Case Studies Examining the Discovery Sequence and Gas Accumulations in Tight-Gas Sandstones
Published:December 01, 2009
- PDF LinkChapter PDF
James Coleman, Emil Attanasi, 2009. "Case Studies Examining the Discovery Sequence and Gas Accumulations in Tight-Gas Sandstones", Unconventional Energy Resources: Making the Unconventional Conventional, Tim Carr, Tony D’Agostino, William Ambrose, Jack Pashin, Norman C. Rosen
Download citation file:
An examination of the geologic characteristics and discovery history of seven plays, which were originally classified by the U. S. Geological Survey (USGS) in 1995 as continuous-type gas sandstone plays, shows that these plays have a high degree of similarity with conventional discrete accumulations in terms of reservoir continuity, sand body geometry, and trapping configurations. The general decline in discovery size with increasing numbers of discoveries suggests a means to put limits on volumes of resources assessed in un-drilled areas of a particular play.
Routine time-series analyses of conventional plays typically show a decline in field discovery size as each subsequent discovery within the play trend is announced. If gas accumulations in low-permeability sandstone plays occur in trap settings typical of discrete conventional accumulations, then modeling of the discovery sequences within plays may provide an effective way to constrain regional estimates of remaining recoverable resources. At the other extreme, if the play is regarded as a single homogeneous continuous entity (albeit, with some “sweet spots”), only the play boundary constrains the number of un-drilled sites that could contribute to remaining recoverable resources, and there should be no general decrease in discovery size.
The seven continuous-type gas sandstone plays selected for this study had a sufficient number of observations to test whether discovery size correlates with sequence of discovery. These showed that discovery size tends to decline with sequence of discovery and in three of the seven the trend was statistically significant. The discovery size rank and sequence relationship was found to be similar to several well known conventional plays.
The recent literature carries an ongoing discussion regarding the nature of gas accumulations in lowpermeability sandstones. In the 1995 National Assessment of Oil and Gas Resources (Gautier et al., 1996), the U.S. Geological Survey (USGS) are classified a number of low-permeability sandstone gas plays1 as “continuous” (see Fig. 1). These plays have been assessed to contain 265 trillion (1012) cubic feet (TCF) (7.42 trillion cubic m) of recoverable gas in untested areas. Individual plays are typically assessed as a single regional gas accumulation described as “pervasive throughout a large area, not significantly affected by hydrodynamic influences and for which the standard methodology of sizes and numbers of discrete accumulations is not appropriate” (Gautier et al., 1996). Shanley and Cluff (2004) and Shanley et al. (2004), however, suggest that “continuous” gas accumulations in the low-permeability sandstones of the Rocky Mountains may be more accurately described as being contained in discrete conventional stratigraphic, structural, or combination traps.
Procedures for assessing undiscovered oil and gas in conventional (or discrete accumulations) commonly rely upon historical discovery sequences to constrain estimates (Arps and Roberts, 1958; Drew et al., 1980; Kaufman, 1993). Attanasi (2005) has analyzed discovery sequences of accumulations in selected low-permeability sandstone plays classified as continuous by the USGS for the 1995 National Oil and Gas Assessment and has compared them to discovery sequences of conventional sandstone plays. In that report, he has found a pronounced degree of similarity between the two discovery sequences. This report reviews the geologic characteristics of these plays and their exploration histories that produced the discovery outcomes in the context of those statistical findings.
The following paragraphs describe the plays originally classified as continuous, describe the method used to examine discovery order, and review the discovery sequence trends. For each of the seven plays, we describe the reported discovery history, reservoir continuity, sand body characteristics, and trapping configuration. In the concluding section, implications of the findings are considered and suggestions about future research are presented. Summaries of the geologic interpretations are offered, and an appendix provides additional summary information for each of the plays examined.
Selection of Seven “Continuous” Gas Plays and Examination of their Discovery History
Twenty-seven confirmed and hypothetical gas plays in low-permeability sandstones were classified as “continuous” in the USGS 1995 National Oil and Gas Assessment (Table 1; Gautier et al., 1996). Only seven of these plays (in boldface type in Table 1) had a sufficient number of accumulations for which there were complete records as reported in the reservoir file of Significant Oil and Gas Fields of the United States (NRG Associates, 2003) to be statistically analyzed for discovery sequence. In 1995, the USGS assessed these seven plays to contain 134 TCF (3.75 trillion cubic m) of recoverable gas in undeveloped areas (Gautier et al., 1996). These seven “continuous” tight-gas sandstone plays (with USGS play number in parentheses) show similar discovery and development histories: the Tight Gas Piceance Mesaverde Williams Fork Play (2007), the Central Basin Mesaverde Play (2209), the Northern Great Plains Biogenic Gas High Potential Play (2810), the Greater Green River Basin Cloverly-Frontier Play (3740), the Greater Green River Basin Mesaverde (Almond) Play (3741), the Greater Green River Basin Lewis Play (3742), and the Cotton Valley Blanket Sandstones Gas Play (4923) (Fig. 1, Table 1).
Geological summaries of each of these plays are reviewed in Appendix I (text and Figures A1–A8). Of these seven, the Central Basin Mesaverde, Greater Green River Basin Cloverly-Frontier, and Cotton Valley Blanket Sandstones Gas plays show a decline trend that is statistically significant. These three will be reviewed first.
Central Basin Mesaverde gas play (2209)
The Central Basin Mesaverde Gas Play of the San Juan Basin in New Mexico and Colorado is developed primarily in the Upper Cretaceous Point Lookout and Cliff House sandstones (Figs. 1 and A2) and secondarily from sandstones within the intervening Menefee Formation. Production depths are mostly from 4,000 to 5,300 ft (1.2 to 1.6 km) from updip pinchouts of marine sandstones into finer grained sediments. Until recently, Mesaverde gas resources within the San Juan Basin have been considered basin-centered continuous. However, Fassett and Boyce (2005) illustrate that conventional geological trapping mechanisms may control gas accumulation within the interval. Natural fractures are usually required for economic rates of gas production.
The primary exploration target in recent years for the San Juan Basin was the Upper Cretaceous Fruitland Coal. Prior to that, structurally deeper and stratigraphically older horizons were the primary targets. With the advent of artificial stimulation technology, relatively low rate, “unconventional” reservoirs could be developed. The presence of existing surface gathering and distribution infrastructure meant that almost all newly found gas could be economically delivered to market.
