Assessment of the Oil and Natural Gas Potential of the East Coast Mesozoic Synrift Basins, Onshore and State Waters of the United States
Published:December 01, 2015
- PDF LinkChapter PDF
James L. Coleman, Jr., Robert C. Milici, Paul J. Post, 2015. "Assessment of the Oil and Natural Gas Potential of the East Coast Mesozoic Synrift Basins, Onshore and State Waters of the United States", Petroleum Systems in “Rift” Basins, Paul J. Post, James Coleman, Jr., Norman C. Rosen, David E. Brown, Tina Roberts-Ashby, Peter Kahn, Mark Rowan
Download citation file:
Immediately prior to the opening of the Atlantic Ocean in the Mesozoic Era, numerous extensional and transtensional basins developed along the eastern margin of North America from Florida to Canada and from the Appalachian Piedmont eastward to the edge of the present-day continental shelf. Using a petroleum system-based methodology, the U.S. Geological Survey examined 13 onshore Mesozoic synrift basins and estimated a mean undiscovered natural gas resource of 3.86 trillion cubic feet (TCF; 109 billion cubic meters, BCM) of gas and a mean undiscovered natural gas liquids resource of 135 million barrels (MMBNGL; 21.5 million cubic meters, MMCM) in continuous accumulations within five of these basins: the Deep River, Dan River-Danville, Richmond, Taylorsville basins, and the southern part of the Newark Basin. The other eight basins were examined, but not assessed due to insufficient data. An additional 26 basins in the East Coast Mesozoic synrift basins trend were examined here for further insights into the development and evolution of a large, but short-lived set of petroleum systems in Mesozoic synrift basins.
An individual composite total petroleum system is contained within each of the assessed basins. Small amounts of oil and natural gas have been recovered from many of the basins, yet no commercial production has been established. Potential and identified source rocks are present as shale and (or) coal. Potential reservoir rocks are low porosity and permeability sandstones as well as shale, siltstone, coal, and fractured igneous rocks. Examination of data indicates that many of these rift basins have undergone substantial uplift (greater than 4,000 ft, 1200 m), and one or more episodes of water washing have affected oil accumulations. Drilling for conventionally trapped structural and (or) stratigraphic prospects has not been successful. Remaining potential appears to be in continuous (unconventional) gas and natural gas liquid accumulations in a variety of reservoir types.
The Mesozoic synrift basins of the eastern United States contain one of the largest, yet enigmatic, set of petroleum systems in the United States. The trend extends across 1,300 miles (2,092 km) and 11 states. It is composed of at least 55 individual rift basins, of which 17 are exposed at the modern surface of erosion and at least 38 are buried beneath the Atlantic Coastal Plain and offshore continental shelf (Fig. 1; Table 1). Historic onshore drilling and testing for oil and gas have occurred in 14 of these basins, and 10 have oil and/or natural gas shows (Table 1). Yet, after 127 years of exploration since 1888, no commercial fields have been discovered. Because of its proximity to population and manufacturing centers in the eastern United States, and as part of its routine assessment of the technically recoverable, undiscovered oil and gas resources of the United States, the U.S. Geological Survey (USGS) undertook a review and assessment of the petroleum potential of these basins in 2012.
The Mesozoic synrift basins of the eastern United States are mostly a series of half grabens that formed following the cessation of tectonic compression associated with the Appalachian Alleghenian Orogeny, which culminated by the end of the Early Permian (Cisuralian) (Hatcher, 2002; Secor et al., 1986). Pangean breakup and extension/transtension created the rift basins, which began to form soon after continental plate collision ceased, potentially as early as late Middle Permian (Guadalupian) in Nova Scotia (Hmich et al., 2006; Olsen and Et-Touhami, 2008) and continued from the Middle Triassic (Anisian) until possibly the late Early to early Middle Jurassic (Pleinsbachian to Aalenian) (Olsen and Et-Touhami, 2008; Wade et al., 1996; Withjack et al., 1998), when ocean floor spreading jumped eastward to the current Atlantic Ocean basin (Benson, 2003; Janney and Castillo, 2001; Müller et al., 2008).
A set of complementary, syntectonic, rift basins to the North American Mesozoic synrift basins extends along the continental margins of northwest Africa and southwest Europe. In the United States and Canada, the synrift basin-fill strata are collectively termed the “Newark Supergroup” (Froelich and Olsen, 1985; Olsen, 1978; Van Houten, 1977). Even though the American set of Atlantic margin Mesozoic synrift basins is collectively called the “Newark Rift Basins,” each basin has its own tectonostratigraphic history. As such, a comprehensive assessment of potential oil and gas resources requires each basin to be examined individually, but in context with the other basins in the set. In this report, this set of North American Atlantic margin Mesozoic synrift basins is called the “East Coast Mesozoic synrift basins.”
