Find E-Lectures



Seismic Acquisition

A Novel Source-Over-Cable Solution to Address the Barents Sea Imaging Challenges

E-Lecture by Per Eivind Dhelie


This EAGE E-Lecture "A Novel Source-Over-Cable Solution to Address the Barents Sea Imaging Challenges " presents a full overview of the TopSeis source-over-cable solution. The challenging geologic setting in the Barents Sea is discussed and legacy seismic data is scrutinized in an effort to understand the fundamental challenges related to improved seismic imaging in the Barents Sea. The acquisition design behind the new technology is presented in detail, from the wide-towed small triple source setup to the split-spread deep towed streamers. Deblending of triple sources is presented as well as a detailed overview of the challenging demultiple problem. A number of seismic image comparisons are also presented illustrating the benefit achieved with the new source-over-cable solution. The final part of the talk looks into the future and proposes possible solutions to obtaining similar acquisition setups, but using only a single vessel towing both the sources and the streamers.

Making the transition from discrete shot records to continuous wavefields – Data examples

E-Lecture by Tilman Kluver


In this presentation, a novel seismic acquisition and processing methodology is demonstrated. The method retrieves the response of the earth from data acquired with continuous source and receiver wavefields. The ideal source wavefield would be band-limited white noise. Ways of generating a source wavefield which approaches the properties of white noise using existing air-gun equipment will be discussed using real data examples. Seismic data acquired by triggering individual air-guns with short randomized time intervals in a near continuous fashion will be compared to seismic data acquired with large source arrays triggered every 25m. The continuous source wavefield improves the sampling of common-receiver gathers compared to conventional acquisition methods. Spreading the source energy out in time results in reduced peak sound pressure levels with the new method.

METIS, a field-proof innovative method to revolutionize onshore seismic acquisition

E-Lecture by Bruno Pagliaccia


METIS (Multiphysics Exploration Technology Integrated System) is an R&D project aimed to make seismic acquisition 4.0 a reality. From logistics to real-time data QC, the entire process has been redesigned with an holistic approach. An hybrid airship to replace helicopter, wireless/real-time/biodegradable sensors to automatize data acquisition and a Command & Control center to operate the system from the base camp. All these elements will allow low-cost, safe and efficient seismic operations in hard-to-reach places like the tropical rainforest of Papua New Guinea (PNG). End of 2017, the main technological bricks have been tested on PRL 15 concession, located onshore PNG, with a demonstration pilot where more than 60 DARTs were successfully dropped from a drone, through the canopy, to the ground with a good coupling. After few shots made with a mud-gun, real-time data acquisition has been validated in one of the most challenging environment to acquire seismic data. This first step is paving the way to the next pilot where a full 3D seismic volume will be acquired after thousands of sensors dropped by an autonomous swarm of drones.

Virtual Seismology: monitoring the subsurface with virtual sources and receivers

E-Lecture by Kees Wapenaar


This E-Lecture introduces virtual seismology: monitoring the subsurface with virtual sources and receivers, which is a new methodology to create virtual sources and virtual receivers in the subsurface from reflection measurements at the earth's surface. Unlike in seismic interferometry, no physical instrument (receiver or source) is needed at the position of the virtual source or receiver. Moreover, no detailed knowledge of the subsurface parameters and structures is required: a smooth velocity model suffices. Yet, the responses to the virtual sources, observed by the virtual receivers, fully account for multiple scattering. This method is not only useful for reflection imaging but is has also large potential for monitoring induced seismicity, characterizing the source properties and forecasting the response to potential future induced earthquakes. This will be demonstrated with numerical models and preliminary real-data results.

Making the transition from discrete shot records to continuous wavefields - Methodology

E-Lecture by Stian Hegna

In this presentation a novel seismic method is introduced. The method utilizes continuous wavefields on both the source side and on the receiver side. It requires continuous seismic recording, and a continuous or near continuous source wavefield approaching the properties of band-limited white noise. The main operational benefits are reduced environmental impact of marine seismic sources, and potentially improved acquisition efficiency. The main geophysical benefit is improved spatial sampling on the source side both inline and potentially cross-line.

Shooting over the Spread

E-Lecture by Vetle Vinje

In this lecture, I present the motivation, the evolution and the benefits of a new marine acquisition concept developed in close cooperation between Lundin and CGG over the last few years. We call the concept TopSeis. It addresses the lack of near-offset data recorded in conventional towed-streamer acquisition by enabling the recording of short- and zero-offset data with the seismic sources located above the streamers. In addition, TopSeis significantly increases the illumination density (number of times a specific depth point is recorded) for both shallow and deep targets. This benefits imaging, AVO and inversion as shown by several synthetic and real examples.

Wave Equation Receiver Deghosting

E-Lecture by Craig Beasley

Current solutions to receiver deghosting of marine seismic data generally involve making complementary measurements of the wavefield or, alternatively, involve estimation of data not recorded due to ghost interference. In this talk, we introduce a new approach that rigorously deghosts single measurement data (p-wave only, for example) without relying on estimation of missing data, which was previously thought to be mathematically impossible. We use the wave equation to compute directly the up and downgoing wavefields between the receiver and the surface. This migration-like approach is possible given that the upcoming wavefield is causal with respect to the downgoing wavefield, a good assumption if certain types of noise such as the direct arrivals are removed from the recorded wavefield. Moreover, our method exposes the underlying physics of the problem and thereby naturally allows for complexities such as variable water velocity, sea surface, complex surface reflection conditions and arbitrary receiver configurations.

Increasing Acquisition Efficiency by Acquisition of Data During Turns, Using a Multimeasurement Streamer

E-Lecture by Sneha Biswas

In this E-Lecture Sneha Biswas talks about the ability of multi-measurement streamers in conjunction with the acquisition of data during turns for an efficient marine seismic acquisition. Acquiring data during turns is an approach to improve productivity of marine seismic surveys. Data results post noise attenuation compares the quality of this data against straight line acquisition and demonstrates the usability of this turn data.

Dip Correction for Convolutional Modelling and Elastic Inversion

E-Lecture by Adam Cherrett

Convolutional seismic modelling is a very convenient and fast technique, widely used in geoscience. It is usually applied in the vertical dimension only, which is one of its shortcomings, leading to inaccurate modelling and restricting its applications. This talk concerns a simple pseudo-3D extension to the convolutional model which improves modelling of dipping events and the lateral resolution of seismic data. This same 3D operator, when used to invert images for elastic properties, can reduce artefacts and improve resolution.

High Resolution 3D Tunnel Seismic Reflection at Olkiluoto, Finland

E-Lecture by Calin Cosma

Calin Cosma (Vibrometic) describes the high-resolution seismic techniques, used for rock characterization ahead and around the access tunnel of the spent nuclear fuel disposal facility, currently being built at Olkiluoto, Finland. Long fractures and deformation zones were identified and mapped up to hundreds of meters from the tunnel. The methodological novelty has been the introduction of the 3D Image Point migration, which proved to be very effective for the imaging of low-aperture rock features with various orientations. The results are compared with the current site model and observations in tunnels and boreholes.

Triple-Source Simultaneous Shooting, a Future for Higher Density Seismic?

E-Lecture by Jan Langhammer

Jan Langhammer’s E-Lecture has the following subject: reducing the crossline bin-size in marine seismic streamer exploration can be performed by using more sources without compromising the efficiency of the operation. Triple-source in marine seismic streamer acquisition has been tested in the past, but with limited commercial success compared to dual-source acquisition. With the introduction of new and better low noise streamers, in addition to the ability to record and deblend simultaneous source data, it is time to revisit the use of triple-sources in marine seismic exploration for decreased crossline bin-size leading to better spatial resolution. A triple-source configuration can find its application in shallow and deeper water areas for imaging of targets where reduced crossline spacing and higher fold may be required.

Well Tie: Principles & New Advancements for Broadband Seismic Data

E-Lecture by Ehsan Naeini

In this presentation, Naeini discusses a quantitative approach to do well tie and to QC the outcome. This covers the basic principles all the way to the latest developments for broadband seismic data. He shows some synthetic and real data examples and highlights the impact of time lag, phase, bandwidth and broadband wavelets for broadband seismic.

