Improving frequency domain deconvolution at low frequencies

Sina Esmaeili, Gary F. Margrave

Finding the method to obtain the best reflectivity is always challenging. The accuracy of estimated reflectivity directly depends on how precise embedded wavelet is estimated. The frequency domain deconvolution algorithm, which is one of the deconvolution methods, estimates the reflectivity function via extracting wavelet from seismic data in frequency domain. This algorithm can estimate the wavelet by smoothing amplitude spectrum of data. It should be noted that the smoothing process in the deconvolution algorithm can affect the resulted reflectivity.

The purpose of this study is to investigate the performance of standard deconvolution and its ability to estimate reflection coefficient content directly from seismic data. We found that there could be different ways to smooth standard deconvolution. Three different smoothers will be used to smooth the spectrum of data to estimate the reflectivity. A synthetic zero-offset and shot gather data will be created and each of the deconvolution algorithm will be applied to them.