Greater Green River Basin Cloverly-Frontier play (3740)
The Cretaceous Cloverly-Frontier play extends across the greater Green River Basin of Wyoming (Figs. 1 and A4). Both stratigraphic units are comprised of fluvial-deltaic to marginal marine sandstones. Production depths range from about 5,000 ft (1.5 km) along the axis of the Rock Springs uplift to about 12,000 ft (3.7 km) in the western Green River Basin and as deep as 21,000 ft (6.4 km) in the eastern Green River Basin. Significant production is associated with structural closure and high fracture intensity, such as that on the Moxa arch in the western Green River Basin. A recent down-structure test of the Frontier in the eastern Green River Basin is significant because it produced substantially increasing volumes of water coincident with substantially decreasing volumes of gas (Coleman, 2008; Wyoming Oil and Gas Conservation Commission web site, accessed 12/23/07). Previous production and drilling histories suggest that the Frontier Formation of the Greater Green River Basin is not uniformly a basin-centered continuous gas reservoir (Coleman et al., 2003). The Cloverly (Dakota) sandstone does not have enough widespread production in the Greater Green River Basin to determine if it, also, might be more accurately characterized as a conventional, low-permeability gas reservoir.
The Frontier and Cloverly (Dakota) gas sandstones were drilled on the Moxa arch in the western Green River Basin and developed as reservoirs having porosity and permeability that is lower than typical, conventional sandstone reservoirs. As development activity proceeded down structural dip to the north, industry continued to complete wells as gas producers. Similar, but less extensive, production was established on the Rock Springs uplift and other structures in the eastern Green River Basin.
Cotton Valley blanket sandstones gas play (4923)
The Upper Jurassic-Lower Cretaceous Cotton Valley Blanket Sandstones Gas Play extends across northern Louisiana (Figs. 1 and A8). Schenk and Viger (1996) originally considered it to be a continuous gas reservoir play because of its low porosity and permeability, pervasive gas shows while drilling, and requirement for artificial fracture stimulation for economic production. However, a reevaluation by Bartberger et al. (2002) concluded that gas-water contacts were present across the trend, as well as areas of high porosity and permeability that did not require fracture stimulation for economic production.
The discovery history of the Cotton Valley Blanket Sandstones Gas Play indicates that development drilling pursued the gas resources of the Cotton Valley “D” Sandstone downdip on the Sabine uplift (Fig. A8). As more reserves were added by drilling and water production was minimal or non-existent, more and more wells were drilled farther downdip. Eventually, wells that produced more water than gas were drilled, and an effective “no-go” limit was established that equated economically and effectively to a gas-water contact. Tight-gas sandstone production was also extended east of the Sabine uplift and downdip from conventional gas reservoir pinchout into the Hico Shale Formation of the Cotton Valley Group (Fig. A8). In some cases, failure to achieve gas production from the Cotton Valley Group tight-gas sandstone reservoirs were mitigated by establishing production from the overlying Hosston Formation; however, this opportunity was not present everywhere. Rarely, operators would make new Cotton Valley discoveries after failing to produce deeper targets (i.e., Bossier shale or Smackover limestone reservoirs).
Tight gas Piceance Mesaverde Williams Fork play (2007)
The tight-gas Piceance Mesaverde Williams Fork Play (2007), in Colorado, is developed primarily in fractured, lenticular, fluvial channel sandstones of Late Cretaceous age (Figs. 1, A1). Recent studies have concluded that this play is still considered a basin-centered, continuous gas accumulation (Johnson and Roberts, 2003; Hood and Yurewicz, 2008; Yurewicz et al., 2008); however, the preponderance of high discontinuity reservoirs (Kuuskraa and Ammer, 2004) suggests that conventional stratigraphic trapping geometries that are smaller than legal spacing units might control gas accumulations in the Williams Fork Play in a pervasively gas-charged basin. In other words, trap sizes smaller than legal-spacing units may retain original regional reservoir pressures, leading to an interpretation that the gas resource accumulation is continuous across the play area, because each new well encounters original reservoir pressures until drilling density is such that two or more wells are drilled into a single accumulation trap and start to affect individual well performance.
Northern Great Plains biogenic gas play, high potential (2810)
The Northern Great Plains Biogenic Gas Play, High Potential, in Montana, is developed in an Upper Cretaceous shallow, moderate to high porosity, low-permeability, complex mixture of siliciclastic and carbonate facies in which biogenic methane has been trapped (Figs. 1, A3). These facies are adjacent to conventional, water-drive reservoirs (Condon, 2000; O’Connell, 2003; Ridgley et al., 2001). This play is unique among the seve plays discussed in this paper because a significant portion of the gas has been biologically generated in the vicinity of the reservoirs. Also, until recently, no exploration drilling has targeted the play (Rice and Spencer, 1996). Recent reassessment of the play reduces the resource potential, but does not reinterpret it to be a conventional, low-permeability gas reservoir play (Ridgley et al., 2001). The complex depositional and preservational conditions of these reservoirs and the potentially complex nature of their trapping configurations make it difficult to discern the true nature of this play.
Greater Green River Basin Mesaverde (Almond) play (3741)
The Upper Cretaceous Mesaverde (Almond) Play (3741) in the Green River Basin, Wyoming, is developed in a mixed continental, paludal, and marginal marine facies mosaic, in which relatively higher porosity and permeability barrier bar and fluvial-tidal channel sandstones host most of the production (Figs. 1, A5, and A6). Generally production is associated with either structural closure or stratigraphic facies changes. Most current production lies beneath a regional pressure seal in the overlying Lewis shale (Charpentier and Cook, 2005; Coleman, 2008; Norris et al., 2005; Shanley et al. 2004).
Greater Green River Basin Lewis play (3742)
The Upper Cretaceous Lewis Play, also located in the Green River Basin, Wyoming, is developed in submarine channel and channelized lobe facies downdip from shale-prone, deltaic margins (Figs. 1 and A7). Post-depositional, structural tilting set up combination structural-stratigraphic traps in many fields. Complex facies associations in the slope channel facies potentially created stratigraphic trapping geometries. Most initial Lewis discoveries have been made while drilling for deeper targets (Hettinger and Roberts, 2005; Steinhoff et al., 2001).
