Study Area

The Appalachian basin is approximately 500 kilometers wide and 1,000 km long. It encompasses a broad area between the Canadian Shield to the north, the Allegheny front to the east and the Cincinnati arch to the west (UTBEG, 2008). It represents part of an ancient foreland basin in the eastern United States that contains complex geology formed by a series of continental plate collisions. This deformation resulted in the formation of the Appalachian Mountains and large areas of stretched, faulted, and deformed ridges and valleys (USGS, 2008; UTBEG, 2008). The axis of the Appalachian basin is underlain by a succession of strata greater than 3000 meters thick (UTBEG, 2008). (Figure 1) Overlying a major interregional unconformity in the Appalachian basin is the Oriskany Sandstone, a widespread gas reservoir and saline formation The Oriskany Sandstone of the Appalachian basin represents the Deerpark stage of the Early Devonian (Diecchio, 1985). The sandstone is typically a fossiliferous quartz arenite cemented with quartz or calcite. It can be traced continuously through New York, Pennsylvania, Ohio, Maryland, West Virginia, Virginia, and Kentucky (Diecchio, 1985; Bruner and Smosna, 2008). The Oriskany typically overlies strata of Helderberg-age limestone or equivalents, and is overlain by Onondaga-age strata, which vary from limestone to chert to shale and are locally sandy (Diecchio, 1985). Since the Oriskany is a major deep gas producer within the basin data such as pressure, temperature, porosity, permeability, and brine composition are available (Diecchio et al., 1984). The data indicate that there exists intergranular and fracture porosity within the Oriskany, and thick low-permeability zones within the Appalachian basin with the potential for containment (Diecchio, 1984; Gupta et al., 2005).

Saline Formations

As part of the effort by the National Energy Technology Laboratory (NETL) of the US Department of Energy and the Regional Carbon Sequestration Partnerships (RCSPs), data was assembled to address questions of carbon capture and storage (CCS). Data on sources and potential geologic storage sites were collected and organized for the construction of the Carbon Sequestration Atlas of the United States and Canada (http://www.netl.doe.gov/technologies/carbon_seq/refshelf/atlas/index.html).

Saline formations are natural salt-water bearing intervals of porous and permeable rocks that occur beneath the level of potable groundwater (<10,000 mg/l by SDWA standards). A number of the saline formations in the Appalachian basin are used for gas storage and waste-fluid disposal. Saline formations are much more extensive than coal seams or oil- and gas-bearing rock, and represent an enormous potential for CO₂ geologic storage. However, much less is known about saline formations, because they lack the characterization experience that industry has acquired through resource recovery from oil and gas reservoirs and coal seams. Therefore, there is a greater amount of uncertainty regarding the suitability of saline formations for CO₂ storage. In order to maintain the injected CO₂ in a supercritical phase (i.e. super-critical liquid) the geologic unit must be approximately 2,500-feet or greater in depth. (Figure 2) Maintaining the CO₂ in a liquid phase is desirable because, as a liquid, it occupies less volume than when in the gaseous phase. Deep sequestration depths also help insure there is an adequate interval of rocks (confining layers) above the potential injection zones to act as a geologic seal. (Figure 3 )

While not all saline formations in the U.S have been examined, the RCSPs have documented the locations of such formations with an estimated sequestration potential ranging from 3,300 to more than 12,200 billion metric tons of CO₂. (Figure 4) As a result of the inclusion of new evaluations from the Gulf Coast and other areas, identified potential for CO₂ geologic storage in saline formations has increased approximately 2,000 to 9,000 billion metric tons from the previous version of the Atlas.

