Application of X-band radar to sense hydrocarbon seepage

Thomas C. Bailey
Premier E&P Co.
Houston

Experiential data

There are two published surveys, U.S. Department of Energy (Fig. 1 [ bytes]) and Sun Exploration & Production Co. (Fig. 2 [ bytes]), in the literature,^4-5 but these are inadequate.

The DOE survey is a small area with few anomalies, few wells productive or dry, and one oil and gas field. The Sun survey is better. It is a larger area with a large number of wells and several fields but with few radar anomalies.

Airborne Petroleum Surveys of Dallas agreed to make available a 240 sq mile proprietary survey flown in the Hardeman basin in 1990. The survey, in the eastern part of the basin, covers a very large area and contains a large number of anomalies, wells dry and productive, and oil and gas fields (Fig. 3 [ bytes]). It has a shortcoming: Only one of the radar anomalies has been drilled since the survey.

One thing apparent from the survey is that anomalies detected by airborne radar display characteristics that seem distinctive from other seepage tools. Most seepage tools, soil gas being an example, map seepage of one magnitude or another almost everywhere. Consequently, maps show widespread seepage of varying intensity. Radar anomalies to the contrary show discreet seepage of limited extent and numbers of areas (Figs. 1, 2, and 3). Anomalies are the exception rather than the rule. This is more akin to actual oil and gas fields. Radar anomalies appear to coincide well with traps.

Fig. 4a [ bytes] is the radar anomaly of Bluebird J sand field in Lincoln and Elbert counties, Colo., just west of the Sun survey. It is a stratigraphic trap formed by a bar sand encased in shale. The radar anomaly matches the trap, the sand, surprisingly well. The northwest limb of the radar anomaly is undrilled.

Fig. 4b [ bytes] is the radar anomaly at Sorrento field taken from the Sun survey. It is a stratigraphic trap formed by a valley-fill, i.jpg. sand-filled, channel. There is only weak nosing structurally. The radar anomaly encompasses most of the productive wells.

Radar anomalies, depending on the surveying company, are usually categorized as Minor, Moderate, or Significant in intensity. Areas of no anomaly are called Background. Minor is best interpreted as a transition into an anomaly or weak seepage or "noise." Moderate and Significant are considered to be bona fide seepage-related anomalies.¹⁶

There are two obvious caveats in the statistical data presented here.

First, it is assumed that in any anomaly with a dry hole, the well penetrated deep enough to condemn the anomaly. This is obviously not always true. A hydrocarbon accumulation could be deeper than the well drilled.

Second, active producing fields are still seeping hydrocarbons to the surface and thus should be anomalous. This is also not necessarily true. It has been shown that seepage decreases and disappears as production approaches depletion.¹⁷ It is not uncommon for very mature fields to appear as 'minors' when surveyed with radar, or have no anomaly.¹⁶

Two sets of statistical data were compiled from the Hardeman basin. One set is post-drill, a "benchmark" set, and other pre-drill, a "benchgauge" set. The former measures how well radar works under known situations while the latter measures how well it can be used to make drill-or-not-drill decisions.

The oldest well control map the author could find for the Hardeman basin was 1992. On the 1992 map there were 705 wells in the survey area, 408 dry holes and 297 producers. The survey mapped 48 radar anomalies, 18 minor, 29 moderate, and one significant (Fig. 3). Twenty-seven radar anomalies were associated with producing fields. Twenty-one radar anomalies were undrilled.

Since 1992, 27 wildcats have been drilled; 21 were dry holes, and seven were producers. One radar anomaly was drilled; it was a producer. None of the dry holes was located in a radar anomaly.

Benchmark data

Two measures were tabulated. One is how many active fields were anomalies at the time of the survey (Table 2 [2902 bytes]) and the other is how many anomalies encompassed only dry holes (Table 3 [3110 bytes]).

The benchmark data hint at a correlation of hydrocarbon accumulations and radar anomalies and that anomalies are not random. The coincidence of radar and existing fields is 42%. The real coincidence is probably higher. Some fields are one-well (too small for radar surveying), some fields are near depletion (seeping very little), and others may not be depleted (not seeping at all). If you just exclude the one-well fields shown in Fig. 3, the coincidence in Table 2 would rise significantly. There were no radar anomalies that contained only dry holes, yet there are a large number of dry holes distributed throughout the area.

Benchgauge data

There are two measures to determine how successful radar is in indicating hydrocarbon accumulations. One, if a moderate or significant anomaly is drilled, was it drilled successfully? Corresponding to this, how many wells drilled off anomalies were successful?

Radar success was 100%, but only one anomaly was drilled between 1992 and 1995 (Table 4 [3678 bytes]). This is a major shortcoming of this survey. There were six discoveries classified as "off anomalies." Three of the six called off anomalies were drilled in minor radar anomalies. Only moderate and significant intensities were considered radar anomalies. The remaining three wells are in background areas but appear to be only one-well fields.

It has been suggested that correlations with existing fields can be explained because of the steel or increased seepage created by the well bores or surveyor bias? These seem unlikely or all fields would be anomalies. Fields that do not show up are best explained as depleted or nearly depleted or are too small for the radar's resolution.

