FURTHER EXPLORATION FOR GAS WARRANTED IN COLUMBIA BASIN

Newell P. Campbell
Consulting petroleum geologist
Yakima, Wash.

Stephen P. Reidel
Washington State University
Pasco, Wash.

Nearly 5 years have passed since the last wildcat well was drilled in the Columbia basin of central Washington.

Although drilling ceased, exploration activities continue on a limited scale. In this report we assess the drilling results (now public information) and describe geological, geophysical, and geochemical programs proceeding in the Columbia basin. Finally we target where more exploration is warranted.

GEOLOGIC HISTORY

Prior to exploratory drilling in the 1980s, little was known of the deep subsurface geology in the Columbia basin. Mid-Miocene lavas of the Columbia River Basalt group obscure older rock covering about 100,000 sq miles in southeastern Washington and northeastern Oregon (Fig. 1). Part of the basalt is folded into east-west anticlinal ridges (Yakima fold belt).

The geology and structure of older rock at the basalt margins are complex and diverse but in places these rocks contain coal beds and clean fluvial sands. Obvious questions that concern explorationists are:

• Do source beds and reservoir rocks exist under the basalt?

• Are the source beds thermally mature?

• Do anticlines of the Yakima fold belt extend downward into pre-basalt rocks and do other traps exist?

• Can standard geophysical and geochemical methods be modified to work in basalt?

We address these questions in the following summary.

WELLSITE DATA

Eight exploratory wells penetrated the Columbia River Basalt group during the 1980s exploration surge (Fig. 1).

Our generalized stratigraphy for each well and an overall prebasalt stratigraphy, for the Columbia basin are given in Fig. 2. The sub-basalt stratigraphy in these wells is significant because it shows the nature of the rocks crossing two major structural features, the Hog Ranch-Naneum Ridge anticline axis and the Palouse slope boundary (Figs. 3, 4).

Six wells were drilled on surface topographic highs-anticlines of the Yakima fold belt-but two wells, 1 Quincy and 1 Darcell, were spudded on geophysical anomalies.

Two wells, 1 Yakima Minerals and 1-9 BN, though noncommercial, produced significant gas during testing. One well, the 1-9 BN, also produced a small amount of condensate.1 Gas production was from nonmarine arkosic sands of early to mid-Tertiary age.

The other six wells had small gas shows but no sustained gas flow.

The source of the gas is almost certainly from coals in Eocene and Oligocene fluvial rocks that underlie the Columbia basin. The Columbia River Basalt along with clayey or tuffaceous zones in the underlying rocks form the seals for the reservoirs.

The wells proved that source and cap rocks are not a problem but reservoir rocks are limited.

Zeolites, tuft, and clay (from decomposition of volcaniclastics) plug voids in many of the sandstones. However, fluvial rocks with good porosity and permeability do exist in surface outcrops along the basalt margin and in well cuttings. These are thought to be stream channel sands deposited as streams meandered across what was once a low coastal plain in early Tertiary.

A well encountering one of these old channels could produce large volumes of gas. For example in 1993 a 1,100 ft water well drilled near Wenatchee, Wash., just off the basalt margin into Wenatchee formation, cut approximately 60 ft of clean channel sand. Water production (and a little gas) were initially estimated at several thousand gallons per minute with pressure at nearly 100 psi. The well remains uncapped.

RECENT INVESTIGATIONS

Detailed geological mapping along the Columbia River Basalt group margin coupled with subsurface data give new insights on the nature of prebasalt sediment and structures under the basalt.

The Columbia basin, considered a back-arc basin undergoing subsidence during eruption of the basalt, can now be shown to have had ongoing subsidence from early Tertiary.

Columbia River Basalt group flowed westward down the Palouse slope and onto the pre-Tertiary sedimentary basin in Mid-Miocene times. Continued subsidence allowed the thickest CRB to be emplaced over the center of the pre-Tertiary basin.

This structural basin, open to the west and filled with Eocene to mid-Miocene fluvial and volcaniclastic sediment, lies beneath that part of the Columbia basin known as the Yakima fold belt.

