- PDF LinkChapter PDF
A. Chaouche, 2009. "Unconventional Seals for Unconventional Gas Resources: Examples from Barnett Shale and Cotton Valley Tight Sands of East Texas", Unconventional Energy Resources: Making the Unconventional Conventional, Tim Carr, Tony D’Agostino, William Ambrose, Jack Pashin, Norman C. Rosen
Download citation file:
Assessment of undiscovered oil and gas resources is based on geological elements and processes of a petroleum system. Application of the petroleum system in the oil industry varies largely on how the processes of hydrocarbon generation, migration, and entrapment are described. It is often depicted as a relationship between source and reservoir rocks connected by fluid paths (e.g., carrier beds, faults, etc.) through geological time. When appropriate conditions (time, temperature, and trap formation) are reached, the effort is focused on secondary migration from source to reservoir. Secondary migration efficiency is a function of the distance between source and reservoir rocks. Tertiary migration (or ‘dismigration’) refers to fluid movement from reservoir to reservoir and involves migration pathways (fault or sand beds and/or unconformities).
There has been much research on source rock quality and its relationship to hydrocarbon potential. Much less has been documented about the rate, mechanisms, and pathways by which gases migrate through kilometer-scale sequences of fine-grained sediments. Mass balance calculations supported by laboratory experiments on good quality source rocks show that significant volumes of hydrocarbons can be generated and expelled from the source rock, but exploration results show that only a small fraction (<10%) is trapped within conventional reservoirs. Dispersion in the carrier beds (10 to 20%), retention in the source rock (30 to 40 %), ‘dismigration’ (10 to 20%), and biodegradation (10 to 20%) are commonly assumed to be the altering mechanisms of the bulk fluid generation. The proximity of source rock and reservoir rock becomes critical to fluid preservation and accumulation.
The unconventional Barnett Shale and Cotton Valley tight sands of East Texas are no different from other petroleum systems. The Barnett Shale is a classic shale gas system that includes the elements of source, reservoir, and seal. The Cotton Valley Formation exhibits an inter-fingering shale/sand system that juxtaposes source and reservoir, offering preservation and high migration efficiency.
Occurrences of sweet spots in Barnett Shale are related to the original source rock richness, maturity, and confinement of the source beds. The Fort Worth basin of East Texas is asymmetric and has a polyphased burial history. Its western part along the Washita high has undergone uplift and erosion at the Miocene. The resulting liable asphaltenes precipitation has created a permeability barrier within the shale preventing gas from escaping laterally to the west. The lower Barnett encased between the Marble Falls Limestone and the Chappel Limestone has limited gas leakage to the top and the bottom, creating an optimum seal for the Newark Field where the highest gas production per well has been observed. Laminated carbonates and chemically induced carbonate nodule deposits in the early organic diagenesis provide vertical and lateral baffles to fluid flow thus enhance the confinement within the most productive Barnett Shale.
In the Cotton Valley Formation, significant permeability reduction occurs within the interfingering shale and tight sands. The migration of oil from shale to sand has accumulated a significant volume of oil that ultimately has cracked to gas when burial reached the gas window in the Cotton Valley. This secondary cracking has resulted in high pressures extending far beyond the source rock, flushing the interstitial water to overlaying formations. Different chemical water mixes has lead to mineralization and thus diagenetic seals enhancing confinement, which has result in stair-step pressure offsets occurring independently of lithology profiles.
The increased costs of finding and developing new conventional reserves and the push for energy independence challenged the oil industry to find new ways of meeting the energy need, inducing a shift in strategy for many oil companies that moved from exploration and development of conventional to unconventional resources. The tendency is echoed by environmentalists seeking cleaner energy, as well as political leaders, providing support through incentives to unconventional gas plays. As a result the energy mix is set for future growth in the U.S.
The drilling technology has moved quickly from conventional reservoirs to focus on unconventional reservoirs such as tight sands and shales. Gas shale represents 7 % of the total US domestic gas supply in 2008 as a consequence to the unprecedented effort in creating and sharing knowledge among the shale gas players. It started with the most prominent the Barnett Shale play in 2000 and spilled over to all potential shale plays creating a huge investment opportunity across North America. As of today, more than 28 shale formations are targeted by small and large oil companies.
The Energy Information Administration estimates that 44% of the current gas needs is provided by the energy mix; tight gas sand is the most active unconventional natural gas play supplying up to 29% of the US production.