The East Coast Mesozoic synrift basins were developed on Appalachian thrust belt terrane from Nova Scotia to Georgia and extend beneath the Atlantic Coastal Plain to the continental shelf margin. Approximately one-third of these basins are exposed and can be mapped from surface exposures and shallow wells (Brown, et al., 1985; Calver and Hobbs, 1963; Gray, et al., 1980; King and Beikman, 1974; Weaver, 1968). The general size and shape of basins buried under the Atlantic Coastal Plain may be inferred from seismic, magnetic, and gravity data. East Coast Mesozoic synrift basins are commonly represented by elongate regions of low magnetic gradients and intensity and generally small gravity lows (Fig. 2). Narrow magnetic highs are associated with Mesozoic diabase intrusions or structurally tilted tholeiitic flow complexes. The broad magnetic highs of North Carolina are known from drill holes to be associated with crystalline rock. In South Carolina, the high amplitude signature of the Charleston Magnetic Terrane has been related to a Jurassic basalt flow complex (Daniels and Leo, 1985; Daniels et al., 1983; Gottfried et al., 1983).
Much of the regional geologic and geophysical basis for this study came from Froelich and Robinson (1988), Gohn (1983), Horton et al. (1991), LeTourneau and Olsen (2003), Manspeizer (1988), Manspeizer et al. (1989), Margolis et al. (1986), Olsen et al. (1989), Rankin (1977), Robinson and Froelich (1985), Schlische (2003), and Withjack et al. (2012, 2013).
The prevailing structural style of the onshore East Coast Mesozoic synrift basins is that of a half-graben, in which one major normal fault forms along a previously existing Alleghenian (Pennsylvanian–Permian) thrust fault (Petersen et al., 1984; Ratcliffe and Burton, 1985). Secondary compensation faults having less vertical throw are common.
Structural complexity varies from basin to basin and across a spectrum of relatively simple half-grabens (e.g., Deep River and Culpeper basins), to more complex basins having both a major basin-bounding fault system and minor faulting on the opposite basin hinge zone (e.g., Hartford basin), to highly complex wrench-faulted rhombgrabens (e.g., Richmond and Newark basins; Biddle and Christie-Blick, 1985; Ziegler and Cornet, 1985).
Prospective four-way closing structures are more likely to occur in those basins dominated by wrench tectonics. Basin orientation, and to a certain degree basin shape, are controlled by the Appalachian tectonic grain and structures. Where these are normal to primary extension stresses, extensional structures (i.e., faulted ramps) predominate; wrench folded structures are rare. However, where the tectonic grain is oblique to extension stresses, wrench fault-induced folds are most common (Root, 1986).
Surface and subsurface mapping suggests that faults associated with early Mesozoic basin development may have continued to be active into the Tertiary, producing low-relief anticlines and associated faulting (McLemore, 1983; Mixon et al., 1984; Schlische, 2003). Uplift and inversion of the onshore basins may have occurred from the Early Jurassic to the Miocene (Blackmer et al., 1994; Hubert et al., 1992; Mixon and Newell, 1977; Tseng et al., 1996).
The amount of post-Early Jurassic uplift is of concern when modelling each basin for hydrocarbon generation. There are two ways to estimate postrift uplift: (1) plotting of well sample depth vs sample vitrinite reflectance (Ro) and extrapolation of a trend line upward to its intersection with a hypothetical Ro paleosurface value of 0.2% (Dow, 1977), and (2) examination of the geometry of the surface expression of igneous intrusions into the synrift sedimentary accumulation.
Diabase intrusions typically change from nearvertical dike emplacement at depth to horizontal sill emplacement near the surface, where reduced overburden stresses permit lateral emplacement. Approximately 10,000 ft (3,050 m) of sediment are needed to produce a trumpet-shaped dike-sill complex. Surface flow basalts would not be extruded until after the underlying sills have been emplaced (Froelich and Gottfried, 1985). The diabase igneous activity took place in the Early Jurassic (Liassic) between 190 and 200 Ma (Sutter, 1985). With these concepts in view, and paleontologic dating of basin sediments, a generalized picture of the total basin synrift fill can be estimated. The presence of basalt flows and sills at the present-day surface suggest little erosion and potentially a deep (>10,000 ft, 3,050 m) basin. A surface predominance of dikes, on the other hand, normally indicates a more deeply eroded basin. This trend is supported by the predominance of dikes within rift basins containing only Triassic sediments, and dikes, sills, and basalt flows being present within basins containing both Jurassic and Triassic synrift strata (Froelich and Gottfried, 1985).
Each of the East Coast Mesozoic synrift basins has its own stratigraphic succession, which is a function of local tectonic activity, drainage system characteristics, and regional and seasonal climate variability. All of the onshore basins have continental to lacustrine deposits; many have intervals containing potential oil and natural gas source rocks (Table 1). Lateral facies changes may be abrupt (Gore, 1988), such that potential source beds may be stratigraphically, paleogeographically, and areally restricted within each basin. Rapid basin subsidence has accentuated depositional facies thickness and variation from basin to basin. Further, syndepositional fault compartmentalization produced has increased intrabasin stratigraphic variability.
