Deep gas poses opportunities, challenges to U.S. operators

May 4, 1998
The previous article in this series on emerging natural gas resources introduced deep gas-natural gas in deep onshore sedimentary basins (below 15,000 ft)-by presenting a 1996 U.S. Geological Survey assessment for this resource. The USGS estimated that 114 tcf of technically recoverable conventional and nonconventional deep gas remains to be discovered in the Rocky Mountains (57 tcf), Gulf Coast (27 tcf), Alaska (18 tcf), West Texas/New Mexico (4 tcf), and Midcontinent (3 tcf), among others.


S.R. Reeves,J.A. Kuuskraa,V.A. Kuuskraa
Advanced Resources International Inc.
Arlington, Va.
The previous article in this series on emerging natural gas resources introduced deep gas-natural gas in deep onshore sedimentary basins (below 15,000 ft)-by presenting a 1996 U.S. Geological Survey assessment for this resource.

The USGS estimated that 114 tcf of technically recoverable conventional and nonconventional deep gas remains to be discovered in the Rocky Mountains (57 tcf), Gulf Coast (27 tcf), Alaska (18 tcf), West Texas/New Mexico (4 tcf), and Midcontinent (3 tcf), among others.

This article, third in this series and the second on deep gas, takes a closer look at this large and challenging resource by addressing the following key questions:

  1. Where are the locations and what are the differences among the major deep gas basins?
  2. How successful and active have the deep gas plays been? and,
  3. What obstacles and rewards are likely for developers of deep gas?
This article concludes with reviews and case studies of three specific deep gas basins: the mature Anadarko basin, the emerging Green River basin, and the frontier Wind River basin. Reviews of these basins highlight the challenges in finding and producing deep gas, as well as the results and rewards.

Deep gas overview

Active deep gas plays

An update of Dyman and others1 showed that about 5,600 productive deep gas wells have been drilled in the U.S. as of 1997 (Fig. 1a [101,213 bytes]).

More than half of these are in the Gulf Coast, with the Anadarko and Permian basins accounting for much of the rest. Based on 1988 data, the latest year for which such information has been published, cumulative gas production from deep wells exceeded 21 tcf, primarily from the Permian (12.4 tcf), the Gulf Coast (6.2 tcf), and the Anadarko (2.4 tcf) basins (Fig. 1b).

The Gulf Coast is currently the most active region and has the largest remaining proved reserves (6.6 tcf) of deep gas (Fig. 1c).

Noticeably absent from these active deep gas regions are the Rockies and Alaska. Historically the Rockies have seen lower gas prices than the other regions, and both regions lack sufficient pipeline outlets. As a result, deep gas resources in these regions are underdeveloped and represent important emerging opportunities.

Deep gas plays are diverse as implied by the geologic age and location of the existing deep gas basins and wells (Fig. 2 [110,600 bytes]). Some deep gas plays are merely downdip extensions of shallower plays where drilling has progressed beyond 15,000 ft, for example the Wilcox sandstone and other Cenozoic plays in the western Gulf of Mexico. Plays of this type generally require "incremental" technology advancements for successful development.

"True" deep plays, where the bulk of the gas play occurs at depths greater than 15,000 ft, are usually geologically older, either Mesozoic or Paleozoic in age, represent a different reservoir setting than the geologically younger plays and require a unique assemblage of technologies for successful development. These plays include the Tuscaloosa, Smackover, and Norphlet plays of the eastern Gulf of Mexico (sandstones and carbonates), the Fusselman and Ellenburger of the Delaware/Val Verde basin (carbonates), and the Hunton, Simpson, and Arbuckle of the Anadarko basin (carbonates).

In the Rockies, "true" deep gas plays include reservoirs in the Madison and Bighorn carbonates in the Green River basin, and in the Cody (Sussex/Shannon) and Frontier sandstones in the Wind River basin. It is also in the Rockies that new deep gas plays continue to emerge, such as the Madison carbonate in Madden field and the Muddy/Dakota sandstone at Waltman (Cave Gulch) field, Wind River basin.

Deep gas drilling

Deep drilling for oil and gas, both offshore and onshore, peaked in 1982 at over 1,200 wells and has steadily declined since then.2 A resurgence followed, with over 300 deep gas wells drilled in 1995 and 1996, up from about 150 wells/year in the early 1990s.3

Along the Texas Gulf Coast the number of deep gas wells drilled jumped to 137 in 1996, up from 38 in 1990. South Louisiana and Oklahoma also have seen increased activity, to 78 wells and 51 wells respectively in 1996, versus 67 and 44 respectively in 1990.