Tight gas sands and gas shales are low permeability-porosity reservoirs storing significant gas reserves in their matrix porosity and natural fractures network. Economic gas production is achieved through induced hydraulic fracturing aimed to open connections to existing natural fractures or to the present geo-mechanical brittleness of the formation unlocking the stored gas. Induced hydraulic fractures are contained in the sweet spots with care taken in avoiding their propagation into the aquifer that underscores the successful development of the play.
This presentation is about the petroleum system that is a key requirement for most the oil and gas bearing provinces. It is aimed at outlining some specific attributes of the petroleum system of the Lower Cotton Formation and emphasizes on the fluid containment within the system that makes it favorable for simulation with hydraulic induced fractures. The Barnett Shale of the Forth Worth basin shows through its burial history an unroofed basin, in which the uplifts, followed by erosional events, has brought the Barnett Shale to much more shallow levels, causing great gas expansion and probably an important source of gas leak to the system.
Cotton Valley Formation of East Texas and Northwest Louisiana
Petroleum system of the Cotton Valley Formation
In the “Assessment of Undiscovered Conventional Oil and Gas Resources–Upper Jurassic–Lower Cretaceous Cotton Valley Group, Smackover Interior Salt Basins Total Petroleum System, in the Texas Basin and Louisiana-Mississippi Salt Basins Provinces” (2006), the USGS stated that the lower part of the Cotton Valley Group of East Texas, the Bossier Shale, has not been assessed and will be treated separately. This presentation deals exclusively with the Lower Cotton Valley that includes the Bossier/Haynesville shale-sand sequence that expands more over south-central and East Texas to Louisiana (Fig. 1).
Much has been published on the occurrence of oil and gas in the Cotton Valley Formation: Coleman and Coleman (1981) suggested hydrocarbon migration from the neighboring sourced beds. Dutton (1987) asserted that Travis Peak Reservoir in East Texas that overlies the lower Cotton Valley received the bulk of its fluids from both the Smackover Limestone source and the Bossier Shale source. Presley and Reed (1984) offered an alternative to the Smackover by suggesting that the interbedded black shale within the Cotton Valley sandstones and the basal marine black shale of Bossier are the main source rocks for the Cotton Valley Formation. Wescott and Hood (1994) demonstrated that the main source rock in the East Texas shale/sand in the lower Cotton is the Bossier Shale. Sassen and More (1988) attributed most of the hydrocarbons found in the Mississippi Alabama reservoirs to sources from the Smackover mudstone carbonate. Lewan (2002) agreed that multiple source rocks and oil sources may have contributed to Cotton Valley accumulations. Rushing et al. (2004) showed that the marine Bossier source rock quality diluted by the clastic shed from the deltas improves basinward to the ancestral depocenter of the Gulf of Mexico.
If the main source rocks in the Cotton Valley Formation have been identified and some source rock/fluid correlations have been carried over time, a comprehensive study that integrates hydrocarbon fluid generation, mineralization due to transport of minerals in solutes interacting with surrounding fluid charges, diagenetic processes that inhibit quartz grow, and preserve porosity or the pressure build up of hydrocarbons that helps preserve porosity at greater depth has yet to be disclosed.
Environment of deposition in the Lower Cotton Valley
The Cotton Valley Formation of the Upper Jurassic is a fluvial deltaic system lying conformably across northeast Texas to northwest Louisiana and progrades from north to northwest to the ancestral Gulf of Mexico depocenter located during the Jurassic along the Louisiana/Mississippi border (USGS, 2006).
The Cotton Valley consists of thick interfingered shale/sand sequences, in which the Bossier Shale is the basal member that includes transgressive foreset beds over prograding delta sediments throughout most of eastern Texas and northwestern Louisiana.
Bossier shales are typically black, organic-rich, calcareous, fossiliferous, marine deposits and are the primary source rock for much of the Upper Jurassic and Lower Cretaceous reservoirs. Condensed and thinner shale intervals in the western part of the basin expand basinward to thicker units, which typifies the general depositional patterns in the East Texas Basin (Fig. 2). Source rock quality is generally poor in the western and northwestern parts of the basin because of the dilution by the clastics, but improves significantly towards the basin center.
