- Copyright © 2002. The American Association of Petroleum Geologists. All rights reserved.
Basin-centered gas systems (BCGSs) are potentially one of the more economically important unconventional gas systems in the world; in the United States they contribute as much as 15% of the total annual gas production. These regionally pervasive gas accumula tions are different from conventionally trapped accumulations in several respects. The basin-centered gas accumulations (BCGAs) associated with BCGSs are typically characterized by regionally pervasive accumulations that are gas saturated, abnormally pres sured, commonly lack a downdip water contact, and have low-permeability reservoirs. The accumulations range from single, isolated reservoirs a few feet thick to multiple, stacked reservoirs several thousand feet thick. Two types of BCGSs are recognized; a direct type, characterized by having gas-prone source rocks, and an indirect type, characterized by having liquid-prone source rocks. During the burial and thermal histories of these systems, the source rock differences between the two types of BCGSs result in strikingly different characteristics that impact exploration strategies. The majority of known BCGAs are the direct type. Exploration activity for BCGAs is in the early stages and thus far has been focused in North America. In other parts of the world, concepts of basin-centered gas systems are poorly known, and exploration activity focused on basin-centered gas accumulations is minimal.
Ben Law is a consultant and sole proprietor of Pangea Hydrocarbon Exploration LLC. His research interests include basin-centered gas and coalbed methane systems. Prior to his consulting position, he was a member and chief of the U.S. Geological Survey Western Tight Gas Sand Project and regional coordinator of South Asia for the U.S. Geological Survey World Energy Project. He received B.S. and M.S. degrees from San Diego State University, California.
The global distribution of gas is not uniform. Some regions, like Russia and the Middle East, have extremely large gas resources to meet their energy demands, whereas other regions, like Japan and Western Europe, have limited amounts of gas and must rely on importing gas to meet their energy demands (DOE/EIA, 2002). The increasing demand for energy in many parts of the world has made it imperative to explore for and exploit unconventional oil and gas resources. One of the larger and more economically viable unconventional gas resources occurs in basin-centered gas accumulations (BCGAs). The BCGAs constitute a realistic, near-term energy resource that has only recently been the focus of exploration. However, with few exceptions, there is a generally poor understanding of BCGAs. Consequently, exploration efforts for this huge gas resource are not as effective as they might be. To develop more effective exploration strategies for BCGAs, it is necessary to modify traditional concepts of petroleum systems and include concepts of nontraditional, unconventional petroleum systems.
In light of the extremely large gas resources contained in BCGAs and the need to satisfy increasing energy needs, the primary objective of this article is to provide a comprehensive overview of BCGAs. More specific objectives include clarification of gas-system nomenclature, providing the elements and processes of basin-centered gas systems (BCGSs), and discussing the origins, geographic and stratigraphic distribution, the gas resource, exploration strategies, and formation evaluation. To accomplish these objectives, BCGAs are discussed in the context of a petroleum system.
Historical Development and Classification
Of all the different types of unconventional gas systems, none have been more poorly defined than BCGSs. The problem of definition has led to misconceptions that have, in some cases, impeded exploration efforts. When regionally pervasive gas accumulations, like BCGAs, became known is uncertain; however, Silver (1950) alluded to pervasive gas accumulations in Cretaceous rocks in the San Juan basin of New Mexico and Colorado and recognized the gas-saturated nature of the reservoirs and the downdip absence of water. Later, the nuclear stimulation experiments conducted in the United States from 1967 to 1973 seem to have implied knowledge of the presence of regionally pervasive gas-charged reservoirs, although there are no geologic reports confirming this assumption. Nuclear detonations conducted in Cretaceous rocks in the San Juan and Piceance basins of New Mexico and Colorado were unsuccessful, and eventually, because of concerns about environmental and radioactive contamination issues, the tests were abandoned (Randolph, 1973, 1974a, b, c).
The first published, unmistakable reference to this type of gas accumulation was by Masters (1979). In his article, Masters (1979) identified the basic concepts of basin-centered gas accumulations, referring to them as "deep basin gas," and provided several defining characteristics of gas-saturated reservoirs in the Deep basin of Alberta, Canada, and in the San Juan basin of New Mexico and Colorado as examples of such accumulations. Later, in a publication edited by Masters (1984), the various geologic aspects of the so-called deep basin gas accumulation in the Elmworth field of Alberta, Canada, were provided. Other significant early articles concerning BCGAs include those by Law et al. (1979, 1980) and Law (1984) in the Greater Green River basin of Wyoming, Colorado, and Utah, and McPeek (1981) in the Great Divide basin of Wyoming. Spencer (1985, 1989a) and Law and Spencer (1993) described many of the attributes common to BCGAs. Several examples of so-called tight gas reservoirs (in most cases equivalent to BCGAs) in the United States are provided in a volume edited by Spencer and Mast (1986). Finley (1984) and Dutton et al. (1993) also described many additional low-permeability reservoirs.
When the term "basin-centered gas accumulations" came into use is uncertain; however, the first published reference to the term was by Rose et al., (1986) in a study of gas accumulations in the Upper Cretaceous Trinidad Sandstone of the Raton basin. It is likely, however, that the term " basin-centered gas accumulations" had been informally used by industry people prior to the first published reference of the name.
The term "tight gas sands" has been widely used to describe BCGAs for many years, and many exploration geologists still use that term. In many cases "tight gas sands" is an appropriate term; however, it is somewhat ambiguous and may include gas accumulations that are trapped as conventional, buoyant accumulations. The use of the term "deep basin gas" (Masters, 1979) has some problems also because all BCGAs do not occur at great depths. For example, much of the gas production in the San Juan basin is from BCGAs at depths as shallow as 3000 ft (914 m). More recently, the term "deep basin gas" has been defined as those gas accumulations deeper than 15,000 ft (4572 m) (Dyman et al., 1997); it is an economic definition and is not based on geologic processes. Finally, the term "continuous gas accumulation" (Schmoker, 1996), although accurately portraying the pervasive nature of BCGAs, is too broad and includes such gas systems as coalbed methane and shale gas. In the absence of other suitable names, the term "basin-centered gas accumulation" is used in this article, although there are some BCGAs that appear, at first glance, to contradict the definition. Gas fields such as the Jonah field in the northern part of the Green River basin in Wyoming and the Natural Buttes field in the Uinta basin in Utah are examples of BCGAs with downdip water contacts. These fields, in my opinion, are gas chimneys rooted in deeper, regionally pervasive BCGAs.
Basin-Centered Gas Systems
A petroleum system, as defined by Magoon and Dow (1994, p. 3), "includes all the elements and processes needed for an oil and gas accumulation to exist." In Magoon and Dow's (1994) definition, the elements include source rock, reservoir rock, seal rock, and overburden rock. Relevant processes include trap formation and the generation, expulsion, migration, and accumulation of petroleum. A BCGS contains all of these components; however, the magnitude and function of some of the components interact to form a unique type of hydrocarbon accumulation.