Even with the increase and despite the large underlying deep gas resource, less than 5% of all gas wells drilled in 1996 had deep gas as a target.

Deep gas is viewed by industry as a high risk venture-deep wells are expensive, and exploring and even delineating deep gas prospects entails high dry-hole rates.

At over $5 million/deep gas well on average (including allocated dry-hole costs) and with a 27% dry-hole rate (almost double that of shallower wells), one can easily understand industry's caution. Still, in 1996, industry spent $1.3 billion (22%) of its $5.7 billion onshore drilling budget on deep wells.

In the Texas Gulf Coast, the heightened activity in recent years has reduced well costs by 19% from 1990 to $3.9 million in 1996. Oklahoma deep well costs were already low but still declined 9% during this period. The extensive past deep drilling experience in Oklahoma has taught operators how to efficiently handle basin-specific problems such as drilling through alternating high/ low pressure formations and how to optimize bits, casing points, and mud systems.

The most expensive deep gas wells in currently active deep gas areas are in South Louisiana. There overpressured zones, high bottomhole temperatures, and corrosive reservoir fluids require special practices. In the Permian basin, well costs have actually risen since 1990 as operators moved from the geologically less complex Delaware portion of the Permian basin to the more challenging and costly deep gas prospects in the Val Verde portion of this basin.

Deep gas development wells cost $4 million-5 million each (including allocated dry-hole costs) and have success rates of 80%. Deep gas exploratory wells, costing nearly $10 million each (including allocated dry-hole costs that account for about 50% of total costs), still represent a significant barrier.

Rewards, challenges

While many barriers exist to deep gas development, particularly in new areas where costs and dry-hole rates are high, rewards are handsome for the successful.

A recent study performed for the Gas Research Institute (GRI) on lower-48 gas development (Table 1 [64,124 bytes]) put an economic measure on current barriers.4 It showed that even though (on average) onshore deep gas (15,000 ft) provides 6.58 bcf/well of reserves, five times more than for a shallower onshore well at 5,000-10,000 ft, these deep wells cost nearly 10 times more to drill.

As the basin reviews and case studies (presented later) show, however, gas recoveries and costs per well vary widely among deep plays. In the superdeep plays of the Rocky Mountains, where well costs can easily exceed $10 million, recoveries of over 20 bcf/well are needed to justify drilling. In Oklahoma with well costs of less than $3 million, reserves of 6 bcf/well may be adequate.

Four technical challenges confront deep gas:

  1. Reducing well costs;
  2. Reducing dry hole rates;
  3. Improving completion efficiency; and
  4. Reducing sour gas production/processing costs.

Cutting well costs

In a study funded by GRI, rig time (day rate times days of drilling) was determined the largest single cost item for a deep gas well.5 The study found that the bulk of rig time was spent drilling (49%) followed by tripping (22%).

Opportunities to reduce these costs include hiking penetration rates through improved bit designs and the development of more powerful and temperature-tolerant downhole motors. A combination of a doubling of both the average deep well penetration rate and bit life was projected to yield about a 20% savings in total well costs, or over $1 million/average U.S. deep gas well. Multiparameter changes to the deep drilling process such as underbalanced or slim-hole drilling may yield as much as a 35% savings over today's deep well costs.

Reducing dry-hole rates

High dry-hole rates for exploratory wells remain a significant barrier to deep gas development and may call for modifications to traditional gas exploration methods.

For example, understanding the role of early hydrocarbon emplacement for preserving reservoir porosity and permeability and identifying areas of intense natural fracturing for enhancing permeability could be employed for improving deep gas exploration success.

One approach for finding naturally fractured settings involves combining satellite-platform imagery for near-surface analysis with aeromagnetics and gravity surveys for basement-feature analysis. This macro-exploration process can be augmented on a prospect level with high resolution 3D seismic, a technique that has been proven successful in other tight gas sand settings.

Completion efficiency

Accumulating evidence shows that deep well completion methods are not particularly effective.

Some needed completion advances for deep, hot, and high-stress formations include improving acid-reaction times during matrix stimulation, maintaining hydraulic fracture fluid stability, and preventing proppant crushing/embedment in hydraulic fractures. Stimulation costs can also be extremely high, requiring a heightened level of understanding with respect to treatment selection/design.

Sour gas costs

Managing H2S and CO2 is troublesome and expensive, beginning with increased safety precautions.

