The seismic data analysis applications developed by G&W Systems are a part of the G&W Systems Geosciences Suite™. They allow integrated and interactive management, processing, and rapid visual inspection of multi-volume seismic data.

All data if a project are available for 2d and 3d visualization in an integrated, multi-windowed environment. Data analysis or processing results are instantly available for visual inspection and QC trough a comprehensive set of 2d/3d data display tools and flexible, multi-dimensional data overlays.

The seismic data analysis toolkit incorporates a well data management and display system that lets the user create all kinds of combined analysis displays and allows for combined geostatistical evaluation of the data.

1.- Structural Attributes:

This application allows to generate volumes of local dip and dip direction attributes using a full 3d semblance-weighted averaging algorithm, that creates small analysis surfaces through the sample of interest in all directions at all dip angles and measures amplitude semblance. The results of this measurement are averaged to obtain estimates of the prevailing dip and dip direction. In parallel this process yields an estimate of the local coherency character of the seismic data.

Structural analysis results can be used to apply steered multi-trace alpha-trim filters to seismic data. The application allows to incorporate coherence information into the filtering process to create structurally smoothed datasets with an improved resolution of fault cuts.

Textural seismic attributes describe the variability of the amplitude content of seismic data that occurs on small spatial scales. Amplitude co-occurrence is measured at small correlation distances (1-3 samples) throughout a seismic volume and the co-occurrence character of seismic amplitudes is referred to as the ‘texture’ of the data. Textural analysis of seismic data is based on the statistical evaluation of amplitude co-occurrence matrices.

4.-Neural network-based multi-attribute classification

G&W Systems’ neural network-based multi-attribute classification technology can be used to generate classification attribute volumes. Based on a supervised learning procedure the system extracts user-defined feature classes from seismic volumes. As the classification is carried out using multiple estimators, it is possible to simultaneously create standard deviation volumes that give a measure of certainty in the transformation.

Data analysis capabilities also include:

• Single trace basic processing, like gain, filter, or deconvolution applications
• Single trace seismic attributes (complex trace analysis)
• Spectral decomposition
• 2d attribute crossplot classification.