This interfingering shale/sand sequence provided a juxtaposition of source rock-reservoir doublets in which most of the oil generated by the Bossier shale was retained in the source rock or migrated to the subjacent tight sands of the Lower Cotton Valley Group. This sequence is well displayed by the pseudo-seismic dip section of Cotton Valley (Taylor Group), Jurassic, Panola County, Texas, in Williams and Mitchum (2001)
Reservoir rock of the Lower Cotton Valley Bossier
The Cotton Valley Formation has been identified as gas play by Anadarko Petroleum Corporation in the nineties. Dew/Mimms fields of Freestone County in East Texas were developed as part of a broader effort in exploring and developing unconventional gas resources. These fields produce from two or three primary sands separated by organic- rich black marine shales. They are characterized with a range of typical sand properties from 5% to 20% porosity and 0.001 to 0.1md rock permeability, and water saturation in reservoir-quality ranged from as low as 5% in the most permeable rock to as high as 50% in the lower-quality reservoir rock (J.A Rushing et al. 2004). These fields of large in place hydrocarbons (1 Tcf) with low recovery have become economical when stimulation with hydraulic fracturations (Emme and Stanil 2002) was applied (Fig. 3).
Attributes of basin-centered accumulation
Rock reservoir and fluid as well as source rock properties, determined from a very rigorous and comprehensive description and characterization program, suggest that the lack of trap/seal associated with abnormal pressure in this relatively low matrix permeability reservoir rock is perhaps part of a bigger scale basin-centered gas system (Newsham and Rushing 2002; Rushing et al., 2004).
The USGS (2002) reevaluation of basin-centered gas accumulation in the Cotton Valley Group sandstones concludes the absence of characteristics of basin-centered gas accumulation within the Cotton Valley low permeability massive sandstones.
Basin-centered gas accumulation is defined as continuous gas involving very low permeability reservoirs, and sand/shale sequence that includes beds of rich organic source rock generating gas when adequate burial/temperature is reached: Law (1984a, 1984b, 1995, 2000). The overpressure develops because of thermal generation of gas occurring at a rate that exceeds the rate at which gas is lost updip by migration through the low permeability reservoirs. Any free water remaining in the pores space is pushed out by the gas expansion to the overlaying formations and only irreducible water remains in the tight-sand reservoir. If the concept doesn’t hold across the entire Cotton Valley Formation, it does show that some of these attributes are persistent locally in Freestone County Texas; wells have very low water saturation, lack water-gas contacts, and show an occurrence of abnormal geopressure that changes with depth independently from the lithology and corroborating the gas generation window (Fig. 4). The concept of basin-centered gas accumulation may need to be reexamined in the light of the new data disclosed from deep wells targeting gas accumulations.
Organic and mineral diagenesis in the Bossier Sand-Shale section of LCVF
Clay minerals are deposited in the ocean at different distances from the coastline and their relative abundance reflects the sediment source and its proximity. Detrital kaolinite indicates coastal proximity, whereas illite and smectite suggest deep marine environment. The nearshore sand and low-stand fan deposits of the lower Cotton Valley are oriented north-west-southeast and are usually encased in the marine Bossier shale. The regressive and transgressive episodes of Bossier sequence have created a juxtaposition of source rock and reservoir, in which fluid expulsion from shale at the oil widow interacted with the adjacent sandstones. Carboxylic acids, organic acids, and carbonic acids have changed the pore-water pH that in turn affected the solubility of clay minerals. The potassium and aluminum exported from the sandstones to the shale influenced the pore fluid composition. These fluid fluxes ultimately have induced mineral precipitation, cementation, and change in the rock texture. Cementation and mineral precipitation occur in the pore media at different depth of burial. The chemical affinity and concentration of the components involved in pore fluid composition determine the nature and the intensity of the cementation and mineral precipitation.
Composition of the sandstones and their specific mineral content control the diagenetic pathway. The lower Cotton Valley assemblage of East Texas and northwest Louisiana are fluvial/deltaic sediments consisting of rich quartz arenites and arkoses that predispose their diagenetic pathway to be different from their counterparts in the more distal part of the basin, were dirtier quartz in the deep water of the ancestral Gulf of Mexico may have preserved their partial primary reservoir rock properties.
