Seismic processing and imaging

Seismic imaging and characterisation rely on estimates of the subsurface velocity models. These estimates are often produced in a separate time-consuming manner. CRGC seeks to reduce the amount of effort needed on the velocity model building that can make the seismic methods more time and cost efficient.

Seismic migration methods rely on a background velocity model of the subsurface. Methods that can provide velocity models automatically along with the migrated seismic data can form an important part of the imaging and characterisation workflows. The team is developing various pre-stack imaging methods that result in simultaneous estimation of the subsurface velocity models and the subsurface image. Currently, the focus is on time migration algorithms. The methods generally rely on the estimation of extra attributes, such as local slopes or curvatures, or use of stationary phase in a summation over the velocities.

Researchers: Prof. Andrej Bona, Dr. Konstantin Tertyshnikov

Diffractions can be used to determine details about the small-scale features that generate them, such as karsts, voids, pinch outs, faults, fractures, and salt flanks. CRGC is developing specialised migration methods and reservoir characterisation techniques for application with diffraction images. Diffraction imaging can have resolution below the typical seismic wavelength.

Many features that are interesting for exploration, such as voids, caves, fractures, and faults are usually not well imaged using methods developed for reflection seismology. However, these features cause diffractions and can be resolved with better focus in diffraction images than in conventional images derived from specular waves. Diffraction imaging is particularly useful with geologically complex carbonate systems. Carbonates are strongly heterogeneous, making them difficult to image with conventional methods.

One of the research topics of the team is imaging of the diffractors along with their seismic characterisation using their amplitude response. This approach is closely linked to improving reflection imaging in low signal to noise environments since only small parts of the diffraction hyperbolae used for summation in standard migration algorithms contain the energy from the reflection. Therefore, the inclusion of all the traces from such diffraction hyperbola may not contribute constructively to the final image. One way of identifying the traces that do not contribute constructively to the final image is by using the amplitude/energy distribution information used in the abovementioned diffraction imaging/characterisation. The team uses a modified 3D Kirchhoff post-stack migration algorithm that utilises coherency attributes obtained by diffraction imaging algorithm in 3D to weight or steer the main Kirchhoff summation.

Researchers: Prof. Roman Pevzner, Prof. Andrej Bona

Source and receiver ghost waves distort the seismic image and have to be accounted for during processing. However, standard deghosting techniques usually ignore variations in sea surface roughness which are caused by ocean surface waves.  This degrades resolution, especially for broadband data. To address these issues, CRGC seeks to understand the behaviour of ghost waves in presence of water surface roughness.

The Curtin team has presented a comparison of rough sea surface reflection modelling results with ultra-high-resolution field seismic data acquired with deep-towed sources and receivers. The Kirchhoff approximation was used to model the sea surface response.  Such a comparison is essential in establishing the validity of the modelling approaches used in data processing, such as deghosting and the development and application of surface-related multiple elimination techniques.. Deep-towed sources and receivers permit the separation of sea surface reflections from primary events.

Researchers: Prof. Andrej Bona, Dr. Stanislav Glubokovskikh, Prof. Roman Pevzner