What's new in exploration

I have heard that “good” is the enemy of “best,” and “good enough” is the road to failure. Since we are often taught the 80% solution, and interpreters are never 100% correct anyway, how often then do we accept “good enough” for oil work? This could be why we left behind so many reserves, and ignored shale oil all of the last century.

Compulsory skills. I know folks who graduated from college and have never opened a book on their own since. I suggest taking this time to refresh or reinvent yourself. It is ridiculous that most courses offered by service companies and professional organizations cost in excess of $2,000, when the people who could benefit the most are out of work. There may be a path around: volunteering for a professional organization.

Let’s look at my wish list of skills still needed by E&P:

Geology: I wish paleontology was still taught. Microfossils that were studied 40 years ago would have limited the discussion on global warming and “new” mean seal level rising. It would still help in identifying rock parameters for the analytical reduction of reservoir characteristics. I suggest a focus on paleo-reconstruction. What is missing is just as important as what is present.

Geophysics: I wish we would move away from proclaiming geophysical expertise, because we use 3D seismic on a workstation. Geophysicists (or cross-over geologists) should know that there is no such thing as perfect field data. Assuming that what you are looking at on your workstation is without error is a big mistake. Almost no one graduating from today’s colleges will sit a full-scale seis crew or ride a boat. Most will never data process. It is up to you to guide the entire data collection and reduction methodology.

New land seismic acquisition and processing: I wish we could show all the possibilities that are showcased in the SEAM Phase 2 land project. One cannot have great data to interpret on that color workstation without knowing how to design the data survey in the first place. Knowing the perfect data (simulation) solution will then show you the limits of your interpretation (http://seg.org/News-Resources/Research-Programs/SEAM). New marine seismic was showcased in SEAM (RPSEA) Phase 1.

New petrophysical methods: I wish we could get more input from engineers, when we have multiple choices on which approach to guesstimate reservoir parameters, and subsequent calibrations to surface derived data. Included here are my paraphrased minutes from a subcommittee of the RPSEA-SEAM Pore Pressure Project, ongoing in spite of the downturn. Look at the complex decision-making involved in a “simple” model. Most of this discussion is from feedback from petrophysicists who work with their reservoir engineers, one on one.

Rock physics. We discussed two approaches for assigning rock properties from basin simulation outputs. One is to use an empirical approach like the one using the Eberhardt-Phillips et al relations. Currently, there are questions about this approach near the mud line, where the relations may not be valid, and extrapolating them to shallow depths could cause unrealistic velocities (they are negative). The general agreement is that the forms of the relations in Eberhardt-Phillips et al are appropriate (for GOM deep sands), but that coefficients used in the relations need to be calibrated for a given region.

The second approach to assign rock properties is to use a model-based approach. For either case, we need well logs to calibrate the results.

The Model Design Committee needs to address these topics:

1. Clay diagenesis: implement using an approach followed in previous 2D simulations.
2. Mineral precipitation: do not include this in the model. The sense is that the correct approach for modeling mineral precipitation is poorly understood.
3. Number of Vshale (volume of shale) classes to use: 10.
4. Mixing law to define properties of Vshale classes: use the simple arithmetic method. There is interest in the Thomas-Steiber mixing law, but the results obtained, by MEMBER, when they used it in their simulations, were considered to be suspicious. It generated much higher pore pressure compared to when arithmetic averaging is used. It also gave pore pressure results with less features than the centroid model predicted. However, the committee would like to explore this in 3D, so they recommended that, of the first two 3D simulations conducted, one be done using arithmetic averaging and another be done using Thomas-Steiber mixing. Note that the first two simulations will be done without implementing any changes to permeability along faults; e.g. fault properties will be the same as the material adjacent to the fault.
5. Stress to use in calculation of compaction: mean stress should be calculated from 3D stress tensors, calculated at each time step.
6. Permeability anisotropy: include this in the model using the documented approaches recommended by the vendors and as used in the previous 2D simulations.
7. Petroleum generation: do not include in the calculations. To have great impact, the petroleum source region would need to be thick, and the migration of petroleum would need to be included.
8. Heat flow: use values that have been used in the previous 2D basin simulations.
9. Fault permeability: we would like to evaluate models that have both open and closed faults.

I have learned in our numerous meetings, some of this is now—not later—technology. The rest of us have to catch up.

About the Authors

William (Bill) Head

Contributing Editor

William (Bill) Head is a technologist with over 40 years of experience in U.S. and international exploration.