Impact of Acquisition Geometry on AVO/AVOA Attributes Quality

E-Lecture by Amine Ourabah

With hydrocarbon exploration moving to more complex plays, understanding the resolution limits of seismic imaging and attributes is more important than ever, especially with the availability of new acquisition methods like ISS® and DSSS allowing acquisition of very high density surveys at a reduced cost. Questions as: ‘What are the benefits of acquiring dense surveys, how do they impact the frequencies, AVO/AVAO attributes’ can be difficult to answer and yet have a significant impact on cost, quality and decision making. Attempts to answer some of these questions using the theory of resolution or synthetic modelling are available. However, we rarely have the opportunity to see the effect of decimating a real dataset on a full range of conventional and azimuthal attributes. In this study we are decimating the Risha high density survey to 20 different geometries. Each one of these geometries were then separately processed and an extensive list of pre-stack conventional and azimuthal attributes were extracted on them. By analysing and comparing these products, we attempt to answer the above questions, using the densest geometry as a benchmark for quality.

Wide-Azimuth Acquisition With Radial Domain Interpolation for Fluvial Morphology Interpretation

E-Lecture by Anastasia Poole

The topic of Anastasia Poole’s E-Lecture is Wide-Azimuth Acquisition with Radial Domain Interpolation for Fluvial Morphology Interpretation. During this presentation she will demonstrate how an integrated approach helps to unlock the fluvial morphology reservoir potential using seismic. Poole will cover the following topics: reservoir geology and economics evaluation, seismic survey design, seismic acquisition, processing and she will finish off with examples of inversion and interpretation. Key technologies discussed in this presentation are: point-source point-receiver full azimuth/broad band acquisition, azimuth aware processing with radial domain interpolation to enable advanced seismic interpretation and rock properties extraction.

Seismic Multiple Removal Techniques: Past, Present and Future

E-Lecture by Eric Verschuur

Eric Verschuur offers a short version of his EET 1 Tour, explaining how multiple reflections influence seismic measurements. An overview is provided of the techniques that were developed to remove them, from the (high-resolution) parabolic Radon transform to the full 3D SRME technique of today. He also discusses the extension to internal multiples and, finally, looks into the future, where multiples could be used as signal in imaging procedures.

Learn more about the EET programme.

Data-Driven Green's Function Retrieval From Reflection Data

E-Lecture by Kees Wapenaar

This E-Lecture introduces new reflection imaging methodology, which deals with internal multiples. First, an iterative scheme retrieves focusing functions from reflection data at the surface, which focus onto virtual sources in the subsurface. Next, the responses to these virtual sources, the Green’s functions, follow from the focusing functions and the reflection data. Once these Green’s functions are obtained, reflection imaging can be carried out by which the primaries and internal multiples are mapped to their correct positions, with correct reflection amplitudes and without the occurrence of false images at wrong positions.

Seismic Processing

Notional ghosts

E-Lecture by Gary Hampson

In this E-Lecture, Gary Hampson, Principal Research Geophysicist at DownUnder Geosolutions, will explain an extension to the airgun Notional Source concept called Notional Ghosts. These are conceptual devices that describe the source ghost in terms of virtual monopole sources each of which has its own signature. As with notional sources, although notional ghosts are never directly observed, they are derived by inversion of the pressure wavefield recorded by the near field hydrophones. In this lecture Gary explains the motivation for this new concept, its theoretical basis and shows real examples.

Read the Abstract
Theory for Marchenko Imaging of Marine Seismic Data with Free Surface Multiple Elimination

E-Lecture by Evert Slob

This E-lecture: Theory for Marchenko Imaging of Marine Seismic Data with Free Surface Multiple Elimination; introduces the theory for reflector imaging from marine seismic data that include ghosts, free surface and internal multiple reflections. First, we look at building the necessary redatuming operators from the data when the free surface would not be present. The redatuming operators are found as the focusing functions that are not influenced by the presence or absence of the free surface. Next, a matrix equation is found from which the focusing functions are computed using the measured data decomposition into up- and downgoing waves. No other processing steps are required. The scheme does not need information about the source time signature or the actual character of the free surface. A macro-model is necessary to construct an estimate of the direct part of the focusing function. The ghost, free surface and internal multiple reflections do not create false images and an artefact free image is obtained.

A 2nd-order Adjoint Truncated Newton Approach to Time-domain Multiparameter Full Waveform Inversion in Viscoacoustic Medium

E-Lecture by Pegliang Yang

This E-Lecture introduces a 2nd-order Adjoint Truncated Newton Approach to Time-domain Multiparameter Full Waveform Inversion in Viscoacoustic Medium. A truncated Newton method for time-domain full waveform inversion in visco-acoustic medium is developed using 2nd-order adjoint state formulation. Time-domain gradient and Hessian-vector product are built by recomputing the incident and adjoint wavefields.  After the design of an efficient preconditioner, the importance of the inverse Hessian for mitigating inter-parameter trade-off  is validated on a toy example. A realistic 2D synthetic mimicking North-sea real data application demonstrates that considering Hessian influence significantly improves the multi-parameter reconstruction, for a reasonable increase of the computational cost compared to standard quasi-Newton method.

Reducing project turnaround by optimizing the model building workflow using full-waveform inversion and reflection tomography: A North Sea case study

E-Lecture by Shruti Gupta

Two of the key challenges for velocity model building in the North Sea are a heterogeneous overburden (for example, the presence of fluvial and sub-glacial channels) and the presence of strong velocity contrasts (most often introduced by a chalk layer). Two model building techniques have been developed in recent years which address these challenges. Full-waveform inversion (FWI) has been shown to be highly effective in resolving overburden heterogeneity and multi-layer tomography allows us to preserve sharp contrasts in our velocity models. In this EAGE E-Lecture, we present how the combination of these two techniques provides an effective and efficient model building workflow for the North Sea environment.

Read now the paper

Understanding Spectral Decomposition

E-Lecture by Victor Aarre

Spectral Decomposition is a standard tool in the seismic interpreters toolbox. It is also a complex tool to use. Many implementations are provided as “black box” technology by vendors, and presented results are often spectacular. The scope of this lecture is hence to give the viewers a thorough scientific understanding of what Spectral Decomposition is, outline how it works, and explain what it can and cannot do. It has been an aim to keep the mathematical terminology simple, and properly explain the terms as they are introduced, such that the content becomes more accessible for an audience outside the hard-core mathematical community.

Using Digital Music Technology for Geophysical Data Analysis and Interpretation

E-Lecture by Paolo Dell'Aversana

In this E-Lecture Dell’Aversana discusses how digital music technology can support geophysical data analysis and interpretation: we apply accurate mathematical transforms for extracting the spectrograms of geophysical data, such as seismic, electromagnetic and gravity data, well logs and so forth. Then, the spectrograms are translated into sounds in standard musical formats and are imported into a software platform of digital music. We apply two complementary workflows to our converted “geo-musical” data: first, we use musical pattern recognition algorithms for automatic data mining and classification; second, we use interactive audio-video display in selected portions of the data set for allowing multisensory perception and advanced cognition. During the E-Lecture we introduce the fundamentals of the approach. Finally we show some applications to real data.

A New Take On FWI: Wavefield-Reconstruction Inversion

E-Lecture by Felix Herrmann

Full-waveform inversion relies on accurate starting models to avoid local minima. We remove this reliance by solving an augmented wave equation, designed to fit the data as well as the physics, in combination with imposing constraints. As a result we end up with an inversion scheme that produces good results in situations where conventional full-waveform inversion is known to fail.

Did you like this E-Lecture? Learn here how to request a webinar with Felix Herrmann.

Maximising the Benefits of Full Waveform Inversion

E-Lecture by Fabio Mancini

Full waveform inversion (FWI) has recently emerged as one of the most exciting new techniques in the seismic industry thanks to its potential to deliver very detailed velocity models. It is also a paradigm shift in the way we process seismic data since it allows us to start the velocity model building on raw shots, with little, if any, need for pre-processing. In this e-lecture, Fabio Mancini will show examples on how these characteristics of FWI can be leveraged to maximising its impact in the exploration and development cycle. In particular he will focus on reducing the turnaround time of seismic processing projects and on the use of the high frequency velocity models as an attribute for direct interpretation.

Efficient 3D Frequency-Domain FWI of OBC Data

E-Lecture by Stéphane Operto

Wide-azimuth long-offset OBC/OBN surveys provide a suitable framework to perform computationally-efficient 3D frequency-domain full waveform inversion (FWI) with a few discrete frequencies. Frequency-domain seismic modeling is performed efficiently with moderate computational resources for a large number of seismic sources with the sparse multifrontal direct solver MUMPS. The relevance and the computational efficiency of the frequency-domain FWI performed in the visco-acoustic VTI approximation is shown with a real 3D OBC case study from the North Sea. The subsurface models built by FWI show a dramatic resolution improvement relative to the initial model built by reflection traveltime tomography down to the base cretaceous reflector below the reservoir level. The relevance of the FWI model is assessed by frequency-domain and time-domain seismic modellings and source wavelet estimation, which might reveal the footprint of attenuation on the imaging results.

