Making seismic diffraction usable in practice

What is it about?

Diffraction, due to practical limitations, is a process that is still largely ignored in both controlled-source and earthquake seismology. In controlled-source studies, the diffracted wavefield usually interferes stronly with reflected contributions and has significantly lower amplitudes making it barely recognizable in the data. Based on enhancing coherence arguments borrowed from optics, we suggest an efficient and unsupervized approach for diffraction extraction by eliminating the stronger interfering wavefields.

Featured Image

Why is it important?

The diffracted wavefield focuses at localized subsurface discontinuities and, therefore, theoretically, permits imaging of structures smaller than the dominant seismic wavelength. In addition, due to its uniform radiation, diffraction samples the subsurface just like a passive source would, if placed at the localized discontinuity, making it an ideal cancidate for the inversion of laterally resolved subsurface properties such as seismic velocities. This work helps to overcome the problem of detectability and separation of diffracted signals, thereby potentially opening the door to similar developments as in x-ray imaging of atoms and crystals.