FWI: Future Perspectives Without the Hype
Friday 11 December, 15:00-16:00 (CET)
FWI has quickly gone from an ambition to reality in recent years thanks to larger computers and significant advances in software. Whilst the original published ambition of Tarantola was to invert for the impedance model of the earth, most implementations until now are based upon the acoustic assumption of the Earth, applied to marine data, and recover a velocity model.
The greatest technical challenges are related to inherent instabilities in the FWI method that can cause it to fail spectacularly; in spite of our latest developments in robustness, efficiency, and flexibility. The only robust claim we can make today about a successfully inverted model is that it fits our imperfectly sampled data, and is an effective and average representation of the Earth; considering the simplified assumptions we make about the physics of the Earth.
The greatest excitement has always been whether FWI can be universally applied to land and marine data, include more accurate physics, recover realistic and detailed elastic earth models including anisotropy, completely bypassing traditional time-consuming processing flows, and be done on a time scale that essentially follows the rate of data collection, or slightly slower.
Are we getting there, how might we get there, or are these ambitions misguided? Indeed, is FWI only a part of an even larger vision? And where can AI take everything? Will machine learning replace FWI altogether? The sparks are sure to fly in a high voltage clash of opinionated personalities. The three energetic communicators promise to make this an engaging hour of insights, debate and entertainment. The must-see event of the entire EAGE program.
MEET THE HOSTS
Andrew Long
Chief Geoscientist, PGS
Andrew Long is Chief Geoscientist at PGS, with interests in most applications of seismic technology and the interpretation of geophysical data. He began his career with a few years in land seismic acquisition and processing, completed a Ph.D. at UWA in Australia, and worked as a Post-Doctoral Research Affiliate at Stanford University before joining PGS in 1997.
Andrew has presented courses for the EAGE, SEG and ASEG; received best presentation/paper awards from the ASEG (1994 and 2004), IPA (2009) and PESA (2014); and is an Honorary Life Member of the ASEG.
Sheng Xu
Lead Researcher, Equinor
Sheng Xu has a BS from Peking University in geophysics 1990; a MS from Institute of Geophysics, Chinese Academy of Sciences in 1993; a DEA from Institut Français du Pétrole in 1998; a ph.D from École des Mines de Paris in 2001. After graduating in 2001, Sheng joined Paradigm in Houston and initiated CRAM development, a successful commercial package of angle domain Kirchhoff migration.
Sheng is now responsible for designing the high-end depth imaging package, his recent achievements include an efficient TTI RTM algorithm, dispersion free finite-difference schemes, and joint local tomographic inversion. He also wrote a GPU version fast Fourier transform, which is many times faster than the commercial ones for wave propagation. Sheng is current working on machine learning applications. Sheng earned best poster award from SEG for 2003 and 2016; Reginald Fessenden Award from SEG for 2016; Bonarelli Award from EAGE for 2007.
Tariq Alkhalifah
Professor of Geophysics, KAUST
Tariq A. Alkhalifah is a professor of geophysics in the division of Physical Sciences and Engineering at King Abdullah University for Science and Technology (KAUST). He assumed his duties there in June 2009. Prior to joining KAUST, Tariq was a research professor and director of the Oil and Gas Research Institute at King Abdulaziz City for Science & Technology (KACST).
Main Sponsor