In general, BCGAs are regionally pervasive accumulations that are gas saturated, abnormally pressured (high or low), commonly lack a downdip water contact, and have low-permeability reservoirs. In the context of a petroleum system, there are two types of basin-centered gas systems: a direct type and an indirect type (Law, 2000). The attributes of these two types of systems are provided in Table 1. Direct and indirect types of BCGSs are distinguished on the basis of source rock quality; a direct BCGS has a gas-prone source rock, and an indirect BCGS has an oil-prone source rock. This fundamental difference, oil-prone vs. gas-prone source rocks, leads to significantly different characteristics, as shown in Table 1. In addition to the two types of systems, there may be hybrid systems in which gas-prone and liquid-prone source rocks have contributed to the development of a BCGA.
The developmental history of a BCGS may be viewed as four reservoir pressure cycles. As a consequence of the dynamic nature of geologic processes and the response to those processes, the phases discussed here are geologically ephemeral. Figure 1 is a diagrammatic representation showing these pressure phases and the development of direct and indirect BCGSs. Meissner (1978) and Law and Dickinson (1985) discussed these phase changes for gas accumulations in low-permeability reservoirs.
Direct and Indirect Systems
During the early burial and thermal histories of direct and indirect systems, the reservoirs are, for the most part, normally pressured, and the fluid phase in the pore system is 100% water saturated (Figure 1). Compaction of framework grains during this phase is an important process. The defining processes for each system, however, are different. For direct systems, phase I terminates with the initiation of thermal gas generation, whereas the termination of phase I in indirect systems occurs with the initiation of thermal cracking of oil to gas. Reservoir quality in indirect systems during phase I is assumed to be relatively better than reservoir quality in direct systems because buoyant accumulations of oil require better porosity and permeability.
During phase I there may be some cases in which reservoir pressures are overpressured. Law and Spencer (1998) suggested that in the early burial stages of a BCGA sequence, prior to the development of a recognizable BCGA, and in some depositional settings of rapid sedimentation, compaction disequilibrium may have been the initial overpressuring mechanism. In this scenario, the pressuring fluid phase is water. However, as the sequence experiences further burial and hotter temperatures, the compaction disequilibrium pressure mechanism may be replaced by hydrocarbon generation and the development of abnormally high pressures characterized by pore fluids composed of gas and little or no water. A possible example of the transition of pressure mechanisms from compaction disequilibrium to hydrocarbon generation may be present in Miocene and Pliocene rocks in the Bekes basin (Spencer et al., 1994) and the Mako trench (B. E. Law, 2000, unpublished data) of Hungary. In these areas, Miocene and Pliocene rocks are overpressured and possess many of the distinguishing characteristics of a BCGA. The overpressures in Miocene rocks appear to be caused by hydrocarbon generation, whereas overlying, overpressured Pliocene rocks appear to be in a transitional pressure phase between compaction disequilibrium and hydrocarbon generation. In this case, a knowledge of pore fluid composition (mainly gas or mainly water) in the Pliocene sequence would offer considerable insight in resolving the problem.
Direct systems require gas-prone source rocks and low-permeability reservoirs in close proximity to each other. As the source and reservoir rocks undergo further burial and exposure to increasing temperatures, the source rocks begin to generate gas (Figure 1). Concomitant with increased gas generation, expulsion, and migration, gas begins to enter adjacent, water-wet sandstones. Because these sandstones have low permeability, the rate at which gas is generated and accumulated in reservoirs is greater than the rate at which gas is lost. Eventually, as newly generated gas accumulates in the pore system, the capillary pressure of the water-wet pores is exceeded, and free, mobile water is expelled from the pore system, resulting in the development of an overpressured, gas-saturated reservoir with little or no free water. Examples of BCGA systems exhibiting this overpressured phase include the Greater Green River (Law, 1984), Wind River (Johnson et al., 1996), Big Horn (Johnson et al., 1999), and Piceance basins (Johnson et al., 1987) in the Rocky Mountain region of the United States and the Taranaki Basin in New Zealand (B. E. Law, 2000, unpublished data) (Table 2).
In contrast to direct systems, indirect systems require a liquid-prone source rock (Figure 1). Reservoir quality in indirect systems is assumed to have been better than in direct systems. In this case, oil and gas are generated and expelled and migrate to reservoirs where they accumulate in structural and stratigraphic traps as discrete, buoyant accumulations with downdip water contacts. With subsequent burial and exposure to higher temperatures, the accumulated oil undergoes thermal cracking to gas, accompanied by a significant increase of fluid volume and pressures (Barker, 1990). The level of thermal maturity at which oil is transformed to gas is commonly thought to be about 1.35% vitrinite reflectance (Ro) (Tissot and Welte, 1984; Hunt, 1996); however, some evidence, discussed in a following section, indicates that the transformation may occur at higher levels of thermal maturity. Alternatively, gas derived from thermally cracked oil within a source rock may subsequently be expelled and migrate to low-permeability reservoirs (Garcia-Gonzales et al., 1993a, b; MacGowan et al., 1993; Hunt, 1996). Under these conditions of changing fluid volume and pressure, the capillary pressure of the water-wet pore system is exceeded, and, like pore pressures in direct systems, the high pressures forcibly expel mobile, free water from the pore system, replacing water with gas, and the development of an overpressured BCGA ensues. An additionally important aspect of this phase is the necessity for the presence of an effective lithologic top seal in reservoirs formerly occupied by discrete oil accumulations.
At the point where direct and indirect systems are in the overpressured phase (phase II), the processes involved in the transition to phase III are identical for both systems (Figure 1). Phase III occurs when the overpressured phase of direct and indirect systems evolves into underpressured conditions. Both systems, subsequent to the phase II history of overpressure, may experience a period of uplift and erosional unloading and/or heat flow perturbations. During, or subsequent to, these burial and thermal history disruptions, some gas is lost from the accumulation, and the overpressured gas reservoirs are subjected to reduced temperatures. The loss of gas, in conjunction with reduced temperatures, effectively results in the development of an underpressured BCGA (Meissner, 1978; Law and Dickinson, 1985). During this pressure transition, Meissner (2000) emphasized the importance of gas loss over temperature reduction as the dominant process.
Conjectural evidence concerning the integrity of seals in direct vs. indirect systems implies that gas is lost more easily from direct BCGAs than from indirect BCGAs. Johnson et al. (1994) have shown that gas in conventionally trapped accumulations in several Rocky Mountain basins originated from BCGAs, demonstrating that loss of gas through relative permeability, capillary pressure seals does occur. Examples of underpressured, phase III direct systems include Cretaceous rocks in the San Juan, Raton, and Denver basins, and examples of underpressured, phase III indirect systems include Lower Silurian reservoirs in the Appalachian basin, Ordovician reservoirs in the Risha area of eastern Jordan, and Cambrian and Ordovician reservoirs in the Ahnet basin of Algeria (Table 2).
Phase IV is theoretical and may be more applicable to direct systems because of the perceived, relatively better quality of seals in indirect systems than seals in direct systems. During phase IV, continued loss of gas from capillary pressure seals in BCGAs is accompanied by water slowly reentering underpressured, gas-bearing reservoirs. Under these conditions, Meissner (1978) and Law and Dickinson (1985) hypothesized that the underpressured, gas-bearing reservoirs would eventually evolve into normally pressured, water-bearing reservoirs, thus completing the pressure cycle.