Producing hot, sour gas also requires special-alloy tubulars and sophisticated gas processing. Developing lower cost acid resistant coatings for tubulars and advanced membranes for gas separation may also help lower costs.

Deep gas case studies

Each of the three deep gas case studies, presented next, has been selected with a specific purpose in mind.

The mature Anadarko basin was chosen because of its historical significance and the controversy surrounding the size of the undiscovered deep gas resource.

The Greater Green River basin (plus the Overthrust belt) represents an emerging deep gas basin, judged to hold the largest volume of future deep gas.

Finally, the Wind River basin is a frontier basin where the deep gas resources are yet to be fully assessed but where early successes indicate considerable promise.

Deep Anadarko basin

The Anadarko basin was the center of deep gas drilling and the setting for record-breaking deep wells in the late 1970s-early 1980s.

A gas supply surplus and price drop in the mid-1980s brought this boom to a halt. Still, the strong underlying geologic potential and the lure of big wells remained.

With the aid of technological advances-high resolution 3D seismic, downhole mud motors, PDC bits, underbalanced drilling, and more efficient hydraulic stimulation-a number of savvy veterans, such as Apache Corp. and Sanguine Ltd., and a few new entrants, such as Chesapeake Operating Co., are leading a comeback for deep gas.

The Anadarko is a Paleozoic basin that covers nearly 60,000 sq miles and holds up to 40,000 ft of sedimentary section. Deep gas is in Cambro-Ordovician through Pennsylvanian-age carbonate and clastic reservoirs along the southern edge of the basin.

The first deep gas discovery was in 1956 in Carter-Knox field (Fig. 3 [149,236 bytes]). The well tapped the Ordovician Bromide (Simpson) formation at 15,300 ft, had an open flow potential of 31 MMcfd, and showed prolific gas wells could be developed in reservoirs below 15,000 ft.

Other deep gas discoveries soon followed, for example, Aledo field in Custer County, Okla., and Washita field in the Texas Panhandle.

These early discoveries led to a series of deep, high pressure wells drilled to test the limits of deep gas: GHK's 1-1 Green in Beckham County into the Morrow/Springer at 22,600 ft; Lone Star's 1 Baden, also in Beckham, to a then-record 30,050 ft, but (unfortunately) dry in the deep target; and in 1974, Lone Star's 1 Bertha Rogers, in Washita County to a new record 31,441 ft in the Arbuckle formation.

The drop in gas prices and the failure to find sufficient reservoir quality brought deep gas drilling to a halt in the 1980s and left behind a "graveyard of iron."6

So far in the 1990s, over 200 deep gas wells have been completed as producers in the basin. Verden field (Grady County) has been the most active with 37 new deep gas wells, followed by Knox, Chitwood, Cement, and Reydon fields.

Chesapeake drilled 32 deep gas wells in the 1990s and is the basin's dominant deep gas developer now. Apache had 22 deep wells and Sanguine 23.

One reason deep drilling has resumed is that deep well development success rates have improved steadily, reaching 90% in recent years, while well costs have remained relatively stable (Table 2 [63,844 bytes]).

Prominent deep gas fields covered in the Anadarko basin case study are grouped and further discussed by the four primary deep formations: Morrow/Springer, Hunton, Simpson, and Arbuckle (Table 3 [130,279 bytes]).


The Morrow and Springer formations provide the bulk of deep gas production in the Anadarko basin.

The productive reservoirs include late Mississippian and early Pennsylvanian-age strata above and below 15,000 ft. Historically, the Morrow has also been used as a "bail-out" zone when deeper Hunton or Arbuckle formations are unproductive.

The Morrow/Springer deep gas in this basin is subdivided into two plays:7

  • The Upper Morrow chert-conglomerate alluvial fan play in Roger Mills, Beckham, and Washita counties of western Oklahoma and Hemphill and Wheeler counties of the Texas Panhandle. The deep Morrow distributary channel and stratigraphic trap reservoirs are erratically distributed, posing a challenge to exploration and efficient development.
  • The Springer marine sandstone play in west-central Oklahoma, where the Springer is below 15,000 ft in parts of Caddo, Washita, Beckham, Grady, and Comanche counties. The Springer, a fluvial-deltaic system, reworked in places by longshore currents, provides some of the most prolific gas reservoirs in the Anadarko.