Summary of discovery and development patterns
In the cases reviewed above, historical development patterns were not simple. Initial wells, which in an historical context might be deemed “the discovery wells,” were generally drilled for a deeper, structural objective having an anticipated higher flow rate and higher volume reserve potential. In a few cases, wells were drilled just deep enough to prevent the expiration of an exploration lease. Stratigraphic intervals having high gas saturations were recognized by drilling effects, including unexpected gas to surface or stuck drill pipe, as well as mud logger detections. The recognition that intervening, potentially diagenetically or hydrodynamically trapped lower production rate gas might be economic was not an instantaneous, “eureka” moment by the oil and gas industry but rather evolved over the course of time as gas market economics improved. With the rise in the demand for natural gas and associated apparent improvement in economics of these types of reservoirs, improvements in completion technology were developed and those accumulations near existing conventional gas infrastructure were developed. Tax incentives helped create improved economic scenarios that facilitated unconventional gas reservoir development and exploitation. More remote areas were developed as the gathering and distribution systems reached out sufficiently to access these resources. Thus, the lower rate gas production could be integrated into the local gathering and regional distribution systems, especially as higher-rate, structurally- and stratigraphically-trapped, conventional production declined (see Coleman, 2008 and 2009, this volume, for additional details and references on the history of tight-gas sandstone reservoir development).
Plays were developed that coupled the potential for co-mingled conventional and unconventional gas reservoirs in order to lower the economic risk of both of these types of plays (e.g., Green River Basin Mesaverde (Almond) and Lewis plays; Appendix I). Only relatively recently did operators attempt to develop a continuous sandstone resource play exclusively within the stratigraphic horizon(s) that contain that play. Even then, geologic and economic surprises caused some areas to be abandoned and some areas expanded as newly discovered reservoirs were developed (e.g., Gulf Coast Basin Cotton Valley Sandstones Gas and Green River Basin Mesaverde (Almond) Plays; Appendix I, Coleman (2008)). At all times, the risk that a reservoir would be encountered that would not live up to pre-drill expectations governed the decision where to drill the next well. Selecting drill locations legally spaced adjacent to a high performing well was always preferred over selecting drill locations even marginally close to a poorly performing well. Consequently, development of a “continuous” reservoir play may proceed in a manner similar to the development of a conventional play. Despite these generalizations, detailed examination of exploration and development files and permit, drilling, and completion records would be required to reconstruct a detailed, plausible history for the full development of these fields.
Statistical Analysis of Discovery Sequence
Resource assessors may commonly impose constraints on the resources assessed in un-drilled areas on the basis of the historical sequence of discovery sizes over time. A sequence associated with declining discovery sizes often signals that future finds will be progressively smaller, assuming no new technology is developed and implemented. Alternatively, a sequence associated with increasing discovery sizes, on average, could indicate that the industry was in a learning process or new technology was implemented that improved search or extraction procedures. Discovery sequences having no discernible size-order relationship may be the result of homogeneity in endowment sizes.
For each play, we tested the discovery size2 order and the discovery order (as represented by discovery year). Discovery sequences for each play that shows the accumulation size rank versus discovery order (rank) of the accumulations is displayed in Figures 2A though 8B. Size is measured in terms of surface area or the estimated recoverable gas of the accumulation. The line represents the least squares fit of the size variable rank against order variable and is drawn only to highlight the direction of relationship, because the actual data are present only at discrete points. Notice that all the figures show the discovery history to be downward sloping.
To formally test the statistical significance of the observed relation, Spearman’s rank correlation coefficient3 was computed. For this statistic the stated null hypothesis was that size rank and discovery sequence (order) was not linearly correlated. As shown in Table 2, the computed statistic was negative for all the continuous plays where there were sufficient data to test the hypothesis. Furthermore, three of the seven plays (2209, 3741, and 4923) had negative correlations that led to rejection of the null hypothesis at the 5% significance level4 (where p-values are less than 0.05) for either area or volume or both.5
In some of the cases, the null hypothesis is not rejected because noise (random or non-randomly distributed events in the discovery sequence) obscures the underlying discovery pattern. Noise may be introduced when a highly prospective but un-drilled area is opened to exploration late in the discovery process, leading to large late discoveries. For very deep plays, noise may also be introduced when initial discoveries result from deeper pool tests of wells already located in shallow producing fields. In such cases, the prospects tested are influenced more strongly by access to shallow field infrastructure than by geologic potential. A similar situation occurs for new shallow plays developed with established, deeper pay infrastructure.
The assessed volumes of undiscovered resources in conventional plays typically depend on how quickly discovery rates decline. To judge the results in Table 2, the Spearman’s rank correlation coefficient was computed for several well known conventional plays. For all the conventional plays examined (Table 3), each had negative rank correlation coefficients, and the magnitudes of the Spearman’s rank correlation coefficient are similar to those of Table 2. In only one case is the pvalue greater than 5%. In the case of the Muddy play (3307) in the Powder River basin, the discovery of two of the play’s largest accumulations during 1968 resulted in the high p-value6 (Attanasi, 2005).
Simulation experiments presented in Attanasi (2005) were used to help interpret the empirical data. These experiments showed that sampling variation introduces noise that obscures the clarity of the trend, even in cases where one expects a strong trend. Furthermore, exploration “surprises” occur even in the actual discovery histories of conventional plays to introduce such noise (Table 3, the Muddy Play, play 3307).
Summary and Implications
The discovery sequences of the accumulations in these seven low-permeability sandstone plays have been analyzed statistically to examine their field discovery history. The USGS estimates that these plays contain an expected value of 134 TCF (3.75 trillion cubic m) of recoverable gas in untested areas (Gautier et al., 1996). Spearman’s rank correlation coefficient indicates that, on average, gas accumulation sizes decline with order. For three of the seven plays (Central Basin Mesaverde Gas Play 2209, Greater Green River Basin Cloverly-Frontier play 3740, and Cotton Valley Blanket Sandstones Gas Play 4923), the null hypothesis (of no size order relation) can be rejected at the 5% level of significance. Both the drilling record and the discovery sequences should be studied when choosing geologic models and assessment methods to estimate gas resources in untested areas.
This statistical overview provides, in an analytical way, another view on the discovery history of “continuous- type” gas plays. The statistical similarity of the discovery history of conventional and continuous sandstone gas reservoirs suggests that both types of reservoirs may have more similarities than differences. Reservoirs in the seven previously defined continuous plays examined in this study have strong geometric similarities to conventional gas reservoirs in terms of potential stratigraphic or combination stratigraphic-structural trapping conditions. Where accumulations are not obviously controlled by structural dip or major depositional facies changes, complex stratigraphic or diagenetic facies changes between probably highly sinuous, channelized fluvial-deltaic or marginal marine sandstone bodies and adjacent co-eval fine-grained facies probably may be sufficient to explain the presence or absence of producible natural gas and formation water.
