N066 An Objective Approach to Seismic Processing

Seismic technology is changing continuously in the drive towards better imaging and better amplitude preservation but of course turnaround should be quicker and at lower cost.

The course is divided into two main themes which are not extensively covered in the geophysical literature. Following an overview of relevant theory, the first part addresses in a non-prescriptive way the type of questions geoscientists have to answer to meet objectives. A set of geophysical tools and rules are provided to help assess the quality and value of seismic data in a given area. We then use the methodology to assess the benefits (and costs) of re-processing and re-acquisition using the latest technologies. Seismic modelling provides a central theme to the methodology.

The second main aim of the course is to provide a demonstration of current geophysical and processing techniques that can be applied to seismic data. From a basic knowledge of seismic processing, the attendees will be given an overview of standard practice, and then for each method we will introduce the basic theory (with few equations), parameterisation, benefits, limitations, pitfalls and likely value. Where possible we will focus on the applications of the techniques using real examples and common terminology. The whole processing sequence is extensively covered, including data preparation for land and marine data, noise and multiple suppression, velocity analysis and model building, statics, time and depth imaging including anisotropy and issues affecting resolution. Interactive processing, using land and marine synthetic and real data examples, is provided throughout the course presentation to enable the attendees to grasp the importance of key parameters. We will also cover the generation of specialised outputs used in quantitative interpretation including AVO and inversion volumes.

Day 1: Workstation Based Workflow: Objective Setting

• Geophysics refresher: an acquisition and processing primer including a brief overview of basic wave theory, noise suppression, velocity analysis, stacking, imaging and factors affecting resolution.
• Seismic data formats: seismic and navigation formats, pitfalls and quality control.
• Data loading: dynamic range (bit precision), reconciling navigation and seismic data, common pitfalls when loading depth data, land data and gathers with recommended quality controls. Seismic display and colour-bars.
• Understanding the data you have: setting objectives
• Meeting project requirements and defining objectives
• Can longer offsets be obtained?
•  Can the bandwidth be improved?
• Will amplitude versus offset analysis be worthwhile?
•  Will depth imaging be required?
• Re-shoot versus re-process using modelling
• Cost-benefit analysis: measure data against objectives
•  Scaling: trace scaling, automatic gain control, offset scaling.
• Data enhancement: frequency, FK, FX filtering.
• Processing tenders: a brief overview of tendering of processing contracts.

Day 2: Basic Building Blocks

• Basic convolution, Fourier Transform
• Sampling and temporal aliasing
• Spatial aliasing
• Inverse theory

Technical Aspects of Survey Design

• Basic survey design, workflow and rules of thumb
• Sampling, deciding survey orientation or azimuthal coverage
• Broadband methodologies and simultaneous shooting strategies
• The effect of field parameters on processing route and vice versa
• Differences between land, marine streamer and OBN
• Case histories: Scott Field, cost-benefit analysis

Signal Processing Workflow

• Amplitudes, frequency and wavelet processing
• Designature, phase and wavelet processing
• Attenuation compensation
• Obtaining broader bandwidth: combining acquisition and processing solutions

Day 3: Noise and Multiple Suppression Workflow

• Noise: types and suppression in Marine and Land seismic data
• Geometry: shot, receiver, offset, midpoint domains; sorting, statics
• Mathematical domains: FK, tau-p analysis
• Aliasing of data, regularisation and interpolation
• FK and Radon: Basic theory, analysis and QC displays, interactive demonstration.
• Multiple suppression. This important area is covered in some detail, comprising:
• Predictive methods: deconvolution, shallow water demultiple
• Moveout methods: FK, Radon, hi-res radon domains
• Wavefield: free surface multiple removal (2D and 3D SRME)
• Interbed multiples - identification and removal.
• Interpreting data affected by multiples
• Basic prestack inversion analysis: Angle gathers and AVO compliant gather conditioning, generation of angle stack and fluid/lithology prediction volumes

Day 4: Imaging Workflow Including Velocity Model Building

• Basic migration: correcting for velocity variation and complex sub-surface.
• Prestack time migration and gather generation. Prestack depth migration:
• Algorithm choice: Kirchhoff single/multi arrival, Beam versus wavefield methods (including reverse time migration), least-squares migration.
• Imaging with multiples, Elastic imaging and Future developments.
• Anisotropy including VTI, TTI, orthorhombic cases.
• Velocity model construction for depth imaging
• Role of interpreter in velocity model building and quality control
•  Full-Waveform Inversion
• Data preparation and initial quality controls
• Diving wave and reflection based inversion schemes
•  Quality control and common pitfalls.
• Tomography – grid or layer based, hybrid methods
• Case histories: Scott Field, North Sea Gas Basin, Sub-salt Gulf of Mexico.
•  Additional uses of seismic velocities: e.g. imaging, pore pressure, depthing.

Day 5: Land Processing Workflow

• Statics: elevation, refraction, tomographic based statics are compared using a series of synthetic and real data examples. Marine statics solutions are also discussed as are statics and layer replacement schemes versus full depth imaging.
• Case history: complex topography

Amplitude and Amplitude Extraction Workflows

• Processing Requirements and Techniques for reservoir characterisation
• Data conditioning and resolution enhancement and attribute extraction.
•  Conventional and Machine Learning based classifications.
• Attributes for conventional and unconventional reservoir characterisation.

Conclusions and Recap

• Processing sequence (includes an overview of processing styles between 1990 to present day) Rules of thumb for data assessment and novel less subjective data comparisons
• Specialised processing: Single sensor, OBC, Elastic and 4D methodologies
• Summary and the Future: Recap. Acquisition and processing methods on the horizon