Multi-Azimuth Streamer Seismic in The Nile Delta: Depth Imaging

E-Lecture by Walter Rietveld

BP has been acquiring Multi-Azimuth (MAZ) Streamer Seismic in the Nile Delta since 2003. The initial results showed that MAZ data greatly improves general image quality, signal-to-noise ratio and lateral resolution, and suppresses diffracted multiples effectively. E-Lecture part 1 discusses the reasons why MAZ was selected and discusses the first results after time processing and PSTM imaging. E-Lecture part 2 discusses the workflow that combines the MAZ data in a multi-parameter reflection tomography approach to build a detailed velocity model that shows variations which correspond very well to the geology interpreted from the seismic, and which yields improved PSDM images. The results and learnings from both presentations are relevant and applicable to wide-azimuth acquisition and processing in general.

An Introduction to Migration without a Single Equation

E-Lecture by Claudio Strobbia

This E-Lecture is a primer on seismic migration: it introduces the principles of imaging, the assumptions and limitation, the artefacts. It discusses also the differences between time and depth migration, and examines the requirements of the simple time migration. Claudio Strobbia here explains these concepts in an easy and understandable way, with lots of illustrations and animations, and without a single equation. A simple and entertaining initiation to seismic imaging.

Watch this E-Lecture on Youku

Quantitative 4D Analysis Using Business Analytics

E-Lecture by Mark Thompson

In Geophysical Reservoir Monitoring (GRM) seismic data, reservoir data, production data, and borehole data must be integrated to get an understanding of what is going on in the reservoir over time. The information used in GRM is a typical example of Big Data, as it can be characterized by volume, velocity, and variety. The learnings of Big Data from other industries have been leveraged and applied to quantitative analysis in GRM.

Did you like this video? You can request a webinar with Mark Thompson to learn more!

Least Squares Reverse Time Migration

E-Lecture by Bin Wang

Bin Wang (TGS) briefly introduces a new imaging algorithm called Least Squares RTM (LSRTM). LSRTM is an inversion-based imaging algorithm, which aims to derive a better reflectivity image. Similar to Full Waveform Inversion, LSRTM is trying to minimize the data residual between a field-recorded seismogram and a synthetic modeled seismogram. Like Reverse Time Migration (RTM), LSRTM is based on the two-way wave equation. Compared to a regular RTM, LSRTM has the following benefits: 1) it gives a high-resolution and broadband seismic image, especially the low frequency end; 2) it reduces migration artifacts due to acquisition foot print and non-uniform illumination; 3) the migration image has more balanced amplitudes, as LSRTM is towards true amplitude imaging.

Water Velocity and Tide Measurement in Marine Seismic Acquisition

E-Lecture by Kanglin Wang

Kanglin Wang (Shell) discusses how to improve time-lapse (4D) seismic with novel water statics processing involving a seafloor device called PIES. A well known challenge in marine 4D processing is the non-repeatability introduced by water velocity and tidal variations during a survey and between different surveys. PIES provides direct measurement and continuous monitoring of these water properties and greatly reduces the uncertainties in timing and positioning of seismic data.

Surface Imaging

Depth Domain Inversion: a least-squares migration approach to quantitative interpretation

E-Lecture by Claudio Leone

QI workflows assume that seismic amplitudes are only linked to contrasts in rock properties, and that other wave propagation effects such as illumination, absorption, etc. have been addressed during seismic data processing and imaging. This assumption is often not met. The consequence is that 1D wavelets can’t adequately relate seismic amplitudes to reflectivity contrasts, and conventional time-domain inversion approaches inevitably struggle to accurately estimate the elastic rock properties. The effects of irregular illumination can be modelled by Point Spread Functions (PSFs), and removed from the seismic image with an inversion directly in the depth domain (Fletcher et at., 2012). This technique, called Depth Domain Inversion (DDI), improves imaging and inversion results by correcting for the effects of irregular illumination caused by the geological structure and overburden velocity variations. A North Sea case study is presented where DDI is enhances amplitude fidelity and resolution beneath cemented sand injectites.

Viscoelastic Full Waveform Inversion; a Symmetrization Strategy

E-Lecture by Gabriel Fabien-Ouellet

Seismic propagation may exhibit very complex physics, especially on land where elastic and attenuation are dominating effects. In this context, full waveform inversion (FWI) must move from the acoustic approximation to the more challenging viscoelastic wave equation.
In this video, Gabriel Fabien-Ouellet explains the theory behind viscoelastic full waveform inversion. Using a symmetrization transformation of the problem, he shows how the adjoint state method can be modified to simplify the numerical implementation of viscoelastic FWI, and then presents the main challenges of inverting for attenuation.

FWI with optimal transport: a 3D implementation and an application on a field dataset

E-Lecture by Jeremie Messud

We present the application to a 3D real dataset of full waveform inversion (FWI) with optimal transport (OT) using the Kantorovich-Rubinstein (KR) distance as proposed by Métivier et al. (2016). This approach involves an efficient numerical implementation for OT in time and space directions, allowing the lateral coherency of the traces to be taken into account; this has an important impact on the quality of the results. The approach also exhibits a reduced sensitivity to local minima compared to least squares (LSQ) misfit. Moreover the iterative method used for the computation of the KR distance allows the production of a set of intermediary solutions that span progressively from LSQ to OT. We recall the main components of the approach and present its numerical implementation in 3D. We show the improvement of the results compared to LSQ FWI on real datasets.

Structural Dip, Migration and Quantitative Interpretation

E-Lecture by Ehsan Z. Naeini

It is relatively well known that migration stretches the waveforms of dipping reflections. It is less well known how this impacts quantitative interpretation of the data where, at the heart of such workflows, the seismic is tied to the well and subsequently inverted to obtain acoustic and/or elastic properties. It is shown here how migration stretch distorts well tie, coloured inversion, and model-based inversion. A post migration first-order zero-offset deconvolution is introduced to remedy this drawback, and the result is demonstrated on a synthetic and a reverse time migrated image. Finally, it is argued that full-waveform inversion would inherently handle the dips without any need for stretch correction. (references: Cherrett 2013 and Naeini 2018).

‘Online’ Marchenko Focusing and Target-oriented Modeling

E-Lecture by Patrick Elison

With the Marchenko method one can derive an incident wavefield (a so-called focusing function) that focuses at an arbitrary location inside a medium and subsequently continues as a diverging wavefield. This is often denoted as creating "virtual sources” inside a medium. In this talk you will learn how such a focusing function can be obtained by iteratively sending a wavefield into a medium and simultaneously recording its reflection response. This requires control over densely sampled sources and receivers at the surface of an object in combination with an a-priory known smooth velocity model. Further, this talk teaches you how the result of the Marchenko scheme can be used for target-oriented seismic modeling, which is of use for localized inversion and time-lapse reservoir monitoring.

A misfit function based on an optimal transport distance for FWI

E-Lecture by Ludovic Métivier

"In the field of seismic imaging, full waveform inversion has become one of the key techniques to provide high resolution quantitative estimation of subsurface mechanical parameters such as wave velocities, density, attenuation, or anisotropy parameters. Recent success stories have led both the academy and industry to investigate this technique. Its applicability is however still limited for exploration targets, mostly because of the lack of low frequencies in the data which makes it difficult to correctly interpret the kinematic attributes of the wavefield. Mitigating this issue is the motivation of the work proposed in this presentation. The strategy we present uses a new distance function between observed and calculated data, based on the optimal transport theory. This new distance provides a more convex misfit function: we show on several synthetic examples how this could help to make full waveform inversion a more robust and powerful tool."

The effect of shale activation on the 4D Seismic Interpretation of a UKCS Field

E-Lecture by Ricardo Rangel

Seismic Facies Classification Using Multimedia and Machine Learning

E-Lecture by Paolo Dell'Aversana

Paolo Dell’Aversana*1, Gianluca Gabbriellini1, Gabriela Carrasquero1, Alfonso Amendola1, Alfonso Iunio Marini1 1Eni S.p.A. Upstream and Technical Services

As it happens for natural intelligence, also artificial intelligence can be improved if it is able to analyze and interpret multimodal information. In this e-lecture, we show that training a computer with multimodal data, increases the possibility of seismic facies classification through machine learning algorithms. We use a new class of attributes in geophysics, representing ‘musical’ properties implicitly included in the data. Together with traditional geophysical attributes, our multimedia machine learning system uses also melodic, harmonic and rhythmic patterns extracted from the data. All these new features show high classification power. In particular, they allow distinguishing low-gas from high-gas saturated sands, as showed in the real example discussed in this lecture.