This analysis suggests that so-called continuous gas sandstone reservoirs may in fact have a high degree of similarity to conventional, low-permeability gas sandstone reservoirs, in terms of their reservoir continuity and sand body geometry. The component that appears to distinguish them from conventional reservoirs is their presence within a pervasively gas-charged stratigraphic interval or basin. However, this conclusion may be premature, because there are very few data to compare the gas saturation levels of non-reservoir intervals encompassing conventional gas reservoirs with those non-reservoir intervals encompassing continuous gas sandstone reservoirs.
If the so-called continuous gas resources in sandstones are found predominately in intervals in which conventional-type traps predominate, significantly disrupt the continuity of reservoir bodies across a basin, and control production volumes, then the effects of this predominate trap style and resulting heterogeneity should greatly determine production volumes. Therefore, estimates of technically recoverable resources in a basin or play area of this type should be based on such a geologic model. If most continuous gas sandstone plays have discovery histories similar to those discussed here, then they would show, on average, declining returns with additional drilling effort. Assessments of continuous gas sandstone plays should reflect a geologic model, which incorporates the physics, hydrodynamics, and geometry of in situ reservoir conditions as well the possibility of, on average, declining returns in order to reflect what might be truly technically recoverable and therefore, what society might reasonably expect to see produced.
Geologic Overview of “Continuous” Plays Examined in this Study
Basin setting: The Upper Cretaceous Williams Fork Formation is the primary gas reservoir interval, is in the upper part of the Mesaverde Group in Colorado, and contains significant intervals of coal that are considered the source for much of the gas in the eastern Mesaverde Total Petroleum System (TPS) (Fig. A1; Dubiel, 2003). It is present as a stratigraphic unit in the Piceance Basin and is the general equivalent to the Almond Formation to the north in the Green River Basin. The gas reservoirs have been deposited in mostly coastal plain and fluvial depositional environments. It is capped by a significant unconformity representing the Cretaceous-Tertiary boundary, which has removed most of the Maastrichtian interval that is present in the Green River Basin. A thick section of paleosols caps the Williams Fork Formation, a complex of generally eastward prograding coastal clastics composed primarily of fluvial deltaic channelized deposits (Hettinger and Kirschbaum, 2003; Johnson and Roberts, 2003).
Geographic extent: The Williams Fork Formation extends from zero limits on the basin margins on the north, east, and south, and thins onto the Douglas Creek arch to the west (Fig. A1). The USGS Piceance Basin Continuous Gas Assessment Unit (AU) is defined as the volume of rock defined by the geographic and stratigraphic extent of vitrinite reflectance (Ro) greater than or equal to 1.10% for units within the Williams Fork and younger, including the Tertiary (Johnson and Roberts, 2003).
Age: The Williams Fork Formation is Late Cretaceous Campanian (about 75 Ma). Peak gas generation occurred about 47 Ma from coal and organically rich shale of approximately equivalent age.
Field characteristics: The primary gas reservoirs of the Piceance Mesaverde (Williams Fork Formation) are lenticular, fluvial channel sandstones (Fig. A1). Stratal seals are relatively impermeable mudrock that surrounds many of the sandstone units, and are augmented by capillary pressure seals within the accumulation (Johnson and Roberts, 2003). Shoreface sandstones are also present in the reservoir intervals of the deep basin accumulations. Kuuskraa and Ammer (2004) illustrate gas reservoirs in the Mesaverde as highly discontinuous, lenticular sandstones, having less than 28 acres (0.1 km2) of stratigraphic continuity. Mesaverde gas sandstone reservoirs “typically require enhanced permeability from natural fractures within the reservoirs to obtain viable gas production” (Johnson and Roberts, 2003, p. 30).
Depth: Mesaverde Group production extends between the depths of about 1,600 ft (0.5 km) to more than 14,000 ft (4.3 km), with an average of about 7,400 ft (2.3 km) (Johnson and Roberts, 2003).
Thickness: Mesaverde Group gas reservoirs typically range from 20 to 60 ft thick (6 to 18 m) (Tremain, 1993), although amalgamated coastal plain fluvial channel sandstones associated with conglomerates can reach hundreds of feet in thickness (Johnson, 1989). Cumulative sandstone thickness in the Mesaverde Williams Fork Formation ranges from about 500 ft (150 m) to over 1900 ft (580 m) in the deepest portions of the Piceance Basin. The total interval thickness increases from about 2000 ft (0.6 km), on the western margin of the basin to over 4500 ft (1.4 km) in the eastern depths (Johnson and Roberts, 2003).
Porosity and permeability: Porosity within the Piceance Basin Mesaverde Group reservoirs varies from less than 5% to 12%. Permeability generally ranges from 0.01 to 0.1 millidarcies (mD); however, permeability may be as low as 0.0006 mD (Pitman and Spencer, 1984; Johnson, 1989; Tremain, 1993; Spencer, 1996; Johnson and Roberts, 2003).
Pressure regime: Cumella and Scheevel (2005) indicate that the pressure gradients range from as high as 0.8 psi/ft in the lower Williams Fork to 0.43 psi/ft near the top of the gas saturated zone.
Discovery and production history (including possible adverse conditions for “normal” development): Significant gas production began in the 1950’s, with greater than 70% of the gas production localized in the Rulison, Parachute, Grand Valley, Mamm Creek, and Sulphur Creek fields. In Rulison, Grand Valley, and Parachute Fields, extensive fracture systems are considered essential for the successful gas production there (Johnson and Roberts, 2003).
Geological evaluation as possible continuous gas sandstone play: Spencer (1996, p. 13) identified the Piceance Basin Mesaverde Williams Fork play (number 2007) as a continuous, tight-gas play, and considered it “a typical basin-centered … gas accumulation containing gas downdip and water updip.” Johnson and Roberts (2003) interpreted the Mesaverde Group and Williams Fork Formation as continuous and transitional gas accumulations in the Piceance Basin. Cumulla and Scheevel (2005), Hood and Yurewicz (2008), Yurewicz (2005), and Yurewicz et al. (2008) also concluded that the Piceance Basin Williams Fork Formation is a continuously gas-saturated reservoir interval.
Basin setting: The Upper Cretaceous Mesaverde unconventional gas reservoirs of the central San Juan Basin are in “sandstone buildups associated with stratigraphic rises in the Upper Cretaceous Point Lookout and Cliff House Sandstones” (Bureau of Indian Affairs Division of Energy and Mineral Resources, 2000b, p. 22). The Mesaverde is comprised of the regressive marine Point Lookout Sandstone, the nonmarine Menefee Formation, and the transgressive marine Cliff House Sandstone. The Mesaverde is overlain and underlain by marine shales (Fig. A2).