Typically, regardless of geologic age, each basin has (or has had) one or more generally cyclic sequences of lacustrine clastic sedimentation, varying from coarse-grained feldspathic (to arkosic) quartz arenites to finely laminated organic-rich mudstones and shales, capped by coarse-grained arenites. Some basins (and probably most) have thin intervals of biochemical limestones and evaporites (e.g., El-Tabakh et al., 1997; Gore, 1986; Smoot and Olsen, 1994).
Olsen (1985) extended the concept of the “Van Houten cycle” (Van Houten, 1962, 1964) relating the stratigraphic sequence that records the transgression, highstand, regression, and lowstand facies to paleo-lake level variations as observed in the Late Triassic Lockatong Formation in the Newark basin. Each cycle can be divided into three divisions.
Division 1, which rests on Division 3 of an older cycle, is a relatively thin, platy to massive siltstone to conglomerate. Division 1 has a higher total organic carbon (TOC) content (0.5% to 1.0%) than the older Division 3 and shows few signs of desiccation. Shallow water current and subaerial exposure sedimentary structures may be present.
Division 2 is commonly a laminated black siltstone, claystone, or carbonate that shows few or no signs of desiccation and is commonly rich in organic carbon (TOC locally more than 20%). It usually contains well preserved fossil remains.
Division 3 is generally thicker than Divisions 1 and 2, and consists of massive, commonly red mudstones and shales, grading to a coarsening upward sequence of sandy, inclined beds showing current bedding and deceleration flow structures. Desiccation structures are pervasive. Root and burrow zones and reptile footprints are common.
Postrift basin fault block tilting in the basins commonly exposes fault boundary debris-flow fanglomerates; nearshore lacustrine sandstones and siltstones; lacustrine mudstones, shales, and carbonates; and fluvial complex sediments from the opposing (i.e., non-faulted, ramp) basin margin. Tectonics may uplift, rotate, and bury complete basin sequences beneath syndepositional unconformities (Cornet, 1985). In general, however, the finer grained, frequently higher porosity and higher organic-content rocks are farthest from the basin margins; whereas, the coarsest rocks are adjacent to basin border faults or intrabasin horst blocks.
Provenance for most of the East Coast Mesozoic synrift basins siliciclastic facies is the metamorphic terrane of the Appalachian Piedmont province, generally composed of metamorphic ultramafic igneous and pelitic Paleozoic and Proterozoic rock. Paleozoic and Proterozoic granitic areas are relatively minor and dispersed, with concentrations only in North Carolina, South Carolina, and Georgia (Fig. 3).
Igneous rocks are an integral part of most East Coast Mesozoic synrift basins. In almost every instance, they are described as non-porous and non-permeable, intruded diabase, and only a few examples of extrusive basalt or andesite (Chowns and Williams, 1983). As such, a discussion of the genesis and distribution of these rocks as part of the overall petroleum system evolution is not presented here.
In most of the basins examined, potential source rocks occur near the base of the synrift sequence. Potential conventional reservoir rocks are below, interbedded with, and overlying the source interval. Where present, source rocks contain a current level of organic richness suggesting that they were likely highly effective zones of hydrocarbon production at the time of peak generation. Conventional reservoir rocks composed of sandstone, conglomerate, and siltstone typically have low porosity and permeability values, although streaks of higher porosity and permeability are recorded in most basins where such data have been collected.
Unconventional (continuous) reservoirs in the form of coal beds and organic-rich shales are present as self-sourcing reservoirs in many basins. Sealing strata, primarily mudstones and shales, are present throughout much of the section. Igneous intrusive or extrusive bodies may also be local seals if impermeable or possibly reservoirs where fractured. Evaporite beds are present only locally in some basins and are usually thin and discontinuous.
Well and seismic data are generally too sparse to delineate potential conventional traps, but the data that are available suggest that geometries that define both structural and combination structural and stratigraphic traps are present. In many exposed basins, surface geologic mapping also indicates that potential structural, stratigraphic, and combination traps may be present.
Thermal maturity measurements and modeling indicate that probably all of the rift basins have undergone at least one period of petroleum generation, expulsion, and migration primarily due to rapid subsidence and penecontemporaneous basin filling with fluvial-deltaic and fluvial-lacustrine strata. Thermal halos around intrusions generally are narrow and show little regional impact to source rock maturation.
Present-day overburden thickness varies substantially from near zero to several thousand feet and is usually not sufficient to have produced the thermal maturation state of the synrift section. Approximately 1,000 ft to 10,000 ft (305 m to 3050 m) of late synrift and possibly some postrift section were deposited and eroded by the beginning of the Middle Jurassic. The unconformity at the current synrift/postrift boundary is typically of 43 million years duration (Early Jurassic Pliensbachian to Early Cretaceous Valanginian) and greater where the basins are currently exposed to modern day erosion. The preservation potential of any synrift-generated petroleum is, therefore, considered the greatest petroleum system risk for the basins (Figs. 4A to 4I).