This fluvial basin extends westward (and northwestward) beyond the basalt margin and perhaps passes completely under the PlioPleistocene volcanics of the southern Cascade Range. 2 3 4

5However, the eastern edge of the prebasalt basin ends abruptly along a north-south line that marks the termination of the Palouse slope and the Ice Harbor Dike Swarm 6 7 (Fig. 4).

Well stratigraphy and gravity data indicate that this abrupt termination marks the edge of the North American craton. East of this line, rocks under the basalt are composed of Precambrian and early Paleozoic meta-sediments and granite.

The abrupt edge of the prebasalt basin suggests that a major fault or graben-like edge marks this zone. Whether a fault or not, the lack of early Tertiary rocks east of this line would seem to severely limit the hydrocarbon potential in the eastern Columbia basin.

CROSS-BASIN STRUCTURES

In addition to the line marking the edge of the Palouse slope, two other major cross-basinal structures influence both basalt and prebasalt rocks: The Hog Ranch-Naneum Ridge anticline axis and the Cle Elum-Wallula lineament, abbreviated CLEW (Fig. 4).

The Hog Ranch-Naneum Ridge axis is a broad structural upwarp, topographically aligned with the Leavenworth fault, that passes under several east-west trending surface anticlines of the Yakima fold belt. It plunges to the south but extends at least as far as the Horse Heaven Hills anticline (Fig. 4). The Columbia River Basalt group and certain prebasalt rocks thin over this axis (Fig. 3).

The CLEW, that part of the Olympic Wallowa lineament crossing the Columbia basin, is a well known northwest-southeast alignment of anticlinal ridges in the Yakima fold belt.

Recent mapping, water well data, and geophysical data suggest that a single feature alone the southeast edge of the CLEW (known as the Naches fault zone) displaces both prebasalt and basalt rocks and is a major boundary between two "sets" of early Tertiary rocks.3

Interestingly, no wildcat test wells have been drilled southwest of this feature although thermogenic methane is present in groundwaters along this zone.8

SURFACE ANTICLINES

Recent seismic and gravity studies9 10 11 12 confirm our findings that:

• The Hog Ranch-Naneum Ridge anticline axis represents a basement high that was active well into the eruptive phase of the Columbia River Basalt group.

• The CLEW passes over but does not offset the Hog Ranch-Naneum Ridge structure. if the CLEW is a zone of strike-slip faulting, then movement must have occurred prior to the eruption of the basalt.

• The Chiwaukum graben probably does not extend southeast much beyond the basalt margin at Wenatchee.

Do prebasalt rocks beneath the Yakima fold belt show the same structural trends as in the Columbia River Basalt group, and do surface anticlines continue downward into pre-basalt rock? All present data indicate that they do not!

Mapping of prebasalt rocks along the northwest basalt margin show that, with the possible exception of the CLEW, folds and faults in prebasalt rocks do not mimic those in the Columbia River basin. 2

Drill hole data are somewhat inconclusive. While the 1-29 Bissa well shows prebasalt rocks elevated over the Hog Ranch-Naneum Ridge anticline axis, other drill holes seem to show a flat basalt-sediment contact, and two wells show thicker than expected sections (BN 2-35, basalt; 1 Hanna, Clarno formation) under an anticlinal ridge.

Geophysical data, and in particular gravity studies,12 show no sedimentary rock thrust into the cores of surface anticlines and an undulating basalt-sediment contact unrelated to surface structural expression. However Jarchow9 calculates that closure on some of the undulations (sediment highs) may exceed any closure of surface anticlines.

Folds of the Yakima fold belt usually contain high-angle, thrust faults exposed on their steepest limb. All present data now suggest that the faults are steep through the basalt and the basalt/sediment interface; they probably flatten and do not extend into prebasalt rocks. Thus, most of the structures seen at the surface of the Yakima fold belt, excluding the CLEW and Hog Ranch-Naneum Ridge anticline axis, appear to reflect "thin skinned" tectonics.

BASIN THERMAL MATURITY

Vitrinite reflectance and total organic carbon studies show that sedimentary rocks beneath the Yakima fold belt are mature and capable of producing gas.

