New method of aerial and surface radiometric prospecting for oil, gas

Donald F. Saunders, K. Ray Burson, Jim F. Branch, C. Keith Thompson
Recon Exploration Inc.
Dallas

• Mean Natural Radioelement Content of Sedimentary Rocks [34,614 bytes]

A new petroleum exploration method has been developed using surface and aerial gamma-ray spectral measurements. Formerly troublesome lithologic and environmental variables are suppressed by correcting potassium and uranium readings using a process of thorium normalization.

Normalized potassium shows characteristic low concentrations above petroleum deposits. Normalized uranium shows higher values than normalized potassium over petroleum and generally lower values elsewhere. These anomalies are attributed to effects of microbial consumption of microseeping light hydrocarbons.

Studies of National Uranium Resource Evaluation (NURE) program aerial gamma-ray spectral data covering portions of six states have shown characteristic normalized potassium and uranium anomalies above 72.7% of 706 oil and gas fields.

Additionally, an average of 27 similar untested anomalies were found for each 1,000 sq miles (2,600 sq km) covered. 1 Similar aerial gama-ray spectral data are available over large portions of potential petroleum areas of the U.S. including Alaska, and Australia.

Preliminary tests in two basins in Australia showed positive correlation between radiometrically favorable areas and known oil and gas regions. ²

Ground-based gamma-ray spectral measurements found the same types of potassium and uranium anomalies over all 12 fields evaluated. Since 1988, our research of surface radiometric data coupled with soil gas hydrocarbon and soil magnetic susceptibility surveys has resulted in discovery of three oil and gas fields in Concho County, Tex.

lntroduction

Petroleum explorationists have been experimenting with gamma radiation measurements is a prospecting method since the early 1950s.

Much of the work involved total-count gamma-ray measurements that did not identify the gammaemitting radioelements and seriously impeded the development of adequate chemical models for anomaly formation.

Availability of reliable field and aerial gamma-ray spectrometers since the late 1960s solved that problem and has allowed recent development of credible models based on the geochemical characteristics of the radioelements.

Potassium and uranium are more mobile than thorium and have been found to reflect the presence of petroleum at depth. Thorium is contained mostly in insoluble minerals and remains relatively fixed under most natural chemical conditions. Data interpretation often involved little or no compensation for several interfering variables including lithologic type (or soil) being sampled, soil moisture content, the presence or absence of any shielding vegetation, and counting geometry. These uncontrolled variables have been responsible for questionable results in many surveys.

Thorium normalization

During the past 10 years we have developed a method of thorium normalization of potassium and uranium measurements to suppress the effects of these variables for gamma-ray spectral data.

The thorium content is used as a lithologic control to define "ideal" potassium and thorium values for each sample. The basic assumption is that whatever happens to influence the apparent concentration of thorium also affects uranium and potassium in similar and predictable ways.

For example, changes in lithology result in roughly parallel increases or decreases in the average concentrations of all three radioelements simultaneously ³ (table). Soil moisture or masses of vegetation cause similar amounts of absorption of gamma rays from thorium, uranium, and potassium.

Also, differences in counting geometry because of topographic irregularities will cause parallel variations in measurements of all three elements. Normalizing to thorium will suppress all of these effects. Thorium concentrations may be used to roughly predict uranium and potassium concentrations by determining their regional relationships.

Significant differences between predicted (or "ideal") uranium and potassium amounts and actual measured values must be due to factors other than lithology, soil moisture, vegetation shielding, or counting geometry. By measuring these secondary effects, one can better define possible petroleum or uranium prospect leads.

This article summarizes:

1. New developments in methods of thorium normalization of potassium and uranium data and
2. Their initial application in petroleum prospecting as described in more detail in a forthcoming article. ¹

Aerial radiometric databases

The NURE program sponsored by the U.S. department of Energy to assess U.S. uranium ore potential.

