The challenges faced in finding new hydrocarbon deposits or optimizing production from existing fields have never been greater. Industry is confronted by ever increasing complexity in the subsurface and by new environments characterized by high risks and high costs. Therefore, it is essential that we fully maximize the information available in existing and future geophysical data sets.
While tremendous effort is directed to the design of new seismic acquisition systems, current seismic processing methods typically utilize only a subset of the data collected due to limitations in both software algorithms and computer capacity. The advent of a new generation of peta-scale super computers is enabling seismic processing researchers to implement a technology called full wavefield inversion (FWI). Industry is currently applying a variation of FWI to build low frequency velocity models to be used in more traditional seismic imaging; however, these applications only scratch the surface of the benefits FWI could provide.
The paradigm shift envisioned occurs when FWI is employed to construct detailed, high-resolution property models of the subsurface that exploit the full bandwidth of the seismic data. High-resolution FWI products, for example Vp, Vs, porosity or lithology, will change the way we interpret seismic data. Interpretation of the subsurface directly in a rock property domain will replace traditional mapping of seismic reflections, at boundaries between rock formations.
We have developed new concepts and algorithms that allow for the rapid inversion of multiple subsurface parameters at frequencies that encompass the full seismic bandwidth. This presentation will highlight successes and remaining challenges in the application of FWI using both synthetic and real data from exploration and production environments.