Amoco-NRL program

Amoco contracted with the Naval Research Laboratory in 1991 to assess radar for hydrocarbon mapping.¹² The purposes of the Amoco-NRL field program were:

1. To quantify the phenomenon;

2. To find out the effects of environment; and

3. To learn if the NRL theory had validity.

The NRL with its more sophisticated ground-based radar was to characterize and quantify airborne radar anomalies. An atmospheric physicist was to gather environmental data. Lastly, an analytical company was to determine if water and hydrocarbon gases were present in the radar anomalies.

The effort quantified the phenomenon for the first time and gathered some environmental data but failed to gather sufficient data important to proving the NRL hypothesis.

The Naval Research Laboratory obtained time-variant records of a radar anomaly. They confirmed time-varying turbulence-like anomalies first seen with Robert Owen.⁶ These are shown in Figs. 5a [ bytes] and 5b [ bytes] as a series of stacked traces covering one second of radar-echo amplitude versus range.

Fig. 5a shows time-variant activity over an undrilled airborne anomaly. Fig. 5b shows time-variant activity over a producing field. The region over which anomalous echoes fall lies between 750-1,500 ft on both figures. Left undetermined was what the atmosphere contained, hydrocarbons, some other gas(es), or water vapor. Scheduling problems prevented the analytical company from being present, so no chemical measurements were made.

Conclusions

There is a plausible hypothesis to explain how radar is applicable to oil and gas exploration. The hypothesis is yet to be scientifically proven. Empirically, the data from a large survey point to a relationship between radar and the presence of hydrocarbons. Further assessment is needed.

An important and often overlooked aspect is tying the radar anomaly to the stratigraphic depth of the seeping trap. It should not be assumed that the anomaly is coming from the stratigraphic interval being explored for or typically productive in the area. Also, because water vapor may be the reflecting medium, there are scenarios for water vapor entering the atmosphere not related to hydrocarbon accumulations, buried decomposing organic material creating false anomalies such as a landfill.

Acknowledgments

Thanks to: Ken Montgomery and Ed Hodges of Airborne Petroleum Surveys, Dallas, for sharing their knowledge and experience and providing access to the Hardeman basin proprietary survey; Don Hemenway and Pete Hansen of NRL for sharing their knowledge; Dr. Owen Phillips of Johns Hopkins University for recognizing the importance of atmospheric turbulence.

References

1. Klusman, Ronald W., Soil gas and related methods for natural resource exploration, John Wiley & Sons, New York, 1993, 483 p.

2. Owen, R.L., and Busby, J.M., Method for exploring the surface of the earth with electromagnetic energy including comparing reradiation characteristics of gases to locate escaping hydrocarbon gases at the surface emitted by deposits of petroleum and/or natural gas at depth, U.S. Patent 3,651,395, Mar. 21, 1972, 5 p.

3. Gournay, Luke S., Hydrocarbon exploration with display of re-radiated and reflected microwave energy, U.S. Patent No. 4,100,481, July 11, 1978, 5 p.

4. Heemstra, R.J., Ray, R.M., Wesson, T.C., Abraham, J.R., and Moore, G.A., A critical laboratory and field evaluation of selected surface prospecting technologies for locating oil and gas, DOE Bartlesville Energy Technology Center RI-78/18, 1979, 84 p.

5. Heimberg, George, and Leach, Richard A., Results of an integrated unconventional exploration program in southeastern Colorado, American Society of Photo Grammetry American Congress on Surveying and Mapping (San Antonio), Convention Technical Papers, 1984, pp. 587-605.

6. Skolnik, Merrill I., Radar detection of hydrocarbon gas seepage associated with underground oil and gas deposits, Naval Research Laboratory Memorandum Report 6245, 1988, 45 p.

7. Skolnik, Merrill I., Introduction to radar systems, McGraw-Hill, New York, 1980, 581 p.

8. Cook, Robert L., Molecular microwave spectroscopy: Encyclopedia of Physical Science and Technology, Academic Press, San Diego, Vol. 10 (Mete-Nond), 1992, pp. 231-277.

9. Dunford, H. Brian, Elements of diatomic molecular spectra, Addison-Wesley (Reading), 1957, 160 p.

10. Ottersten, Hans, Radar backscattering from the turbulent clear atmosphere, Radio Science, Vol. 4, No. 12, 1969, pp. 1,251-55.

11. Tatarski, V.I., Wave propagation in a turbulent medium, McGraw-Hill, New York, 1961.

12. Hemenway, Donald F., Hansen, James P., and George, Eugene G., Radar sensing of petroleum seepage gases, Naval Research Laboratory Memorandum Report 5300-93-7177, 1993, 25 p.

13. Reitsema, R.H., Gas composition in a well column (abst.), AAPG Bull., Vol. 69, No. 2, 1985, p. 300.

14. Biryulin, V.P., Golubev, O.A., Mironov, V.D., Popov, A.I., Nazarov, I.M., and Fridman, Sh. D., Geochemical prospecting for oil and gas by remote laser spectrometry of methane in the air at ground level, International Geology Review, Vol. 23, No. 6, 1981, pp. 679-682.

15. Price, Leigh C., A critical overview of and proposed working model for hydrocarbon microseepage, USGS Open-File Report 85-271, 1985, 83 p.

16. Hodges, Ed., Airborne Petroleum Services, personal communication.

17. Tedesco, Steven A., Surface geochemistry in petroleum exploration, Chapman & Hall, New York, 1995, 206 p.

Copyright 1996 Oil & Gas Journal. All Rights Reserved.