Prediction and removal of internal multiples under strong reflectors by using inverse scattering series algorithm

Abstract

The inverse scattering series (ISS) internal multiple attenuation (IMA) algorithm is modified and improved by identifying certain multiple generators to remove internal multiples under the strong reflectors. The ISS IMA method is a fully datadriven method that can predict all internal multiples for all horizons at once, without requiring any subsurface information. However, data-driven method requires high-quality data and it’s also high computational cost. Therefore, by identifying certain strong multiple generators, the IMA method can reduce computational cost and improve efficiency to predict and remove the internal multiples under the strong reflectors in the target reservoir. The modified IMA method provides added value for attenuating the target internal multiples. It is applied on synthetic and field data to remove internal multiples for some specified strong generators. Test results demonstrate that the modified IMA method can remove effectively internal multiples step by step for the target reservoir.