System Elements and Processes
Source rock quality is the fundamentally most important element distinguishing direct from indirect BCGSs and sets the stage for all subsequent differences between the two systems. The source rocks for direct BCGSs are most commonly humic-type coal beds and carbonaceous shale, such as occur in Cretaceous rocks in most Rocky Mountain basins or Carboniferous rocks in Europe. Source rocks for indirect BCGSs are hydrogen-rich shales such as those in the Ordovician shale in the Appalachian basin or in Silurian shales in the Middle East and North Africa. Garcia-Gonzales et al. (1993a, b), MacGowan et al. (1993), and Surdam et al. (1997) concluded that some of the coal beds in the Greater Green River basin of Wyoming (Upper Cretaceous Almond coal beds) generated liquid hydrocarbons that were subsequently thermally cracked to gas, while still in the coal beds. They further speculated that, because of the increased fluid volume associated with the oil to gas transformation, high pressures created fractures within the coal beds, facilitating the expulsion of gas. The gas then migrated and accumulated in low-permeability reservoirs. Law (1984) concluded that all, or most, of the gas in low-permeability reservoirs in the Greater Green River basin was sourced from humic, type III organic matter contained in coal beds and carbonaceous shale in several coal-bearing Upper Cretaceous intervals. The relative contribution of gas to BCGA reservoirs from these different processes is not known. In the Greater Green River basin BCGA, the gas likely is dominantly sourced directly from gas-prone, humic coal beds in the Lance, Almond, and Rock Springs formations, with a minor contribution from the cracking of oil to gas in Almond Formation coal beds in the very deepest part of the Great Divide and Washakie basins.
Gas-charged reservoirs in direct and indirect BCGSs are regionally pervasive, commonly encompassing several thousand square miles, and may consist of single, isolated reservoirs a few feet thick or vertically stacked reservoirs several thousand feet thick. Multiple, stacked reservoirs are common in direct BCGSs, whereas single, discrete reservoirs are common in indirect BCGSs. Direct and indirect reservoirs always exhibit low porosity (<13%) and low, in-situ permeability (<0.1 md) (Spencer, 1985, 1989a). They are composed of sandstone, siltstone, and, to a much lesser degree, carbonates; the only occurrence of a BCGA carbonate reservoir known to me is in the Sichuan basin of China (Da-Jun and Yun-ho, 1994). The environments of deposition of BCGA reservoirs range from marine to nonmarine. Reservoirs are gas-saturated, with little or no producible water, and are downdip from water-bearing reservoirs (Figure 2), a reversal of conditions found in conventional gas systems (Masters, 1979; Law, 1984; Spencer, 1985, 1989a).
The BCGS reservoirs can be divided into lenticular and blanket reservoirs (Finley, 1984; Spencer, 1985, 1989a). Lenticular reservoirs, such as small, fluvial channel sandstones, typically have a limited pore volume and very low permeability. In contrast, blanketlike reservoirs, such as braided stream, delta front, and eolian sandstones, typically have very large pore volumes and relatively better permeability than lenticular reservoirs. The distinction between these types of reservoirs becomes important when attempting to distinguish between gas- and water-bearing reservoirs and is an important factor in the design of drilling and completion programs.
In thick, vertically stacked direct BCGA reservoirs, interbedded water-bearing reservoirs are not uncommon. For example, the blanketlike Upper Cretaceous Ericson Sandstone of the Mesaverde Group in western Wyoming is a water-bearing reservoir interbedded with a regionally pervasive BCGA. Additional examples in western Wyoming of interbedded, water-bearing reservoirs occur in the Upper Cretaceous Frontier and Blair formations, Almond sandstone, and Lewis Shale. Examples of water-bearing reservoirs also exist in the Elmworth field BCGA, Alberta basin, Canada. The Rollins and Trout Creek members of the Mesaverde Formation in the Piceance basin of Colorado are blanket reservoirs that are water bearing (Johnson et al., 1987). The occurrence of water in thick, BCGA sequences is also possible through the introduction of water along fractures and faults.
Where detailed work has been conducted in direct BCGAs, gas-saturated reservoirs grade vertically, across stratigraphic boundaries, as well as updip into transitional, water- and gas-bearing zones that, in turn, grade into normally pressured, water-bearing reservoirs (Figure 2). In indirect BCGAs, gas-saturated reservoirs grade updip into transitional, water- and gas-bearing zones; however, vertical transitional zones across bed boundaries do not occur, and there is an abrupt, distinct boundary between the abnormally pressured BCGA and normally pressured, water-bearing reservoirs (Figure 2). The nature of these fluid boundaries is related to seal integrity. Seals in BCGAs range from lithologic to relative permeability, or water-block, seals, referred to in this article as capillary pressure seals. Capillary pressure seals generally occur in reservoirs that have very small pore throats and two or more fluid phases. Under these conditions, the permeability to each fluid phase is effectively reduced.
Because of the nature of seals in direct BCGAs, a question arises concerning the integrity of the seal. Based on burial and thermal history reconstructions, capillary pressure seals in Cretaceous and Tertiary BCGAs in the Rocky Mountain region are effective for periods of time ranging from 25 to 40 m.y., the lapsed time since formation of most BCGAs in the region. However, as a consequence of the nature of these seals, there is a perception that the seals are leaky and, given sufficient time, will degenerate and become ineffective. If the perception of a leaky seal over significantly long periods of time is correct, then one might expect to see a predominance of direct BCGAs in rocks that have experienced the formation of a BCGA within a few tens of million years. Also, in a more general sense, one would expect to observe a higher frequency of direct BCGAs in younger rocks than in older rocks. Observations of known BCGAs are skewed toward Cretaceous systems, largely because most of the work conducted on BCGAs has been in Cretaceous and Tertiary rocks. There are no detailed studies of seal integrity in pre-Cretaceous BCGAs, although Ryder and Zagorski (forthcoming) have concluded that the updip seal in the Lower Silurian Clinton-Medina BCGA in the Appalachian basin is a water block. In indirect systems, it is important to distinguish between vertical seals, top seals, and updip seals; in the Clinton-Medina reservoir, there is an apparently effective updip, capillary pressure seal (Ryder and Zagorski, forthcoming), whereas the upper, top seal is a lithologic seal composed of evaporite, shale, and carbonate (Drozd and Cole, 1994).
Hydrocarbon Generation, Expulsion, and Migration
There is a large body of literature concerning hydrocarbon generation, expulsion, and migration (see Hunt  for detailed discussions). As depicted in Figure 1, the generation of hydrocarbons from source rocks in direct and indirect BCGSs occurs at levels of thermal maturity exceeding 0.6% Ro (Hunt, 1996). According to Meissner (1984), thermal generation of gas from humic coal beds begins at 0.73% Ro. Peak generation may occur at levels of thermal maturity between 0.8-0.9% Ro (Tissot and Welte, 1984). In the Greater Green River basin, measured levels of thermal maturity at the top of direct BCGAs range from 0.7 to 0.9% Ro (Law, 1984), implying that source beds for the gas would have levels of thermal maturity equal to or greater than 0.7-0.9% Ro.
