Offset-Dependent Predictive Deconvolution

Predictive deconvolution depends on the two important parameters, prediction distance (alpha) and operator length (N), and is commonly used to attenuate reverberation, short path and also long path multiples in seismic data. These parameters are generally used as constant along offset in routine processing. In this study, an offset-depended application procedure is developed in choice of the two parameters. It was observed that the use of the parameters based offset increased the performance of the predictive deconvolution rather than the constant parameters, leading to better stack and interpretable sections.