An iterative workflow for facies modeling on the Alvheim Field, Norwegian Continental Shelf

E-Lecture by Andor Hjellbakk

In this E-Lecture, Andor Hjellbakk will take you through some simple but effective principles for building a deterministic facies model for the deep-marine deposited Tertiary reservoir of the Aker BP operated Alvheim Field. The facies modeling workflow is combining state-of-the-art modeling techniques with input data from 100 km of penetrated reservoir, spectral decomposition (seismic) data and impedance data. Further, the datasets from the subsurface is linked to observations from outcrop data to make sure realistic depositional trends and dimensional data are used in the modeling process.

Read the paper
Q-Compensation Through Depth Domain Inversion

E-Lecture by Maud Cavalca

Compensation for seismic attenuation inherent to the dissipative and dispersive nature of the Earth remains a challenge in seismic imaging of complex media. In this E-Lecture, Maud Cavalca briefly reviews various deterministic approaches that tackle the problem, and focuses on a scheme that compensates for Q within depth domain inversion. She shows that this type of approach constitutes a viable and efficient alternative to ‘Q-migration’ techniques that attempt at compensating for Q during migration.

Passive Seismic Surface-Wave Interferometry for Reservoir-Scale Imaging

E-Lecture by Sjoerd de Ridder

In this E-Lecture Sjoerd de Ridder provides an introduction to passive seismic interferometry using surface waves with applications to reservoir geophysics. First, he addresses the nature of seismic noise typically recorded at low frequencies by marine seismic arrays. Then, he explaines the basic principle of passive seismic interferometry and shows virtual seismic sources obtained by cross-correlating microseism noise. Lastly, he demonstrates example applications of imaging anisotropy and time-lapse velocity changes using virtual seismic sources.

Automatic Gas Pockets Detection by High Resolution Volumetric Q-Tomography Using Accureate Frequency Peak Estimation

E-Lecture by Fatiha Gamar

In this E-Lecture, Fatiha Gamar explains how a high-resolution volumetric Q tomography can be developed to attain an accurate volumetric estimation of the attenuation model. A key component of the workflow is the estimation of effective attenuation in the pre-stack data domain through accurate picking of the frequency peak. Finally she presents a case study where this approach has been used to reveal shallow gas pockets and compensate for absorption in the migration.

Reservoir Elastic Parameters Estimation from Surface Seismic Data using JMI-res: A Full-Wavefield Approach

E-Lecture by Aayush Garg

Traditionally, reservoir elastic parameters inversion suffers from the unaccounted overburden multiple scattering and transmission imprint in the local input data used for the target-oriented inversion. In this E-lecture, we present a full-wavefield approach, called reservoir-oriented joint migration inversion (JMI-res), to estimate the high-resolution reservoir elastic parameters from surface seismic data. In the JMI-res, we reconstruct the fully redatumed virtual source-receiver data (local impulse responses) within the earth subsurface, while correctly accounting for the overburden internal multiples and transmission losses. Then, we apply a localized elastic full waveform inversion on the local impulse responses to get the elastic parameters. We show that JMI-res provides much more reliable local target impulse responses, thus yielding high-resolution elastic parameters, compared to a standard redatuming procedure based on time reversal of data. Moreover, this approach avoids the need to go full elastic for the whole subsurface, as within JMI-res elastic full waveform inversion is only restricted to the reservoir target domain.

Wave-Equation-Based AVO Inversion for High-Resolution Reservoir Characterisation

E-Lecture by Dries Gisolf

Wave-equation-based AVO inversion aims at getting high-resolution property models from seismic data. These property models should allow the building of reservoir models and the estimation of reserves. Wave-equation-based AVO inversion fully utilises the non-linear relationship between seismic data and the properties to be inverted for, leading to a wider spatial bandwidth than could be expected on the basis of the temporal bandwidth of the data. Also, it inverts directly for elastic parameters, giving a higher sensitivity to pore-fill than conventional impedances.

Watch this E-Lecture on Youku

The following courses by Dries Gisolf are available via EAGE:

Full-Waveform Inversion for High-Resolution Reservoir Characterization
The Principles of Quantitative Acoustical Imaging

Consult our Calendar or Events for the next deliveries or request an in-house training proposal!
Barents Sea Case Study: Integrated Depth Velocity Model Building

E-Lecture by Marit Guttormsen

In this E-Lecture, Marit Guttormsen presents the results of a 3D OBS dataset acquired in the Barents Sea. The result was a significant uplift in the structural image by imaging the Pressure energy converted to Shear energy (PS). In order to achieve the image an integrated Velocity Model Building (VMB) flow, using several datasets and a variety of techniques, was implemented. The lecture attempts to summarize the key steps in such an integrated VMB workflow.

Accurate Modelling and Inversion With Structural Models Using The Finite-Difference Method

E-Lecture by James Hobro

In this E-Lecture James Hobro (Schlumberger Gould Research) describes a new approach to representing complex 3-D structural features (e.g. salt bodies) in gridded modelling and inversion applications. This approach removes the model parameter aliasing commonly present in gridded models and allows structural boundaries to be positioned to sub-cell accuracy. It can be linearized to enable inversion methods that operate on gridded models to update the shape parameters describing model structure directly. It is shown in synthetic modelling tests that this method significantly improves the accuracy of seismic data modelled using the finite-difference method in the presence of high-contrast structural features such as salt bodies. An application to full waveform inversion is also demonstrated using synthetic data, in which bulk shifts and fine structure are resolved in a complex top-salt Gulf-of-Mexico model.

Resolving Near-Surface Velocity Anomalies in Marine Data

E-Lecture by Ian Jones

Unresolved velocity anomalies in the near surface degrade deeper imaging. As a consequence, great care needs to be taken to ensure that all significant near-surface effects have been dealt with before attempting to build the deeper parts of a velocity-depth model. In order to incorporate velocity anomalies into the model, a range of options can be used, depending on whether the geobody geometry alone is discernible, or whether its velocity distribution is also known. Here I describe current industrial practice for building complex near-surface models, which is based on a range of approximate techniques, as well as the more complete solution offered by the emerging technology of waveform inversion. Although building complex near surface models is a painstaking process, a suitable near-surface velocity model can usually be obtained.

The following course by Ian Jones is available via EAGE:
An Introduction to Velocity Model Building

Consult our Calendar or Events for the next deliveries or request an in-house training proposal!

In-Situ Local Angle Domain Sata As An Ideal Representation For Directivity Driven Imaging

E-Lecture by  Zvi Koren 

In this E-Lecture Zvi Koren presents a novel imaging system for enriching (maximizing) information from the available recorded seismic data. The system is based on two stage imaging process composed of:

1. Mapping into the Local Angle Domain (LAD): Internal optimal beam forming and in situ full wavefield mapping and decomposition (binning) of the 5D recorded seismic data into the subsurface 7D LAD using a bottom-up ray-based migration operator. The resulted imaged dataset is organized and stored as 5D angle-domain common image gathers, where each depth point consists of two directional angles (apparent dip and azimuth) and two scattering angles (opening angle and opening azimuth).

2. LAD-based Conditioning and Processing: mute, data reconstruction, Q compensation LAD-based Imaging: Ability to selectively enhance structural model continuity (specular energy imaging) and high resolution discontinuous objects: small faults and fracture systems (diffraction imaging) or near vertical walls (corner-waves imaging) directly from the 5D LAD gathers It is shown that the method enhances the data components associated with the seismic modelling used for the migration (e.g. primary P-waves) and simultaneously attenuates all other data characteristics (“noise”; “multiples”) in both data and space-angle domains.

Recent Advances in Ray Based Tomography

E-Lecture by Gilles Lambaré

In the field of velocity model building there has been a strong focus over the last years on full waveform inversion methods, and much less on ray-based tomographic approaches. Tomographic approaches however remain the standard model building tool for seismic depth imaging, and have made very significant progress in recent years. They have improved in accuracy, flexibility, and turnaround with the development of industrial non-linear multi-layer approaches based on slope tomography. Good progress has also been made in joint inversion and the use of a priori information and constraints, mitigating the problem of lack of structural consistency and resolution of ray-based tomography. With tomography results currently giving an apparent frequency in the velocity model of up to 6 Hz and broadband acquisitions of down to 2 Hz, we have moved from a mid-frequency gap to an overlap of the resolution we can expect from ray-based tomography and imaging.

