Improving Complex structure Image by Reverse Time Migration in Mountainous Areas

Abstract

The Reverse Time Migration (RTM) has the advantage of combining both the benefits of oneway wave equation migration for multi-arrivals and the Kirchhoff migration for steep and overturned reflectors. Another advantage of RTM is the usage of amplitude handling. However, all migrating algorithms are generally applied to marine seismic data, or shifting data from rugged acquisition topography onto a reference flat surface. In areas of smooth topography where the near-surface velocity is much slower than the subsurface velocity, a static shift is adequate for the transformation with small distortions onto the imaging. However, if the near surface velocity is comparable to the subsurface velocity, the static shifts are big and this approximation becomes inadequate. Specifically, the image distortions become severe when geological structures with rugged acquisition topography have much steeper dips. Under these circumstances, a static shift distorts the wavefield and degrades the velocity analysis and imaging. Therefore, topographic RTM is necessary. However, other considerations are also required, such as vertical grid size, migration apertures, and source wavelet peak frequency for suppressing artifacts, since these factors have an effect on image quality in terrestrial RTM. In this paper, we first describe the errors resulting from the static correction for rugged topographic changes. We then show the imaging distortions by migrating from the redatumed planar surface by using the prestack RTM migrations. Also, we demonstrate that RTM from topography is required in rugged acquisition topography in mountainous regions. Furthermore, it is important that vertical grid size, migration apertures and source wavelet peak frequency need to be optimized when using RTM in land seismic data for improving image quality in the shallow layers. In our experience, the parameters using RTM in land differs greatly from those used in marine environments.