Source wavelet and its angular spectrum from plan‐wave seismograms

Abstract

In a typical seismic experiment (land or marine), the generated source wavelet is either unknown, or known only approximately. In order to solve many geophysical problems, an accurate wavelet estimation is crucial. This paper demonstrates a fast numerical algorithm which allows not only estimating the source wavelet but also its angular spectrum (radiation pattern) from plane‐wave decomposed seismograms. Surprisingly, literature on this subject is virtually missing, although a nonuniform angular spectrum of the generated wavelet can substantially affect the recorded data. The technique presented here is based on the downward continuation (DC) and reverse‐time extrapolation (RTE) of the recorded data (both pressure and its vertical derivative). The proposed method is iterative: The rate of convergence does not depend on phase characteristics of the generated wavelet. We demonstrate that both the wavelet signature and its angular spectrum can be estimated accurately from plane‐wave seismograms. We also illustrate the technique on a variety of wavelets with different amplitude/phase characteristics.