data processing. In the post-stack seismic profile, the most obvious feature is the structural feature and non-stationary feature. Inspired by it, we propose an improved non-local means (NLM) method with adaptive search window (ASWNLM). This method includes determining the direction of the search window by a correlation algorithm and the height of the search window by the instantaneous center frequency. Tests on synthetic data and real post-stack data prove our method can remove random noise as much as possible while preserving signal detail well even without a proper filtering parameter h.

Main Objectives

The method considers the directional information contained in the seismic data when using total variation(TV)-based method to suppress the noise. Besides, high-order derivatives are introduced to overcome the staircasing effect of TV regularization.

New Aspects

a high-order directional total variation method for seismic data denoising

Summary

Noise attenuation is one of the key problems in seismic data processing. Total variation (TV) has played an important role in seismic data denoising and reconstruction. We develop a high-order directional total variation method for seismic data denoising that considers the structural direction of the seismic data. It involves a parameter to balance higher-order derivatives, thereby reducing the staircasing effect of the bounded variation functional. We test the method on a model where the data are contaminated by different types of noise. The corresponding denoising performance is compared with the TV and conventional directional total variation method from two aspects of signal-to-noise ratio and effective signal leakage degree. The model test and field data application illustrate the advantage of this functional as a regularization term for seismic noise attenuation.

Main Objectives

random noise attenuation

New Aspects

A random noise reduction method in f-x domain

Summary

Random noise attenuation is a persistent problem in seismic exploration. As we known, the l2-norm regularization is always used to improve inversion stability. However, in the presence of noise, the result could be unstable. We have developed a novel random noise attenuation method for seismic data based on inversion in the frequency-space (f-x) domain. In this approach, a prediction error filter (PEF) is calculated from the noisy data. Then we use this filter as another constraint to the seismic data in order to get the de-noised data by inversion. The choice of trade-off parameters also influences the performance of this method. This method overcame the problem of the generation of spurious events and reduction of the signal amplitudes in the conventional f-x prediction method. Tests on synthetic example and field data example demonstrated that, the proposed method can recovery the signal effectively and decrease the influence of noise.

Main Objectives

Velocity modeling, tomography, forward modeling

New Aspects

Combination of tomographic inversion and forward modeling by ray tracing

Summary

A Vp velocity model of the sediments, crust and upper mantle of the Dodecanese area, Greece, was obtained by evaluating wide angle reflection refraction seismic (WARRS) data, using Ocean Bottom Seismographs and land stations along a 140 km line. Seismic energy was generated by air gun shots spaced at 120 m intervals fired by a tuned air gun array of 49 l volume. Velocity modeling by seismic tomography combined with forward modeling by ray tracing revealed the structure of the crust and sediments. Four sedimentary layers were mapped with velocities ranging from 1.7 to 5.4 km/s. Lateral velocity variations and existence of several faults indicate intense fragmentation and the development of several individual basins. Sediments are only 2 to 2.5 km thick. The upper crust is 6.5 km thick with Vp velocities between 5.8 to 6.5 km/s. The lower crust is 10 to 12 km thick with Vp velocities between 6.8 to 7.2 km/s. Moho discontinuity is identified at 20 to 22 km depth. A low velocity zone of 7.8 km/s in the uppermost part of the mantle is probably high temperature asthenosphere that was uplifted below the southern Aegean Sea due to the subduction of the Ionian oceanic lithosphere.

Main Objectives

Use horizon automatic picking technology to obtain the spatial location of seismic profile horizons, and use this to constrain the ray tracing and tomographic equations to obtain more accurate tomographic results

New Aspects

Automatic layer picking， anisotropic Ray Tracing

Summary

Tomographic inversion is a widely used velocity modeling method in exploration seismic field, which can effectively invert the velocity of underground media velocity. However, in the conventional reflected wave tomography inversion method, it is difficult to pick up the traveling time information, which ultimately leads to low accuracy of the inversion result. Adding horizon position as a regularization constraint in the tomography process can solve the above problems, but the existing horizon picking relies on manual picking, which is time-consuming, laborious and low precision. This paper proposes a new strategy for automatic horizon picking. First, adaptively stacks the ADCIGs (angle domain common image gathers), then filters the profile in data domain, finally uses a seismic DNA algorithm based on cluster analysis and multiple connection algorithms to automatic pick horizons. The horizon is used to regularize the ray tracing and tomographic equations, and different parameters are updated according to the residual residuals of ADCIGs in different angle ranges to achieve effective inversion of the anisotropy parameters of VTI media.

