High Resolution Full Wavefield Inversion: Impact on Imaging and Reservoir Characterisation

Abstract

Abstract Seismic imaging has been advancing rapidly over the last two decades, transitioning from post-stack time migration to pre-stack depth migration and from ray-theoretic algorithms (Kirchhoff migration) to one-way wave equation solvers (wave equation migration) and recently to two-way wave equation methods like Reverse Time Migration (RTM). The most recent step in this evolving sequence is Full Wavefield Inversion (FWI), which has rapidly become a focal area of research, development, and application within the geophysical community and industry. FWI products, unlike migration results, are not depictions of seismic reflectivity, but rather high resolution images of subsurface properties such as compressional and shear velocities and density. The high resolution velocity models generated with FWI can be used to address challenging imaging problems that are difficult to solve using conventional processing tools. High fidelity velocity models will assist in imaging below fault shadows and shallow channels and may also better identify and quantify overpressured drilling hazards. FWI estimates of other elastic paramters, in particular the Vp/Vs ratio, are promising lithology indicators useful in reservoir characterization applications. In addition to imaging challenges, the broadband nature of FWI is ideally suited to address reservoir monitoring problems with the ability to provide high resolution updates of property changes in the subsurface due to production. FWI products, appropriately integrated into an interpretation strategy, will allow us to better understand subtler subsurface features such as stratigraphic traps or non-DHI plays. In our presentation, we will use FWI examples from several real data studies to explain the nature of the process and illustrate some of the potential applications. Introduction Full Wavefield Inversion (FWI) is the most advanced and ambitious seismic imaging technology ever conceived. This technology produces the dream of every geoscientist: broadband images of sub-surface rock properties ahead of the drill bit. Seismic imaging has advanced rapidly over the last two decades in lockstep with advances in High Performance Computing (HPC). Traditional seismic imaging methods, such as Beam Migration or Reverse Time Migration (RTM), focus seismic reflections from layer interfaces that the geoscientist can use to map structure and amplitudes that relate to the change in rock properties across layer boundaries. By contrast, FWI simultaneously focuses the image of the subsurface and produces amplitudes that represent actual elastic rock properties within the layers (velocity, density, etc).