CO₂ Sources

There are two types of carbon dioxide (CO₂) emission sources: stationary sources and non-stationary sources. Non-stationary source emissions include CO₂ emissions from the transportation sector. Stationary source emissions come from a particular, identifiable, localized source, such as a power plant. CO₂ from stationary sources can be separated from stack gas emissions and subsequently transported to a geologic storage injection site. The “North American CO₂ Sources” map displays the location and relative magnitude of a variety of CO₂ stationary sources. (Figure 5)

According to the EPA, in 2006, total U.S. GHG emissions were estimated at 7,054.2 million metric tons CO₂ equivalent. This estimate included CO₂ emissions as well as other GHGs such as methane (CH₄), nitrous oxide (N₂O), and hydrofluorocarbons (HFCs). Annual GHG emissions from fossil fuel combustion primarily CO₂ were estimated at 5,637.9 million metric tons with 3,781.9 million metric tons from stationary sources. While not all potential GHG sources have been examined, NETL’s RCSPs through the NatCarb effort have documented the location of more than 4,796 stationary sources with total annual emissions of 3,276 million metric tons of CO₂. The concentration of major CO₂ sources in the Appalachian basin is one of the highest in North America. The Midwest Regional Carbon Sequestration Partnership has estimated that the area generates almost 21 percent of our country’s electricity, 78 percent of which is from coal.

Structure

The northern Appalachian basin is an elongate, asymmetric foreland basin with a preserved northeast-southwest trending central axis that extends through Pennsylvania, western Maryland, and West Virginia. The eastern margin of the basin is concealed beneath thrust sheets in the Blue Ridge Province of the Appalachian Mountains. The western margin of the basin occurs in east-central Kentucky and central Ohio. The Cincinnati and Findlay arches separate the Appalachian basin from the Illinois and Michigan basins. Following Cambrian (Iapetan) rifting, the basin was enlarged by periodically reactivation of geologic structures that developed in response to collisional tectonics along the eastern margin of North America during the Taconic (Upper Ordovician), Acadian (Middle to Upper Devonian), and Alleghany (Upper Carboniferous) orogenies of the Paleozoic Era (Tankard, 1986; Quinlan and Beaumont, 1984; Thomas, 1995; Shumaker, 1996). The structure on the Oriskany Sandstone forms a surface that dips toward the center of the Appalachian basin forming a northeast-southwest trend. At the center of the study area the Oriskany Sandstone reaches a subsea depth in excess of 2000 meters. The surface shallows towards the outcrop area to the east along the Allegheny Front and Eastern Overthrust Belt. The surface shallows toward the Cincinnati Arch to the west where the Oriskany Sandstone subcrops or outcrops at ground level.  (Figure 6)

Depth to Oriskany

The Oriskany Sandstone reaches depths of over 2,500 meters along a northeast-southwest trend in the center of the Appalachian basin. It shallows toward the Eastern Overthrust Belt which coincides with the know outcrop area at the east of the basin. The Oriskany gradually shallows toward the Cincinnati Arch to the west in Ohio where it subcrops or outcrops at ground level. (Figure 7)

Temperature

Bottom-hole temperatures (BHTs) are recorded during logging of the borehole and commonly are not at equilibrium with formation temperature and require correction. In general, BHTs from shallow boreholes are too high, and BHTs from deep boreholes are too low. One method to correct the BHT values is to plot them versus depth (Forster et al., 1999). This same type of crossplot was constructed for the Oriskany Sandstone. The average surface temperature for the Appalachian basin (12 °C) provides an intercept, and a correction factor can be generated by comparing the regression equations between the unconstrained and constrained intercepts. For the Oriskany (based on n = 57 wells), the corrected gradient for BHT is:

Oriskany temperature = 0.0195 x (depth in meters) + 12°C

Corrected BHTs were used to construct a map of subsurface temperature of the Oriskany Sandstone. (Figure 8) The Low Plateau and High Plateau of the central Appalachian basin are the deepest and hottest areas.

Isopach

The thickness of the Oriskany Sandstone varies across the Appalachian basin from zero in the “no sand” area to a thickness of over 75 meters. It is typically thickest in the High Plateau and Eastern Overthrust Belt regions. The Oriskany thins to the west, northwest and to the south, where it is generally less than 12 meters thick. (Figure 9)

Reservoir Pressure

Under hydrostatic groundwater conditions, pressure increases with depth at the rate of 9.74 kPa/m for freshwater and roughly 10.71 kPa/m for brine with 145,000 mg/L TDS and is related to both depth and burial history. Final shut-in pressure recorded from gas wells of the Oriskany Sandstone (based on n = 258 wells) is significantly below the expected hydrostatic gradient for fresh groundwater. Pressure plotted against depth indicates that maximum recorded formation pressure follows a gradient of 6.8 kPa/m. This underpressuring may be due to drawdown from nearby wells and lack of sufficient shut-in time to reach pressure equilibrium. The Low Plateau and High Plateau of the central Appalachian basin are the deepest and higher pressured areas. (Figure 10)

