Imaging of time-lapse structural changes with linearized inverse scattering

Shahin Jabbari, Kristopher A. Innanen, Mostafa Naghizadeh

Inverse scattering theory has been used widely in many applications in seismology including time-lapse problems. The difference data during the change in a reservoir from the baseline survey to monitor survey is determined using the linear approximation of the Born series. The linear Born approximation is used to derive a forward operator, mapping the model or perturbation to the measured data, and an adjoint operator, mapping the measured data to the perturbation based on the work of Kaplan (2010). The reference medium in time-lapse problem is the baseline survey medium and the perturbed medium is the monitor survey medium. A difference data is formed by applying structural change in the baseline survey and subtracting it from monitor survey. As the reference medium is as complicated as the perturbed medium, some difficulties such as spurious multiples in the difference data are encountered. To eliminate these unwanted events extending the full version of the Born series is strongly suggested. This paper reviews the earlier work of Innanen and Naghizadeh (2010) to establish a basic work to investigate the role of higher order terms in removing the spurious terms in the reference data.