Oil and Natural Gas Exploration Overview
The Mesozoic synrift basins of the eastern United States have been the object of oil, gas, coal, uranium, and other mineral exploration since man first found indications of their occurrences there. Sometimes, one man's prospect was another man's nuisance. Mesozoic synrift coals were mined from 1701 until 1923. The underground mines were closed, usually not because the coals were exhausted, but instead, because of deadly mine explosions. Because of the gaseous nature of the coals, many lives were lost, and several mines were abandoned due to persistent fires, some burning more than 25 years (Roberts, 1950; Woodworth, 1902).
Reports of oil sands in coal drill holes date to 1878. Coal field extension drilling in 1931 found oiland gas-bearing zones between 1,200 ft (366 m) and 1,725 ft (526 m) in the Richmond basin of Virginia. Infrequent wildcat drilling on the Atlantic Coastal Plain discovered the presence of buried Mesozoic synrift basins from Maryland to Georgia. Some wells found oil and gas shows, but commercial production was never established.
In the 1970s, Chevron led all companies in the recurring Mesozoic synrift basin play by drilling deep wells to “basement” (typically prerift Paleozoic igneous or metamorphic units) in North Carolina and Georgia. Again, oil and gas shows were found; however, no completions were made. Southeastern Exploration Company (SEPCO) in conjunction with silent partners CNG and Texaco followed Chevron and began leasing drilling rights, acquiring two-dimensional (2D) reflection seismic data, and drilling in most of the southern and central basins. SEPCO operated their wells under a variety of names in various states (Shore, Seaboard, Essex). In 1980, using U. S. Department of Energy (DOE) funding, Merrill Natural Resources drilled several coal degasification wells in the Richmond basin. These wells found oil shows, which adversely affected their hopes for establishing coal-bed methane (CBM) production in the basin. No completions were made at this time, because of the tie-in with Federal funds and stipulations surrounding their use.
Exploration continued as SEPCO (Essex) plugged and abandoned a 12,750-ft (3,890 m) South Carolina well following mechanical drilling failure in 1984. Shore plugged and abandoned a 4,570-ft (1,390 m) basement test well in the Richmond basin (Virginia) after successful drilling operations in 1985. Also that year, North Central Oil, a company unaffiliated with SEPCO, drilled and plugged a 10,500-ft (3,200 m) Newark basin (Pennsylvania) test well (Table 2).
In 1985 and 1986, Texaco and Exxon completed a six-well, 5,500-ft (1,676 m) continuous core hole program and a three-well conventional new-field wildcat (NFW) drilling program in the Taylorsville Basin of eastern Virginia, although Exxon did not participate in the last two NFWs. These companies also conducted 2D reflection seismic programs in Maryland, Virginia, and Georgia. Texaco similarly acquired 2D seismic data in South Carolina, Pennsylvania, and Connecticut. A speculative group 2D seismic acquisition program was completed in Chesapeake Bay, the James River, and Potomac River estuaries. Ziegler (1983) summarized the state of exploration for the East Coast Mesozoic synrift basin in the early 1980s.
During this period of high exploration activity in the eastern United States, the most touted Triassic–Jurassic age rift basin discovery along the conjugate margin of North Africa occurred in 1981 at the Onarep No. 101 Meskala well in the Essaouira basin of coastal Morocco. This well flowed at rates of 6 million cubic feet of gas per day (MMCFD) (170 thousand m3 per day) plus 510 barrels of condensate per day (BCPD) (81 m3 per day) for its initial potential test on ¼-inch (0.6 cm) choke, at flowing tubing pressure (FTP) of 4,800 pounds per square inch (PSI; 337.6 Kg/cm) from Triassic sandstone at 11,020 to 11,129 ft (3,359 m to 3,392 m) depth. Shut-in bottom hole pressures were approximately 9,000 PSI (632.9 Kg/cm) (Anonymous, 1986). However, following the enthusiasm of the discovery, nine subsequent wells were drilled at Meskala with overall disappointing results that downgraded the estimated reserves from 300 BCFG to 35 BCFG (8.5 to 1.0 m3), and from 20 million barrels of liquid (MMBbl) to 1 MMBbl (3.2 to 0.2 million m3) condensate (Joint UNDP/World Bank (ESMAP), 1986).
Exposed Onshore East Coast Mesozoic Synrift Basin Review and Evaluation
Hartford basin (Connecticut and Massachusetts)
The Hartford basin is the northernmost exposed Mesozoic synrift basin in the Atlantic coastal states of the United States. The geology of the Hartford basin is summarized in Hubert et al. (1992) and Schlische and Olsen (1989). The Newark Supergroup sedimentary rocks of the Connecticut Valley of New England are preserved in the Deerfield (Massachusetts) and Hartford subbasins (Connecticut) (Basin 1, Table 1; Fig. 1). For this discussion the term “Hartford basin” includes both the Deerfield and Hartford subbasins.