The level of thermal maturity marking the transformation of oil to gas in indirect systems (initiation of phase II on Figure 1) is uncertain. Conventional wisdom indicates that thermal cracking of oil to gas occurs at about 1.35% Ro (Tissot and Welte, 1984; Hunt, 1996). Price (1997) questioned this value and concluded that the transformation of oil to gas occurred at much higher levels of thermal maturity. More recent kinetic studies by Tsuzuki et al. (1999) using hydrous pyrolysis experiments also suggest that oil is stable over higher levels of thermal maturity than previously thought. Applying these kinetic parameters to burial history curves in the United States Gulf Coast indicates that oil cracking to gas starts at vitrinite reflectance values of 1.75% Ro (M. D. Lewan, 2002, personal communication).
In general, hydrocarbon migration distances in direct BCGSs are short, perhaps on the order of a few hundred feet or less. The exception to short hydrocarbon migration distances may occur in cases where the regional top of a BCGA has been ruptured, facilitating vertical migration of gas along faults and fractures for distances far greater than a few hundred feet, such as in the Jonah field of western Wyoming, discussed in a following section.
In indirect BCGSs, hydrocarbon migration distances are highly variable, similar to migration distances in conventional petroleum systems. Approximately 1000 ft (305 m) of vertical migration is proposed for the Clinton-Medina BCGA (Ryder and Zagorski, forthcoming). In direct BCGSs, gas is the dominant migrating hydrocarbon phase, and in indirect BCGSs, oil and gas may be expected to be the migrating fluid phases.
The development of a trap in the conventional sense of a structural or stratigraphic trap is an important process in a petroleum system. In a direct BCGS system, however, it is of secondary importance, whereas in indirect systems it is very important. In direct systems, the top of gas accumulations cuts across structural and stratigraphic boundaries (Law, 1984; Spencer, 1985; Law and Spencer, 1993) and is, therefore, not normally dependent on the development of structural or stratigraphic traps. The Jonah field of western Wyoming (Figures 3, 4) is a good example of a direct BCGA in which structural and stratigraphic aspects are important. The lateral boundaries of the field are defined by faults (Montgomery and Robinson, 1997; Warner, 1998, 2000). The top of the accumulation is defined by a silty shale seal in the Upper Cretaceous Lance Formation.
In the development of indirect BCGAs, a conventional structural or stratigraphic trap is necessary for the accumulation of oil and gas, much the same way as oil and gas accumulate in conventional, buoyancy-driven accumulations. The development of indirect BCGAs occurs at a later burial stage than direct systems, when conventionally accumulated oil is thermally cracked to gas, accompanied by a significant increase in pore fluid volume and pore pressure (Figure 1). Oil, however, does not always accumulate in discrete accumulations and may be disseminated throughout a reservoir. In such cases, the amount of oil in the accumulation may not be present in sufficient quantity to develop pore pressures high enough to form a BCGA during the thermal conversion of oil to gas. Thus, the formation of a suitable trap and the temporal relationships among trap formation and gas generation, expulsion, migration, and entrapment are critical processes in indirect systems.
Examples of Gas Systems
To illustrate the elements and processes of direct and indirect BCGSs, an example of each system is included in the following discussion. Additional examples are provided in Table 2.
Direct Type: Greater Green River Basin
The Greater Green River basin, located in southwestern Wyoming (Figure 3), is one of several foreland basins in the Rocky Mountain region containing BCGSs. The stratigraphic interval containing the BCGS includes all of the Cretaceous sequence, locally extending into lower Tertiary rocks. Stratigraphic correlations of lower Tertiary and Cretaceous rocks in the Greater Green River basin are shown in Figure 5. For a comprehensive discussion of the stratigraphy and structure of the basin see Ryder (1988). Estimates of in-place gas resources contained in the BCGA within Cretaceous and Tertiary rocks are as large as 5063 tcf (Law et al., 1989), and the mean estimate of recoverable gas is 119.3 tcf (Law, 1996). Additional references to a BCGS in the Greater Green River basin include publications by Law et al. (1979, 1980), McPeek (1981), Davis (1984), Law (1984), Keighin et al. (1989), Law and Spencer (1989), Spencer (1989b), Surdam (1992), Garcia-Gonzales et al. (1993a, b), MacGowan et al. (1993), and Surdam et al. (2001). General characteristics of the Greater Green River basin BCGS are as follows:
Area: 19,700 mi2 (51,000 km2)
Source rocks: Upper Cretaceous and lower Tertiary coal beds and carbonaceous shales in the Fort Union, Lance, Almond, and Rock Springs formations. Organic matter is largely gas-prone type III kerogen (Law, 1984) with additional contribution from thermally cracked oils sourced from sapropelic coal beds (Garcia-Gonzales et al., 1993a, b; MacGowan et al., 1993; Surdam et al., 1997).
Generation-expulsion-migration: late Eocene-late Oligocene (40-25 Ma)
Reservoir rocks: Cretaceous to lower Tertiary sandstones. Multiple, stackedreservoirs occur in rock intervals as thick as 14,000 ft (4267 m) (Figure 6). Individual reservoirs range in thickness from 15 to 125 ft (4.6-38 m). Gas reservoirs are saturated and contain water at irreducible levels. The gas-bearing interval does not commonly contain interbedded, water-bearing reservoirs.
Permeability: <0.1 md (in-situ)
Environments of deposition: mainly fluvial dominated and, to a lesser degree, marginal marine deltaic and barrier bar
Reservoir pressure: overpressured, with gradients ranging from 0.5 to 0.9 psi/ft (Figures 7, 8) ( Law et al., 1979, 1980; McPeek, 1981; Davis, 1984; Law, 1984; Spencer, 1987, 1989b; Surdam et al., 1997)
Seals: Regional seals are capillary pressure seals. Locally, structural andstratigraphic seals are important.
Gas accumulations: downdip from normally pressured, water-bearing reservoirs (Figure 2) (Law, 1984; Spencer, 1985); lacks a downdipwater contact (Law, 1984). The level of thermal maturity at top of accumulation ranges from 0.7 to 0.9% Ro (Law, 1984) (Figures 7, 8), commonly 0.8% Ro (Law, 1984).
Depth to accumulation: ranges from 8000 to 11,500 ft (2438-3505 m)
Gas quality: Gas is of a thermal origin and generally composed of >90% methane, <5% ethane and higher homologs, <5% carbon dioxide, and negligible nitrogen. Condensate ranges from <5 to 70 bbl/mmcf gas.
Sweet spots: structural and stratigraphic
Indirect Type: Lower Silurian Clinton-Medina-Tuscarora, Appalachian Basin
The Lower Silurian Clinton-Medina-Tuscarora BCGS, located in the Appalachian basin (Figures 9, 10), is one of the better documented examples of an indirect BCGS. Estimates of recoverable resources range from 8.0 to 30.3 tcf (Gautier et al., 1996; McCormac et al., 1996). For additional discussions of the Clinton-Medina-Tuscarora, refer to investigations by Davis (1984), Law and Dickinson (1985), Laughrey and Harper (1986), Zagorski (1988, 1991), Law et al. (1998a), Ryder (1998), and Ryder and Zagorski (forthcoming). The critical elements in this system include the following:
Area: Clinton-Medina part is 45,000 mi2 (116,550 km2); Tuscarora part is 30,000 mi2 (77,700 km2).