As no marine rocks are likely to exist and because gas has been found to date, the Columbia basin must be considered mainly a gas province. However, where source beds are buried to a depth of 4 km, the geothermal gradient places the temperature at about 195 C.8 This would place any source beds in the oil window.

RECENT GEOPHYSICAL DATA

Many early seismic and magnetotelluric surveys across the Columbia basin remain proprietary. Several new surveys provide limited but highly useful data. They include:

• An east-west cross-Cascade seismic survey to determine the depth of the early Tertiary basin and examine the Morton anticline;13

• A north-south high explosive, large offset seismic survey on trend with the Hog Ranch-Naneum Ridge anticline axis; 9 and

• Several seismic reflection lines around the 1 Hanna well in Oregon.14

These surveys combined with better quality regional gravity and magnetic surveys support our observations that:

1. In spite of unpublished (and published) claims of good data obtained from conventional seismic and magnetotelluric surveys, errors in locating the depth to the base of the basalt can easily exceed 10%-enough to mask large structural traps. Results at the 1 Quincy and 1 Darcell wells support this.

2. High explosive, long offset seismic surveys coupled with regional gravity data provide far better results than conventional seismic data, at least in locating large structures-but these have not been tested by drilling.

3. All geophysical surveys suggest that prebasalt structures (those within the sub-basalt sediments) do not mimic those of the overlying basalt, therefore drilling the tops of surface anticlines may not be the correct approach.

4. At present, no geophysical methods are capable of locating stratigraphic or structural traps within the sedimentary basin under the basalt, specifically buried stream channels containing good reservoir rock. In fact, locating highs on the basalt-sediment contact is presently the best use for geophysical methods.

GEOCHEMICAL STUDIES

Methane is common in water Wells drilled into the Columbia basin.

The only commercial gas produced in the Columbia basin came from a low pressure gas field associated with the Rattlesnake Hills anticline. Johnson and others8 wrote that gas from 19 water wells in the Columbia basin Was found to be deep, coalbed-generated methane derived from coal sediments beneath the Columbia River Basalt.

The natural gas was found to be 99% methane, and isotopic compositions of methane in groundwater from basalt aquifers indicated that the methane is a mixture of biogenic components [see formula] (d 13 C-CH4 to -88 ), and d 2 H-CH4 to -265 ) and thermogenic (d 13 C-CH4 to -35 and d 2 H-CH4 to - 134 ).8

Thermogenic methane dominates in deeper zone, and biogenic methane dominates in shallower zones. Chemical and isotopic data are consistent with entrainment of deep, coalbed-generated methane in upwelling groundwater from below the Columbia River Basalt group that mixes with near-surface groundwater. The areal distribution pattern of methane suggests that fault intersections are necessary for vertical migration of deep methane through the basalt.

Although most methane in groundwater is undersaturated, the amount of gas can be significant. A gas explosion partially destroyed the water tower at Mabton, Wash., and deep water wells often require degassing equipment.

Fig. 5 shows the location of maximum gas production associated with deep water wells in the Yakima fold belt. No wildcat wells have been drilled in this area to date.

Some gas must also reach the surface. It is unknown whether other geochemical investigation (i.e., vegetation anomaly studies) were previously, conducted, but such studies might have great value in location of drillable sites.

RECOMMENDATIONS

Despite the lack of good seismic data and drilling success, one promising area remains untested.

Given that historically wells were drilled initially on oil and gas seeps, it is surprising that in the Columbia basin not a single well has been drilled in the vicinity of the greatest accumulation of gas in groundwater (Fig. 5).

An exploration program targeting the area southwest of the CLEW, including parts of Rattlesnake ridge, Ahtanum ridge, and Horse Heaven Hills anticlines have the following advantages:

1. By far the greatest concentrations of gas found in water wells.

2. Only 6,000-8,000 ft of basalt thickness (low considering the 10,000-12,000 ft in the area of the Saddle Mountains 1-9 BN well).

3. The area lies along and south of the Naches River fault zone, a through going feature associated with the CLEW, where no exploratory wells have been drilled. The shallow gas anomalies suggest that gas may be trapped against or over the Naches River fault and seep upward into overlying basalt along faults.