Beginning in 1974, the U.S. government collected reconnaissance aerial gamma-ray spectral data covering all of the conterminous U.S. and much of Alaska. A similar continuing program was initiated in Canada and Australia. Although these programs were directed specifically to evaluate uranium resources, much of the data cover areas of interest to petroleum explorers.

The data are available at relatively low cost compared to costs of acquisition. The method presented in this article may play a major role in making this large database an effective contributor in finding new onshore oil resources.

Methods

Surface GR spectral measurements

Measurements of surface gamma-ray spectra were made using an Exploranium Model GR-410 four-channel spectrometer equipped with a 344 cc (21 cu in.) sodium iodide (NaI[T1]) crystal detector.

The detector was placed on a plastic support with the crystal elevated about 8 in. (20 cm) above the ground surface. Measurement locations over known oil or gas fields were chosen away from possible areas of contamination from production operations.

Each surface station included measurements of equivalent uranium concentration (eU = U measured as ²¹⁴Bi), the equivalent thorium concentration (eTh = Th measured as ²⁰⁸TI) and the potassium concentration (K measured as ⁴⁰K). In this article, both aerial and surface gamma-ray spectrometer measurements of these elements will be referenced by their names or symbols: U, Th, and K, respectively.

Counting times were 4 min at each station to obtain a statistically adequate number of counts. The total count and the number of counts in each of the windows (potassium, uranium, and thorium) were recorded at each station. The windows were centered on the following photopeaks: 2.26 Mev for thorium, 1.76 Mev for uranium, and 1.46 Mev for potassium.

Arial GR data

NURE and Australian aerial gamma-ray data were obtained on magnetic tapes from the EROS Data Center, Sioux Falls, S.D., and the Australia Geological Survey Organisation (formerly the Australia Bureau of Mineral Resources), Canberra, ACT, Australia, respectively.

These data had been processed to:

1. remove aircraft background radiation,
2. compensate for atmospheric radon,
3. correct for cosmic radiation,
4. correct for higher energy contributions to lower energy windows by spectral stripping,
5. correct for altitude, and
6. convert the corrected data to ground concentrations.

The data were presented on the tapes as parts per million uranium and thorium and percent potassium.

Data processing

Uranium and potassium data for each surface or aerial gamma-ray spectral profile were normalized by plotting the field-measured Ks vs. Th _s, and U _s vs. Th _s values for all stations.

Various linear, logarithmic, and second order curve fitting procedures were tried, and the simplest effective equations relating these variables were determined:

K_i - f₁ (Th_s) (1)

U_i = f₂ (Th_s) (2)

where K_i is the "ideal" thorium-defined potassium value for the station with a real thorium value of Th_s, and U_i is the "ideal" thorium-defined uranium value for that station.

Deviations of the real values from the calculated ideal values for each station were obtained using equations of the form

KD% = (K_s - K_i)/K_s (3)

UD% = (U_s - U_i)/U_s (4)

where K_s and U_s are the measured values at the station, and KD% and UD% are the relative deviations expressed as a fraction of the station values. KD% generally shows negative values over petroleum, and UD% shows less negative or positive values. The difference, UD%-KD%, is called DRAD, which usually shows anomalous positive values over petroleum and negative values elsewhere.

Comparative profiles of K_s, U_s, Th_s, KD%, UD%, and DRAD were plotted for the examples presented in Fig. 1 [135,021 bytes] and Fig. 2 [159,305 bytes] to show their relationships. For the aerial data example, Fig. 3 [134,714 bytes], only KD% and UD% profiles are provided to illustrate typical "crossover" anomalies over petroleum accumulations.

Surface profiles

Summary

Profiles over Leona and OSR fields (Leon County, Tex.) and Lonesome Dove II field (Concho County, Tex.) are presented as examples to show how this new method suppresses lithologic variables and consistently reflects the presence of subsurface petroleum in stratigraphic traps.

Results have been generally similar for additional profiles measured over Alabama Ferry-Glen Rose and Alabama Ferry-Woodbine (Leon County, Tex.), ¹⁴ and several other unpublished examples.