Hydrocarbon generation impacts also the mineral diagenesis through the pressure it delivers to the system. The Bossier Shale of the lower Cotton Valley has generated through time significant volume of oil and gas, inducing abnormal pressure that has balanced back the lithostatic pressure exerted by the sedimentary column, attenuating the effect of the chemical compaction dissolution. The counter-reaction exerted by hydrocarbons can be sensed through the limited number of stylolites we have observed in some of the Cotton Valley cores cut in deeply buried sandstones (15,000 ft and greater).
In the lower Cotton Valley system, the Bossier Shale is encased within lower porosity/permeability sands. Occurrence of oil and gas from the cracking of the kerogen has migrated into the system. Bredehoeft et al. (1994) show that pore pressures of oil generation could be maintained for long periods in low-permeability source rocks and lithostatic pressure values can be attained.
Pyrobitumen have been encountered in many reservoir rocks of the Cotton Valley, suggesting an “in situ” thermal cracking of oil that has converted oil to gas. Such an event has contributed significantly to the pressure build-up of the system. The oil cracking to gas requires higher activation energy and likely postdates the bulk of gas generation involving the kerogen.
The lower Cotton Valley system has been, therefore, under hydrocarbon pressure lift since the oil generation, occurring around 90 My ago, to gas generation, then oil to gas cracking. As a consequence of the state of abnormal pressure that prevailed in many locations of East Texas and northwest Louisiana, rock reservoir properties have been relatively preserved (Figs. 5A and 5B).
Lower Cotton Valley petroleum system model
The lower Cotton Valley of East Texas and northwest Louisiana is a very efficient system, in which fluid generation from the encased Bossier sourced rock migrated to the subjacent sandstones. The system has built pressure through time partly because of the tight reservoir rock but more importantly because of the diagenetic seals created by the upward displacement of the interstitial water to the overlaying beds. Many well-bore penetrations show abrupt increase mud weight with depth (Fig. 4). The architectural result is pressure/depth profile showing a stair-step change in pressure that is characteristic of multiple stacked-pressure compartments.
The observed trend of offset pressure in Mims Creek area seems to be around 12,000 ft, corroborating the gas generation window. This occurrence, independently from the lithology, suggests that fluid fluxes and incompatible water chemistry mixing have been the controlling factors.
The Cotton Valley petroleum system as a whole incorporates source rock and reservoir rock, and can be considered as a single container where the system is self-sourced, self-pressured and self-sealed. The confinement conducted by both the thick overburden section of sediments and the diagenetic seals resulting from the fluid movements preserved most of the hydrocarbons generated over its geological history. The Laramide uplift and the removed section that followed had little, if any, effect on the hydrocarbon dispersion/leak.
Barnett Shale of Forth Worth Basin
The Barnett Shale is a black, organic-rich shale associated with occasional limestone and deposited during the Late Mississippian in the Forth Worth basin of north-central Texas. The Barnett Shale is the primary source rock for oil and gas for many Paleozoic conventional clastic and carbonate reservoirs. The Barnett Shale expands from west to east in the Fort Worth Basin from less than 100 ft to 800 ft in the Ouachita structural front to 1000 ft in the north-east, towards the Muenster arch.
The present-day asymmetrical thickness from west to east and north to south is related to the structural movements of the Bend arch/Lampasas arch to the west and to the Llano uplift to the south, while an active tectonic subsidence took place along the Ouachita fault trend in the east end of the Forth Worth Basin. The Ouachita trend fault is not only a weak basement zone favoring the emplacement of a thick shale section, but also plays a major role in the thermal history of the basin.
Barnett Shale source rock
With an organic content averaging 5% TOC and a type II marine kerogen, the Barnett Shale Formation possesses a strong hydrocarbon potential for sourcing conventional traps and retained enough gas that made it ultimately a great unconventional shale gas play. Source rock maturation assessment shows a polyphased history consistent with the structural events that shaped the Forth Worth basin.
Hydrogen indexes increased from less than 50 mg HC/gr TOC in the eastern part to 400 mg HC/gr TOC and up in the western part of the Forth Worth Basin. The Barnett Shale shows high Gamma response up to 400 API, low transit time, and high resistivity; this is a typical log response to many world class source rocks. The source rock maturity of this unit decreases from east to west: values of vitrinite reflectance in the range of 1.3 Ro in the vicinity of the Ouachita trust fold belt decline to 0.6 in the Bend arch. This trend of maturity is supported by the BTU values of the gas, as pointed by Jarvie et al. (2005).
