Target-oriented, Least-squares Imaging of Blended Data

Summary In traditional seismic acquisition, temporal overlap between the recordings of cons ecutive shots is avoided. In this paper we will investigate the proposal to abandon this acquisition constraint by introducing dense source distributions in an overlapping fashion (’blended acquisition’). The result is an improved source space with a favorable azimuth distribution. We will show that blended measurements can be directly used in a target-oriented migration process with a multi-shift least-squares imaging condition, such that a densely sampled illumination is obtained in the ray parameter domain. As a deblending step is avoided, it enables an efficient imaging process for blended data. The method will be illustrated with numerical examples.