EXPLORATION

Using gas chimneys as an exploration tool

Part 2 – Field case histories are discussed that use the principles of gas chimney identification and interpretation that were presented in Part 1.

F. Aminzadeh, dGB-USA, T. Berge, Forest Oil Co., P. de Groot, dGB BV, and G. Valenti, AGIP

he use of gas chimneys for hydrocarbon exploration in southern Africa and the Gulf of Mexico is illustrated. By using new neural-network-based software, gas-migration pathways can be discerned and probable hydrocarbon seeps located on the seafloor can be characterized. Results from a successful exploration campaign are presented, as are several views of gas chimneys.

South African Example

Africa has three extensive, thick-sectioned deltaic systems on its west coast. Of these, the Congo and Niger deltas have major hydrocarbon resources. The Orange River delta is less explored, but has gas and oil discoveries that indicate its potential. These discoveries point to at least three hydrocarbon systems in the area: 1) in-graben deposits of the early rift succession, e.g., the AJ1 oil well (discussed below); 2) late rift, Kudu-type (gas); and 3) the Albian-Aptian drift succession (gas). The deepwater deltaic portion has not been drilled, nor has its potential been assessed.

The focus here is on a 312-km², 3-D seismic survey in Block 2A, centered on a gas discovery, the AK1. This well allowed a detailed, first look at a small part of the delta. The original discovery well was plugged and abandoned, since it was thought to be a noncommercial structural trap. The 3-D survey showed that the field, designated Ibhubesi, is in fact a giant stratigraphic trap. The 3-D area might only cover a small southerly portion of the field, which may ultimately produce 15 Tcf of gas.

Attribute processing and gradient analyses of the chimney volume clearly show gas accumulations in meandering, fluvial channels and other component facies of fluvial-deltaic systems. Fluvial channels, meander belts, crevasse splays, overbank deposits, tributary systems and deltas all can be identified in a 3-D view of the reservoir, Fig. 9.

Fig. 9. 3-D view of Ibhubesi Field reservoir interval. Visualization of chimneys within the structure. Click for enlarged view

A map view of this same 3-D volume is shown in Fig. 10, but this time showing (in red) the chimney cube attribute with seepage anomalies along the main fault. When viewed in a vertical section, these gas seepage anomalies are seen originating at the reservoir level and ascending to the surface, Fig. 11.

Fig. 10. Secondary migration of gas. Click for enlarged view

Fig. 11. Vertical section: Gas seepage anomalies (red streaks) are seen originating at the reservoir level and ascending. Click for enlarged view

A four-well drilling program was undertaken to evaluate the field and prove up enough reserves for development. Three different anomalies were targeted; each well will test individual compartments with 300 to 900 Bcfg, yielding total reserves of 3.1 Tcfg.

To date, the first three wells have been completed, and results have been highly successful. The AK2 well tested 30 MMcfgd and over 600 bpd of condensate from a 20-m-thick pay sand. FTP was 2,200 psi on a 3/4-in. choke. Neither significant pressure drawdown nor produced water was observed during the 12-hr test. Reservoir characteristics were better than expected: a clean and well-sorted matrix with average porosity of 22 to 27%.

A deeper, 15-m gas-bearing sand of similar quality to the AK2 reservoir rock was targeted with well AV1, but the drillstring twisted off; that sand was subsequently penetrated in a sidetrack. This hole is interesting, because (besides confirming additional reserves) the lowest gas sand in AV1 is deeper than the lowest proven gas and highest known water in the AK1 well, clearly showing that it is a stratigraphic trap and separate reservoir. Thus far, results indicate that the initial vision of a giant, regional stratigraphic trap is being proven.

Chimney analysis made a significant contribution to the interpretation and validation of earlier work. Fig. 12 shows how chimney analysis is integrated with conventional seismic processing and how seepage chimneys appear on a conventional reflection display.

Fig. 12. Conventional seismic display showing gas chimney with a time pull-down (along arrow). Click for enlarged view

Results after training the neural network with known and hand-picked chimneys and nonchimneys obtained the chimney probability shown in Fig. 13, which displays chimney output for inline 2800, compared against the original seismic. Fig. 14 displays the distribution picks from different time gates. RMS error for training and test data sets is shown in Fig. 15.

Fig. 13. (A) Conventional seismic display for inline 2800. (B) Same inline; displayed is chimney-probability from neural network. Click for enlarged view

Fig. 14. Distribution of selected training locations (positive and negative picks) vs. time-depth.

Fig. 15. For chimney determination: neural network RMS error for the training and test set versus training cycles.

Fig. 16 shows time slices and highlights major chimney-like features. These features reach the seafloor along the fault trace, where there is a 40-m-high, ridge-shaped mound formed by a chemotropic community of organisms that live on methane seeps.