Reydon gas field

Reydon field in Roger Mills County was discovered in 1962 by Gulf Oil Corp. with the 1 Hartley well, perforated in Lower Morrow at 17,414 ft and Upper Morrow at 14,980 ft.

The well and field remained shut in until mid-1970, when gas sales and step-out drilling began. The step-out well, Texas Pacific's 1 Libby, about 1 mile from the discovery well, put Reydon field and the Upper Morrow gas play on the map. The well's CAOF of 174 MMcfd set a basin record.

The Morrow reservoir at Reydon produces gas from three NW-SE trending lenticular sands at 14,022 ft on the field's northern edge to 17,067 ft on the southern edge. It has 20 ft of net pay, 12% porosity, and 15 md permeability. The high reservoir overpressure of 12,600 psi (0.8 psi/ft of depth) provides strong gas flow rates and high gas storage. The gas is 97% methane with 1,000 BTU/cf heating value.

Through 1996, 111 Morrow wells have been completed and about 744 bcf of gas produced, making Reydon the basin's largest deep gas field. EUR is 900 bcf, about 8.5 bcf/well.

Enron Oil and Gas, Samson Resources, and Sanguine (among others) are the field operators. Completed wells cost $1.5-2.5 million. Based on these results and costs, the Morrow formation at Reydon represents an economically attractive gas play if dry holes and marginally productive areas can be avoided.

Elk City gas field

Elk City (Morrow and Springer/Cunningham) field sits atop a NW-SE trending anticline fold in one of the deepest parts of the Anadarko basin. Shell Oil's 5 Rumberger, TD 24,014 ft in 1957, was first to penetrate deep sediments at Elk City.

The Rumberger was abandoned, but its subsurface data laid the foundation for superdeep gas production in the basin. Eventually labeled the Elk City Springer discovery well, the 1-1 Green, was completed in 1969.

Elk City field has produced 316 bcf from Morrow and 88 bcf from Springer. EUR is about 11 bcf/well in Morrow and 20 bcf/well in Springer.

The Morrow and Springer sands at Elk City were 4,000 ft deeper than any previous productive horizon in the basin and helped disprove two widely held beliefs at that time: that deep reservoirs in the Anadarko would lack sufficient porosities for gas production and that high reservoir pressures would harm production.

The deep gas development at Elk City also triggered ultradeep drilling, including the 1 Bertha Rogers at 31,441 ft (into the Arbuckle). This well established that methane was stable at extreme pressures (over 26,000 psi) and high temperatures (485° F.), and that porosity, permeability, and reservoir potential could exist below 29,000 ft.

Verden gas field

Apache has focused its recent deep Springer drilling in Verden field, Caddo County, Okla. Apache began developing this prospect in the 1970s, but as gas prices slumped the 3-mile-deep wells became uneconomic.

Apache reactivated the area in 1990, applying to the deep Springer frac methods learned in other formations. It was able to substantially boost well production, raising productivity in one well from 300 Mcfd to 10 MMcfd.

Apache drilled seven deep Springer wells in 1996 that averaged 7 MMcfd each. Over the last 3 years, it has drilled 28 successful deep Springer wells out of 30 tries, expecting 6+ bcf/well.8

Hunton Group

The Hunton Group is a shallow-water carbonate reservoir complex that lies stratigraphically below the Morrow/Springer.

Diagenesis, paleokarst, and natural fractures are important for Hunton reservoir quality and play a significant role in reservoir development. Reverse faults, resulting from compressional and transpressional tectonics, often underlie the Hunton reservoirs and provide the tectonic overprint for the natural fracture system.

Along the northern edge of the deep Anadarko basin, the Hunton produces at Aledo, North Custer City, and Putnam fields at depths of 14,000-16,400 ft. Cumulative production from these fields is nearly 800 bcf.

In the southern edge of deep Anadarko basin, along the Wichita fault zone, Hunton produces at West and Northeast Mayfield fields in Oklahoma and a series of prolific fields in Texas, including Washita Creek and Mills Ranch. Together these fields have produced over 1.6 tcf of deep gas.

Aledo gas field

J.C. Banner Co. made the deep Hunton discovery at Aledo in 1967. The 1 Walker, completed at 15,385-420 ft, cut 217 ft of pay with greater than 6% porosity, with the top 87 ft averaging 17% porosity. CAOF was 106 MMcfd. Three confirmation wells had CAOFs of 10-100 MMcfd.

Hunton limestone at Aledo is structurally controlled by faults in the north and east, with gas productivity enhanced by natural fractures. EUR is expected to be about 270 bcf-of which 250 bcf has been produced-for an average of 38 bcf/well.