Porosity

Previously analyzed core data were used to estimate the porosity of the Oriskany Sandstone. The data was limited to areas of existing oil and gas fields, and therefore mapping based on these data was spatially extrapolated across data gaps within the study area. The core scale porosity measurements were scaled up to well scale and then to basin scale using the method described by Bachu and Underschultz (1992, 1993). The method states that the formation-scale porosity index (Φ) of the unit is the arithmetic average of the core-scale values weighted by the thickness of the unit. A total of 803 wells were assigned values and a mean of 8.08% porosity was calculated. The Low Plateau and High Plateau of the central Appalachian basin are the deepest and contain the area with the lowest porosity values.  (Figure 11)

Total Dissolved Solids (TDS)

Existing brine geochemical data were gathered from published and unpublished state and federal geological surveys, as well as local oil and gas companies. Additional brine samples were collected from existing oil and gas wells distributed throughout the Appalachian basin, with an attempt made to locate the sample sites where know data was lacking. Across the extent of the Oriskany Sandstone the TDS ranges from freshwater (TDS < 10,000 mg/L) to dense brine (TDS > 300,000 mg/L). The brine samples were characterized by large differences between the reported TDS concentrations from neighboring wells within the same gas field. The cause of these differences may be related to areal, vertical, and temporal variability, errors introduced from sampling procedures, or to varying methods of chemical analysis (Jorgensen, et al., 1993). The

dense brines are concentrated in the Oriskany structural lows at the center of the basin. The relatively lower TDS concentrations are associated with the recharge area along the outcrop area to the east and the discharge area to the west. (Figure 12)

Ca/Mg Ratio

Formation waters associated with microbial gas have low Ca/Mg ratios (<2) associated with high alkalinity values from calcite precipitation, induced by microbial methanogenesis (Martini et al., 1998). The ratio of calcium concentration to magnesium concentration for the Oriskany Sandstone across the Appalachian basin is generally less than ten. The ratio is typically lower at the recharge area to the east and the discharge area to the west, with values less than five. The ratio is generally higher in the northeast area of the basin, reaching values of up to 30. (Figure 13)

Sequestration Area

Temperature and pressure conditions must be adequate to keep injected CO₂ in the dense supercritical or liquid phase (at its critical point for CO₂, its temperature Tc is 31.1°C and Pc is 7.83 megapascals, or MPa), this is usually interpreted as a depth greater than 800 meters. An Oriskany Sandstone depth map was constructed to determine areas suitable for CO₂ sequestration suing the 800 meter determining factor. (Figure 14) Within the study site only the areas to the west and northwest were deemed too shallow for sequestration, as well as the outcrop belt to the east. This excluded area consists of the central and western counties of Ohio, as well as the northwest section of Erie and Crawford counties of Pennsylvania.

Potential Energy

Map of contours reflecting levels of potential energy from high (Red) to low (Violet). (Figure 15)  In a fluid environment, the controls on surface are hydraulic head values calculated from water saturated formation pore pressures. From the isopotential contours, flow paths and sinks can be inferred. Modified from Dahlberg, 1995.

Hydraulic Head

Formation pressure was used to calculate equivalent freshwater hydraulic head H_o according to the formula

H_o = z + p / ρ_og

where z is elevation, p is pressure, ρ_o = 1000 kg/m3, and g is the gravitational constant (Bachu and Undershultz, 1993, 1995; Anfort et al., 2001). The density of formation waters in sedimentary basins, a function of temperature and salinity, is not equal to the density of freshwater and has the potential to introduce error (Bachu and Undershultz, 1993, 1995). An indication of the significance of this introduced error is given by the dimensionless driving force ratio (DFR) defined (Davies, 1987; Bachu, 1995) as:

DFR = Δρ|ΔE|/|ρ_o|ΔH_o|h

where |ΔE| is the magnitude of the aquifer slope, |ΔH_o|h is the magnitude of the horizontal component of the freshwater hydraulic-head gradient, Δρ is the difference between formation-water and freshwater densities. If the DFR value is greater than 0.5, neglecting buoyancy effects will introduce significant errors in flow analysis (Davies, 1987). Within the Oriskany Sandstone the values were generally much less than 0.5, with the exception of two areas located in south-central Pennsylvania . The distribution of freshwater hydraulic head shows the expected trends of northwestward and southeastward flow from the basin center toward the western subcrop area and the eastern outcrop area. (Figure 16) Use of freshwater hydraulic heads in the flow analysis of variable density formation waters may introduce significant errors, depending on interaction between the potential and buoyancy forces driving the flow. Hydraulic heads range from over 1000 m in the deeper part of the basin to less than 250 m at the recharge area to east and the discharge area to the west.

Summary

The Oriskany Sandstone of the Appalachian basin is a widely distributed saline aquifer which has produced large quantities of hydrocarbons. Using published and unpublished data of rock characteristics, pressure, temperature, and formation water geochemistry

along with new brine samples were used to map the regional-scale hydrogeological regime and its relation to the migration of hydrocarbons and geologic CO₂ sequestration potential. Basin-scale fluid flow of the Oriskany Sandstone formation waters is generally controlled by salinity differences and by differences in structural elevation. The flow pattern is substantiated by the salinity distributions and water geochemistry (Ca/Mg), with relatively lower salinity at recharge to the east and discharge to the west due to mixing with fresh meteoric water and higher salinity between the recharge and discharge zones. The basin-scale flow pattern is also substantiated by the distribution of oil and gas fields that occur in the Central Appalachian basin; the major productive gas fields occur at the boundary between lower salinity and are typically absent in areas of higher salinity. It is believed that hydrocarbon distribution is influenced by basinal variations in buoyancy and entrainment by the formation water flow. Improved containment of large-scale CO₂ injection appears to be associated in the Oriskany with convergent flow located in the eastern Appalachian basin. Storage capacity for the Oriskany saline formation is estimated by the equation

GCO₂ = A hg фtot ρ E

GCO₂ = estimate of total saline formation storage capacity in grams A = area of basin

greater than 800 meters in depth (151,072,049,688.5 m ), hg = average thickness of formation at depths greater than 800 meters (17.01 meters), фtot = average formation-scale porosity for thickness hg (8.08 %) ρ = density of CO₂ at pressure and temperature that represents storage conditions for saline formation averaged over hg (800 kg/m at P = 18.01 mPa and T = 43.29 °C), and E = storage efficiency factor that reflects a fraction of total pore volume filled by CO₂ (USDOE estimations of E are a low of 0.01 to high of 0.04). The result is a storage resource estimate of 1.6665 to 6.666 gigatonnes of CO₂.

References

Anfont, S.J., S. Bachu, and L.R. Bentley, 2001, Regional-scale hydrogeology of the Upper Devonian-Lower Cretaceous sedimentary succession, south-central Alberta basin, Canada: American Association of Petroleum Geologists, v. 85, p. 637-660.

Bachu, S., 1995, Flow of variable-density formation water in deep sloping aquifers: review of methods of representation with case studies: Journal of Hydrology, v. 164, p. 19-38.

Bachu, S., and J.R. Undershultz, 1992, Regional-scale porosity and permeability variations, Peace River Arch area, Alberta, Canada: American Association of Petroleum

Geologists, v. 76, p. 547-562.

Bachu, S., and J.R. Undershultz, 1993, Hydrogeology of formation waters, Northeastern Alberta basin: American Association of Petroleum Geologists, v. 77, p. 1745-1768.

Bachu, S., and J.R. Undershultz, 1995, Large-scale underpressuring in the Mississippian-Cretaceous succession, Southwestern Alberta basin: American Association of

Petroleum Geologists, v. 79, p. 989-1004.

Bruner, K., and D. Smosna, 2008, A trip through the Paleozoic of the Central Appalachian basin with emphasis on the Oriskany Sandstone, Middle Devonian shales, and Tuscarora Sandstone: Dominion Exploration and Production, Inc.