Source rock: Ordovician Utica Shale (Cole et al., 1987; Drozd and Cole, 1994; Burruss and Ryder, 1998; Ryder et al., 1998). The Utica Shale contains type II kerogen and is thermally overmature (>1.3% Ro).
Generation-migration-accumulation: Late Devonian-Early Mississippian (370-320 Ma) (Drozd and Cole, 1994; Laughrey and Harper, 1996; Nuccio et al., 1997; Ryder et al., 1998; Ryder and Zagorski, forthcoming)
Reservoir rocks: Lower Silurian Clinton-Medina in eastern Ohio and western Pennsylvania and Tuscarora Sandstone in central Pennsylvania. The reservoir interval ranges in thickness from 100 to 600 ft (30-183 m) (Ryder and Zagorski, forthcoming). The thermal maturity of the reservoir ranges from 1.1 to 2.0% Ro (Wandrey et al., 1997).
Reservoir pressure: Reservoirs are normally pressured in the updip part of the Clinton-Medina in eastern Ohio, producing oil, gas, and water (Figures 9, 11). In western Pennsylvania reservoirs are underpressured and produce mainly gas with very small amounts of water (Figures 9, 11). Ryder and Zagorski (forthcoming) reported pressure gradients of 0.39-0.25 psi/ft in the underpressured part of the system. In central Pennsylvania, the Tuscarora Sandstone, equivalent to the Clinton-Medina, is overpressured and produces gas with small amounts of water (Figures 9, 11). Ryder and Zagorski (forthcoming) reported pressure gradients ranging from 0.50 to 0.60 psi/ft in the overpressured Tuscarora Sandstone incentral Pennsylvania. The variable pressure gradients within the stratigraphic interval are shown in Figure 12.
Seals: The top seal is interpreted to be the shales, carbonates, and evaporites in the overlying Upper Silurian (Drozd and Cole, 1994). The updip seal has been identified as a water block (Zagorski, 1988, 1991; Ryder and Zagorski, forthcoming).
Gas accumulations: Downdip from normally pressured, water-bearing reservoirs; lacks downdip water contact (Figure 11)
Depth to accumulation: 6500 ft (1981 m) in western Pennsylvania to 12,000 ft (3658 m) in central Pennsylvania (Ryder and Zagorski, forthcoming)
Gas quality: Gas is interpreted to be a product of thermally cracked oil (Law and Dickinson, 1985; Law and Spencer, 1993; Law et al., 1998; Ryder and Zagorski, forthcoming). Gas in the Clinton-Medina sandstone is generally composed of 79-94% methane; 3-12% ethane, propane, and C4+ hydrocarbon; and 3-9% nitrogen and carbon dioxide (Burruss and Ryder, 1998; Ryder and Zagorski, forthcoming). In the Tuscarora Sandstone, gas is commonly dry (C1/C1-5 = 0.98-0.99), with nitrogen and carbon dioxide contents of 4-22% and <1-83%, respectively (Ryder and Zagorski, forthcoming).
Sweet spots: structural and stratigraphic
The underpressured reservoirs in the Clinton-Medina are interpreted to haveundergone an earlier overpressured phase caused by the thermal transformation of oil to gas (Law and Dickinson, 1985; Law and Spencer, 1993; Law et al., 1998a; Ryder and Zagorski, forthcoming). Later, during a period of regional uplift accompanied by loss of gas and reservoir cooling, the overpressured, gas-bearing Clinton-Medina underwent a transition to an underpressuring phase. The overpressured, gas-bearing Tuscarora Sandstone reservoirs in central Pennsylvania are interpreted to be pressure remnants of the earlier overpressured phase in the Clinton-Medina (Law et al., 1998a; Ryder and Zagorski, forthcoming).
Globally, no resource data are available for BCGAs; however, where estimates of in place gas have been made the resources are very large: in-place gas resource estimates in the United States for a given BCGA are generally greater than 10 tcf (Table 3). Unfortunately, no comprehensive gas resource data exist for all BCGAs in the United States, in large part because BCGAs are not recognized as a distinct type of gas accumulation. However, an appreciation for the magnitude of the resource can be determined from estimates of in-place and recoverable gas in selected areas of the United States. Previous assessments of in-place gas in so-called tight and basin-centered accumulations in the United States are shown on Table 3. Using a volumetric methodology approach, in-place gas resources for the Piceance (Johnson et al., 1987), Greater Green River (Law et al., 1989), Wind River (Johnson et al., 1996), and Big Horn basins (Johnson et al., 1999) in the Rocky Mountain region were estimated at 6788 tcf. In-place gas estimates in other basins range from 334 to 777 tcf (Table 3).
There are even fewer published estimates of recoverable gas. In the United States, the National Petroleum Council (1992) estimated 232 tcf of recoverable gas from so-called tight reservoirs with current technology and 349 tcf of gas with advanced technology. In 1995, the U.S. Geological Survey (Gautier et al., 1996) included as part of their National Assessment several plays in seven basins that were determined to contain BCGAs. In that assessment, in-place gas resources were not estimated, and a methodology developed by Schmoker (1996) for estimation of recoverable gas was used. Estimates of recoverable gas from those basins are 223.55 tcf (Table 4). If all of the basins in the United States containing BCGAs would have been assessed, it is highly probable that the total recoverable gas in BCGAs in the United States would exceed 400 tcf.
Production figures for the United States, like gas resource assessments, are uncertain; however, the estimate that most accurately reflects gas production from BCGAs in the United States is provided by the Gas Research Institute (GRI). Gas production from so-called tight gas sands in 1996 (the last year for which there are records) was 3.35 tcf (Hill, 2000), approximately 17% of total United States production. This figure, however, may be misleading because all so-called tight gas sands are not necessarily BCGAs; discrete, buoyancy-driven gas accumulations may occur in some tight gas sands. With this caveat under consideration, I estimate 15% of annual gas production in the United States is from BCGAs, the largest, gas-producing contributor of all unconventional gas accumulations.
Spatial and Temporal Distribution
The global distribution of BCGAs is poorly known, and knowledge of the stratigraphic distribution of BCGAs is incomplete. Even in North America, where most of the exploration activity for BCGAs has occurred, the geographic distribution is not well known. Figure 13 shows the locations of known and suspected BCGAs in the United States. A tabulation of these areas, as well as areas outside North America, is shown in Table 2. The geographic distribution of BCGAs is probably best known in the Rocky Mountain region, where a considerable amount of research has occurred.
Worldwide, there are only a few references available alluding to the presence of BCGAs (Table 2). Many more areas undoubtedly contain BCGAs, but because of the poor understanding of the concepts of BCGSs in countries outside North America, the global distribution of BCGAs is poorly known. For example, in North America, many Rocky Mountain basins contain direct BCGAs. By analogy with Rocky Mountain basins, it is likely that many of the Andean foreland basins of South America also contain BCGAs. Several of the basins in the Middle East and North Africa probably contain indirect BCGAs similar to those in Jordan and Algeria.