Main Objectives

Bayesian elastic seismic inversion

New Aspects

Ensemble Kalman Filter, Uncerntainty quantification

Summary

An ensembled-based method in the Bayesian framework is used for probabilistic seismic inversion. The goal is to estimate 1D elastic models and their uncertainties without having to perform explicit sampling of the posterior distribution. The problem is parameterized as a collection of 1D elastic (Vp, Vs, and density) profiles represented by multi-variate Gaussian distributions. At the centre of the algorithm is the Ensemble Transform Kalman Filter method implemented in an iterative least-squares optimization loop, in which linear weights of the ensemble members and the transform matrix are optimized in order to update the ensemble members. Two numerical examples are presented. The first one is pre-stack seismic inversion for elastic model building and the second one is a reservoir-targeted elastic inversion of post-stack seismic data. In both cases, a reasonable posterior estimate of the elastic parameters is obtained and the uncertainty associated with the inverted parameters is quantified.

Main Objectives

The paper aims to developing new method for velocity inversion/calibration when the source is also unknown. It is potentially useful in passive data survey, where the reconstructed velocity by the proposed method can facilitate subsequent source reconstruction.

New Aspects

There are two main differences between the proposed approach with existing ones. 1. existing methods assume sources to be a linear combination of separated point sources, while our method allows more general sources including those lying on a line singularity (representing rock cracks), as long as the activation time is relatively brief. 2. existing methods use alternative minimization procedures between the source and the velocity, whereas ours separately recover the two using an FWI like procedure, which makes it faster and more memory efficient.

Summary

We consider the problem of velocity inversion/calibration in passive survey, where the seismic source is also an unknown. In earthquake detection or microseismic localization, the major task is to reconstruct the passive seismic sources, but due to the source-velocity coupling, source reconstructions are inherently affected by inaccurate knowledge of the velocity, bringing the need of velocity calibration. We propose a source independent velocity calibration method that recovers the velocity without the source information, thus providing a better ground for source inversion. Unlike existing methods that assume sources to be a linear combination of separated point sources, the proposed method allows sources to lie on a line singularity (representing rock cracks), as long as the activation time is relatively brief. The proposed approach is based on the observation that the spatial distribution of the source is separable from the velocity model after a proper Helmholtz domain projection.

Main Objectives

joint inversion of natural source data

New Aspects

correspondence map joint inversion

Summary

Accurate characterization of the subsurface require the integration of multiple geophysical information, which can be done through joint inversion. The basis of any joint inversion method is the existence of functional links between the multiple model parameters. The two main joint inversion approaches are through structural or petrophysical relationships. Joint inversion using correspondence maps combines the advantages of jointly estimating geophysical models and parameter relationships. In this work, we use a correspondence map approach to invert jointly surface-wave dispersion curves and magnetotelluric data for subsurface shear velocity and resistivity but also for a possible relationship between them. By inverting synthetic data, we show that when this relationship is linear, we can recover both the geophysical models and the relationship parameters. We also show that the approach is successful when seeking higher-order relationships.

Main Objectives

Temperature-dependent model for analyzing joint effects of pore fluid and micro-cracks on rock anelasticity behaviors.

New Aspects

Modeling and experimental data show fluid property dependence of temperature dominates velocity-temperature relations in rocks.

Summary

Understanding the effect of temperature on the physical properties of rocks is important to develop deep oil, gas and geothermal resources. However, relatively little effort has thus far been made to capture the temperature effect on velocity in fluid saturated rocks. To overcome this limitation, a double-porosity medium model is developed, which incorporates the Batzle-Wang empirical equations for the pressure and temperature dependence of the pore fluid, as well as the David and Zimmerman models to quantify temperature-dependent micro-crack density. This model is tested on samples of carbonate, for which the variation of ultrasonic P-wave velocity with temperature has been measured. Modeling and experimental data show that the proposed model is versatile enough to quantitively describe the velocity-temperature relation. It turns out that for these rock samples the temperature-dependent fluid properties dominate the velocity-temperature relations.

Main Objectives

In order to approximate the traveltime in an elastic orthorhombic (ORT) medium for converted waves, we define an explicit rational-form approximation for the traveltime of the converted PS1, PS2 and S1S2 waves. By using the effective model parameters for PS1, PS2 and S1S2 waves, the coefficients in the converted-wave traveltime approximation can be represented by the anisotropy parameters defined in the elastic ORT model. From the results in the numercal examples that for converted PS1 and PS2 waves, the proposed rational-form approximation is very accurate regardless of the tested ORT model.