Fig. 16. Neural network chimney prediction at 2,100 ms. Compare with Fig. 10. Click for enlarged view

Gulf of Mexico Example

In this case study, the prospect area is the continental shelf edge, western offshore Louisiana. Here, thick sand / shale sequences were deposited in OCS-to-upper-slope paleo-environments. Prospective targets range in age from early-to-late Pleistocene at about 13,500-ft subsea and shallower (C. Macintyrei to Trimosina A datum). Both gas and oil discoveries are predicted. Several producing Pleistocene fields are within about six mi of the area.

The depositional setting is that of a mini-basin that was created by evacuation of a deeply buried salt sequence. One margin of the mini-basin is cut by a north-south trending fault. This fault system cuts the seafloor and, as seen on seismic, has raised zones on the seafloor in several places along its extent. Such features have been attributed to active seafloor gas vents.

Chimney cubes were used to document the locations and subsurface extent of probable active gas leaks. The chimney cube method appears to provide good resolution of vertical distribution of gas and gas-saturated fluids within the 3-D volume. This distribution information, combined with structural mapping of prospective horizons, defines hydrocarbon-migration pathways and potential traps. A number of targets have been defined that lie updip to these zones of active, vertical hydrocarbon migration.

Regionally, oil and gas production has been documented as taking place adjacent to active fluid-conductive faults. For that reason, documentation of active gas conduits using chimney cubes can help define gas prospects. In this GOM example, seismic chimneys – related to hydrocarbon-migration paths – are evident in the raw seismic data as a vertical disturbance of the response, Fig. 17A. The chaotic reflection pattern is characteristic of chimneys. In such areas, both energy and trace-to-trace similarity decrease. These chaotic reflections, attributed to chimneys, are highlighted in Fig. 17B. Fig. 18 shows the neural network chimney prediction at 1,000 ms.

Fig. 17. (A) Chaotic character of seismic section. (B) Neural network chimney prediction of same section. Click for enlarged view

Fig. 18. Neural network chimney prediction at 1,000 ms.

Chimney shapes vary considerably. Some are cylindrical (e.g., above a mound), while others are elongated or curved (e.g., along fractures, faults and paleo-channels). Applying the trained network on a trace-by-trace and sample-by-sample basis yielded the chimney prediction cube, Fig. 19. The detailed display reveals a possible mud volcano at the seafloor above a seismic chimney.

Fig. 19. (A) 3-D display of a chimney cube in the Gulf of Mexico. (B) Close-up view shows a possible mud volcano on seafloor above a chimney-like pattern. Click for enlarged view

Acknowledgment

The authors appreciate the contributions of Statoil, specifically those of Roar Heggland and Paul Meldahl to the development of this technology and extensive insight and collaboration on earlier projects. The authors also acknowledge Herald Ligtenberg of dGB for his contributions to gas chimney processing. Forest Oil and Agip are thanked for granting permission to present their examples and data.

The authors
	Fred Aminzadeh, President and CEO of dGB-USA (faz@dgbusa.com) worked for Unocal for 17 years. He has authored many books and articles on different aspects of geophysical technology, including modeling, pattern recognition, seismic attributes and seismic data processing. He also holds three patents on AVO modeling, seismic-while-drilling and hybrid reservoir characterization. Mr. Aminzadeh earned his PhD from the University of Southern California, and has served as chairman of SEG’s research committee for two years.
	Paul de Groot is co-founder and director of dGB (de Groot-Bril Earth Sciences B.V.), which specializes in quantitative seismic interpretation, stratigraphic analysis, seismic inversion, neural networks-based reservoir characterization and gas chimney / fracture detection. He began his career in 1981 with Shell, then became director of Quest Geophysical Services and subsequently, served as senior geophysicist for TNO Institute of Applied Geoscience before co-founding dGB in 1995. Mr. De Groot holds MSc and PhD degrees in geophysics from the University of Delft.
	Timothy Berge, chief geophysicist at Forest Oil International, has a Masters degree in geology from the University of Texas at Austin. He has worked for Exxon for 17 years in Denver, Midland, Bogota, Colombia, and Houston; for Corpoven in Puerto La Cruz, Venezuela; and with Apache in China. His technical expertise includes the areas of structural geology, sequence and seismic stratigraphy, workstation applications and 3-D interpretation. Mr. Berge is the author of numerous publications, and is SEG’s Global Affair’s committee’s regional coordinator for Africa.
	Giuseppe Valenti holds an MS degree in geology from the Universitá of Parma, Italy, and joined Eni, AGIP Division, in 1988. After a few years spent in geophysical research and seismic stratigraphy application, he moved to West Africa (Angola) and then to the U.S. (Gulf of Mexico) to concentrate on deepwater E&P.

Part 1