Geographic extent: The Mesaverde Group extends from the basin margin in the north, east, and west to a Quaternary erosional surface in the south of San Juan Basin of northern New Mexico and southern Colorado (Bureau of Indian Affairs Division of Energy and Mineral Resources, 2000 a, b).
Age: The Mesaverde Group is Late Cretaceous Santonian/Campanian in age.
Field characteristics: The principle gas reservoirs in the Mesaverde are the marine Point Lookout and Cliff House sandstones; smaller amounts are from thin, lenticular fluvial channel sandstones and coal beds of the Menefee Formation. The gas appears to be trapped by updip pinchouts of marine sandstone into finer grained paludal or marine sediments; seals are present in shale or coal (Huffman, 1987, 1996). Economic flow rates require intersecting natural fracture systems.
Depth: Production depths are mostly from 4,000 ft (1.2 km) to 5,300 ft (1.6 km), with average depth of 4,500 ft (1.4 km).
Thickness: The Mesaverde ranges in thickness from about 500 ft (150 m) to over 2500 ft (762 m) and net sandstone to gross thickness of between 20 to 50%. Total pay thickness is from 20 to 200 ft (6 to 61 m).
Porosity and permeability: Porosity ranges from 4% to 18%, with an average of 10%. Permeability ranges from 0.5 mD to 400.0 mD, with usually values less than or equal to 2 mD. Smaller fields have higher values.
Pressure regime: The Mesaverde pressure regime ranges from subnormally pressured (0.22 psi/ft) to normally pressured (> 0.45 psi/ft).
Discovery and production history (including possible adverse conditions for “normal” development): Two adjacent fields, encompassing over 1,000,000 acres (4000 km2), contain the majority of the historic production from the San Juan Mesaverde Group. The Blanco Mesaverde field discovery well was completed in 1927, and the Ignacio Blanco Mesaverde field discovery well was completed in 1952. They have produced almost 7,000 BCFG (200 BM3) and more than 30 MMB of condensate, approximately half of their estimated total recovered volume. Most of the recent gas discoveries range in areal size from 2,000 to 10,000 acres (8.1 to 40 km2) and have estimated total recoveries of from 10 to 35 BCFG (0.28 to 0.98 BM3) (Huffman, 1996).
Geological evaluation as possible continuous gas sandstone play: This play has been routinely described as basin-centered continuous gas sandstone reservoir play until recently, when Fassett and Boyce (2005) stated that it was a conventional, stratigraphically-trapped, low-permeability reservoir system.
Basin setting: The North Great Plains biogenic gas play is a shallow play developed in predominantly siliciclastic rocks of the upper Upper Cretaceous Montana Group and in similar reservoirs in the lower Upper Cretaceous Colorado Group (Fig. A3). Biogenic gas is also produced from marine chalk in the lower Upper Cretaceous Greenhorn Formation.
Geographic extent: The boundaries of the play have been defined as the distribution of Late Cretaceous predominantly marine sandstone and siltstone (Rice and Shurr, 1980). The play is limited to the northwest by erosional truncation on the Sweetgrass Arch and to the west by a north-south line defining the western limit of the Late Cretaceous strata. The play is limited eastward by depth and loss of shelf sandstone and siltstone. It is limited on the west by facies change into conventional reservoirs (shoreline sandstones and continental deposits) and updip water (Rice and Spencer, 1996). Locally, potential reservoir sandstones may be absent or thin beneath regional unconformities (Ridgley and Gilboy, 2001). This gas resource has been extensively developed in Canada over an area of more than 8,000 sq mi on the Alberta Shelf in Alberta and Saskatchewan and estimated ultimate recoveries (EUR) average about 2 BCFG/sq mi (0.02 BM3/sq km) (Suffield Evaluation Comm., 1974; Rice and Shurr, 1980; Rice and Spencer, 1996). In Montana, the play had been extended for greater than 600 sq. mi. by 1979.
Age: Late Cretaceous
Field characteristics: Stratigraphic trapping conditions are present “within both clastic and carbonate reservoirs due to permeability barriers related to facies changes, and to the distribution of fracture systems” (Dyman, 1996, p. 12). In 1995, seven significant fields produced biogenic gas from Late Cretaceous reservoirs in north-central Montana. Of these, four had poorly developed to well developed water drive systems.
Ridgley et al. (1999) described the non-carbonate reservoirs as “thinly laminated sandstone, siltstone, and shale.” Rice and Spencer (1996) characterized the field traps as “micro-stratigraphic” over much of the play area. “The gas is trapped by changes in capillary pressure and pore sizes. Once the gas was exsolved out of the pore water, it was preferentially trapped in the relatively coarser clastics and cannot migrate out of the reservoir except by diffusion” (Rice and Spencer, 1996, p. 16). Henry (2000) characterized the play in Canada as stratigraphically trapped. Structural modifications to this trap style are present at Bowdoin dome and Cedar Creek anticline. Large areas between structures are generally devoid of drilling and, consequently, no production.
Depth: Gas reservoirs range from less than 1,000 ft to about 4,000 ft deep (0.3 to 1.2 km).
Thickness: Late Cretaceous strata vary in gross thickness from about 1,000 ft to more than 3,000 (0.3 to more than 0.9 km) within the defined play area. Specific pay intervals approximate 120-150 ft (37 to 46 m).
Porosity and permeability: “The best gas accumulations are in late Cretaceous reservoirs in permeable shoreface and shelf sandstone” (Dyman, 1996, p. 12). Improved production occurs where natural fracturing or faulting are present. The best reservoir quality occurs in structural “sweet spots,” which were paleohighs where higher quality sandstones reservoir characteristics were better developed (Rice and Spencer, 1996). Bowdoin dome has some conventional and “near-conventional ‘sweet spot’ sandstones” that have produced since 1929 from near its structural crest (Rice and Spencer, 1996, p. 15). The Bowdoin Field is larger than the Bowdoin Dome structure and does not have gas/water contacts. Sandstone porosity ranges between 6 and 20%, with averages of about 12% and high values of 15 to 17%. Based on thin section analyses, Rice and Spencer (1996) note that individual sandstone and silt-stone laminae may have porosity as high as 25%. Permeability to gas is calculated at less than 0.1 mD (Rice et al., 1990; Rice and Spencer, 1996). Fractures are probably essential for production from the carbonate reservoirs (Rice et al., 1990).
Pressure regime: Throughout the play area these reservoirs are underpressured (25-75% of hydrostatic; Rakhit and Hume, 2004).
