Attributes is designed to create and extract a collection of advanced, multi-trace seismic attributes to enhance the seismic interpreter’s ability to analyse frequency content, reduce noise, detect fractures and other discontinuous features in seismic volumes. Single trace attributes are available in other HampsonRussell products.
Benefits of Attributes Include:
The attributes extracted in the package include a variety of curvature attributes, the phase congruency and energy-ratio attributes, and both peak and constant frequency analysis using empirical mode decomposition (EMD) and its extensions. The Attributes package contains a complete range of multi-trace edge-preserving filters, such as the Kuwahara and alpha trimmed mean filters for an optimal pre-conditioning of the seismic data.
Additional single trace attributes such as instantaneous envelope, phase and frequency, trace integration and differencing, and frequency slices are available in other parts of the HampsonRussell suite of programs.
Curvature is defined as the rate of change of the direction of a curve in two dimensions. It can be extended to a 3D surface as shown in the figure. The two curves shown here are the maximum curvature, Kmax, and the minimum curvature, Kmin. Both are available within Attributes.
Other curvature attributes include:
Associated with the dip and strike curvature attributes, the dip and azimuth angles can also be computed.*
The phase congruency algorithm, developed by Kovesi (1996) was initially used in robot vision to detect edges. The fundamental idea behind this algorithm is to find the point at which the phase of the various frequency components in a dataset are equal, or congruent.
This can be visualized most easily in a 1D Fourier analysis, but it is more rigorously done in this approach using a 2D Fourier analysis.
The key steps are:
The phase congruency attribute applied to the time slice of the “Example dataset”. Notice the definition of the discontinuities shown by the phase congruency attribute.
The energy ratio attribute provides a different way of revealing geological discontinuities such as faults on the seismic volumes. Like phase congruency it can reveal subtle fractures. It is computed as follows:
Left: The amplitude time slice at 1000 ms. Right: The 1000 ms time slice of the energy ratio attribute computed for the entire data volume.
The empirical mode decomposition (EMD) algorithm is a spectral analysis technique which has been found to give better results than the instantaneous frequency, short-time Fourier transform (STFT) and wavelet transform techniques currently in use within our industry for time-frequency analysis. The EMD algorithm decomposes the seismic data into a series of intrinsic mode functions (IMFs), which in turn can be interpreted as the localized frequency content that we wish to visualize.
Limitations in the EMD algorithm inspired the ensemble empirical mode decomposition (EEMD) and complete ensemble empirical mode decomposition (CEEMD) algorithms, both of which are found in the Attributes package. These methods improve and stabilize the original EMD algorithm. The figures below show the CEEMD algorithm (left), short-window Fourier transform (middle) and instantaneous frequency (right) results from the same time slice, all computed within HampsonRussell software. Although all three show a similar overall response, the CEEMD result shows more realistic detail in the bottom part of the map. Han, J. and Van der Baan, M. 2013. Empirical mode decomposition for seismic time-frequency analysis. Geophysics, 78 (2), p. 9-19.
* Roberts, A., 2011. Curvature attributes and their application to 3D interpreted horizons. First Break, 19 (2), February 2001, p 85-99.
Contact us by opening an online inquiry.