The seismic aberrancy attribute measures the lateral change of curvature along a surface, which complements other seismic geometric attributes such as coherence and curvature. Aberrancy provides a means to map the trace of faults that appear as flexures that may have insufficient offset to delineate with coherence. Generally, we first compute the vector dip using the full-bandwidth seismic amplitude volume. Aberrancy is then computed as the second derivative of the vector dip in the three different directions. Due to either seismic data quality or to the underlying geology, certain spectral components of the seismic amplitude volume often appear higher quality than others, which further result in higher quality geometric attributes. We develop a multispectral aberrancy method to further improve the imaging quality of small-scale geologic features. We first decompose the full-bandwidth seismic data after data-conditioning into different spectral voices, to build the multispectral covariance matrix. Next, we compute the eigenvectors and eigenvalues from the multispectral covariance matrix, followed by the generation of inline and crossline dip volumes. Finally, we provide the multispectral aberrancy by computing the second derivative of the vector dip and rotating the coordinate system. We evaluate the proposed multispectral aberrancy method using a 3D seismic dataset imaging complex channel reservoirs.
Presentation Date: Tuesday, October 13, 2020
Session Start Time: 1:50 PM
Presentation Time: 4:45 PM
Location: 361A
Presentation Type: Oral