Agaritta and Dark Horse oil fields and Selden gas field, Concho County, Tex., were discovered using gamma-ray spectral data in conjunction with soil gas hydrocarbon measurements, soil magnetic susceptibility studies, and subsurface geology. ⁵

Preliminary wildcat results are available for an exploratory play by Indigo Oil Inc. in Concho County, Tex., over the period from 1987 to 1993. ⁶

Fifteen wildcats were drilled on significant surface gamma-ray spectral low-potassium anomalies. Six resulted in production records or initial production tests that promise economic success. This yields a probable economic wildcat success ratio of 40%

Leona, OSR fields

Leona field produces oil from thin-bedded sandstones of the Sub-Clarksville

member of the Upper Cretaceous Eagle Ford formation at a depth of about 7,000 ft (2,100 m).

The oil is stratigraphically trapped where the bars grade updip into marine shale. In plots of the radiometric data over Leona field, Leon County, Tex., and part of OSR field (Fig. 1), unprocessed field counts for potassium (K_s), uranium (U_s), and thorium (Th_s) all show a high centered on the field. This is ascribed to a higher clay content in the surface sediments.

After thorium normalization to suppress this lithologic effect, KD% shows negative values in this same region. Corresponding UD% is less low with a consequent positive DRAD anomaly where crossover occurs (indicated by hachured patterns).

The radiometric anomaly at stations 18 through 25 is ascribed to the edge of OSR field located just east of Leona field. OSR field produces from the Woodbine formation.

Lonesome Dove II field

Lonesome Dove II field produces from the lower King sand of Pennsylvanian age at a depth of about 2,200 it (670 m). The reservoir has characteristics of both fluvial point bars and distributary channels.

In contrast to Leona field, Lonesome Dove II field shows low values in the unprocessed data for all three of the radioelements over the field (Fig. 2). There is considerable caliche in the surface soils that appears to be diluting the radioactivity in that region.

The thorium-normalized values yield UD% and KD% crossover (hachured) and high DRAD anomalies that are consistent with those found over other petroleum accumulations.

The high UD% anomaly at station 19 may be associated with leakage of radon up an unmapped fault causing a subsequent high gamma-ray reading from its daughter, ²¹⁴Bi, at the surface.

Aerial gamma-ray spectral profiles

Summary

NURE data studies. Our proprietary studies of NURE program aerial gamma-ray spectral data since 1986 covered 19,608,000 acres in Texas, Oklahoma, Arkansas, Louisiana, Mississippi, and Florida.

One study completed in 1988 in conjunction with follow-up aerial hydrocarbon sensing and surface geochemical studies resulted in a 1990 discovery of Northeast Nacogdoches Travis Peak gas field, Nacogdoches County, Tex.

Australia data studies. Recent additional studies, using Australian aerial gammaray data, included test areas in the Canning and Otway basins. ²

In total these tests covered 7,400 sq km (2,850 sq miles) and included five oil fields (Blina, Lloyd, West Terrace, Boundary, and Sundown) in the Canning basin and a commercial gas field (PPL-1) in the Otway basin.

All producing fields were associated with significant negative normalized potassium and higher normalized uranium anomalies. Based on a total of 69 wells in three test areas, it is estimated that the chance of encountering hydrocarbons (economic production or shows) in wells within radiometrically anomalous favorable zones is about 2.6 times greater than in wells outside favorable areas.

A recent proprietary study of a seismic prospect in the Canning basin showed significant low potassium anomalies. Subsequent drilling resulted in the Point Torment gas discovery.

Blackfoot field

Productive formations in Blackfoot field include the Rodessa limestone, the Pettit limestone, and the uppermost Travis Peak sandstone at depths ranging from 8,900 ft (2,700 m) to 9,920 ft (3,024 m).

The Blackfoot structure is a small, faulted, elongate, domal closure in Anderson County, Tex.

Fig. 3 shows thorium-normalized KD% and UD% curves and the location of the NURE flightline with respect to the producing wells. The regions of significant crossover and low KD% values are over Blackfoot field or the relatively narrow area between Holly Spring and Owens Creek gas fields.