Hunton gas at Aledo is dry with 93% methane, 5% CO2, 2% nitrogen and other components, and has a heating value of 928 BTU/cf. Oxy USA and Phillips Petroleum are the major field operators.

Mayfield gas fields

Apexco established Hun- ton production at West Mayfield in 1976 with the 19,000 ft 1-19 Mills. Its CAOF was 103 MMcfd.

Five miles north, Union Oil completed the 1-33 Annie Bruner at Northeast Mayfield into the Hunton at 24,548 ft, making it the world's deepest producing well at the time. After losing the record to the Chevron 1 Ledbetter (completed into the Arbuckle at Mills Ranch) in 1982, Northeast Mayfield reclaimed the record with the Mesa 2-29 Tipton well, productive at 24,969 ft.

Hunton cumulative gas from these two fields is about 400 bcf, with Northeast Mayfield still producing about 50 MMcfd. The 32 Hunton wells (24 active) in these fields have produced 15 bcf each and are expected to reach 20 bcf each. The gas is high quality (1,100 BTU/cf), with 1.2% CO2 and 0.3% N2. Amoco, Exxon, Helmerich & Payne, and Union are main operators.

Washita Creek gas field

Phillips spudded the 1-C Bowers (Hemphill County, Tex.) in 1965 to test a seismically defined deep, dome-shaped anticline with 1,000 ft of closure. The well was completed in Hunton at 19,474-952 ft and had a CAOF of 265 MMcfd after a 25,000 gal acid treatment.

Since then, Washita Creek has become the largest Hunton gas field, with 11 completions (6 active) and 437 bcf of gas recovery through 1994. Washita Creek also has some of the basin's most prolific gas wells, with the best expected to produce 130 bcf (average 44 bcf). Phillips is the main operator.

Simpson Group

Below the Hunton in the southeast portion of the basin is a series of Ordovician-age sandstones, shales, and carbonates referred to as the Simpson Group. The group includes the Viola, Bromide and Oil Creek formations and has produced over 250 bcf of deep gas.

Knox, Chitwood fields

Knox field began producing deep gas in 1956 from Bromide at 15,251-310 ft. It has produced over 163 bcf of gas from Bromide plus 15 bcf from Oil Creek.

Chitwood field to the north has 23 Bromide wells, of which 15 are still producing, and has yielded 62 bcf of gas and 5.6 million bbl of condensate since 1961. Average EUR is 3 bcf/well, and produced gas is 99% hydrocarbons.

Chesapeake Operating targets four deep carbonate reservoirs-the Viola through Sycamore sequence (the Sycamore is a Mississippian-age limestone above Bromide)-in Knox field. As of end 1997, Chesapeake had drilled 34 of as many as 300 wells projected on 160 acre spacing.

Using Chesapeake's estimate of 3.5 bcf/well, that would make the Simpson carbonates at Knox a 1 tcf field. Chesapeake estimates a cost of approximately $2 million/well.9

Arbuckle formation

The Arbuckle formation, a Cambrian-Ordovician age carbonate sequence, reaches depths greater than 25,000 ft in the deep southern and western parts of the basin and produces some of the deepest onshore gas in the world.

Since the early 1980s, development has slowed almost to a halt; only one new well targeted this formation in 1997. Because of high well costs, limited data on productive extent, and few discoveries, industry perceives the Arbuckle as a high risk play.

West Mayfield field

Helmerich & Payne Inc. completed the first productive Arbuckle well at West Mayfield, 1 Cupp, in mid 1974. The well was perforated in 175 ft of pay at 16,850-17,476 ft and, without treatment, registered a CAOF of 32 MMcfd.

The discovery well led to five more prolific producers. The 1-C Cupp, a northern offset completed in 1975, notching a 120 MMcfd CAOF from 235 ft of pay at 17,280-18,433 ft. The Arbuckle in this field has 100-200 net ft of pay with an average porosity of 6%.

The field's seven completions are spaced at one well per square mile and have produced 70 bcf in total. EUR is put at 83 bcf or 14 bcf/well. The Arbuckle gas is 97.8% methane with 1.5% CO2 and 0.7% N2. Energy content is near 1,000 BTU/cf.

Overthrust-Green River

The Overthrust belt and Greater Green River basin are in southwestern Wyo- ming. Three deep gas producing regions exist within this area: the Overthrust belt itself (in the south), and in the Green River basin the Moxa arch (both north and south), and the Washakie sub-basin (Fig. 4 [194,029 bytes]).