Reverse Time Migration: How Does It Work, When To Use It

E-Lecture by Etienne Robein

Building an accurate image of the subsurface in complex geological settings remains a serious issue for geophysicists. If the first challenge in this context remains to build a model of the velocity at which waves propagate in the subsurface, the second one is to compute the seismic image itself and Reverse Time Migration or RTM, is one answer to this challenge. In this E-Lecture Etienne Robein explains how RTM works and why it provides generally superior results to those of other migration methods in complex areas, but he discusses also its main limitation: RTM is a computer intensive tool to migrate high frequencies.

Etienne Robein is also the instructor of two short courses offered via EAGE:

Seismic Imaging: A Review of the Techniques, their Principles, Merits and Limitations (EET 4)
Seismic Depth Imaging and Anisotropic Velocity Model Building

Consult our Calendar or Events for the next deliveries or request an in-house training proposal!

On the Use of Epsilon and Delta in Anisotropic Velocity Model Building and Depth Migration

E-Lecture by Etienne Robein

The objective of seismic imaging is to get a sharp and accurate image of the elastic reflectivity in the subsurface, especially in complex geological settings. By ‘accurate’, we mean that reflectors should be at the right depth, but also correct lateral position. The accuracy of a Pre-Stack Depth Migration image goes together with our ability to estimate the velocity at which waves propagate in the subsurface. To build such a Velocity Model remains a challenge because the earth is not only heterogeneous (waves propagate at different velocities at different locations), but also anisotropic (propagation velocity is a function of propagation direction). The presentation discusses the physical meaning of Thomsen’s parameters epsilon and delta included in the Model to describe anisotropy. It illustrates a popular method to estimate them in a Pre-Stack Depth Migration project to conclude that borehole calibration is a necessary step to achieve the best accuracy possible.

Watch this E-Lecture on Youku

Etienne Robein is also the instructor of two short courses offered via EAGE:

Seismic Imaging: A Review of the Techniques, their Principles, Merits and Limitations (EET 4)
Seismic Depth Imaging and Anisotropic Velocity Model Building

Consult our Calendar or Events for the next deliveries or request an in-house training proposal!

Demystifying Marchenko Imaging

E-Lecture by Roel Snieder

Marchenko imaging is an imaging method that makes it possible to image and focus waves in an unknown medium. I show how Marchenko imaging makes it possible to create an image at a fixed location inside the subsurface, without knowing the subsurface between the acquisition surface and the imaging point, while taken taking all multiples into account. I also describe a number of strategies for creating an image.

Roel Snieder is the instructor of "Art of Science", a short course offered by EAGE. Consult our Calendar or Events for the next deliveries or request an in-house training proposal!

Reverse-time Imaging of Dual-Source 4C Marine Seismic Data Using Primaries, Ghosts and Multiples

E-Lecture by Ivan Vasconcelos

In this video, Ivan Vasconcelos covers the physical principles and practical approach of a wave-equation-based, reverse-time seismic imaging method designed to handle dual-source, four-component acoustic data. Vasconcelos et al. show that the method can naturally make use of source- and receiver-side ghosts in imaging, while bringing additional advantages such as handling finite-frequency directivity and, to some extent, multiple scattering. The concepts are illustrated with both simple and geologically complex models.

Reservoir Characterization

3D Seismic Imaging of Volcanogenic Massive Sulphide Deposits in Metamorphosed and Deformed Rocks - Two Case Studies from Canada

E-Lecture by Gilles Bellefleur

This E-Lecture: "3D Seismic Imaging of Volcanogenic Massive Sulphide Deposits in Metamorphosed and Deformed Rocks - Two Case Studies from Canada" presents two case studies on the application of 3D seismic data for mineral exploration in the Snow Lake (Manitoba) and Bathurst (New Brunswick) mining camps, Canada. Both 3D data sets were acquired over deep volcanogenic massive sulphide (VMS) deposits hosted in metamorphosed and deformed rocks. The two deposits investigated are the Lalor (Manitoba) and Halfmile Lake (New Brunswick) VMS deposits. At Lalor, collocated petrophysical and lithogeochemical measurements, and a 3D geological model are used to explain strong and continuous reflections observed in intensely hydrothermally-altered and metamorphosed rocks in the footwall of the deposit. The strong reflections are found at contacts between metamorphosed rocks with felsic and mafic protoliths, some being a proxy for the most prospective horizon in the mining camp. At Halfmile Lake, a 3D seismic survey acquired in the late 1990s led to the discovery of a deep mineralized zone at a depth of 1.2 km. A new anomaly identified on recently re-processed data is the possible continuation at depth of the deep mineralized zone discovered on the original data. Reflections suggest that the deep mineralized zone and its possible continuation are offset by a shallowly dipping fault. The two case studies demonstrate the value of 3D seismic data for mineral exploration, even in areas where geological conditions are complexified by deformation and metamorphism.

Basic Well Log Data Acquisition

E-Lecture by David Sendra

This e-lecture consist of a basic overview of the methods used to acquire well log data. Without this data, exploration, appraisal and development phases on any field would not be possible.  This data yields very important parameters used by all explorationists around the world.

Watching Gas Move: Seismic 4D Reservoir Monitoring of the Pluto Gas Field, North West Shelf of Australia

E-Lecture by André Gerhardt

This E-Lecture: "Watching Gas Move: Seismic 4D Reservoir Monitoring of the Pluto Gas Field, North West Shelf of Australia", whilst Seismic 4D monitoring is a mature and established technology commonly used in the reservoir management of oil fields, not much has been published on its use for monitoring gas fields. In this presentation I describe in detail the various steps undertaken in the feasibility studies that underpinned the 4D seismic monitoring of the Pluto Gas Field in the North West Shelf of Australia. I also present some of the results of the first monitor survey which show remarkable similarity with the expectations from the feasibility studies. We postulate that the production-related (pressure and saturation) effects are much easier to predict and quantify when sufficient aquifer support is present.

Quantitative prediction of injected CO2 at Sleipner using wave-equation based AVO

E-Lecture by Peter Haffinger

Wave-equation based amplitude-versus-offset inversion is a fundamentally new approach towards converting seismic pre-stack data into elastic subsurface properties. By solving the full elastic wave-equation, the routinely made assumption of a linear relationship between seismic amplitudes and reservoir properties is overcome. As a consequence interbed multiples, mode conversions and transmission effects over the inversion interval are properly accounted for. Additionally, the method directly solves for compressibility and shear compliance, which are approximately three times more sensitive to fluid and lithology changes when compared to impedances. This makes the technology suitable for quantitative interpretation even in geologically highly complex scenarios. In this e-Lecture, a high level introduction to wave-equation based AVO inversion will be given and an application to the Sleipner Carbon Capture and Storage (CCS) project will be presented.

Calibration of Distributed Acoustic Sensing (DAS) VSP Data

E-Lecture by Mark E. Willis

Vertical seismic profiling (VSP) is a technology used to characterize seismic properties of the earth in a large volume surrounding a well. The goal of this lecture is to familiarize the viewer with a new technology, distributed acoustic sensing (DAS), that makes VSPs quicker to acquire and easier to repeat. DAS uses fiber optic cables as sensors, replacing geophones for recording the seismic wave field. First a brief introduction of the DAS technology is presented. Then VSP data from DAS and geophones are compared and calibrated to demonstrate the big uplift possible using this new technology.

Evaluating the Gap between Seismic Scale and Well Scale Observations of Structure

E-Lecture by Ryan Williams

Detecting structure at reservoir scale on seismic data has been a long-standing issue due to vertical resolution constraints. Commonly results from seismic data are downscaled or well based fracture studies are upscaled in an attempt to predict fault/fracture position. In this study, both seismic and well analysis methods were combined in a cognitive based workflow, allowing for accurate fault/fracture prediction not only at the well location, but laterally within the 3D volume.

Time-Lapse Observations from PRM at Snorre

E-Lecture by Mark Thompson

Snorre utilizes water alternating gas (WAG) to optimize the sweep efficiency of the injection program, where the WAG cycles in a WAG injector can typically vary between two and 12 months. Prior to PRM the traditional seismic time-lapse surveys on Snorre were acquired with a three year interval between the surveys. During the three year interval between 4D seismic surveys the production history can become very complex and the time-lapse signal can be equally complex with both constructive and destructive signal effects making it difficult to interpret the 4D data and relate the 4D signal to production effects. Permanent Reservoir Monitoring (PRM) was installed at the Snorre to improve 4D data quality and, with more frequent data acquisition, simplify a complex 4D story. By October 2014 the PRM system had been installed and the first seismic acquired. Time-lapse seismic has since been acquired once or twice a year, with the aim to geophysically monitor the field frequently. Through observations, this e-lecture with demonstrate that frequent acquisition and faster turnaround from acquisition to interpretation using PRM has been realized value at Snorre.

