Target-oriented adaptive subtraction in data-driven multiple removal

Abstract

Data-driven multiple removal methods have proven a valuable addition to the demultiple toolbox for three main reasons. Firstly, such methods do not make use of any a priori information about the subsurface geology and, therefore, such information cannot bias the solutions. Secondly, data-driven methods can be applied in either 1D, 2D, or 3D mode and can therefore account for the full, multidimensional complexity of the earth. Finally, because no a priori information is used, the required user interaction is minimized. These methods are now applied in production-style processing environments with comparable speed and turn-around as conventional techniques.