Super-virtual Interferometry

Intermediate steps of denoising via super-virtual interferometry (SVI) for core-mantle boundary diffractions. Every iteration of SVI consists of a)—c) a cross-correlation and summation of the data to generate records with virtual arrivals, followed by d)—f) a convolution of the records with the virtual arrivals to create records with super-virtual arrivals.
In this project, I have established a novel physics-based denoising model for seismic refraction imaging. It was carried out in close collaboration with marine geophysicists at The University of Texas at Austin and King Abdullah University of Science and Technology. The experimental goal was to understand the sub-seafloor geology using refracted seismic energy from the geological boundaries. Curstal-scale ocean-bottom-seismometer (OBS) surveys were carried out with sufficiently large offset i.e., distance between the OBS stations and the airgun sources. Large offsets facilitate the recording of the refracted signals radiated from deep geological boundaries, such as the MOHO. However, as the energy propagates over large distances, the signal strength decays resulting in poor signal-to-noise ratio. Consequently, an accurate tomography and imaging of the crustal structure required denoising of the refracted energy.

Signal processing methods based on seismic interferometry principally extract energy, which is coherent across the source and/or receiver dimensions of the measurements. However, the existing interferometric signal models were not reliable when focused on the application to seismic refractions. I have identified a special coherent structure for the refracted waves in seismic data that led to a reliable interferometric model for denoising. This signal model boosted our groups efforts to perform denoising, and the subsequent algorithms resulted in a subfield of seismic interferometry that I spearheaded: super-virtual interferometry (SVI). SVI constructively stacks (sums up) refracted energy from hundreds of weak air-gun sources to enhance its signal-to-noise ratio. My numerical experiments have shown that SVI can potentially boost the signal-to-noise ratio of the refracted seismic energy in the order of $\sqrt{N}$, where $N$ is the number of intermediate stations between a given source and receivers pair. Furthermore, I have also demonstrated the benefits of SVI for a variety of seismic surveys including near-surface geophysical experiments that image bedrock. In all these experiments, without SVI nearly half of the measured large-offset measurements was unusable for refraction traveltime tomography. Later on, I have developed a similar interferometric signal model for denoising core-mantle boundary (CMB) diffractions in collaboration with seismologists. This signal model for CMB diffractions, along with an intuitive sketch of SVI, is depicted in figure above.

Pawan Bharadwaj
Pawan Bharadwaj
Assistant Professor, Center for Earth Sciences

Pawan is an assistant professor in the Center for Earth Sciences at the Indian Institute of Science (IISc). He enjoys developing novel algorithms related to geophysical inverse problems, signal processing and machine learning.