Seismic 4D Inversion for Quantitative Use in Automated History Matching

E-Lecture by Milana Ayzenberg

Milana Ayzenberg (Statoil) discusses the different aspects of using 4D seismic for conditioning reservoir models. Assisted history matching is employed to match the reservoir production history and the 4D seismic inversion data simultaneously. The three building bricks of a conditioning workflow are the forward modelling from the reservoir model to the inverted 4D seismic attributes; the quantitative seismic inversion; and the history matching which closes the dynamic conditioning loop. Milana discusses in detail these three components, with a particular focus on the quantitative aspects and uncertainties in a 4D inversion. The conditioning workflow is demonstrated on a North Sea field which exhibits a complex 4D signal. The seismic data is inverted to 4D changes in elastic parameters. This data is further used for conditioning the reservoir model on equal footing with production and pressure data.

Seismic Surveillance for Reservoir Delivery "A Practitioner's Perspective"

E-Lecture by Olav Inge Barkved

In this E-Lecture Olav Barkved will share examples of value creation from 4D seismic and demonstrate how the technology can be taken a step further through the use of field-wide permanent installed seismic sensors. Permanently installed systems for seismic surveillance offer frequent and high quality seismic time-lapse surveys to be used for active reservoir management, to support the drilling of infill wells, and to assist the production engineers in managing the wells. Olav, currently with Petoro and previously with BP Norway, is sharing from his experience in managing and making the most out of a dedicated seismic surveillance system for reservoir delivery.

Subsalt Time-Lapse Seismic for Reservoir Monitoring Using i4D in Deepwater

E-Lecture by David Chalenski


David Chalenski presents a case study using a low-cost but high-data-quality seismic technique to monitor a waterflooded field. This technique, termed i4D, utilizes simple planning methods to decimate a source and nodal receiver patch from a previous full-field survey to monitor the area around a high-risk, high-yield well (such as a water injector) with high accuracy and confidence. He demonstrates that nodal surveys can be targeted even in a highly complex sub-salt field. If planned appropriately, expected data quality is equivalent to a full-field nodal survey. He shows the 4D results and presents business decisions which were influenced using the results of this survey.

Applied AVO

E-Lecture by Anthony Fogg

AVO (Amplitude Versus Offset) analysis is a method many geoscientists may be aware of, but they perhaps do not know how the techniques are implemented and the best way to apply the technology. This E-Lecture introduces the basics of the AVO theory and how it is used to create attributes from seismic reflection data that reveal the underlying rock and fluid characteristics of the sub-surface. It uses synthetic models to demonstrate the importance of velocities and migration algorithms in obtaining the optimum seismic image for AVO and subsequent pre-stack seismic inversion analysis.

Watch this E-Lecture on Youku

The contents introduced in this E-Lecture are part of a comprehensive course offered by Anthony Fogg via EAGE:
AVO in an Inversion World 
Consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!

Distributed Acoustic Sensing Cable for Surface Seismic

E-Lecture by Kees Hornman

Distributed Acoustic Sensing (DAS) is a relatively novel technology, which uses a fibre-optic cable as sensor. Advantages are that the cable is passive, it does not have separate detectors and the channel spacing can be set by the instrument. DAS has already found applications in monitoring of hydraulic fracturing and VSPs. A downside is that a straight fibre has no broadside sensitivity and can therefore not be used for reflection seismic with horizontal cables. A helically wound cable (HWC) solves that problem; the video describes a field trial, which confirms the theoretically expected isotropic sensitivity of the HWC for P-waves.

The Startup of Permanent Reservoir Monitoring for Snorre and Grane

E-Lecture by Mark Thompson

Statoil has supported the development of geophysical reservoir monitoring (GRM) techniques, and through a focused seismic monitoring (FSM) project, successfully piloted permanent reservoir monitoring (PRM) technologies. In this E-Lecture, Mark Thompson tells that this experience culminated in the implementation of two PRM projects at the Snorre and Grane licenses. In December 2012, Statoil awarded a contract to manufacture and deliver approximately 700 kms of seismic cable for PRM installations, representing the largest PRM installation to date. By October 2014, two full PRM systems had been installed with first seismic acquired at both Grane and Snorre, and a second survey underway at Snorre. In less than two years a major PRM project had been successfully executed and first seismic acquired.

Seismic Characterization of Shallow Gas in The Netherlands

E-Lecture by Mijke van den Boogaard

The Cenozoic in the Dutch offshore is known to host abundant shallow amplitude anomalies related to hydrocarbons. The Netherlands is the first country in the North Sea area in which shallow gas is developed and has proven to be a valuable resource. Amongst 3 successfully producing fields and 5 additional proven accumulations, EBN has identified more than 150 leads from seismic data (bright spots). In terms of volumes several of those leads are expected to rank economically. The success of the producing fields, initial volumes, and the large availability of 3D seismic data has contributed to an increased interest from the industry. The work that is presented includes an overview of the shallow gas play in the northern Dutch offshore and focuses on the seismic characterization system established to help selecting those bright spots that have highest potential for development. The main objectives are to de-risk the play and to improve the understanding of the relation between seismic anomalies and gas saturation.

Quantitative Analysis of Schoonebeek SeisMovie Data

E-Lecture by Paul Zwartjes

A permanently installed SeisMovie™ system was deployed at Schoonebeek oil field in The Netherlands to monitor steam injection in one of the patterns. Qualitatively, the 4D data has given us unprecedented insight into the lateral extent and temporal evolution of the steam front. We attempted to invert directly for steam thickness but the limited areal extent of the system (narrow swath) resulted in a sub-optimally imaged stack that did not match the synthetic data generated from the reservoir model (i.e., we could not “close-the-loop”). Instead, we compared forward modelled data generated from a history matched reservoir model to 4D synthetic and observed time lapse seismic attributes and concluded that we see a slow vertical steam chest growth of approximately 1.7cm/day over a 7 month period.

Integrated Geophysics

Integrated Geophysics and New Methods for Multi-nodal Data Analysis

E-Lecture by Paolo Dell' Aversana

The central subject of this e-lecture is the concept of “expanded integration”. This involves all the key aspects of the process of integration of geophysical and geological data. In fact, expanded integration includes combination of multi-physical, multi-scale, multi-domain and multi-sensory (images and sounds) information. This comprehensive approach is performed using new methods of data analysis. These are called “Brain Based Technologies” because they take into account the key aspects of “high-level” human cognition, such as integrated perception, data fusion, imaging, mapping, combination of multi-sensory inputs and pattern recognition.

Sub-Salt modelling in 3D – Integration of Seismic, Well and Gravity Data, Validated by Drilling

E-Lecture by Antony Price

Offshore sub-salt seismic imaging along the West African margin is a challenge in many areas, and with complex salt geometry, seismic depth imaging alone faces certain limitations. In an effort to further de-risk structures sub-salt, Antony Price et al. integrated with gravity data in 3D. They detail that the incorporation of an independent geophysical parameter, such as density can effectively de-risk these difficult targets. Coincident gravity anomalies and seismic base salt closures were observed in this locality. The gravity anomalies are larger in amplitude and not equivalent to any possible post-salt structure, independent of density. In short, the scenario 3D gravity modelling suggests that these are indeed sub-salt structures, most likely localized basement or some other high density closures - which is consistent with the seismic imaging.

Did you like this E-Lecture? To learn more about Sub-Salt Modelling in 3D with Antony Price, request a webinar via EAGE.


Exploration discoveries and future trends

E-Lecture by Andrew Latham

The oil and gas exploration industry has fixed its broken economics and emerged from the recent oil price downturn in good health. But the industry is now much smaller, with fewer companies drilling fewer wells. New field volumes discovered have fallen to 70-year lows.
New plays and frontiers are often at the heart of the improved profitability. Companies that are prepared to take greater exploration risks are reaping greater rewards. This E-Lecture reviews some of the most important new discoveries behind these trends, and considers the future outlook for the sector.





Paleogeography & Stratigraphy

E-Lecture by Juan Tavella

In this E-Lecture Juan Tavella describes the main aspects of a Case Study for unconventional reservoir characterization of the Upper Jurassic shale in Northern Mexico. It is based on the workflow developed in the past three years to predict key properties to understand the behaviour of a source rock as a reservoir. It integrates well and seismic to provide attributes that in a exploratory scenario will help to outline high potential zones and to generate useful information to design drilling navigation and completion.