New Aspects

For the simplification purpose, The Taylor-series approximation is applied in the corresponding vertical slowness for three pure-wave modes. By using the effective model parameters for PS1, PS2 and S1S2 waves, the coefficients in the converted-wave traveltime approximation can be represented by the anisotropy parameters defined in the elastic ORT model. The accuracy in the converted-wave traveltime for three ORT models is illustrated in numerical examples. One can see from the results that for converted PS1 and PS2 waves, the proposed rational-form approximation is very accurate regardless of the tested ORT model. For a converted S1S2 wave, due to the existence of cusps, triplications, and shear singularities, the error is relatively larger compared with PS waves.

Summary

In order to approximate the traveltime in an elastic orthorhombic (ORT) medium for converted waves, we define an explicit rational-form approximation for the traveltime of the converted PS1, PS2 and S1S2 waves. For the simplification purpose, The Taylor-series approximation is applied in the corresponding vertical slowness for three pure-wave modes. By using the effective model parameters for PS1, PS2 and S1S2 waves, the coefficients in the converted-wave traveltime approximation can be represented by the anisotropy parameters defined in the elastic ORT model. The accuracy in the converted-wave traveltime for three ORT models is illustrated in numerical examples. One can see from the results that for converted PS1 and PS2 waves, the proposed rational-form approximation is very accurate regardless of the tested ORT model. For a converted S1S2 wave, due to the existence of cusps, triplications, and shear singularities, the error is relatively larger compared with PS waves.

Main Objectives

All the existing CPML methods for the second-order wave equation involve adding auxiliary terms and rewriting the CPML wave equations, which will lead to the increase of calculation, more auxiliary variables, and complicate the implementation more than is necessary. We propose a simple and efficient implementation of CPML for the second-order wave equation system.

New Aspects

1: We solve the original CPML wave equations numerically in the stretched coordinate. 2: We do not introduce auxiliary variables or auxiliary equations for transforming the second-order CPML equation. The proposed method is simple and efficient and fewer variables are introduced than other methods.

Summary

The perfectly matched layer (PML) boundary condition has been widely used as a very effective absorbing boundary condition for seismic wavefield simulations. Convolutional PML (CPML) achieved by using a complex frequency-shifted stretch function was the latest development to further improve PML’s absorption performance for near-grazing angle incident waves as well as for low-frequency incident waves. However, the mathematical theory of most PML methods are derived from the first-order equation system; When implementing the PML technique to second-order wave equations, all the existing methods involve adding auxiliary terms and rewriting the CPML wave equations into the original coordinate, which will lead to the increase of calculation, more auxiliary variables, and complicate the implementation more than is necessary. We propose a new implementation of CPML for the second-order wave equation system. It does not need to introduce auxiliary variables or auxiliary equations for transforming the second-order CPML equations into the original coordinate, and furthermore, the implementation is simple and efficient.

Main Objectives

we incorporate the PML technique into LSM and demonstrate the absorption effect of PML with the help of numerical examples.

New Aspects

very few works have discussed PML implementations for LSM. In this abstract, we incorporate the PML technique into LSM and demonstrate the absorption effect of PML with the help of numerical examples.

Summary

The lattice spring model (LSM) combined with the velocity Verlet algorithm is a newly developed scheme for modeling elastic wave propagation in solid media. Unlike conventional wave equation based schemes, LSM is established on the basis of micro-mechanics of the subsurface media, which enjoys better dynamic characteristics of elastic systems. But LSM is still suffering the boundary reflections and little work has been reported on this topic. The focus of the present study was to develop a special form of absorbing boundary condition based on the perfectly matched layer (PML) concept for LSM. The PML formulation is tested using a homogeneous model and the Marmousi model.The perfectly matched layer concept appears to be very well suited for LSM.

Main Objectives

Lattice Boltzmann Method is introduced to simulate acoustic wave propagation with viscous sponge layers (VSL), where the relaxation time, which is related to the kinematic shear viscosity, varies as different functions and the reflected wave absorbing effects are compared. We demonstrate that the situation when the relaxation time in VSL varies as a cubic function has a better absorbing ability for boundary reflection, which is of vital importance for seismic wavefield simulation by LBM.