The stratigraphic distribution of BCGAs extends from the Cambrian through the Eocene (Table 2). However, there appear to be some differences in the stratigraphic distribution of direct and indirect BCGAs. For example, the preponderance of direct BCGAs occur in Cretaceous through Eocene rocks (Table 2), whereas indirect BCGAs more commonly occur in pre-Cretaceous rocks. Although some of the apparent difference in stratigraphic distribution may be attributable to the disproportionate number of studies in Cretaceous and younger rocks compared to numbers of studies in pre-Cretaceous rocks, the question of seal integrity in direct systems arises. As previously discussed, the effective life of capillary pressure seals in direct systems is not known; therefore, because of the perceptions of a leaky seal in direct systems, the occurrence of direct systems in pre-Cretaceous rocks may be less common than in Cretaceous and younger rocks. Some examples, however, of pre-Cretaceous direct BCGAs include Permian rocks in the Timan-Pechora basin, Russia (Law et al., 1996), and the Sichuan basin, China (Da-Jun and Yun-ho, 1994); Pennsylvanian rocks in the Arkoma basin (Meckel et al., 1992); and Carboniferous rocks in the Dnieper-Donets basin, Ukraine (Law et al., 1998b) (Table 2).
Indirect BCGAs occur in rocks ranging from Cambrian through Cretaceous. Examples include Cambrian and Ordovician reservoirs in the Ahnet basin of Algeria, Ordovician reservoirs in Jordan (Ahlbrandt et al., 1997), Lower Silurian reservoirs in the Appalachian basin (Davis, 1984; Law and Dickinson, 1985; Zagorski, 1988, 1991; Law and Spencer, 1993; Law et al., 1998a; Ryder and Zagorski, forthcoming), and Jurassic sandstone reservoirs in the Bossier Shale (Montgomery and Karlewicz, 2001; Emme and Stancil, 2002) in the United States Gulf Coast (Table 2).
All BCGA reservoirs require carefully designed drilling programs and some type of artificial stimulation for commercial production rates. Reservoir continuity is an important consideration in the design of an appropriate drilling and completion program. Single, lenticular reservoirs have limited volume and are generally not commercial, whereas single, blanket reservoirs have much larger volumes and may be commercial, but, because blanket reservoirs commonly have better reservoir quality than lenticular reservoirs, they may be water bearing, as discussed previously.
In lenticular, fluvial-dominated reservoirs, such as those in the Jonah field in the northern part of the Green River basin of Wyoming or the Rulison field in the Piceance basin of Colorado, it is imperative to stimulate as many reservoirs as possible to attain commercial rates of gas production. The completion practices in the Jonah field provide a good example of commingling production from multiple, lenticular reservoirs (Finch et al., 1997; Eberhard, 2001); as many as 28 sandstones are perforated and fractured (Montgomery and Robinson, 1997). In a similar manner, gas production from multiple sandstone reservoirs in the Upper Cretaceous Williams Fork Formation in the Piceance basin of western Colorado is commingled following multiple fracture treatments in an interval about 2400 ft (732 m) thick (R. E. Mueller, 2002, personal communication).
Early attempts to produce from blanket reservoirs were mixed. Massive hydraulic fracturing techniques using 300,000 lb of proppant were used in an attempt to create long fractures. However, the large fracture treatments commonly resulted in shorter fracture lengths than predicted because of fracturing out of the reservoir into adjacent, nonreservoir rocks (Spencer, 1989a). This problem has, in some cases, been modified by adjusting pumping rates of the fracture fluids.
Natural fractures are important factors in successfully completing a well. The probability of a vertically drilled hole intersecting fractures is considerably less than horizontal or slant holes. For example, at the U.S. Department of Energy Multiwell Experiment site in the Piceance basin of Colorado, a slant hole was drilled through lenticular gas reservoirs. The hole was then deviated to horizontal in a blanket reservoir. Fifty-two fractures were reported from 266 ft (81 m) of core taken from the slant hole part of the hole. In contrast, a nearby vertically drilled hole penetrating the same slant hole interval encountered one fracture, and, in the horizontally drilled part of the hole, 37 fractures were reported from 115 ft (35 m) of core (Lorenz and Hill, 1991). In a more recently drilled 14,950 ft (4557 m)-deep well in the Green River basin of Wyoming, more than 400 open fractures were detected on a Formation MicroImager log from a 1750 ft (533 m)-long horizontally drilled leg in the Upper Cretaceous Frontier Formation (Krystinik and Lorenz, 2000). In the same well, approximately 76 natural fractures were recorded from a 78.2 ft (23.8 m)-long core taken from the same horizontal leg (Lorenz and Mroz, 1999). From these two examples, the probability of encountering fractures in slant or horizontal wells vs. vertically drilled wells is well documented. The cost of drilling nonvertical wells, however, is considerably greater than the cost of drilling vertical wells.
Reservoir damage is another important aspect of formation evaluation. Spencer (1985) listed several different types of reservoir damage, including (1) movement of secondary clays causing plugging of pore throats, (2) swelling of smectitic clays, (3) increasing water saturation with consequent reduction of relative permeability to gas, (4) fracturing gel compounds left in the reservoir, and (5) chemical additives causing precipitation of minerals and compounds during acidizing and hydraulic fracturing. The potential problem of swelling clays, in most cases, is minor, because most BCGAs occur in sequences where the level of thermal maturation is sufficiently high to convert swelling clays into nonswelling clays.
The objective of any hydrocarbon exploration program is to progress from coarse, loosely defined ideas to refined, drillable locations. In the case of BCGAs, exploration strategies are no different and may be viewed as a four-step process that includes (1) reconnaissance, (2) confirmation, (3) delineation, and (4) sweet spot identification. The exploration phases are mostly applicable to direct BCGAs; as a consequence of the relatively new classification of BCGAs into direct and indirect types, strategies for indirect BCGAs have not been formulated, although it is obvious that source rock considerations, level of thermal maturation, and temporal relationships among hydrocarbon generation, expulsion, migration, and trap formation are very important considerations in the exploration for indirect BCGAs.
The reconnaissance phase entails the identification of basins that may contain BCGAs. In direct systems, identification of source rocks is critical. For example, the identification of humic, gas-prone coal beds is the most obvious source rock for direct BCGAs; in nearly every country with coal reserves, there are some published data concerning geographic distribution, rank, and thickness. The rank of coal beds must be greater than high-volatile C (greater than vitrinite reflectance values of 0.6% Ro) to initiate thermal generation of gas (Hunt, 1996).
The existence of reservoirs with appropriate quality is another important aspect to consider during the reconnaissance phase. In most cases, coal-bearing intervals are associated with interbedded sandstones that have low porosity and permeability, especially at diagenetic stages commensurate with thermal maturity levels greater than 0.6% Ro. Sandstones deposited in alluvial plain, coal-bearing environments typically have poor reservoir properties. High porosity and permeability in reservoirs are not desirable attributes for the development of a BCGA. In basins where some drilling activity has occurred, gas shows are also very helpful.