The USGS estimated that the Overthrust belt and the Greater Green River basin have 1.3 tcf of conventional, plus 55.2 tcf of continuous-type (unconventional), un- discovered recoverable deep gas. More deep gas in these basins is included in the growth of already discovered deep gas fields (probable resources).

Even more encouraging, a USGS gas-in-place study for the Greater Green River basin estimated that over 3,000 tcf of deep gas exists in the continuous-type Mesa- verde and Frontier/Cloverly tight gas formations, making this a truly massive storehouse of natural gas.

Overthrust belt

The deep formations in the southern Overthrust belt are the Jurassic age Nugget, the Mississippian Madison (Mission Canyon), and the Ordovician Big Horn. The Nugget sandstone occurs at 15,000 ft in Anschutz Ranch and Bessie Bottom fields, considerably deeper than in the majority of the Overthrust Belt area.

Two gas and 10 oil and gas wells operated by Amoco and JN E&P produce sweet, hydrocarbon-rich gas and condensate from Nugget in these fields. The belief has been that Nugget (and equivalent) sandstones will not be productive below 12,000 ft, but these fields are examples of geologic settings where they are productive.10

The majority of deep gas production from the Overthrust belt is from the Mississippian-aged Madison Group (Mission Canyon and Lodgepole formations) that produces sour gas containing 14-19% CO2 and 4-6% H2S. At Whitney Canyon-Carter Creek field, Madison has produced 1,303 bcf of gas from 30 wells since 1980.

Big Horn is the region's deepest gas producing formation. The two active Big Horn wells in Whitney Canyon-Carter Creek field each averaged 9 MMcfd of gas and 197 b/d of condensate the past year.

Amoco in 1995-96 drilled and recompleted three deep wells in this field. The most productive (dually completed in Madison and Big Horn) has made nearly 17 bcf of gas and 140,000 bbl of condensate in 1 year on line.

Moxa arch

The Moxa arch is a long N-S trending thrust-related structure in the western Greater Green River basin.

Gas production exists throughout the structure, but the productive deep gas formations are at its northern and southern edges. In the southern Moxa arch, deep gas is produced from the Cretaceous-age Dakota and Pennsylvanian-age Morgan formations.

Similar to the Nugget in the Overthrust, Dakota sandstone is buried more than 15,000 ft on the southernmost part of the arch. Sweet Dakota oil and gas are produced from four wells in Lucky Ditch and Whiskey Springs fields by Union Pacific Resources, Apache, and Medallion Exploration.

The next tier of southern fields in the Moxa arch-Butcher Knife Springs, Church Buttes, and Bruff-produce deep gas from the Morgan reservoir, an interbedded limestone and dolomite. Like Madison in the Overthrust belt, Morgan gas is mostly sour, high in H2S and CO2. Most Morgan wells are shut-in, but Wexpro has produced 7 bcf of sour gas from these three fields.

On Moxa's northern extreme is the LaBarge platform, where deep Madison natural gas and CO2 are produced in the sub-thrust area. The upstructure Madison wells (still below 15,000 ft) produce 64% CO2 and 4% H2S. In the downstructure play, CO2 content reaches 90%.

Nevertheless, the Madison wells provide impressive reserves. The 12 wells at Fogarty Creek field have already averaged over 120 bcf each. Industry data on this area, submitted to the Bureau of Land Management, estimated average gas in place of 100-300 bcf/sq mile.

Nineteen wells-at Fogarty Creek, Lake Ridge, and a few smaller fields-produce CO2 and methane from Madison.

Washakie sub-basin

The Washakie sub-basin has the best quality deep gas in the Greater Green River basin. The dominant deep formations are the Jurassic-age Nugget sandstone, Pennsylvanian-age Weber sandstone, and Mississippian-age Madison carbonate.

The Madison formation, Texaco's first deep gas objective in the Table Rock Unit, produced 92 bcf from five wells before watering out in the 1980s. Table Rock field now has 18 deep Nugget and Weber wells that have produced over 200 bcf of gas since 1975.

The eight active Nugget wells currently average about 1.2 MMcfd each; the three currently active Weber wells average 4.3 MMcfd each. Gas from the Nugget and Weber reservoirs is high quality with only traces of CO2 and H2S. Wells are usually frac-stimulated to boost capacity, hiking flows and costs.

Wind River basin

The basin covers 8,500 sq miles in central Wyoming.