New Aspects

We try quadratic, logarithmic, square-root, sigmoid and cubic functions to figure out in viscous sponge layers which relaxation time function can better absorb the reflected wave when using lattice Boltzmann method to simulate acoustic wave propagation.

Summary

The lattice Boltzmann method (LBM), a kind of mesoscopic method based on kinetic theory, is widely used in computational fluid dynamics and acoustic wave. Compared with the traditional wave equation, LBM has many advantages, such as great ease of implementation, near-infinite potential for the parallelization, flexible handling of boundary conditions and simulating complex phenomena easily. In this abstract, LBM is introduced to simulate acoustic wave propagation with viscous sponge layers (VSL), where the relaxation time, which is related to the kinematic shear viscosity, varies as

different functions and the reflected wave absorbing effects are compared. Numerical results demonstrate that the situation when the relaxation time in VSL varies as a cubic function has a better absorbing ability for boundary reflection, which is of vital importance for seismic wavefield simulation by LBM.

Main Objectives

To reconstruct elastic source wavefield with high spacing accuracy, tradition methods still cost more computer storage.

New Aspects

we propose a 3D elastic wavefield reconstruction method based on optimal operator boundary storage strategy, which cost less storage and reconstruct source wavefield with high spacing accuracy same as previous methods. Our algorithm is success in accurately reconstruct elastic wavefield and reducing 80% of the storage.

Summary

Reverse time migration has to store both forward- and backward-propagated wavefield which cost a large of memory, especially in elastic world. Such problems can be solved by wavefield reconstruction method. To reconstruct source wavefield with high spacing accuracy, tradition methods still cost more computer storage. In this study, we propose a 3D elastic wavefield reconstruction method based on optimal operator boundary storage strategy, which cost less storage and reconstruct source wavefield with high spacing accuracy same as previous methods. Our algorithm is success in accurately reconstruct elastic wavefield and reducing 80% of the storage.

Main Objectives

Attenuating spurious reflections from unmatched nodes

New Aspects

Using basis functions with different orders inside the model

Summary

Different families of the numerical methods have been used to solve the wave equation. The discontinuous Galerkin Method allows the wavefield to be discontinuous at the element interface. In this study, the accuracy of the Interior Penalty Discontinuous Galerkin Method (IP-DGM) has been investigated. IP-DGM provides virtually equal accuracy to the Spectral Element Method (SEM). Also, we considered the case where the nodal basis functions have different orders inside the model and consequently the nodes on the adjacent elements with different order basis functions don’t fit together. Using basis functions with different orders inside the model makes the modeling more flexible and can reduce the total number of the nodes and thereby the computational cost. The SEM sees this as the discontinuity and reflections appear from those interfaces while IP-DGM can resolve this problem by choosing an appropriate penalty parameter.

Main Objectives

getting pure and correct separated P and S wavefields from coupled elastic wavefield

New Aspects

zero-order pseudo-Helmholtz decomposition for 3D anisotropic media based on eigenform analysis and numerical approximation in wavenumber domain

Summary

A good P and S wavefields separation can improve the reliability of imaging from reverse-time migration. Traditional separation is based on the divergence and curl operator, but this method is unreasonable in anisotropic media. Though separated pure elastic wavefields show correct amplitude and phase by applying vector separation in wavenumber domain, it is not practical because of computation consuming. For the 3D anisotropic media, a modified method based on eigenform analysis and numerical approximation in wavenumber domain can get pure P and S wavefields. It is computational efficiency using 2D Fourier transformation along two horizontal directions. After doing synthetic test with a 3D VTI model, elastic wavefields can be separated well. This suggest that the zero-order pseudo-Helmholtz decomposition method can do decomposition in 3D media both correctly and effectively.

Main Objectives

To provide a more accurate seismic modeling method, and to improve the quality of imaging result.

New Aspects

We derive the one-way wave equation in the ray center coordinate system of VTI medium.

Summary

The imaging precision of complex geological structure depend on an appropriate and accurate seismic wave propagation simulation methods. The methods of seismic wave simulation include two categories in the current seismic exploration. One is elastic wave equation method. It is relatively more accurate, but it is non-selective in the wave field to be simulated, which is slow in calculation, sensitive to velocity model and has relatively high requirements for computer hardware. The other one is ray method. It decomposes the wave field into a series of rays (or bundles of rays) by using the high-frequency approximation solution of wave equations, and then gets the ray path and ray travel time through ray tracing (It also can simulate the beam wave-field), and thus implements the simulation of wave-field propagation. The high-frequency approximation theory occupies an important position in pre-stack depth migration and tomography. Ray method is intuitive, efficient and flexible, but can cause problems such as caustics, ray shadow areas, etc. Therefore, it is a good practice to combine wave equation and ray theory in view of the different advantages of the above two methods, which is not only efficient, but also takes into account the accuracy.

