Schlumberger

Multiple Attenuation

Overview Library

An extensive portfolio to address your multiple-elimination challenges

Reducing multiple contamination in data is one of the greatest challenges in seismic processing, and no single approach is suitable for all scenarios. We offer an extensive portfolio of prediction and adaptive subtraction methods that address surface and internal multiples. Based on your survey challenges and processing goals, appropriate methods from this portfolio can be used to obtain the optimal solution.

Prediction methods

Currently, data-driven algorithms are the preferred prediction methods for modeling internal and surface multiples. When the prerequisites for these algorithms are difficult to satisfy, model information or transform methods are used instead.

3D GSMP General Surface Multiple Prediction

General Deterministic Water-Layer Demultiple (GDWD)

Extended Internal Multiple Prediction (XIMP)

Industry-leading algorithm that can handle any acquisition geometry and uses both seismic data and interpreted horizons to provide a true-azimuth internal multiple model—particularly important for full-azimuth land acquisition.

Inverse-Scattering Internal Multiple Prediction (ISIMP)

Wavefield Extrapolation Multiple Modeling (WEMM)

Weighted Least-Squares (WLS) Radon Transform

Adaptive subtraction methods

Modeling multiples is followed by subtracting the model from the acquired data. Acquisition geometry, sampling imperfections, algorithm assumptions, low signal-to-noise levels, or complex geology can cause prediction methods to produce models of multiple energy that do not precisely match the multiple events in the data. Consequently, these models cannot be subtracted from the seismic dataset without correcting timing, amplitude, and phase errors.

Least-squares simultaneous subtraction
This application performs adaptive matching and subtraction of one or more noise models by using least-squares-derived temporal filters to simultaneously match all the models to the input seismic data. Matching filters are calculated in different time and space windows within a given input gather.

Advanced adaptive subtraction with curvelet transform
This iterative workflow makes use of the ability to separate primary energy from the multiple model to derive an effective matching operator. Combining traditional least-squares adaptive subtraction with the complex-value-based curvelet domain representation of the seismic data makes this an effective method for attenuating complex surface multiples.

Request More Information
WesternGeco