Two anomalously low UD% areas (A and B) are tentatively ascribed to a lack of equilibrium in the uranium series or over-correction of the raw aerial data for atmospheric radon.

Relation of anomalies to petroleum

Alteration model

Hydrocarbon microseepage. Based on all information available to the authors, it is concluded that hydrocarbon microseepage involves buoyant, relatively rapid, vertical ascent of ultra-small (colloidal size) gas bubbles of light hydrocarbons (primarily methane through the butanes) through a network of interconnected, groundwater-filled joints and bedding planes. ⁷

When the bubbles reach the surface of the water table, first they enter the interstitial soil gases where they may be sampled and detected by sensitive gas chromatography, and then they escape into the atmosphere where they may be detected by microwave spectrometry.

Chemical and biochemical hydrocarbon degradation. As hydrocarbons rise toward the surface, sulfate reducing bacteria may consume them and produce hydrogen sulfide (H₂S), carbon dioxide (CO₂), and secondary carbonate mineralization (CaCO₃). ⁸

Carbon dioxide in groundwater forms carbonic acid (H₂CO₃), which can react with clay minerals (calcium silicates, for example) to create more secondary pore-filling calcium carbonate mineralization and silicification (SiO₂). ⁹ This can result in rendering the near-surface materials more dense and resistant to erosion.

The effects may be:

1. increased seismic velocity over the petroleum accumulation,
2. erosional topographic highs and consequent geomorphic anomalies as observed over Alabama Ferry (Glen Rose) field, ⁴ or
3. dilution of the normal radioactive element concentrations with resultant low-thorium (as well as low-potassium and low-uranium) gamma-ray activities over production (as observed over Lonesome Dove II field, Fig. 2).

Potassium in sediments is contained primarily in the clay mineral, illite, and hydrogen ions (H ₃O ⁺) from carbonic acid may replace potassium ions by ion exchange in the little lattice.

The displaced potassium may be removed in groundwater solution to produce radiometrically anomalous low potassium concentrations in the soils. ¹

Observations to date are in general agreement with this model as it pertains to potassium. Thorium should remain relatively fixed in its original distribution in the heavy insoluble minerals, and our observations generally confirm this.

The situation with respect to uranium is more complex than that for potassium. Microbially produced, hydrogen-sulfide related, chemically reducing environments can result in a buildup of uranium concentrations.

Uranium has two valence states. The oxidized form, the uranyl ion (UO₂⁺⁺), is soluble in ground waters. When it is reduced, it converts to UO₂, uraninite, which is relatively insoluble and precipitates. Thus, uranium will tend to migrate from an oxidizing environment to a reducing one where its concentration will build up over time.

If conditions are right, a sedimentary uranium deposit may be formed. Thus, lesser buildups of uranium in surface sediments may be useful clues to petroleum at depth.

The chemically reducing environment in zones over production can convert nonmagnetic iron minerals (hematite) to the magnetic oxide (magnetite) and to magnetic sulfides (pyrrhotite and griegite).

Surface oxidation may convert magnetite to maghemite, the magnetic form of hermatite. This may result in aerial "micromagnetic" anomalies and increased magnetic susceptibility in soils or rocks overlying petroleum deposits. ⁷

Interpretation problems

Application of gamma-ray spectral measurements in petroleum or uranium prospecting is subject to some limitations that must be comprehended for effective results. Two of the most important problems involve radiation shielding and atmospheric radon ( ²²²Rn).

Surface-measured gamma radiation will not reflect radioelement content below 20-25 cm (8-10 in.) of surface soil or rock. If the area is covered by water, recent floodplain alluvium, glacial till, loess, paving material, or other shielding matter, most measured radiation must originate in the shielding material. To provide adequate information on the radioactivity of the bedrock, gamma-ray spectral measurements will work best when made over residual soil or bedrock outcrops.