Our hydrogen indexes plots against depth of the Barnett Shale shows two clusters of kerogen transformation: low and at shallow depth in the Parker and Young counties; high at higher depth in Jack, Wise, Tarrant, and Denton counties (Fig. 6). This appearance in hydrogen indexes reflecting the transformation ratio looks, a priori, normal but in reality has been shaped by two mains source rock maturity phases: pre-Cretaceous and post Miocene times.
The depth of burial reached by the Barnett Shale during the first episode is equivalent to the oil window Ro (0.6 to 0.9). This episode is likely diachronic and has occurred early in the east and north of the basin, reaching the west and south later. The uplift and erosion of about 4,000 ft of sediment that followed during the Cretaceous interrupts the hydrocarbon generation and has unroofed most Fort Worth basin, inducing significant volume loss of light hydrocarbons mainly gas and precipitation of asphaltenes and heavy compounds (NSO) in the pore space.
In the second episode, the burial resumed in the north and eastern flank of the basin along the Ouachita front, deepening the source rock to the gas window Ro (1.2 to 1.6), while the south and western part were kept “frozen” because of low rate of sedimentation combined to rising basement arch limiting therefore any additional burial.
Burial resumed in the east, central, and in the north of the Forth Worth Basin during the Tertiary. Remaining kerogen, heavy compounds, and oil that have been generated in the first phase of burial and trapped in the source rock, when the basin was uplifted, restarting their cracking in the east generating mainly gas (Fig. 7).
The uplift and truncation of the sedimentary cover during the Cretaceous created in the source rock a drop in pressure and gas expansion responsible for most of the heavy compounds and their precipitation in the pores space, plugging the pore spaces and reducing the permeability as it is today in western flank of the basin (Fig. 8).
The multi-phased geological history of the Fort Worth Basin changed dramatically the redistribution of hydrocarbons and their entrapment. Besides the organic carbon content of the Barnett Shale that holds gas by adsorption, as shown on the correlation between TOC and production indexes of the Wise county wells (Jarvie, 2003), the confinement of the Barnett Shale appears critical for the gas entrapment, which consequently improves the development of the pressure relatively to the rest of the basin. As a result, higher gas production is reported in the eastern part of the Fort Worth basin, where the confinement of the Barnett Shale is believed to have been enhanced laterally to the west by the relative impermeability or the “permeability jail” created by the precipitation of the heavy compounds as a consequence to oil generation process interrupted by the Cretaceous uplift.
This permeability barrier to the west combined to the bottom seal and top seal by, respectively, the Chappel Limestone and Frostburg Limestone Formations, enhanced in some locations, such as the Newark field, the confinement of the Lower Barnett Formation, leading to the higher gas production wells. The efficiency of the well completion through mechanical fracturing has been reported by many Barnett players as inherent to the confinement of the rocks.
The similar geochemical feature of Barnett and Bossier resides in the oil-prone marine (type II kerogen) source rock character they share that was able to generate significant amount of gas when depth of burial was appropriate. These two source rocks have had higher hydrogen content at the time of deposition, likely in the range of 350 to 500 mg HC/g TOC before their burial.
The continuous burial during Jurassic and Cretaceous for the East Texas Basin has been interrupted at the Paleogene/Neogene (Laramide event) that removed a section of about 3,000 ft, based on vitrinite reflectance estimates. The important rate of sedimentation during the Mesozoic/Cenozoic, together with thick interbedded shale/sand sequences, results in a significant seal for the system that did not suffer any significant loss for the hydrocarbons that have been generated. The successive pulses of oil then gas have built a continuum in pressure through time that preserved rock properties and enhanced the confinement of the system through the diagenetic seals they created.
The multi-phased burial history of the Fort Worth basin suggests the geometry of an exhumed basin that experienced significant loss of oil and gas and significant pressure dissipation. The remaining fluids in the system have been trapped in the pore space as heavy compounds or residual oil that cracked to gas when the source rock resumed its burial. The Barnett Shale plays the double role of source and reservoir rock. The effectiveness of top and bottom seals was secured by limestones, and laterally by faulting/folding to the east and a permeability barrier to the west. Such confinement seems to have controlled the bulk of gas generated in the last episode of its burial history.
Both the Cotton Valley with its diagenetic seals and the Barnett Shale with its heavy compounds locking the path to the west offer an unconventional trapping mechanism for unconventional resources.