Seismic source reconstruction in an orthogonal geometry based on local and non-local information in the time slice domain

Abstract

3D land seismic acquisition using an orthogonal geometry allows for obtaining high lateral image resolution. One of the main challenges for this kind of seismic acquisition is to capture high-quality images minimizing both cost and environmental impact. This work presents an approach for recovery missing wavefields due to the elimination of seismic sources which is based on local and non-local prior information of the 3D seismic data. More precisely, in an orthogonal arrangement of sources and receivers, one or multiple missing wavefields are recovered by exploiting the local smoothness of the data and the non-local self-similarity exhibited by the 3D seismic information across every time slice. Furthermore, an algorithm based on the alternating direction method of multipliers (ADMM) is developed for solving the recovery problem. Numerical experiments conducted over synthetic and real datasets demonstrate that the proposed recovery method leads to recover reliable versions of the missing wavefields whose traces preserve the seismic events, compared with state-of-the-art recovery approaches. Moreover, the proposed approach yields recovery improvements of around 5 dB in terms of SNR compared to the performances obtained by other methods.