Once a basin containing a potential BCGA has been identified, the task becomes one of confirmation. Because all BCGAs are abnormally pressured, the principal task during this phase is the determination of reservoir pressure and the mechanism of abnormal pressure. Most basins do not have sufficient quantity or quality of pressure data for this determination. Therefore, a combination of attributes listed on Table 1 can provide compelling evidence for the presence of abnormal pressure and a BCGA. Pore pressure indicators such as pore fluid composition (gas with little or no producible water) in conjunction with porosity (<13%), permeability (<0.1 md), thermal maturation (>0.7% Ro) data, and sustained gas shows are very useful. In some cases, sonic velocity data have been used to indicate the presence of abnormal pressures (Surdam et al., 1997, 2000, 2001; Surdam, 1997).
Although the determination of abnormal pressure is important, it is equally important to determine the mechanism of abnormal pressure. For direct BCGAs, the pressure mechanism is hydrocarbon generation (Spencer, 1987), and for indirect BCGAs, the pressure mechanism is thermal cracking of liquid hydrocarbons to gas (Law, 2000). A useful criteria for determining the pressure mechanism is through a knowledge of the composition of pore fluids: pore fluids in direct and indirect systems are composed of gas with little or no producible water (Spencer, 1987; Law and Spencer, 1993), whereas in abnormally pressured reservoirs, where the composition of pore fluid is mainly water, the pressure mechanism may be one of several other mechanisms, thereby precluding a hydrocarbon-generation mechanism and presence of a BCGA.
Formation resistivity and spontaneous potential curves measured on geophysical well logs also have been used to indicate the presence of a BCGA. In Upper Cretaceous rocks in the San Juan basin and Mesozoic rocks in the Alberta basin, resistivities greater than 20 Ω were reported to be gas saturated (Masters, 1979). Zagorski (1988, 1991) noted that the boundary between conventional and BCGA reservoirs in northwestern Pennsylvania could be distinguished at 80 Ω reservoirs with high water saturation were defined by resistivities <80 Ω.m, and reservoirs within the BCGA have resistivities >80 Ω.m. In Upper Cretaceous rocks in the Greater Green River basin, spontaneous potential curves are commonly reversed in abnormally pressured BCGAs (Law et al., 1979, 1980; Law, 1984).
The delineation phase entails mapping the vertical and areal distribution of the gas accumulation. The preferred way of accomplishing this phase is through the use of reliable pressure data. In most basins, however, pressure data are absent or of such low quality that reliable maps cannot be constructed; consequently, some indirect method may have to be used. The selected mapping parameter should be one that has been calibrated to well-documented pressure data. For example, thermal maturity values ranging from 0.7 to 0.9% Ro were determined to be coincident with the top of overpressuring in the Greater Green River basin (Law, 1984). In later work, 0.8% Ro was used to map the depth to the top of overpressuring in the basin (Pawlewicz et al., 1986; Law et al., 1989). Johnson et al. (1987, 1996, 1999) used a value of 0.73% Ro to map the top of the gas- and water-bearing transition zone above gas-saturated reservoirs in the Piceance basin of Colorado and the Wind River and Bighorn basins of Wyoming.
To determine an accurate, reliable mapping method, a detailed study of a small area within the basin is recommended rather than a broad-based regional study. For the detailed study, a small representative area with relatively complete, high-quality data should be chosen. Comprehensive, multidiscipline investigations including stratigraphic, structural, source rock, reservoir rock, pressure, thermal history, petrophysical, and well log analyses should then be conducted within the selected area. The objective of this comprehensive investigation is to establish a type area or analog for the entire basin to which incomplete or fragmentary data from other parts of the basin can be compared. From such analog studies, indirect mapping tools, such as levels of thermal maturity, present-day temperature, and log responses, may be determined. Examples of such analog studies include the Pacific Creek area in the Greater Green River basin (Law et al., 1979, 1980), the Wagon Wheel well in the Greater Green River basin (Law and Spencer, 1989), and the Multiwell Experiment site in the Piceance basin, Colorado (Northrop et al., 1984; Spencer and Keighin, 1984; Law and Spencer, 1989). Regional mapping using some of these indirect parameters can then be used not only to determine the stratigraphic and areal distribution of the BCGA but also to help identify areas of enhanced reservoir quality, or sweet spots.
Sweet Spot Identification
Although a few BCGAs are commercially productive over their entire areal extent, such as the San Juan basin of Colorado and New Mexico, most BCGAs are not commercially productive over their entire area. Consequently, areas within the BCGA of enhanced reservoir quality (sweet spots) must be identified. These sweet spots may be structural or stratigraphic in nature and always occur within the abnormal pressure envelope. In addition, they most likely occur near the upper boundary of the BCGA.
In Figure 6, the top of overpressure and BCGA in the Washakie basin is shown as a fairly smooth, uniform line cutting across structural and stratigraphic boundaries. In this case, if very closely spaced pressure data were available along the line of section, the pressure boundary would most likely not be as smooth as shown but would probably be highly irregular, with significant areas of high relief. The areas of high, positive relief, or bumps, may be indicative of structural and/or stratigraphic sweet spots that occur at or near the upper boundary of the BCGA. In the absence of closely spaced pressure data, it is difficult to identify a sweet spot. However, some techniques can be used to identify and focus more expensive techniques such as three-dimensional (3-D) seismic surveys. Those techniques may include lineament, thermal maturity, and present-day temperature mapping. Aeromagnetic, gravity, and surface geochemical surveys also may be useful in the identification of potential sweet spots. Surdam (1997) and Surdam et al. (1997) described methods employing sonic logs to identify sweet spots in several basins in Wyoming.
The best example of a BCGA structural sweet spot is the Jonah field in the northern part of the Green River basin, Wyoming (Figures 3, 4). As previously discussed, the Jonah field is a gas chimney, rooted in a regionally pervasive BCGA described by Law (1984) and producing from multiple sandstone reservoirs in the Upper Cretaceous Lance Formation. Alternatively, Cluff and Cluff (2001) have interpreted the Jonah field to be a remnant of a larger, much more shallow BCGA than presently identified. The Jonah field is a wedge-shaped area with the north, south, and west boundaries of the field defined by westward converging faults (Figure 4). The eastern boundary is undefined. The geologic characteristics of the Jonah field are given by Montgomery and Robinson (1997) and Warner (1998, 2000). According to Warner (2000) the top of overpressure (top of gas-saturated reservoirs) within the field occurs at depths of 7700 ft (2347 m) at the west end of the field (updip end of field) and 9500 ft (2896 m) at the east end of the field (downdip end of the field). Outside the field, the top of overpressure and gas-saturated reservoirs occur at depths ranging from 11,200 to 11,600 ft (3414-3536 m) (Warner, 2000). Thus, there is 2500-3000 ft (726-914 m) of relief on the top of overpressuring from outside the field to inside the field (Figure 4). The gas chimney has subsequently been identified through the use of sonic velocity data (Surdam et al., 2001).