No oil and gas exploration wells have been drilled in the Hartford basin, and the 2D reflection seismic lines which were acquired there in the 1980s are not available to the public (Drzewiecki, 2011). Using seismic refraction data, Wenk (1984) estimates that the basin is greater than 16,000 ft (4,877 m) at its deepest. Phillips (1988) calculates a depth to Paleozoic prerift basement of approximately 16,400 ft (5,000 m) using aeromagnetic and gravity data. LeTourneau (1985) generally supports this figure, estimating that between 18,000 ft (5,486 m) and 23,000 ft (7,010 m) of synrift section may be present. Kent and Olsen (1988) also estimate as much as 23,000 ft (7,010 m) of section may be present. The Hartford basin is an asymmetric graben, in whhich the major normal fault is on its eastern border (Hubert et al., 1992). It is approximately 101 mi (162 km) long and 22 mi (35 km) wide at its widest (Fig. 1). A continuation southward beneath Long Island (New York) area is speculative (de Laguna and Brashears, 1948; Hubert et al., 1978; Hutchinson et al., 1986; Wheeler, 1938).
The stratigraphy of the Hartford basin is composed of the basal New Haven Arkose (Triassic), consisting of approximately 6,600 ft (2,012 m) of mostly red fluvial arkosic sandstone, mudstone, and conglomerate. It is overlain by approximately 1,000 ft (305 m) of interbedded Jurassic basalt (Hampden, Holyoke, and Talcott basalts) and lacustrine, playa, fluvial, and alluvial fan sedimentary units of the East Berlin and Shuttle Meadow formations, and at least 6,600 ft (2,012 m) of Jurassic Portland Formation, which consists of playa, fluvial, and alluvial-fan red beds and lacustrine gray and black strata that cap the underlying formations (Hubert et al., 1992; Kent and Olsen, 1988; Zerezghi, 2007). Potential source rock shales are found in the East Berlin, Shuttle Meadow, and lower Portland formations (Kruge et al., 1990a, b). These units range in age from Late Triassic (Rhaetian) to Early Jurassic (Sinemurian to possibly Pliensbachian) (Fig. 5; Cornet and Traverse, 1975; Kozur and Weems, 2010). Coalified logs, but no coal beds, have been reported from the top of the New Haven Arkose (Robbins et al., 1988).
The Hartford basin has good to excellent source rocks (up to 3.5% present-day TOC in shales and higher values for coal and bitumen samples) at immature to peak gas stages of maturity (average calculated percent vitrinite reflectance (%Ro) between 0.52 and 1.6) (Pratt and Burruss, 1988; Pratt et al., 1986). Analytical data (Kotra et al., 1988; Pratt et al., 1986; Spiker et al., 1988) indicate that these source rocks contain a mixture of Type I/II (oil and gas prone) and Type III (gas prone) kerogen (Coleman, 2015, this volume). Dickneider et al. (2003) illustrate total ion chromatograms of various fractions from Soxhlet extraction from the Shuttle Meadow and Portland formations. These data indicate the total source rock composition is likely a stratified succession of preserved organic material deposited by allocthonous shallow water (i.e., turbidity currents) and autochtonous deep water (i.e., pelagic deposition). Source rocks extend throughout the eastern half of the basin but are thickest closest to the eastern boundary fault (Olsen et al., 2005; Pratt et al., 1986).
No subsurface reservoir data are available to characterize potential hydrocarbon reservoir facies; however, examination of surface samples indicates that porosity varies from 0 to 21%. Measured gas and liquid permeability values range from 5 to 78 millidarcies (mD). (Huber and LeTourneau, 2006; Wolela and Gierlowski-Kordesch, 2007). Groundwater studies have shown that the synrift rocks of the Hartford basin are capable of producing on pump high volumes of water (8 to 235 gallons per minute [gpm] = 274 to 8,057 oil field barrels per day = 890 liters/minute) from wells 20 to 460 ft (6 to 140 m) deep (Randall, 1964).
Individual potential sealing units within the Hartford basin have not been identified. However, intraformational shale and mudstone beds and nonporous and nonpermeable igneous bodies may provide conventional sealing intervals.
Much of the Hartford basin is buried by Pleistocene glacial deposits, but sufficient natural and manmade exposures reveal Mesozoic synrift strata. These exposures indicate that no pre-Pleistocene formations overlie the synrift strata. The basin is surrounded by Paleozoic and Proterozoic rocks, so estimating the potential postrift overburden thickness that has been removed is difficult. As there is no well-based thermal maturity data, a depth profile cannot be constructed and interpreted to estimate an amount of postrift erosion of the basin. Pratt et al. (1986) used Ro measurements of surface samples and a reconstructed synrift stratigraphic thickness to estimate that no more than 5,000 ft (1,524 m) of section had been eroded from the upper part of the preserved synrift section. Conversely, Roden-Tice and Wintsch (2002) have used apatite and zircon fission-track analyses to indicate that approximately 16,500 ft (5,029 m) of sediments were deposited in and/or over the Hartford basin during the Jurassic and possibly Early Cretaceous and then eroded. They recognize that some of this apparently anomalously thick syn- and postrift sedimentation and subsequent uplift may have a postrift component of Cenomanian to Miocene sedimentation and post-Miocene erosion contributing to the significant difference between interpretations from the two data sets.
