What is it about?
Aside from reflections at continuous interfaces in the subsurface, there exists a second, often forgotten component of the seismic wavefield. These diffracted contributions are caused at small-scale laterally discontinuous features like faults or fracture systems, which have an immediate impact to society. Arguably the biggest complication of dedicated diffraction imaging constitutes in the fact that the respective wavefields heavily interfere and are often orders of magnitude weaker than reflections, which makes them essentially invisible on individual records. To confront these problems, we present a data-driven strategy that automatically detects, extracts and enhances the full prestack diffraction response to make it available for subsequent imaging.
Featured Image
Why is it important?
By many, seismic diffraction imaging is still viewed as a niche discipline, which is at odds with the unique opportunities, in terms of resolution and illumination, this wavefield promises to provide. While previous accounts aim at separating reflections and diffractions during the imaging step, we demonstrate that a separation can already be achieved in the full unstacked data domain, thereby enabling dedicated imaging and inversion strategies.
Perspectives
When large-scale (large-offset) seismic acquisitions are possible, full prestack information on the diffraction response provides the opportunity to amplify and enhance even the faintest signals with striking implications for high-resolution imaging and inversion. This mindset of a full-wavefield extraction of diffractions directly in the data domain, together with subsequent research on the surgical subtraction of coherent wavefields, currently seems like the most promising avenue to me.
Dr Benjamin Schwarz
Fraunhofer Institute for Wind Energy Systems IWES
Read the Original
This page is a summary of: Common-reflection-surface-based prestack diffraction separation and imaging, Geophysics, January 2018, Society of Exploration Geophysicists,
DOI: 10.1190/geo2016-0445.1.
You can read the full text:
Contributors
The following have contributed to this page