Geological Modelling

Geological Well Testing in Fractured Reservoirs

E-Lecture by Patrick Corbett

In this contribution we consider synthetic well test responses generated through numerical simulation of a model derived from an outcrop-based fault/fracture geometry. We consider how the well might connect with the fractures to help understand relationships between the different fracture well test responses.

Volume Based Modeling: Automated Construction of Complex Structural Models

E-Lecture by Laurent Souche

Laurent Souche (Schlumberger) exposes the latest advances in structural modeling. After discussing the conceptual differences between surface-based and volume-based approaches for building 3D faulted structural models of the subsurface, the key algorithms underlying the volume-based technology are described. The main geological and geometrical constraints controlling the interpolation of a 3D attribute representing the relative geological age of the formations are also detailed. Finally, advantages of volume-based methods are illustrated using synthetic examples, physical sandbox models and real field data.


Structural Geology

Giant Regional Play Fairways from Modern Seismic Data - part 1

E-Lecture by Neil Hodgson and Karyna Rodriguez



First part of the E-Lecture “Giant play fairways are play fairways for giants”. Recent developments in exploration ideas: imaging pre-salt syn-rift clastics in Gabon to inorganic carbonate deposition in the Santos Basin of Brazil, from basin floor contourite – turbidite mixed clastic systems of the south Atlantic, isolated carbonate build-ups in Somalia and Guyana, conjugate margins drawn together and understood in a dynamic topography framework, and illustrated with modern seismic examples.

Giant Regional Play Fairways from Modern Seismic Data - part 2

E-Lecture by Neil Hodgson and Karyna Rodriguez


Second part of the E-Lecture “Giant play fairways are play fairways for giants”. Recent developments in exploration ideas: imaging pre-salt syn-rift clastics in Gabon to inorganic carbonate deposition in the Santos Basin of Brazil, from basin floor contourite – turbidite mixed clastic systems of the south Atlantic, isolated carbonate build-ups in Somalia and Guyana, conjugate margins drawn together and understood in a dynamic topography framework, and illustrated with modern seismic examples.


Reservoir Characterization



Simulating Correlated Discrete Fracture Networks constrained by Microseismic Data

E-Lecture by François Bonneau

This e-lecture presents a workflow to simulate correlated Discrete Fracture Networks (DFN) constrained by statistics and microseismic data. We investigate the possibility to analyze microseismic data using the 3D Hough transform and to constraint the Discrete Fracture Network simulation. Then, we detail a sequential pseudo-genetic DFN simulation workflow that produces a hierachical DFN. It uses mechanical proxies inspired from mechanics to sample parameter distribution laws and to efficiently organize simulated fractures in space. This work has been funded by the RING consortium (

The Importance of Overburden Stress Path in Assessment of Stress Dependence for 4D Applications

E-Lecture by Rune M. Holt

In time-lapse (4D) seismic, slow-down in the overburden is often seen as a footprint of depletion in subsurface reservoirs. The work presented in this lecture has demonstrated that for stress changes around the in situ stress, laboratory measured stress and strain sensitivity of a field shale core depends strongly on the stress path. Translated to the field, the stress path describes how horizontal stress varies with changes in the vertical stress when stress arching occurs above a depleted (or inflated) site. Geomechanical simulations are needed in order to identify the correct in situ stress path for each specific field situation, but 4D data may also provide guidelines towards the identification of the overburden stress path.


E-Lecture by Jörg Herwanger

In this compact version of his EAGE Education Tour (EET 5), Jörg Herwanger (Ikon Science, previously Schlumberger) discusses the process of building and calibrating geomechanical models using 3D and 4D seismic data. He analyzes the three main uses that seismic data provide in building geomechanical models: horizon and fault interpretation for building structural models, AVO inversion and rock physics models for creating mechanical property models, and for model calibration. Herwanger elaborates on these three points in two case studies. The first case study presents a 3D exploration geomechanical model. The second study presents a 4D geomechanical model used for field development planning.

Watch this E-Lecture on Youku

If you are interested in the short course, consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!

One 4D geomechanical model - and it's many applications

E-Lecture by Jörg Herwanger

This E-Lecture is a worked case-study in Applied Oilfield Geomechanics: it introduces the elements of a geomechanical model, covers aspects of geomechanical model building and calibration, and works through three applications of the geomechancial model in support of operational decisions. The geomechanical applications presented by Jorg include wellbore stability for drilling an inclined infill well, risk of fault re-activation during gas reinjection and fracture containment in individual reservoirs during hydraulic stimulation.

If you are interested in the short course, consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!


Rock Physics

Rock Physics and Seismic Reservoir Prediction Contrained by Depositional and Burial Trends

E-Lecture by Per Avseth

In this compact version of his Distinguished Lecture Programme presentation, Per Avseth (Norwegian University of Science and Technology) discusses Rock Physics and Seismic Reservoir Prediction Constrained by Depositional and Burial Trends.

Per Avseth is the instructor of a short course and a webinar offered via EAGE. Consult our Calendar of Events to learn about the next deliveries or request one!

Seismic Screening for Hydrocarbon Prospects Using Rock-Physics Attributes

E-Lecture by Per Avseth

A methodology to create easy-to-implement rock-physics attributes that can be used to screen for reservoir sandstones and hydrocarbon pore fill from seismic inversion data is demonstrated. We have honored the physical properties of the rocks by defining attributes that complied with calibrated rock-physics models, including the fluid and rock impedances. We have demonstrated the use of these attributes on well log and seismic inversion data from the Norwegian Sea, and we successfully screened out reservoir rocks filled with either water or hydrocarbons.

Per Avseth is the instructor of a short course and a webinar offered via EAGE. Consult our Calendar of Events to learn about the next deliveries or request one!

Pseudo-Elastic Impedance - A Norwegian Sea Demonstration

E-Lecture by Per Avseth

In recent years, elastic inversion and quantitative interpretation of pre-stack seismic data have become standard procedures in the petroleum industry. In this presentation, Per Avseth (Norwegian University of Science and Technology) compares extended elastic impedance trends at different chi angles, with rock physics template models. He demonstrates that the elastic impedance attribute does not always comply with the complexity and non-linearity of rock physics models related to geological processes. He also shows how one can honour the non-linearity of a rock physics model and create pseudo-elastic impedance as a function of deviation away from a curved water-wet rock physics model itself (CPEI). Finally, Per demonstrates the use of this approach on seismic inversion data from a selected area in the Mid Norwegian Sea, by mapping of lithology and fluid anomalies that are consistent with rigorous rock physics models.

Per Avseth is the instructor of a short course and a webinar offered via EAGE. Consult our Calendar of Events to learn about the next deliveries or request one!

Improved Rock Property Estimation from Joint Inversion of PP and PS Reflectivities

E-Lecture by Ivan Lehocki

The objective of this work was to develop a method for joint inversion of PP and PS data with the goal of unique and accurate determination of important layer parameters, namely y1 and y2 ratios. The method also gives reliable estimates of density ratio, a parameter that can be directly linked to hydrocarbon saturations. The usage of the method is limited to the assumptions used in the derivation of Zoeppritz’s equations, i.e. interface of two horizontal, homogeneous, isotropic, elastic layers. What is new? Easy to understand approach for inversion of all four elastic parameters entering Zoeppritz equations.

Seismic Anisotropy in Shaly Formations…Revisited

E-Lecture by Patrick N.J Rasolofosaon

Seismic anisotropy, that is to say the directional dependence of seismic velocity, is quite common in sedimentary formations and is often linked to the presence of shale. If not correctly taken into account it can strongly affect surface seismic data interpretation, seismic to well tie and azimuth versus offset analysis. From the analysis of two large databases of up to 800 ‘shaly’ samples in a broad sense, including shales but also mudshale, clayshale, siltstone, argillite, claystone, siltshale, mudstone, we demonstrate that seismic anisotropy in such formations is to a large extent determined by factors other than compaction processes, such as depositional environment, chemical composition of fluid, silt fraction, etc. Furthermore, the alignment of the individual clay platelets, main constituents of shales, can explain most of the anisotropy measurements of the databases. Assuming the elastic properties of the individual clay platelets, we propose simple plots for straightforwardly quantifying the Legendre Orientation Distribution Function coefficients and of the clay platelet alignment from the measurement of seismic anisotropy parameters.