Discovery and production history (including possible adverse conditions for “normal” development): The play was originally discovered in Alberta in the 1800’s. In the US, the recognition of the presence of natural gas in shallow water wells in 1913 led to eventual development of the play. Tiger Ridge Field was discovered in 1966 as the result of an offset from a dry hole that reached total depth in the Mississippian Madison Group. Almost no wildcat drilling for the biogenic gas play has occurred (Dyman, 1987, 1996; Rice and Spencer, 1996). Gas chemistry data indicate that the gas in this play is probably a mixture of biogenic, thermo-genic, and atmospheric gases (Ridgley et al. 1999).
Geological evaluation as possible continuous gas sandstone play: The Northern Great Plains Bio-genic Gas play extends across a large area of Montana and Alberta (O’Connell, 2003) and is directly related to geochemical and physical conditions conducive to the generation, migration, and entrapment of shallow bio-genic methane. A reassessment in 2000 (Ridgley et al., 2001a) reduced the potential resources of the play in Montana from 42 TCFG (1.2 TM3) to less than 10 TCFG (0.3 TM3) based on significant improvement in knowledge about the play between 1995 and 2000. An improved understanding that very different sedimentary processes, provenance, tectonics, and groundwater flow distinguish this area from similar stratigraphic assemblages to the north in Alberta led to the revised assessment. This new assessment continued to view the Northern Great Plains Biogenic Gas as a continuous gas play, even though there was recognition that stratigraphic variability and subtle structures would continue to control the presence of sweet spots within the play area (Ridgley et al., 2001b; O’Connell, 2003).
Basin setting: The Cretaceous Cloverly-Frontier sandstones (and their equivalents) extend across the entire greater Green River Basin from outcrops in the Idaho-Wyoming thrust belt; Uinta, Wind River, and Rawlins uplifts; and the Sierra Madre and Park Range mountains as an overall eastward prograding, marine-punctuated deltaic, coastal, and marine shelf sandstone series. The main producing trend is in fluvial-deltaic sandstones in the western Green River Basin on the Moxa arch (Fig. A4). Isolated production from marine shelf sandstones is present in the eastern Green River Basin (Washakie Basin) on basement-cored anticlines. The two formations are separated by a thick, organicrich marine shale interval, which includes the Mowry, Shell Creek, Thermopolis, and Aspen shales.
Geographic extent: Both the Cloverly and Frontier sandstones extend across the greater Green River Basin, cropping out on faulted basin margins.
Age: The Cloverly Formation is used as the name for the equivalent of the lower part of the Dakota Formation and is Early Cretaceous (Albian-Aptian) 100-103 Ma. The Frontier Formation is Late Cretaceous (Cenomanian-Turonian-Coniacian) in age (88-91Ma).
Field characteristics: Production from continuous accumulations is focused on the Moxa arch, where intense fracturing and minor faulting help improve reservoir delivery although depositional environment and diagenetic history are also important controls on reservoir quality (Kirschbaum and Roberts, 2005). Marine shales in the Thermopolis, Mowry/Aspen, and Hilliard shales provide local and subregional seals. Other, coastal- or alluvial-plain mudrock intervals also seal some accumulations. Diagenetic permeability barriers can act as seals or traps in some fields (Muller and Coalson, 1989; Reisser and Blanke, 1989). Diagenetic traps and early clay cementation have been described or insinuated for Dakota and Frontier reservoirs on the Moxa arch (Muller and Coalson, 1989; Reisser and Blanke, 1989; Stands-over-bull, 1999).
Natural, artificial, or combinations of both fracture systems are essential for economic flow rates. To the east of the Rock Springs Uplift, Frontier, and Cloverly (Dakota) sandstones are productive only on structure.
Depth: Frontier production ranges from about 7,000 ft (2.1 km) on the Moxa arch to over 19,700 ft (6 km) in the deep eastern greater Green River Basin. The Cloverly (Dakota) Formation ranges in depth from about 5,000 ft (1.5 km) along the trend of the Rock Springs uplift to about 12,000 ft (3.6 km) in the western greater Green River Basin and 21,000 ft (6.4 km) in the eastern greater Green River Basin.
Thickness: The Frontier Formation varies in thickness from less than 100 ft (30 m) in the southwestern Green River Basin to over 1000 ft (300 m) in the northeastern portion. The Cloverly (Dakota) Formation net sandstone ranges in thickness from less than 100 ft (30 m) to slightly more than 250 ft (76 m).
Porosity and permeability: Across the greater Green River Basin, reservoir porosity has a range of 2.4–28%, and permeability has a range of 0.0008–500 mD (Cardinal and Stewart, 1979; Miller et al., 1992). In the Moxa arch area, tidal, fluvial, and shoreface sandstones have similar values ranging from an average of 9.3 to 11% (Winn et al., 1984; Stonecipher and Diedrich, 1993). High porosity values approach about 17%, and permeability values are as high as 50 mD; most measurements are less than 2 mD for tidal and fluvial sandstones on the arch (Stonecipher and Diedrich, 1993). The fluvial and tidally influenced sandstones tend to be better quality reservoirs than shoreface sandstones because of early quartz cementation in the shoreface sandstones and dissolution of late calcite cements in fluvial/tidal sandstones (Winn et al., 1984). Lower shoreface sandstones have the lowest quality because of compaction and clay cementation (Dutton et al., 1992; Stonecipher et al., 1984; Kirschbaum and Roberts, 2005).
Pressure regime: The Frontier gas reservoirs in the eastern Green River Basin are overpressured, and a pressure transition zone is in the Niobrara-Steele-Baxter interval. The Cloverly (Dakota) reservoirs in the eastern Green River Basin are also overpressured below overpressured Frontier. Frontier gas reservoirs are over-pressured on the Moxa arch in the western Green River Basin below the Second Frontier sandstone (Finley, 1984; Law, 1984; Coleman et al., 2003).
Discovery and production history (including possible adverse conditions for “normal” development): Frontier and Cloverly (Dakota) wells were originally drilled on structures in the basin, such as the Moxa arch, Rock Springs uplift, and later on seismically-defined structures. With the recognition of large gas resources in tight sandstones on the Moxa arch, several development plans were implemented to exploit this gas beginning in the mid-1950’s.
Geological evaluation as possible continuous gas sandstone play: The greater Green River Basin-Cloverly-Frontier continuous gas play was considered hypothetical in 1995, even though field development was considered mature on the Moxa arch (Law, 1996). The Cloverly-Frontier continuous gas play was incorporated into the Mowry Composite Total Petroleum System (TPS) by Kirschbaum and Roberts (2005), which includes both a conventional gas assessment unit (AU) and a continuous gas AU. The conventional AU extends around the southern and eastern margins of the greater Green River Basin and in the area of the Rock Springs uplift. The continuous AU occupies the deeper portions of the basin and the western reaches near the Idaho-Wyoming thrust belt including the Moxa arch.