4. No seismic surveys have been run in this area. Gravity data show a sediment high in the Ahtanum ridge area. Seismic surveys could be used to confirm this structure.

5. TOC and vitrinite reflectance place this area well within the gas window.

6. Land in this area is mostly private, eliminating the leasing problems that plagued other areas of the basin.

In conclusion, although recent studies in the Columbia basin have eliminated some parts of the basin as exploration targets and changed the exploration emphasis, large areas remain untested, including what we consider the locality with the greatest potential. We remain optimistic about the future of gas exploration in the Columbia basin.

REFERENCES

1. Lingley, W.S. Jr., and Walsh, T.J., Issues relating to petroleum drilling near the proposed high-level nuclear waste repository at Hanford, Washington Geologic Newsletter, Vol. 14, No. 3, 1989, pp. 10-19.

2. Campbell, N.P., Structural geology along the northwestern Columbia River basalt margin, Washington, Washington Division of Geology and Earth Resources Open File Report 88-5, 1988.

3. Campbell, N.P., Structural and stratigraphic interpretation of rocks under the Yakima fold belt Columbia basin, based on recent surface mapping and well data, in Reidel, S.P., and Hooper, P.R., eds., Volcanism and tectonism in the Columbia River flood-basalt province, GSA Special Paper 239, 1989, pp. 209-222.

4. Stanley, W.D., Gwilliam, W.J., Latham, G., and Westhusing, K., The southern Washington Cascades conductor-A previously unrecognized thick sedimentary sequence? AAPG Bull., Vol. 76, No. 10, 1992, pp. 1,569-85.

5. Walsh, T.J., and Lingley, W.S. Jr., Coal maturation and the natural gas potential of western and central Washington, Olympia, Wash., Washington State Department of Natural Resources Open File Report No. 91-2, 1991, 27 p.

6. Reidel, S.P., Fecht, K.R., Hagood, NI.C., and Tolan, T.C., The geological evolution of the central Columbia plateau, in Reidel, S.P., and Hooper, P.R., ed, Volcanism and tectonism in the Columbia River flood-basalt province, GSA Special Paper 239, 1989, pp. 21-53.

7. Reidel, S.P., Campbell, N.P., Fecht, K.R., and Lindsey, K.A., Late Cenozoic structure and stratigraphy of south-central Washington, in Cheney, E., and Lasmanis R., eds., Second Symposium on Geology of Washington, Washington Division of Geology, and Earth Resources, Bull. 80, 1993.

8. Johnson, V.G., Graham, D.L., and Reidel, S.P., Methane in Columbia River Basalt aquifer: Isotopic and geohydrologic evidence for a deep coalbed gas source in the Columbia basin, Washington, AAPG Bull., Vol. 77, No. 7, 1993, pp. 1,192-1,207.

9. Jarchow, C.M., Investigations of magnetic underplating beneath the northwestern Basin and Range province, Nevada, seismic data acquisition and tectonic problems of the Columbia plateau, Washington, and the nature of the Mohorovicic discontinuity worldwide, PhD thesis, Stanford University, 1991, 258 p.

10. Jarchow, C.M., Catchings, R.O., and Utter, W.J., How Washington crew got good, thrifty seismic in bad data area, OGJ, Vol. 89, No. 24, 1991, pp. 54-55.

11. Catchings, R.O., and Mooney, W.O., Crustal structure of the Columbia plateau: Evidence for continental rifting, journal of Geophysical Research, Vol. 93, 1988, pp. 459-474.

12. Saltus, R.W., Upper crustal structure beneath the Columbia River Basalt group, Washington: Gravity interpretation controlled by borehole and seismic studies, GSA Bull., Vol. 105, 1993, pp. 1,247-59.

13. Warren, R.K., and Srnka, L.J., Exploration in the basalt covered areas of the Columbia River basin, Wash., using electromagnetic array profiling, Geophysics, Vol. 57, No. 8, 1992, pp. 986-993.

14. Withers, R., Eggers, D., Fox T.P., and Crebs, T., A case study of integrated hydrocarbon exploration through basalt, Geophysics, in press, 1993.