Full Waveform Inversion (FWI) is now a well-established tool in the velocity model building process, having evolved from a specialist academic technique into a mainstream production tool. Continued research and development further increase its capabilities and uses, bringing benefits to other aspects of seismic processing and imaging.
One area of clear progress relates directly to frequency content. The maximum frequency used in current projects may be 1 or even 2 octaves higher than just a couple of years ago. Several factors are responsible for this improvement: valuable experience has been gained from using FWI on many projects of varying acquisition type and data quality; upgrades to the FWI algorithm use a more accurate wave equation and propagation engine; reduced memory and CPU requirements coupled with significantly greater compute power; and greater accuracy of initial models input to FWI.