Gas has been produced there since 1917, but producers have not pursued the basin's large deep gas resources because of its complex geology, lack of pipeline infrastructure, and environmental constraints. Today, with improving pipeline outlets, independents are again drilling deep, expensive wells.

Recently, the USGS estimated that 441 tcf of gas-in-place exists in the basin's Upper Cretaceous and Lower Tertiary-age low-permeability sands in formations below 15,000 ft. Not included were the Triassic-age Phosphoria and Tensleep formations or the Paleozoic-age Madison formation, currently the basin's most prolific deep gas producing horizons.

The USGS gas-in-place estimates are shown by formation in Table 4 [39,473 bytes]. Some of this deep gas is being produced, particularly from the Frontier formation at Tepee Flats and Cave Gulch/Waltman fields and Cody and Frontier at Madden field, discussed later.

In spite of the large volumes of gas in place, the current USGS estimates for deep gas in the Wind River are modest. The 1996 study assigned 131 bcf of technically recoverable natural gas to three deep conventional plays in this basin and did not quantitatively assess the deep, continuous-type resources (tight gas). In fact, USGS is currently assessing the gas plays in the basin, suggesting that a larger and far more promising resource may exist.

Wind River development

Deep gas development began in earnest in the early 1970s at Madden field11(Fig. 5 [88,909 bytes]). The Madden Deep Unit 1 discovery well drilled into the Cody shale had an initial flow of 7.1 MMcfd and has produced 12 bcf of gas. Subsequent wells have targeted the Shannon and Sussex sandstones within the larger Cody interval.

To date, 42 wells have been drilled to the Cody interval leading to 23 producers (55% success). Twelve of the 23 productive wells appear to be economic (expected recoveries of over 8 bcf/well), and nine are expected to produce 22-50 bcf each.

The basin's deepest currently producing formation is the Madison Group. The first well targeting Madison at Madden field was Bighorn 1-5, spudded in 1983 by LL&E. Even though shut-in until the $45 million, 50 MMcfd Lost Cabin gas treatment plant started up in May 1995, the well has already produced 18 bcf of sour gas.

LL&E (now part of Burlington Resources) shot 3D seismic over the Madden area to establish geologic control for its step-out wells, as play economics do not allow for many dry holes. With the recent completion of two Madden step-out wells, including the Bighorn 4-36 that tested at 44 MMcfd, Burlington is looking to book 1 tcf of proved (developed and undeveloped) reserves and bring the Lost Cabin plant to full capacity.

Following earlier deep Frontier gas production at nearby Tepee Flats and Waltman fields, Barrett Resources in 1997 drilled a deep Wind River basin well, the Cave Gulch 16 Deep Test in the Cave Gulch Unit of Waltman field, targeting the Frontier, Muddy, and Lakota formations at 20,000+ ft. The well was tested in the naturally fractured Muddy sandstone and had an initial unstimulated flow rate of 21.7 MMcfd that fell to 6 MMcfd after a week's tests.

Having established that Muddy could produce naturally at 3-4 MMcfd, Barrett set a bridge plug and completed in the Third Frontier sandstone at about 18,900 ft. The well is producing from the Third Frontier at 10.2 MMcfd after hydraulic stimulation. Other deep gas producible zones (First, Fourth, Fifth Frontiers and the Muddy) exist behind pipe and will be added to the production stream once pressure from the Third Frontier reservoir declines.

Barrett connected to a pipeline in March another superdeep well that encountered gas at high pressure while drilling at 18,175 ft.

Well costs

The time and costs to drill a deep well in the Wind River basin have recently decreased but still remain significant (Table 5 [30,559 bytes]).

Below 17,500 ft, well costs increase significantly. Wells drilled to the deep Madison carbonate are the most costly and time consuming. Stainless steel casing and other precautions must be used to deal with as much as 11% H2S.

LL&E spent $24 million on its first Madison well and almost $20 million on the second.

Other barriers

Deep Wind River gas development has many hindrances: environmental rules, high drilling/completion costs, sour gas, and limited pipeline outlets.

Perhaps most significant is the challenge of finding and delineating the productive areas and gas bearing formations. With R&D, many of these barriers may be mitigated. For example:

  • Better geologic understanding and extended resource characterization could reduce the high, 40-50%, dry hole rates.
  • Natural fracture detection technology could help developers more confidently target the "sweet spots" in the extensive deep tight gas formations.
  • Optimizing drilling practices by having detailed data on formation pressures and lithology could cut drilling costs one third.
Beyond this, improved baseline information for environmental submissions, lower cost drilling muds and corrosion resistant tubulars, and advanced gas processing technology all could lower the costs.