Near Surface

Seismic spatial gradient measurements

E-lecture by: Nihed el Allouche



Rotation measurements have found applications in various fields of geophysics ranging from near-surface archaeological mapping to large scale upper mantle tomographic inversion. In the absence of a cost-effective sensor sensitive over the typical seismic frequency bandwidth, rotation measurement can be approximated by finite-differencing the response of the vertical particle velocity over a short distance. Accurate estimates of the horizontal gradient of the vertical wavefield can be obtained when perturbations associated with the measurement are minimised. These perturbations can be sensor related, for example geophone sensitivity and natural frequency, and/or deployment related such as tilt and coupling errors.
In this E-Lecture, we show how perturbations can affect the gradient measurement (the tilt error turns out to be dominant) and discuss the minimum receiver spacing required to avoid the impact of perturbations dominating the gradient estimates. We also discuss a “sensitivity” chart ranking the perturbations according to impact and indicate which one of these needs to be minimised in order to obtain reliable gradient estimates.

Bedrock and fracture zone delineation using Different Near-Surface Seismic Sources

E-lecture by: Bojan Brodic



This E-Lecture "Bedrock and fracture zone delineation using different near-surface seismic sources" discusses a seismic survey conducted using four different small-scale and inexpensive seismic sources to delineate bedrock surface and a fracture zone intersected by a well at c. 50 m depth. The seismic sources analyzed are a 5-kg sledgehammer, a metal I-beam struck laterally, an accelerated weight drop and a prototype EM based seismic source with two hammers striking successively at an adjustable impact rate. From the recording perspective, a high-fold (star-type) acquisition spread was designed combining a three-component microelectro-mechanical system (MEMS) seismic landstreamer and wireless seismic recorders. The performance of every source used is analyzed and reflection seismic sections, along with 3D tomography results shown.

The results indicate well delineated undulating bedrock topography, both on tomography and seismic sections for all sources. Weak reflectivity is observed where the fracture zone is expected. Most of the sources used show similar potential for fracture zone and near-surface imaging and star-type seismic array used enables 3D overview of the shallow subsurface and potential for pseudo 3D reflection seismic processing.
Estimation of velocity distribution in tree trunks and reflection imaging by GPR considering curved raypaths

E-lecture by: Kazunori Takahashi



This E-Lecture introduces "estimation of velocity distribution in tree trunks and reflection imaging by GPR considering curved raypaths". Tree inspection has been conducted in order to assess the risk of tree collapse. Ground penetrating radar (GPR) is a powerful tool for the inspection as it is capable of imaging the internal structure of tree trunks noninvasively. The key of the accurate imaging with the reflection measurement is to set the propagation velocity of electromagnetic waves in imaging algorithms. The E-Lecture presents a simple method to estimate the velocity with the transmission measurement. A challenge of the estimation and imaging is the fact that the velocity inside living trees is usually distributed and setting a constant velocity in the reflection imaging introduces errors. The present method estimates the velocity distribution and images considering curved raypaths by employing a ray tracing algorithm. The method can also be used for estimating the velocity/permittivity of other cylindrical objects.

Characterization of Holocene Sediments using geophysical methods and borehole information: Ebro Delta (Spain)

E-Lecture by Beatriz Benjumea



Beatriz Benjumea shows the integration of geophysical and borehole information to characterize Holocene sediments in the Ebro Delta (Spain). Nowadays, this delta coastline is retreating landward due to a combination of sediment subsidence, sea level rise and sediment deficit caused by river damming. Measuring spatial changes of near-surface sediment facies is critical to evaluate subsidence.

In this presentation, geophysical patterns that help to characterize the Holocene sequence are extracted from borehole and geophysical information at two test sites. The applied methods are: active and passive seismic and electrical resistivity tomography (ERT).

These patterns are then used in an extended survey along the delta with the following targets: to discriminate between surficial sediments facies, to detect the top of the Prodelta marine clays and finally to image the base of the Holocene delta.

3D Inversion of Magnetic Data Affected by Remanent Magnetization

E-Lecture by Yaoguo Li

The inversion of magnetic data in the presence of strong remanent magnetization has long been a challenging problem, because of the unknown direction of the total magnetization that is the vector sum of the induced and remanent components. This E-Lecture presents three different techniques for tackling this problem: (1) direction estimation, (2) amplitude inversion, and (3) fuzzy c-means clustering magnetization inversion. Combined with the existing susceptibility inversion, these techniques form a tool kit that allows one to invert virtually any magnetic data set for the purpose of quantitative interpretation in exploration applications. In particular, the magnetization inversion provides a new opportunity both for imaging source configurations and for carrying out geology differentiation.

The contents introduced in this E-Lecture are part of a comprehensive course offered by Yaoguo Li via EAGE. Consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!

An Electromagnetic Survey Of A Gas Hydrate Vent Offshore Mid-Norway

E-Lecture by Andrei Swidinsky

Andrei Swidinsky describes a transient electromagnetic survey to image the resistivity structure of the hydrate vent CNE03, offshore Mid-Norway (located approximately 10 km north of the Storegga Slide sidewall). The experiment is unique in that the electric dipole transmitter has two polarizations for each transmission station. Furthermore, the newly designed transmitter is deployed in a pogo-style acquisition, which is suitable for a detailed investigation of a small scale seafloor target. The geometry of the experiment makes conventional interpretation methods difficult so that we simplify the data by creating a single rotationally invariant quantity from the original four electric field measurements (two electric field measurements for two transmitter polarizations). This invariant is further reduced to an apparent resistivity, which is useful for rapid resistivity mapping of the seafloor. Results show that CNE03 is characterized by increased apparent resistivities which correlate well with increased P-wave velocities determined from ocean bottom seismometer measurements.



GeoBIM for infrastructure planning

E-Lecture by Mats Svensson

This E-lecture aims at bridging the gap between the deliveries from geophysicists to the infrastructure industry, so that the results from geophysics are used in the best possible way. The key to this is proper communication. For this we all need to know who all the stakeholders are and what interests, what skills and which tools do they have, and we also need to improve the tools we use. This E-lecture is trying to clarify those issues and also suggesting the powerful GeoBIM concept to handle both data, models and communication, exemplified by a large railway project in Sweden.

Read now the paper 


4D Inversion of Continuous Land Seismic Reservoir Monitoring of Thermal EOR

E-Lecture by Laurène Michou

Laurène Michou (CGG) presents 4D seismic inversion results from a continuous seismic monitoring survey of the Schoonebeek reservoir produced by thermal EOR. The 4D simultaneous stratigraphic inversion methodology was driven by the continuous nature of the monitoring. Inversion results brought significant insight about steam distribution.

Satellite InSAR Data Reservoir Monitoring From Space

E-Lecture by Alessandro Ferretti

Satellite radar data for surface deformation monitoring are gaining increasing attention. They provide a powerful tool for remotely measuring small surface displacements that can be applied successfully to many different applications, spanning from sinkhole detection to reservoir optimization. This course provides a step-by-step introduction to satellite radar sensors, SAR imagery, SAR interferometry and advanced InSAR techniques. Rather than a tutorial for remote sensing specialists, the course starts from very basic concepts and explain in plain language the most important ideas related to SAR data processing and why geoscientists and engineers should take a vested interest in this new information source.

The contents introduced in this E-Lecture are part of a comprehensive course offered by Alessandro Ferretti via EAGE:

Satellite InSAR Data: Reservoir Monitoring From Space

Consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!


Training and Development


Soft Skills

The Art of Science

E-Lecture by Roel Snieder

This class gives general professional skills needed to be effective and successful in research. These skills include generating research questions, creating a work plan, scientific publishing, oral and written communication, time management and many other topics. This video is part of EAGE Online Education Programme. The European Association of Geoscientists and Engineers (EAGE) is a global professional, non-profit association for geoscientists and engineers. EAGE strives to promote innovation and technical progress and aims to foster communication and cooperation between those working in, studying or interested in these fields. To learn more about EAGE Education visit

The contents introduced in this E-Lecture are part of a comprehensive course offered by Roel Snieder via EAGE. Consult our Calendar of Events to learn about the upcoming deliveries of request it for in-house training!


Data Science

Seismic interpretation with deep learning

E-Lecture by Anders U. Waldeland

How and why can Deep Learning be used for seismic interpretation? The machine learning technique called deep learning is revolutionizing the field of computer vision. A central part of deep learning is convolutional neural networks (CNN). This E-lecture gives a simple and intuitive introduction to CNNs in the context of seismic interpretation.

Github link:

Paper link:

Paper link: