Extended Internal Multiple Prediction (XIMP)

Overview Library

Address the challenge of multiples that closely resemble the primaries

Internal multiples are usually created by seismic reflections bouncing between closely spaced geological beds, resulting in multiples that have little—if any—moveout or velocity difference compared with the primaries that generated them. These multiples mainly contaminate land datasets. Because they are difficult to differentiate from the primaries, especially in areas with relatively flat geology, they have been more difficult to address than surface-related multiples.

Predict internal multiples at true azimuth

Extended internal multiple prediction (XIMP) is a data-driven multiple-modeling algorithm that predicts internal multiples from recorded events using wavefield extrapolation based on the Kirchhoff integral. The method does not require the prior velocity function to differentiate between primaries and internal multiples. It only requires the onset times of the multiple-generating events, which can be determined using interpretation and VSP and well data.

In media with moderate structural complexity, XIMP is able to handle any acquisition geometry and predict multiples at true azimuth. This is particularly important for full-azimuth land acquisition. XIMP provides a true-azimuth internal multiple model and an accurate estimate of the multiple energy in the data. The multiple energy can then be removed from the data by using adaptive subtraction techniques. Using XIMP in conjunction with 3D surface demultiple techniques (e.g., 3D GSMP general surface multiple prediction) results in a much better representation of the primary energy.

Related services and products

Request More Information

XIMP: A Data-Driven Demultiple Algorithm

Extended internal multiple prediction (XIMP)Poststack time migration (PSTM) Example gathersExample semblances
PrevNextZoom1 of 4