Methodology

As FWI operates directly on shot gathers, it can be deployed early in the model building process and it is an efficient velocity model building tool.

FWI can be used as part of any Prestack Depth Migration (PSDM) velocity model building flow with virtually no time impact, as the input data requires minimum pre-processing and any free-surface effects can be left in the data.

The method begins from an initial starting model which is then iteratively improved using a sequence of linearized inversions, to solve the full non-linear 3D FWI problem. 

 

FWI model building workflow

New Flavors Abound

As computing power has expanded, so has the seismic frequency range on which FWI is run. While the greatest benefit for model building is achieved from access to relatively low frequencies, increasing the frequency content in FWI helps in reservoir characterization.

Obtaining full-bandwidth, absolute elastic-attributes for lithology and fluid prediction requires a low-frequency model. The lower frequency component required for the modeling can be generated from velocities, assuming they contain sufficient resolution. This reduces the emphasis on the well and seismic information. This can make a vital difference in exploration settings. Less a prior input is an advantage in attribute prediction and ensures that the results of pre-stack seismic inversion are primarily data-driven.

Using a broader frequency content to obtain the FWI model can provide the lower frequency component for elastic property generation. Well information becomes a control point. This allows reservoir geoscientists to confidently derive reliable elastic attributes, such as acoustic impedance and Vp/Vs ratios, away from the well locations. 

Higher-frequency content in FWI further improves elastic attribute calculations for reservoir characterization and can be combined with broadband data input to minimize the need for well data
Higher-frequency content in FWI further improves elastic attribute calculations for reservoir characterization and can be combined with broadband data input to minimize the need for well data