Main Objectives

We define the Rational Form (RF) approximations for P-wave traveltime and relative geometrical spreading in elastic ORT model. To facilitate the coefficients derivation in these approximation forms, the Taylor series (expansion in offsets) in the vertical P-wave slowness measured at zero-offset is applied. The same approximation forms computed in the acoustic ORT model are also derived for the comparison.

New Aspects

In the numerical tests, three ORT models with the parameters obtained from the real data are used to test the accuracy of each approximation. The numerical examples yield the results that, apart from the error along the y-axis in the ORT model 2 for the relative geometrical spreading, the RF approximations are all very accurate for all tested models in both traveltime and relative geometrical spreading that can be performed for the practical use.

Summary

We define the Rational Form (RF) approximations for P-wave traveltime and relative geometrical spreading in elastic ORT model. To facilitate the coefficients derivation in these approximation forms, the Taylor series (expansion in offsets) in the vertical P-wave slowness measured at zero-offset is applied. The same approximation forms computed in the acoustic ORT model are also derived for the comparison. In the numerical tests, three ORT models with the parameters obtained from the real data are used to test the accuracy of each approximation. The numerical examples yield the results that, apart from the error along the y-axis in the ORT model 2 for the relative geometrical spreading, the RF approximations are all very accurate for all tested models in both traveltime and relative geometrical spreading that can be performed for the practical use.

Main Objectives

Obtain accurate reflection traveltime and reflection angle through reflection raytracing.

New Aspects

Using the Snell’s law in multi-stage fast marching method to do reflection raytracing.

Summary

Reflection traveltime tomography needs accurate calculation of reflection traveltime and reflection angle. The multi-stage fast marching method (MFMM) is widely used in reflection raytracing, which performed well in reflection traveltime calculation, but the reflection angle has pretty large error since the reflected ray does not follow the Snell’s law strictly nearby the reflector. Here, we developed a modified multi-stage fast marching method that use Snell’s law to compute the reflection traveltime of grid points just above the reflector and the corresponding reflection point at the reflector to improve the accuracy of reflection angle. The numerical examples show the reflection angle improved a large scale.

Main Objectives

Designing more efficient forward modeller for frequency-domain anisotropic full waveform inversion

New Aspects

Reduction of numerical error and improvement of computational cost by a considerable margin

Summary

We present a new finite difference scheme for the frequency-domain quasi-acoustic wave equations for vertical transversely isotropic (VTI) media. Compared to the optimized scheme, the proposed one enhances the overall performance and accuracy of the frequency domain solution as it significantly reduces the numerical dispersion error without any amount of extra cost. It improves upon the discretization rule of 4 grid points per minimum wavelength and reduces it to 2.5. The numerical examples of wave propagation in homogeneous and highly heterogenous models prove the validity and precision of the proposed scheme.

Main Objectives

Compared with the VTI model, the research of the triplication in transversely isotropic medium with a tilted symmetry axis (TTI) is still keeping blank. In order to analyze the triplication for the converted wave in the TTI model, we examine the traveltime of the triplication from the curvature of averaged P and S wave slowness.

New Aspects

By applying the rotation angle, we examine the existence of the traveltime triplication in a homogeneous TTI model by the averaged slowness individually from the P and S waves. We find that the triplication can be encountered in the TTI model and affected by the model parameters and the rotation angle. In addition to the influence of model parameters (strong anisotropy), we can tell from the tested examples that the triplication in TTI model depends on the concave shape of the S wave that is considered for the converted wave calculation, where is the concave shape occurs and what degree of concavity is.

Summary

In seismic data processing, serious problems could be caused by the existence of triplication and need to be treated properly for tomography and other inversion methods. In order to analyze the triplication for the converted wave in the TTI model, we examine the traveltime of the triplication from the curvature of averaged P and S wave slowness. Three models are defined and tested in the numerical examples to illustrate the behavior of the TTI model for the triplicated traveltime with the change of the rotation angle. Since the orientation of an interface is related to the orientation of the symmetry axis, the triplicated traveltime is encountered for the converted wave in the TTI model assuming interfaces to be planar and horizontal. The triplicated region is influenced by the place and level of the concave curvature of the P and S wave slowness.

