Resolution of Faults in the Eagle Ford, Part 3

Resolution of Faults in the Eagle Ford, Part 3

Prior to the analysis described in this and the related three Application Briefs (EF-1, EF-2, and EF-4), the client believed that the faults were not well resolved within the Eagle Ford on the base “amplitude” seismic survey or on views of a single similarity attribute. This brief demonstrates that faults can be readily resolved via analysis of multiple attributes simultaneously.

Principal Component Analysis in Paradise® was used initially to reduce an initial set of 25 geometric attributes to ten from the first two Eigenvectors. Five similarity attributes, which included Chaotic Reflection, Dip of maximum Similarity, Similarity, Smoothed Dip of Maximum Similarity and Smoothed Similarity, contributed the greatest variance to the result set according to the first Eigenvector. Curvature in the Dip and Strike Directions, Maximum Curvature, Mean Curvature, and Shape Index contributed the most variance within the second Eigenvector. For setting up the input to the Self Organizing Map (SOM) process, a recommended work process uses the most prominent 2-5 attributes that contribute the greatest variance among each of the largest Eigenvectors which may number 1-4 or 5. The combined contribution for the select set of attributes should comprise at least a cumulative 60% of the variance within the region of the PCA analysis.

A result illustrated in Figure 1 is from a geometric SOM made from these ten attributes visualized as a “ghost” on the Top Eagle Ford horizon, which has then been pushed down into the High Resistivity Eagle Ford shale objective. The multiplicity of faults that can now be seen defied expectations. Well 3H’s borehole encounter of six faults while drilling, which can be individually seen here, could not have been anticipated from the use of a single similarity attribute display such as Figures 2a and 2b.

seismic interpretation in the eagle ford 01

seismic interpretation in the eagle ford 02

Figure 3 shows results from the same Instantaneous SOM with the 2D Colormap formerly seen in Figure 2 in Paradise Application Brief EF-2 (PAB EF2). The figure is from an inline that transects the downdip portion of the geobody also shown there. The patterns seen in the neuron textures now reveal details in the structure of that body in the vertical section. Numerous faults, including many that are compressional in nature, can be interpreted that were not evident at all in the original seismic data.

seismic interpretation with SOMs in the eagle ford 03

Figure 4 shows a close-up of a single compressional fault with remarkable detail in the offset in the Eagle Ford shale. For this view, the favored 2D Colormap for an 8x8 topology, i.e., 64 neurons, called Map Shade Dark, is in use (see also Figure 1b in PAB EF2). Onlap of high resistivity greyish-gold facies are connoted by white left arrows. An underlying double red downlap is located by the black right arrow. Note that the former is evident as fill in the parting of the latter at the fault in the Eagle Ford. Offset is actually most subtle here in the Upper Eagle Ford marl.

seismic interpretation with SOMs in the eagle ford 04

All Seismic data owned and provided courtesy of Seitel, Inc.

Attribute Analysis in Unconventional Resource Plays Using Unsupervised Neural Networks

Attribute Analysis in Unconventional Resource Plays Using Unsupervised Neural Networks

 

Analysis of Unconventional Resource using Inversion attributes and seismic attributes

 

Key elements in understanding unconventional resource plays encompass the following categories:

  • Reservoir Geology: thickness, lateral extent, stratigraphy, mineralogy, porosity and permeability
  • Geochemistry: Total Organic Content (TOC), maturity (Ro-heat), and kerogen% (richness)
  • Geomechanics: acoustic impedance inversion, Young’s modulus, Poisson’s ratio (Vp/Vs) and pressures
  • Faults, Fractures, and Stress Regimes: coherency (similarity), curvature, fault volumes, velocity anisotropy (azimuthal distribution) and stress maps.

This case study involved a newly acquired 3D seismic volume in a fringe area of the Eagle Ford Shale Trend. The 3D is approximately 10 square miles and four wells had been drilled to date on 2D data previously interpreted. Two wells targeted the Eagle Ford Shale Formation, and another two wells were drilled for the Austin Chalk and the Buda Lime Formations. All four wells were drilled in normal-pressured reservoirs with mixed results when it came to quality shows and commercial production. After processing the 3D volume and the initial interpretation was completed, well results and logs were incorporated by the client to create critical inversion attributes known to assist in the assessment of the shale’s productivity. Attributes contributed by the client to the analysis were: Final Density, Lambda Rho, MuRho, Poisson’s Brittleness, Poisson’s Ratio, Shear Impedance, Brittleness Coefficient, and P-impedance. Additional attributes run for the analysis were: Spectral Decomposition volumes, curvature and similarity volumes, Instantaneous attributes and Amplitude-related volumes (Average Energy and sweetness). The zone of study was confined to roughly the Top Austin Chalk to the Top of the Edwards Lime, encompassing the Austin Chalk, Eagle Ford Shale and Buda Limestone, which was approximately from 1.2 to 1.6 seconds. In addition to the PSTM volume, the generated plus client-provided attributes used to highlight sweet spots included:

  • Attenuation
  • Bandwidth
  • Envelope Slope
  • Instantaneous Q
  • MuRho
  • S-Impedance
  • Trace Envelope
  • Young’s Brittleness

A 12 x 6 topology was used for the analysis, so there were 72 neurons training on the attribute information. Figure 1 is a time slice showing the interpreted “sweet spots” in the Eagle Ford Shale on the 3D from the SOM Analysis. multi attribute analysis for unconventionals Two wells had been drilled, targeting the Eagle Ford Shale Formation. One was drilled prior to the acquisition of the data, and had few shows. It was plugged as a non-commercial well. The second well had good shows in the horizontal section of the hole, but encountered mechanical difficulties during drilling and had to be temporarily plugged. Figure 2 is an arbitrary seismic line through the deviated borehole of the second well showing the anomalous zone in both the Eagle Ford and Buda Formations and the points at which the well encountered the shows. seismic attribute analysis for unconventionals In conclusion, SOM analysis proved to be complementary to the interpretation of the data. The company who owns this 3D is now planning on targeting the area with five additional wells in the coming year. The application of using SOM analysis using selected seismic attributes can dramatically reduce uncertainty and thus decrease exploration risk in unconventional reservoirs.