Surface mapping suggests that four-way and three-way structures within the basin are rare (Wise, 1992). However, pore-throat traps (Vincelette et al., 1999) are evident from the microscopic accumulations of hydrocarbons as described from thin sections (Pratt and Burruss, 1988; Zerezghi, 2007).
Rapid subsidence during the Triassic and Early Jurassic coupled with Jurassic igneous intrusions and hydrothermal fluid-flow caused the source rocks within the basin to quickly reach conditions for hydrocarbon generation and expulsion during the Early Jurassic (Hubert et al., 1992). The critical moment probably occurred during the Early and Middle Jurassic (Fig. 4A).
Rifting probably ceased by the Middle Jurassic, after which uplift began, likely along east-dipping Paleozoic orogenic fault zones reactivated as oblique strike-slip fault zones during the Late Jurassic and Cretaceous (see summary by de Boer and Clifford, 1988) that set up conditions for erosion and exhumation of the basin. Faulting and differential crustal uplift in the area during the Late Cretaceous and Tertiary rotated fault blocks, including those within the Hartford basin (Roden-Tice and Wintsch, 2002).
The USGS did not quantitatively assess the oil and gas resource potential of the Hartford basin during the 2011 assessment, but did recognize that the basin had a composite continuous gas petroleum system1 (Milici et al., 2012). Coleman (2015, this volume) concluded that while the basin contains evidence of generated and migrated liquid hydrocarbons, its potential for unconventional tight gas production is unknown.
Pomperaug basin (Connecticut)
The geology of the Pomperaug basin is discussed in detail in Hobbs (1901) and Huber and LeTourneau (2006). It is one of the smallest exposed Mesozoic synrift basins in the eastern United States, being approximately 8 mi (13 km) long and 3 mi (5 km) wide (Basin 2, Table 1, Fig. 1). Though veneered by Pleistocene glacial deposits, synrift strata are sufficiently exposed to permit detailed mapping. The basin is approximately 1,000 ft (305 m) deep and contains approximately 1,600 ft (488 m) of preserved Mesozoic synrift sedimentary rocks. The main border fault is on the east side of the basin (Burton, 2006).
Combustible black shales were known from the early history of the area and prompted some Welsh miners to attempt mining the shales in the Pomperaug basin as coal in 1831. Explosive gas and water influx ultimately caused mining to cease. In 1888, an exploratory oil well was drilled 20 ft (6 m) from the old mine shaft to a depth of approximately 1,525 ft (465 m). The well encountered bituminous shale and some petroliferous odors in the Mesozoic in addition to two basalt sheets before reaching total depth in gold- and silver-bearing basement rock at approximately 1,246 ft (380 m) (Hovey, 1890; Huber and LeTourneau, 2006). Attempts to complete the well as a producer ended in mechanical failure (Hovey, 1890).
The stratigraphy of the Pomperaug basin is composed mostly of fluvial sandstone and conglomerate (South Britain Conglomerate, approximately 890 ft (271 m) thick) capped by a relatively thin basalt (East Hill basalt, approximately 33 ft (10 m) thick) at/near the Triassic–Jurassic boundary. Overlying the East Hill basalt is the fluvial–lacustrine and aeolian Cass Formation (approximately 115 ft (35 m) thick). The Cass Formation contains red and green silty shale and siltstone overlain by dark gray, silty shale and bitumenrich, laminated limestone. It, in turn, is overlain by the 210-ft (64-m) thick Orenaug basalt, which is succeeded by the fluvial-lacustrine White Oaks Formation (approximately 100 ft (30 m) thick). The White Oaks Formation is poorly exposed, but is thought to consist of organic- and carbonate-rich laminated black shale. Potential source rocks are reported from the Cass Brook and White Oaks formations (Huber and LeTourneau, 2006).
No other petroleum system data are available on the basin. The USGS did not assess the Pomperaug basin in 2011 because of its size and lack of data.
Newark basin (Pennsylvania and New Jersey)
The geology of the Newark basin is discussed in detail by Faill (2003), Froelich and Robinson (1988), Glaeser (1966), Herman and Serfes (2010), Lyttle and Epstein (1987), Olsen et al. (1989, 1996), Schlische (1992), Schlische and Olsen (1988), and Withjack et al. (2013) among others. For this discussion, the Newark basin is limited to the extent of Triassic and Jurassic sedimentary and igneous rocks of the Newark Supergroup that extend from southeastern Pennsylvania through central New Jersey into southernmost New York.