Main Objectives

Find a solution to the corner problem that hampers the generalization of numerically exact non-reflecting boundary conditions to more than one space dimension.

New Aspects

A workaround is proposed that restores translation invariance, so that exact non-reflecting boundary conditions are not only feasible in waveguide problems, but also in seismic simulations, albeit only on two opposing sides. This can be useful in the presence of a shallow sea or when simulating land seismic data.

Summary

Recently introduced non-reflecting boundary conditions are numerically exact: the solution on a given domain is the same as a subset of one on an enlarged domain where boundary reflections do not have time to reach the original domain. In 1D with second- or higher-order finite differences, a recurrence relation based on translation invariance provides the boundary conditions. In 2D or 3D, a recurrence relation was only found for a non-reflecting boundary on one or two opposing sides of the domain and zero Dirichlet or Neumann boundaries elsewhere. Otherwise, corners cause translation invariance to be lost.

The proposed workaround restores translation invariance with classic, approximately non-reflecting boundary conditions on the other sides. As a proof of principle, the method is applied to the 2-D constant-density acoustic wave equation, discretized on a rectangular domain with a second-order finite-difference scheme, first-order Enquist-Majda boundary conditions as approximate ones, and numerically exact boundary conditions in the horizontal direction. The method is computationally costly but has the advantage that it can be reused on a sequence of problems as long as the time step and the sound speed values next to the boundary are kept fixed.

Main Objectives

solution of acoustic wave equation in ω-k domain

New Aspects

dispersion-free numerical approach, efficient solution via iterative solvers

Summary

In this paper we present solution of the acoustic wave equation in ω-k domain. In ω-k domain we can evaluate temporal and spatial differentials accurately therefore the numerical solution is free of numerical dispersion. We present theory and numerical implantation of this approach and discuss about efficient solution of dense system of equations using iterative solvers. 1D and 2D examples are presented to show the performance of the method in terms of accuracy and computational costs.

Main Objectives

We proposed a scheme to estimate the optimal FD parameters, i.e. the number of samples per shortest wavelength and the spatial order of FD operators, to achieve the minimum computational cost, for a given numerical error threshold.

New Aspects

To achieve this objective, (1) a precise way is derived to introduce spatial dispersion into the reference wavefield; (2) a scheme is proposed to find the optimal FD parameters.

Summary

Finite-difference modelling is widely used for seismic wavefield modelling. A finer grid spacing or a high-order finite-difference operator is normally used to suppress spatial dispersion. However, both ways increase the computational cost significantly. We proposed a scheme to estimate the optimal FD parameters, i.e. the number of samples per shortest wavelength and the spatial order of FD operators. Firstly, we derived an equation that can introduce spatial dispersion precisely into the reference wavefield for various scenarios of grid spacing and FD orders. secondly, we measure the RMS error between the reference wavefield and dispersed wavefield. We then can compute the number of samples per shortest wavelength for a specified FD order, given a spatial numerical error. Thus, we can further compute the total multiplication and addition computational cost against the number of samples per shortest wavelength. Finally, the minimum point in the curve of the computational cost against the samples per shortest wavelength indicates the optimal FD parameters. For a Ricker wavelet, we recommend 3.125 grid points per shortest wavelength, which is measured at 2.5 times the peak frequency, and the 16th-order spatial FD operator to achieve an RMS error less than 1%.

Main Objectives

This work provides the robust mathematical approach for taking into account VTI-anisotropy and fractured structure of the upper part of the geological section.

New Aspects

A novel approach for seismic wave simulations in fractured VTI media with the grid-characteristic numerical method was proposed.

Summary

Precise numerical modelling of the wave propagation in geological media is crucially important for seismic inversion and migration techniques. Due to the sedimentation procedure the upper part of geological sections is anisotropic. For this reason, the velocity of the wave propagation depends on the propagation direction. One of the major cases is the vertical transverse isotropic (VTI) media. Different approaches were proposed for precise simulation of wavefields in such media.

In this work we propose a novel approach for simulation of the dynamic behavior of the VTI fractured medium. It relies on the grid-characteristic numerical method on rectangular grids. It continues our activity of extending this method for different types of rheology. The method was successfully implemented as a part of our in-house software RECT and applied to the improved Marmousi2 model.

Main Objectives

fluid identification and saturation calculation on low contrast reservoirs

New Aspects