Spencer (1987) identified an unconventional assessment unit in the Frontier-Cloverly (Dakota) where the following conditions are met:
Overpressure is present.
The bottom hole temperature is greater than 200°F.
The vitrinite reflectance is greater than 0.8%.
Permeabilities are low.
No gas/water contacts have been reported.
These conditions are generally present at depths of 8,000 to 12,000 ft (2400 to 3700 km).
Kirschbaum and Roberts (2005) also included a transition zone within their assessment unit, where reservoirs can be either gas charged or water saturated.
Basin setting: The Upper Cretaceous Mesaverde Group of the greater Green River Basin was deposited as a series of back-stepping barrier bar sandstone overlying a coal-swamp dominated coastal complex, which in turn overlies a coastal fluvial interval (Figs. A5, A6). The eastern Green River Basin is surrounded by basement-cored uplifts and contains a deep, basement-cored and faulted uplift saddle (the Wamsutter arch) near its center. The term “Almond Formation” is frequently used as a stratigraphic synonym for the Mesaverde Formation (or Group) in south central Wyoming.
In the greater Green River Basin, the Mesaverde Group consists of the upper Almond Formation, middle Ericson Sandstone (Allen Ridge Formation), and lower Rock Springs Formation. To the southeast in Colorado, it includes the upper Almond Formation, middle Williams Fork Formation, and the lower Iles Formation.
Geographic extent: The Mesaverde extends from outcrops on the north, east, south, and west into the main portion of the eastern Green River Basin and in outcrops surrounding the core of the Rock Springs uplift. It is also present in the western Green River Basin, west of the Rock Springs uplift.
Age: The Mesaverde is Late Cretaceous Campanian in age (71-72 Ma).
Field characteristics: The main gas production to date from the Mesaverde of Green River Basin has been from the eastern portion of the basin, where reservoirs are developed in marginal marine bar sandstones in the upper part of the Almond Formation along the Wamsutter arch between the Washakie and Great Divide sub-basins. Lesser production has been established from fluvial sandstones within the middle and lower Almond. “Only minor amounts of hydrocarbons have been produced thus far from formations in the Mesaverde Group below the Almond, although much of the mean estimate of 3,347 TCFG (94 TM3) of in-place gas estimated by Law et al. (1989) in the Mesaverde Group is in these formations” (Johnson et al., 2005).
Depth: The depth to the top of the Almond Formation varies from less than 2000 ft to greater than 16,000 ft (0.61 to 4.9 km) in the eastern Green River Basin. To-date, most of the Almond gas production has been from depths less than 11,000 ft (3.4 km).
Thickness: The Almond Formation varies from about 250 to about 500 ft (76 to 150 m) in thickness.
Porosity and permeability: Log-based porosity in the Mesaverde gas reservoirs typically ranges from about 8% to 11%. Intervals with porosities below about 8% are usually not artificially stimulated and completed as producers. Intervals with porosity above about 11%, when completed often flow unacceptably high volumes of water and significantly lower volumes of gas (Coleman et al., 2003; Coleman, 2008). Core porosity ranges from less than 5% to usually no more than 12%, averaging about 10% in productive sandstones; core permeability ranges from 0.001 to less than 0.1mD (Evans et al., 2001; Sturm et al., 2001).
Pressure regime: The Mesaverde gas reservoirs are generally overpressured directly beneath a relatively sharp pressure transition zone near the base of the overlying Lewis Shale.
Discovery and production history (including possible adverse conditions for “normal” development): The Mesaverde of the eastern Green River Basin was originally drilled on the Wamsutter arch, a structural saddle between the Rock Springs uplift and the Sierra Madre uplift, resulting in the discovery of Old Wamsutter Field. Several infill drilling programs and a major acreage capture program have also established productive reservoirs in the middle and lower Almond Formation (Horn and Schrooten, 2001; Norris and McClain, 2005; Norris et al., 2005).
Geological evaluation as possible continuous gas sandstone play: In the greater Green River Basin, the Mesaverde TPS includes the Almond Continuous Gas Assessment Unit, the Rock Springs–Ericson Continuous Gas Assessment Unit, and the Mesaverde Coalbed Gas Assessment Unit, plus one conventional assessment unit, the Mesaverde Conventional Oil and Gas Assessment Unit.
In the western Green River Basin, the Mesaverde is included within the Mesaverde–Lance–Fort Union Continuous Gas Assessment Unit, where it occurs in the deeper part of the Mesaverde–Lance–Fort Union Composite TPS. A conventional AU overlies this continuous AU in this area. The contact between the Mesaverde–Lance–Fort Union Continuous Assessment Unit and the overlying Conventional AU cannot be delineated. Most (and perhaps all) of the wells, which produce from this TPS, are from the younger Lance and Fox Hills Formations.
The Mesaverde gas reservoirs are low porosity and permeability reservoirs that typically are relatively consistent as gas producers across the Wamsutter arch. The gas accumulations can easily be interpreted as a myriad of stratigraphic traps associated with barrier bar, tidal channel, and fluvial sandstone pinch outs. The best reservoirs are porous barrier bar sandstones near the crest of the Wamsutter arch. In areas to the west, one of the barrier bar trends appears to be devoid of producible gas even though it produces gas shows during drilling, possibly because of leakage (Norris et al., 2005). In the spectrum of continuity between discrete conventional and classic continuous fields, Charpentier and Cook (2005) interpret the Mesaverde of the eastern Green River Basin as closer to the conventional side, having components of both conventional and continuous gas reservoirs.
Basin setting: The Upper Cretaceous Lewis sandstones in the greater Green River Basin were deposited as generally west-, north-, and south-prograding slope and basin turbidite complex sandstones downdip from essentially coeval Fox Hills deltaic sandstones (Fig. A7). The sandstones were deposited in coalescing channel and channelized lobe settings and consist of marine sandstone bodies.
Geographic extent: The Lewis sandstones in the greater Green River Basin extend from the Rock Springs uplift to outcrops along the eastern margin of the eastern Green River Basin in south central Wyoming. Their extent as gas reservoirs is constrained by basement-cored anticlinal uplifts on the east, south, and west sides of the basin and by facies change to the north (Figs. A6 and A7; Finn and Johnson, 2005). The Lewis Continuous Gas Assessment Unit of the Southwestern Wyoming Province covers approximately 3,310,000 acres (13,000 km2), where the Lewis Shale has attained thermal maturation levels that exceed 0.8% Ro (Hettinger and Roberts, 2005)
Age: The Lewis sandstones are Late Cretaceous (Latest Campanian-Early Maastrichtian) in age (69-71 Ma).