Dahlberg, E. C., 1995, Applied Hydrodynamics in Petroleum Exploration, Springer-Verlag, 295 p.

Davies, P.B., 1987, Modeling areal, variable density, ground-water flow using equivalent head-analysis of potentially significant errors, in Proceedings, National Water Well

Association-International Ground Water Modeling Center: Solving ground water problems with models, Denver, February 1987, Volume 1: Dublin, Ohio: National Water

Well Association, p. 888-903.

Diecchio, R.J., 1985, Regional controls of gas accumulation in Oriskany sandstone, Central Appalachian basin: The American Association of Petroleum Geologists Bulletin,

v. 69, p. 722-732.

Diecchio, R.J., S.E. Jones, and Dennison, J.M., 1984, Oriskany sandstone: regional stratigraphic relationships and production trends. Morgantown: West Virginia Geological and Economic Survey.

Forester, A., D.F. Merriam, and W.L. Watney, 1999, Problems and potential of industrial temperature data from a cratonic basin environment, in A. Forester and D.F. Merriam, eds., Geothermics in basin analysis: New York, Plenum Press, p. 35-59.

Gupta, N., P. Jagucki, J. Sminchak, D. Meggyesy, F. Spane, T.S. Ramakrishnan, and A. Boyd, , 2005, Determining carbon sequestration injection potential at a site specific location within the Ohio River Valley region, in Proceedings, International Conference on Greenhouse Gas Control Technologies, 7th, Vancouver, September 2004: Amsterdam: Elsevier, v. 1, p. 511-519.

Jorgensen, D.G., J.O. Helgesen, and J.L. Imes, 1993, Regional aquifers in Kansas, Nebraska, and parts of Arkansas, Colorado, Missouri, New Mexico, Oklahoma, South

Dakota, Texas, and Wyoming—Geohydrologic framework: U.S. Geological Survey Professional Paper 1414-B, p. 72.

Martini, A.M., L.M. Walter, J.M. Budai, T.C.W. Ku, C.J. Kaiser, M. Schoell, 1998, Genetic and temporal relations between formation waters and biogenic methane:

Upper Devonian Antrim Shale, Michigan Basin, USA: Geochimica et Cosmochimica Acta, v. 62, p. 1699-1720.

Quinlan, G.M., and C. Beaumont, 1984, Appalachian thrusting, lithospheric flexure, and the Paleozoic stratigraphy of the Eastern Interior of North America: Canadian Journal of Earth Sciences, v. 21, p. 973-996.

Shumaker, R.C., 1996, Structural history of the Appalachian Basin, in Roen, J.B., and Walker, B.J., (Eds.) Atlas of Major Appalachian Gas Plays: Gas Research Institute,

West Virginia Geological and Economic Survey Publication v. 25, p. 8-22.

Tankard, A.J., 1986, Depositional response to foreland deformation in the Carboniferous of eastern Kentucky: AAPG Bulletin, v. 70, p. 853-868.

Thomas, W.A., 1995, Diachronous thrust loading and fault partitioning of the Black Warrior foreland basin within the Alabama recess of the Late Paleozoic Appalachian-

Ouachita thrust belt, in Dorobek, S.L., and Ross, G.M. (Eds.) Stratigraphic Evolution of Foreland Basins: SEPM Special Publication, v. 52, p. 111-126.

USGS, U.S. Geological Survey, Appalachians: Introduction, http://energy.er.usgs.gov/regional_studies/appalachians/, (accessed August 24, 2008).

UTBEG, Bureau of Economic Geology, the University of Texas at Austin, Research Environmental Quality, http://www.beg.utexas.edu/environqlty/co2seq/research.htm,

(accessed August 25, 2008).

Acknowledgments

Data and samples were provided by Dominion Exploration and Production, Chesapeake

Energy, NiSource, Texas Keystone, Inc., Spectra Energy, US Silica, Co., and the geological surveys for West Virginia, Maryland, Ohio and Pennsylvania. Financial support was provided by the Southeast Regional Carbon Sequestration Partnership (SECARB) and the National Energy Technology Laboratory of the US Department of Energy. HighMount Exploration and production provided support to Jamie Skeen.

Return to top.