The future

The outlook for U.S. deep gas is promising and represents a substantial future resource for domestic energy supply.

Despite the recent upsurge in activity, however, significant barriers still exist, constraining its development. Decreasing well costs through improved drilling practices, reducing dry-hole rates with more focused deep gas exploration technologies, optimizing completion practices, and finding more cost-effective methods for producing and processing sour gas are critical technical issues that need to be addressed.

Given the high-cost/high-risk view of deep gas by industry, these issues are unlikely to be adequately addressed through experience alone and will require R&D programs be focused on solving these problems. Due to the cross-cutting technical issues facing existing and potential deep gas developers across all basins, collaborative R&D among many parties may provide the most cost-effective and rewarding solutions. This will aid industry to more rapidly unleash the vast deep gas resource.


The authors express their appreciation to the Gas Research Institute, particularly Tom H. Fate and Charles F. Brandenburg, for supporting this study.


  1. Dyman, T.S., Rice, D.D., and Westcott, P.A., Eds., Geologic controls on deep natural gas resources in the U.S., USGS Bull. 2146, 1997.
  2. U.S. deep well drilling leads worldwide activity jump, Petroleum Engineer International, Vol. 68, No. 9, 1995, p. 43.
  3. Joint association survey on drilling costs, 1990-96, American Petroleum Institute, Washington, D.C.
  4. Kuuskraa, V.A., Challenges and opportunities in deep gas, Gas Research Institute presentation, Chicago, September, 1997.
  5. Andersen, E.E., Maurer, W.C., and Pittard, G.T., Deep drilling cost sensitivity analysis, Topical Report, GRI Contract No. 5088-260-1731, MEI No. TR 89-18, August 1989 (draft).
  6. The deep Anadarko basin, Petroleum Information Corp., Denver, 1982.
  7. Bebout, D., et al., Atlas of major Midcontinent gas reservoirs, GRI 93/0260, 1993, pp. 19, 53.
  8. 1996 summary, Apache Corp., Houston, Tex., annual report 1997, p. 11.
  9. Lyle, D., Oklahoma deep carbonates yield major dividends, Hart's Oil & Gas World, September 1996, p. 52, 54.
  10. Fox, J.E., Lambert, P.W., Mast, R.F., Nuss, N.W., and Rein, R.D., Porosity variation in the Tensleep and its equivalent the Weber sandstone, western Wyoming: A log and petrographic analysis, Deep Drilling Frontiers of the Central Rocky Mountains, RMAG Symposium, 1975, p. 185.
  11. Dunleavy, J.M., and Gilbertson, R.L., Madden anticline, growing giant, Rocky Mountain Oil and Gas Fields, Wyoming Geological Association Symposium, 1986, pp. 107-157.

The series

Part 1-Kuuskraa, Vello A., Outlook bright for U.S. natural gas resources, OGJ, Apr. 13, 1998, p. 92.

Part 2-Dyman, Thaddeus S., Schmoker, James W., and Root, David H., USGS assesses deep undiscovered gas resource, OGJ, Apr. 20, 1998, p. 99.

Scott R. Reeves is executive vice-president of Advanced Resources International Inc., where he provides reservoir engineering and field support to numerous domestic and international private clients and manages E&P research projects for the U.S. Department of Energy, the Gas Research Institute, and others. He also is an international consultant to the United Nations and private industry. He holds a BS degree in petroleum engineering from Texas A&M University and an MBA degree from Duke University.
Jason Kuuskraa is a consulting associate with ARI. He specializes in the evaluation and financial analysis of unconventional and emerging natural gas resources. He prepared the data bases and structure for the new Unconventional Gas Model used by the U.S. DOE/EIA's National Energy Modeling System. He holds degress in mathematics and history from Boston College.
Vello A. Kuuskraa is president of ARI. He was a 1985-86 Society of Petroleum Engineers Distinguished Lecturer, served on the Secretary of Energy's "Assessment of the U.S. Natural Gas Resource Base," and was a member of the National Academy of Sciences' Committee on the National Energy Modeling System. He received an MBA degree (highest distinction) from the Wharton School, University of Pennsylvania, and a BS degree in mathematics/economics from North Carolina State University.

Copyright 1998 Oil & Gas Journal. All Rights Reserved.