Schlische and Olsen (1988) originally divided the Newark basin into three subbasins: the New Jersey, the Delaware River, and the Pennsylvania subbasins based on the style and density of intrabasinal faulting along with the nature of the preserved sedimentary strata. The Newark basin as used here is also equivalent to the Newark remnant of the Birdsboro basin of Faill (2003). The Gettysburg basin is discussed below, and the so-called “Narrow Neck” basin, which connects the Newark basin to the Gettysburg basin, was not examined for petroleum potential because of its small sedimentary rock volume.
The Newark basin is a complex half-graben basin, in which the major basin bounding fault is on the northwest side (Basin 3, Table 1, Fig. 1). It is cut by several generally sinistral oblique slip faults forming three principle sets: west, northwest, and northeast. Foreland-type folds have developed along these fault trends (Lucas et al., 1988). The Newark basin extends northeastward for approximately 140 mi and is approximately 30 mi wide at its widest extent. Down-plunge projections, 2D seismic profiles, and potential field data suggest that the basin may be at least 26,000 ft (7,925 m) deep (Drake et al., 1996).
The New Jersey subbasin, which occupies the northeastern third of the Newark basin, contains the thickest preserved section of Jurassic sedimentary rocks. Extension is concentrated along the border fault creating the largest synrift sedimentary volume of all three subbasins, including the thickest section of Jurassic synrift sedimentary and igneous rocks. The middle portion of the basin, the Delaware River subbasin, has a more shallow-dipping border fault system and extension is distributed along the border fault and two internal, basin-dipping faults. The southern Pennsylvania subbasin has formed through normal faulting along a very shallow-dipping border fault system and an internal system of minor normal faults (Schlische and Olsen, 1988).
Prior to 1985, only two oil and gas new field wildcat wells were drilled in the basin. In 1891, the Eastern Oil Company No. 1 Stern Farm drilled to a total depth of 2,100 ft (640 m), and there were reports of two live oil shows between depths of 940 ft (287 m) and 1,110 ft (338 m) in both sandstone and shale; and penetration of an 8-ft (2-m), anthracite grade coal seam at 1,554 ft (474 m) (Pyron, 1997). In 1948, the Van Horn No. 1 Dolak drilled through 2,380 ft (725 m) of Newark Supergroup strata but there were no reports of oil or gas (Richards, 1948).
In 1985, North Central Oil and Gas offset the No. 1 Stern Farm well by 200 ft (61 m) and drilled the No. 1 Cabot-KBI, a 10,500-ft (3,200 m) well on a seismically-defined, four-way closed anticline (Table 2; Lyttle and Epstein, 1987; Pyron, 1997). The well was drilled near the village of Ferndale between two large diabase sills (Willard et al., 1959). Malinconico (2002) showed that the well was spud in rocks of high thermal maturity (Ro=2.9%) and reached total depth still in Newark Supergroup strata in even more thermally mature synrift strata (Ro=3.4%). Gas shows were plentiful throughout the organic-rich Lockatong Formation. Sidewall core porosities within the prospective Stockton Formation sandstones (beneath the source rock Lockatong) were impressively high, considering the thermal maturity. Porosity values ranged from 12 to 20%, and permeability values were empirically calculated (not measured) at 0.1 to 4 mD. North Central ran a drill-steam test in shale in the Lockatong Formation that resulted only in a weak blow, recovering 626 ft (191 m) of drilling mud. The pressure charts indicated severe plugging of perforations. The well was plugged and abandoned.
In 1987, North Central drilled a second well, the No. 1 Parestis, a 6,712-ft (2,046-m) test, which also penetrated the Newark Supergroup section. No oil or gas shows were reported during the drilling of this well, which drilled within structural closure at pre-Triassic and Upper Triassic Lockatong horizons (Reynolds, 1994). Annotations on the wireline logs give clear indications that several zones flowed high volumes of water from fractured intervals intercepted by this well.
Core-hole drilling in 1985 and 1986 by the USGS and U. S. Army Corps of Engineers produced an abundance of rock data that have yet to be fully reported (Fedosh, 1985; Fedosch and Smoot, 1988; Ratcliffe and Burton, 1985). The Corps of Engineer's work was designed to lead to the construction of a 13.5-mi (21.7 km) flood diversion tunnel for the Passaic River across the strike of the Newark basin. Kerogen-rich shale of youngest Early Jurassic age was identified in the cores (Fedosh, 1985). The Newark Basin Coring Project (1990-1993) followed the Passaic River diversionary tunnel coring project and produced over 22,200 ft (6,767 m) of continuous core from seven locations, resulting in slightly over 17,000 ft (5,182 m) of stratigraphically continuous section through the synrift basin fill (Olsen et al., 1996). Recently, additional stratigraphic tests were drilled in the northern Newark basin to test the potential viability for carbon dioxide geologic sequestration in Newark Supergroup rocks (Slater et al., 2013).
