Adaptive Groundroll Filtering

The attenuation of Surface Waves while maintaining a "friendly" preservation of the Body Waves is a difficult goal to achieve in land data processing. However, Groundroll (GR) characteristics can be extracted from the input data themselves to feed an adaptive filtering in order to remove the Surface and Guided Waves. This approach is based on a cascaded Elastic Modeling of the signal and noise in the FX domain that uses the GR characteristics of several frequency bands. The part of the model corresponding to the noise is then subtracted from the data using a least squares approach. This method has several advantages over techniques such as FK filtering because it better preserves the signal and works even when the noise is aliased, dispersive and has irregular spatial sampling.