Atmospheric radon produces daughters, including ²¹⁴Bi (half-life = 19 min), which may become attached to dust particles in the atmosphere. Rainfall will wash them out to cause relatively large but short-lived increases in the surface ²¹⁴Bi gamma activity that would result in erroneously high apparent uranium concentrations.

Because of this effect one should wait at least 135 min after rainfall ceases before beginning gamma-ray spectral measurements.

Benefits, limitations

This gamma-ray spectral method is subject to the same benefits and limitations as all other surface methods dependent on seeping or microseeping hydrocarbons and related alterations.

The principal benefit is that it indicates the probable presence of hydrocarbons at depth. The data cannot reveal the depth to the source or sources and cannot be used to predict economic success or failure of wildcat tests.

Drilling data have demonstrated that the combined effects of several strata with oil or gas shows may create a surface anomaly as strong as one over a producing field.

The presence of a statistically valid DRAD anomaly can enhance the probability of wildcat success, but it does not eliminate the possibility of a dry hole. Experience has demonstrated that DRAD anomalies may be found only over portions of fields and cannot be used to determine production boundaries accurately.

Conclusions

All results to date demonstrate that thorium normalization of gamma-ray spectral uranium and potassium data effectively suppresses lithologic and environmental effects on both surface and aerial radiometric data. This allows measurement of the special effects related to the presence of petroleum at depth.

Study results suggest that special processing of aerial radiometric data with follow-up surface gamma-ray spectral studies can be practically applied in prospecting for onshore petroleum reserves in both stratigraphic and structural traps.

As presented, this method may be especially effective in surface studies when used in conjunction with aerial or soil gas hydrocarbon measurements and soil magnetic susceptibility studies.

Acknowledgments

We thank Fina Oil & Chemical Co. for permission to use the data and geologic information on Leona and OSR fields. Thanks also to Indigo Oil Inc. for data and information on Agaritta.

Dark Horse, Selden, and new unnamed fields and to Marshall & Winston Inc. and Bowerman Oil & Gas Inc. for permission to sample over Lonesome Dove II field. We also thank the Australia Geological Survey Organisation for furnishing the data used in the Australian tests.

References

1. Saunders, D.F., Burson, K.R., Branch, J.F., and Thompson, C.K., Relation of thorium-normalized surface and aerial radiometric data to subsurface petroleum accumulations, Geophysics, Vol. 58, 1993, pp. 1,417-27.
2. Saunders, D.F., Branch, J.F., and Thompson, C.K., Tests of Australian radiometric data for use in petroleum reconnaissance, Geophysics, in press.
3. Galbraith, J.H., and Saunders, D.F., Rock classification by characteristics of aerial gamma-ray measurements: J. Geochem. Expl., Vol. 18, 1983, pp. 47-73.
4. Saunders, D.F., Burson, K.R., Branch, J.F., and Thompson, C.K., Alabama Ferry field detectable by hydrocarbon microseepages and related alterations, OGJ, Vol. 87, Nos. 53-55, 1989, pp. 108-110.
5. Saunders, D.F., Burson, K.R., Brown, J.J., and Thompson, C.K., Combined geological and surface geochemical methods discover Agaritta and Brady Creek fields, Concho County, Tex., AAPG Bull., Vol. 77, 1993, pp. 1,219-40.
6. J.J. Brown, personal communication.
7. Saunders, D.F., Burson, K.R., and Thompson, C.K., Relation of soil magnetic susceptibility and soil gas hydrocarbon measurements to subsurface petroleum accumulations, AAPG Bull., Vol. 75, 1991, pp. 389-408.
8. Kartsev, A.A., Tabasaranskii, Z.A., Subbota, M.I., and Mogilevskii, G.A., Geochemical methods of prospecting and exploration for petroleum and natural gas, University of California Press, 1959, pp. 294-295.
9. Donovan, T.J., and Dalziel, M.C., Late diagnetic indicators of buried oil and gas, U.S.G.S. Open File Report 77-817, 1977.

THE AUTHORS