Field characteristics: Fields developed in the Green River Basin in the Lewis are generally stratigraphic in nature, having critical closure up structural dip against sandstone pinchout facies change into lower Fox Hills/upper Lewis slope shales. Where fields are developed in the vicinity of the Wamsutter arch, critical dip is developed by post-depositional basin inversion along the arch and stratigraphic trapping geometries at the down-depositional-dip/up-structural-dip facies changes of the channels and lobes into finer grained coeval facies (Steinhoff et al., 2001; Pasternak, 2004). Fields produce higher volumes where wells are artificially stimulated and they intersect higher fracture density or porosity zones.
Lewis sandstone fields typically produce minor volumes of liquid hydrocarbons and liquid water in addition to the gas. A typical development strategy involves drilling a well to the Almond Formation, approximately 1700 ft (520 m) below the top of the Lewis and using the relatively dry gas and slightly higher pressures of the Almond to help lift the liquids from the Lewis sandstone reservoir (Coleman et al., 2003; Coleman, 2008).
Depth: The Lewis produces from depths of about 2,400 ft to 17,200 ft (0.73 to 5.2 km).
Thickness: The overall Lewis Shale ranges from 2,100 ft to about 2,600 ft (640 to 790 m) thick, and thins to the north and west, toward its depositional pinch-out. Net sandstone thickness ranges from about 200 ft to 600 ft (61 to 180 m). Individual reservoirs range from several feet (1 m) to about 100 ft (30 m) in thickness. The ideal production setting involves several stacked reservoirs, where they may accumulate as much as 350 ft (110 m) of reservoir thickness (Hettinger and Roberts, 2005).
Porosity and permeability: Producing sandstones have porosity values between about 8 and 25% and permeability values from about 0.01 mD to 50 mD (Hettinger and Roberts, 2005).
Pressure regime: The Lewis may be slightly underpressured, normally pressured, or overpressured where it is productive. A basin-wide pressure seal at the base of the Lewis and the top of the underlying Almond Formation forms a major, regional pressure seal in the eastern Green River Basin. The overpressured Lewis reservoirs are in the deeper part of the basin. Pressure gradients range across the basin from 0.32 psi/ft to 0.64 psi/ft (Hettinger and Roberts, 2005).
Discovery and production history (including possible adverse conditions for “normal” development): Lewis sandstone reservoirs of the eastern Green River Basin were discovered while drilling for pre-Tertiary potential on the crest of the Table Rock Anticline, Sweetwater County, Wyoming, in 1954. These initial reservoirs were of limited extent and potential. However, discovery in 1954 and 1958 of larger, purely stratigraphically-trapped Lewis sandstones on the southwest extension of the Table Rock Anticline on a phantom geophysical structure in the Desert Springs Unit, approximately 15 miles northwest of the Table Rock discovery, proved that the Lewis sandstone gas accumulations were not controlled by structural culminations, but by facies change pinchout into the essentially equivalent Lewis shale (Ritzma, 1968). Understanding the reversal of basin depositional geometries (downdip depositional pinchout of sandstones up structural dip) and the role overpressure played in resource accumulations led to increased drilling and development (McPeek, 1981; Weimer, 1960).
Geological evaluation as possible continuous gas sandstone play: The Lewis sandstone reservoirs are low porosity and permeability reservoirs in a stratigraphic trap setting. They typically have a down structural dip interval that produces higher volumes of water than the equivalent updip section.
Basin setting: The Cotton Valley blanket sandstones are part of a large, generally southward prograding, marine-dominated deltaic complex of Late Jurassic to Early Cretaceous age along the northern margin of the Gulf of Mexico sedimentary basin (Fig. A8).
Geographic extent: The blanket sandstones developed between two large delta areas in east Texas and east Louisiana as wave-dominated shelf sandstone deposits. They extend from north to south for approximately 10 to 50 mi (16 to 80 km) (and east to west for approximately 100 mi (160 km) (Coleman and Coleman, 1981).
Age: The Cotton Valley Group ranges in age from Late Jurassic Tithonian (~146 Ma) to Early Cretaceous Valinginian (~132 Ma).
Field characteristics: blanket sandstones; marine shelf and coastal deltaic
Depth: Production depth approx. 5,000 ft to 11,000 ft (1500 to 3400 m)
Thickness: The Cotton Valley blanket sandstones range in thickness from individual sandstone thickness of 30 to 70 ft (9 to 21 m) to cumulative gross thickness of about 2,200 ft (670 m).
Porosity and permeability: Porosity values within this trend usually range from 4 to 20%, and averages generally between 4 and 8%. Conventional core permeability values generally range from 0.02 to 0.41 mD, with an average of 0.09 mD.
Pressure regime: The Cotton Valley blanket sandstones extend from a zone of normal pressure in east Texas and west Louisiana into a zone of overpressure in east central Louisiana. The pressure seal, where present, is generally the tight limestone at the base of the lower Knowles Limestone (or other low-permeability units higher up in the overlying Lower Cretaceous Hosston Formation, where salt-cored anticlines are present) (Coleman and Coleman, 1981; Dyman and Condon, 2006).
Discovery and production history (including possible adverse conditions for “normal” development): Cotton Valley tight-gas production resulted from step-out development drilling for the Cotton Valley “D” sandstone on the flank of the Sabine Uplift in east Texas and northwest Louisiana (Coleman and Coleman, 1981; Bartberger et al., 2002; Coleman, 2008).
Geological evaluation as possible continuous gas sandstone play: The Cotton Valley blanket sandstones were originally considered a continuous gas sandstone play because of their low porosity and permeability, which required fracture stimulation to produce gas at commercial quantities (Schenk and Viger, 1996; Bartberger et al., 2002). However, examination by Bartberger et al. (2002) concluded that the Cotton Valley blanket sandstones were a conventional play, because of the presence of gas-water contacts in at least seven fields across the trend and the presence of relatively high permeabilities and production rates without fracture stimulation.
Appendix I References
The authors acknowledge Loring White, Richard Bishop, Robert Milici, Stephen Cumella, and Keith Shanley for their reviews of an earlier version of this manuscript. Later reviews by Tony D’Agostino, Ricardo Olea, Larry Drew, Sharon Swanson, and Kathie Rankin are appreciated.