A good example of a thermal maturity anomaly associated with a sweet spot is the Lower Cretaceous Muddy ("J") Sandstone in the Denver basin of Colorado. Regional thermal maturity mapping in the Denver basin of Colorado (Higley et al., 1992) shows the presence of an anomaly associated with a BCGA (Figure 14). The anomaly, defined by reflectance values greater than 0.9% Ro, is nearly coincident with the field boundaries of production from the Muddy Sandstone in the Wattenburg field. The anomaly is located north of the structurally deepest part of the basin and is coincident with the northeast projection of the Colorado Mineral Belt. The field is also coincident with a temperature anomaly mapped by Meyer and McGee (1985).
Because the top of a BCGA is determined, in part, by permeability variations and the ease with which gas may move through reservoirs, measured levels of thermal maturity at the top of a BCGA may provide indirect evidence of the presence of a sweet spot; relatively low values of thermal maturity (<0.8% Ro) at the top of an overpressured BCGA are indicative of a potential sweet spot, whereas relatively high values of thermal maturity (>0.8% Ro) are indicative of very low permeability in an overpressured BCGA. Based on vitrinite reflectance profiles from two wells within the Jonah field (Warner, 1998), the level of thermal maturity at the top of overpressured, gas-saturated reservoirs is less than 0.7% Ro, compared to 0.8% Ro outside the field. Thermal maturity indices, however, cannot be used to identify potential sweet spots in underpressured BCGAs. The level of thermal maturity at the top of an underpressured BCGA most likely is higher than the level of thermal maturity at the top of an overpressured BCGA because the dimensions, or size, of a BCGA are reduced during the transition from overpressure to underpressure. Consequently, the level of thermal maturity at the top of an underpressured BCGA reflects that size constriction.
Stratigraphic sweet spots are more difficult to discern than structural sweet spots because detailed facies mapping requires close-spaced to moderately spaced subsurface data. An example of a stratigraphic sweet spot includes the Upper Cretaceous Almond Formation in the Washakie basin of southwest Wyoming, where reservoirs in the upper, marginal marine part of the formation are typically much more productive than reservoirs in the lower, fluvial-dominated part of the formation. Additional stratigraphic sweet spots include sandstones within the Upper Cretaceous Lewis Shale in the Great Divide basin and the Frontier Formation along the structural crest of the Moxa arch in the Green River basin.
Finally, based on empirical observations, there appears to be a relationship between producibility and the nature of abnormal pressure; overpressured BCGA reservoirs generally require the identification of sweet spots for commercial production, whereas underpressured reservoirs are regionally productive and do not require the identification of sweet spots. The best examples of regionally productive gas production from underpressured systems occur in Upper Cretaceous reservoirs in the San Juan basin of New Mexico and Colorado and in the Lower Silurian Clinton-Medina reservoirs in the Appalachian basin of Pennsylvania. However, sweet spots, even in underpressured BCGAs, are desirable features to identify. The reasons for this apparent relationship between producibility and the nature of abnormal pressure are uncertain. Perhaps reservoir quality is slightly improved during the transition from an overpressured system to a underpressured system.
In 1978 the Natural Gas Policy Act provided incentive prices and, later, tax credits for gas production from coal, shale, and low-permeability sandstone reservoirs in an attempt to stimulate the development of gas from unconventional, marginally economic reservoirs. Those incentives, along with significant funding from the U.S. Department of Energy for research and development of tight gas sands, were instrumental in unlocking a gas supply that has had and will have a significant impact on the energy needs of the United States and the world. At the end of 1992 the incentives expired, and there was some skepticism in the industry concerning continued gas production without some economic help. However, new technological gains, an improved geologic and engineering understanding of tight gas sands, and higher gas prices have combined to make BCGAs (tight gas sands) a very attractive exploration objective. In the United States and Canada, exploration and exploitation of this huge gas resource has experienced considerable success, and activity should accelerate over the next several years. Internationally, exploration activity is currently minimal but likely will increase in the near future. As the concepts of BCGSs become better known outside North America, there will be an increased focus on the tremendous potential of this gas resource.
As previously discussed, gas production from BCGAs is currently making a significant contribution to the energy needs of the United States, and the future role of BCGSs will be significant; however, some large obstacles must be addressed for this type of unconventional gas system to meet or surpass expectations. In order for BCGAs to play an increasing role in the energy requirements of the United States and the world, the following topics and problems need to be addressed:
An effective global education program is essential to stimulate and expand exploration programs beyond the United States and Canada; traditional concepts of petroleum systems need modification.
In many basins, BCGAs occur at depths greater than 10,000 ft (3048 m). Artificial stimulation at these depths is difficult and expensive. Although there have been significant improvements in drilling and completion technologies within the past 20 yr, continued advances in technologies are essential to tap the very large gas resources at these depths.
In thick, gas-saturated reservoirs containing interbedded water-bearing reservoirs, improved techniques are needed to discriminate between gas-bearing and water-bearing reservoirs.
The integrity of capillary pressure seals over long periods of geologic time needs to be determined.
More geologic research into the occurrence of BCGAs, especially indirect types, is desirable. Essentially no information is available concerning the nature of or exploration strategies for indirect BCGAs.
Methods of identifying and characterizing natural fractures must be improved.
Relationships among kerogen type, thermal maturity, initiation of gas generation, peak gas generation, transformation of oil to gas, and volumetric fluid changes accompanying the transition of oil to gas need additional research.
Basin-centered gas accumulations, a type of unconventional gas accumulation, are typically regionally pervasive accumulations encompassing hundreds or thousands of square miles and may occur as single, isolated reservoirs a few feet thick or as multiple, stacked reservoirs several thousand feet thick. Some of the more important distinguishing characteristics of BCGAs include abnormal pressures (over- or underpressured), low-permeability reservoirs, and a general absence of downdip water. Two types of BCGSs are recognized: a direct type distinguished by having a gas-prone source rock and an indirect type that has a liquid-prone source rock. Direct systems commonly have leaky capillary pressure seals, whereas indirect systems have more effective lithologic seals. Although the two systems have several similar attributes, the fundamental difference between the systems, gas-prone vs. liquid-prone source rocks, results in some dissimilar attributes that require different exploration strategies. The worldwide potential for major gas production from BCGAs has not been fully appreciated, whereas in the United States, gas production from these regionally pervasive accumulations is a significant contributor to the nation's energy requirements and will likely increase in the near future.
I am grateful to my many U.S. Geological Survey and industry colleagues for their support over the years. I am especially indebted to Charles Spencer for his insights and collaboration on aspects of basin-centered gas systems (BCGSs). A large part of the research was funded by the U.S. Department of Energy under the very capable management of Karl Frohne and William Gwilliam. The work also benefited from periodic, constructive discussions and unpublished subsurface data provided by Bill Barrett, Bill Hanson, Greg Anderson, Doug Battin, Jeff Aldrich, John McIntyre, and John Gustavson. Finally, the reviews by Charles Spencer, Dale Leckie, and Bob Ryder significantly improved the manuscript.
- Manuscript receivedJune 21, 2001.
- Final acceptanceJune 6, 2002.