Because of the high concentration of subsurface geologic drilling in the basin, the stratigraphy of the Newark basin is well known. The details revealed by methodical coring have allowed researchers to divide the synrift section into several members in addition to the previously defined formations (Olsen et al., 1996). In general, the Triassic Newark basin synrift fill begins with the Stockton Formation (alluvial and fluvial arkosic sandstone and conglomerate, and mudstone) followed by the Lockatong Formation (lacustrine gray and black mudstone and carbonate and minor sandstone), overlain by the Passaic Formation (equivalent to lower Brunswick Group; lacustrine sandstone, siltstone, and mudstone associated with minor fluvial and alluvial sandstone).
A succession of several Lower Jurassic formations in the upper Brunswick Group overlie the Passaic Formation and consist of interbedded basalt and lacustrine mudstone, sandstone, and conglomerate (Lyttle and Epstein, 1987; Olsen et al., 1989, 1996). Along the eastern margin of the basin, a thin interval of postrift Upper Cretaceous of the Atlantic Coastal Plain caps the rift fill (Lyttle and Epstein, 1987). The synrift formations range in age from Late Triassic (Early Carnian) to Early Jurassic (Hettangian) (Fig. 5; Kozur and Weems, 2010).
As part of a regional geochemical analysis of Newark Supergroup rocks, Pratt et al. (1986) examined potential source rock quality shales within the Newark basin. Of the 37 samples analyzed, 28 samples had values greater than 1.0% TOC. The average (avg.) of all samples was 1.8% TOC, and the average of those samples >1.0% was 2.1% TOC. Four formations were analyzed:
Triassic Lockatong Formation (n=19, min. TOC = 0.2%, avg. TOC = 1.7%, max. TOC = 3.5%);
Upper Triassic-Lower Jurassic Passaic Formation (n=4, min. TOC = 0.4%, avg. TOC = 1.5%, max. TOC = 2.3%);
Lower Jurassic Feltville Formation (n=1, TOC = 1.5%); and
Lower Jurassic Towaco Formation (n=13, min. TOC = 0.8%, avg. TOC = 2.0%, max. TOC = 3.6%).
For regional values, these numbers are indicative of potentially effective hydrocarbon source rocks, as a number of present-day TOC values are in excess of 1.0%. However, when values are compared within a specific sample locality (localities A through K of Pratt et al., 1986), the variability of TOC is also significant:
(n=7, min. TOC = 0.2, avg. TOC = 1.4, max. TOC =3.0),
(n=8, min. TOC = 0.6, avg. TOC = 1.7, max. TOC = 3.2),
(n=4, min. TOC = 0.8, avg. TOC = 2.1, max. TOC = 3.5),
(n=3, min. TOC = 0.4, avg. TOC = 1.6, max. TOC = 2.3),
(n=1, TOC = 1.2),
(n=1, TOC = 1.5),
(n=2, min. TOC =2.6, avg. TOC = 3.0, max. TOC = 3.3),
(n=2, min. TOC = 3.2, avg. TOC = 3.4, max. TOC = 3.6),
(n=1, TOC = 0.8),
(n=3, min. TOC = 0.9, avg. TOC = 2.0, max. TOC = 3.0), and
(n=5, min. TOC = 0.8, avg. TOC = 1.2, max. TOC = 1.6).
Katz et al. (1988) found somewhat similar TOC values in their study:
Lockatong Formation (n=31, min. TOC = 0.1%, avg. TOC = 1.2, max TOC = 9.1%),
Passaic Formation (n=9, min TOC = 0.1%, avg. TOC = 1.2, max. TOC = 3.6%), Feltville Formation (n=20, min. TOC = 0.3%, avg. TOC = 2.4%, max. TOC = 11.2%), and
Towaco Formation (n=6, min. TOC = 0.4%, avg. TOC = 1.1%, max. TOC = 2.2%).
Laughrey et al. (undated) reported additional outcrop sampling and organic geochemical analyses of the Newark Supergroup. This study of 10 samples in Bucks County, Pennsylvania, showed a range of TOC between 0.3% and 1.8%, and having an average of 0.8%. Specific locations or formation identities were not reported in this study.
Spiker et al. (1988) found that the source of kerogen from the Newark basin black shales was a mixture of autochthonous lacustrine algae and allochthonous woody plant detritus. The woody plant component ranged from 30% to 70% of the total kerogen examined. This assemblage highly favored the generation of natural gas over oil.
The Triassic Lockatong Formation is the most likely source interval for the largest volume of hydrocarbons in conventional reservoirs within the basin and the best candidate to be a self-sourcing unconventional reservoir. The Lockatong occurs across most of the Newark basin, but thins to the northeast (Lyttle and Epstein, 1987). Geochemical analyses of drill cutting samples from the Lockatong Formation in the North Central Cabot-KBI No. 1 well, Bucks County, Pennsylvania, range from a present-day 0.1% TOC to 0.9% TOC, with an average of 0.3% (n=114) (Laughrey et al., undated). Restoring these low TOC values using the methodology of Daly and Edman (1987) does not significantly increase them to levels that would result in their being considered viable source rocks capable of expelling hydrocarbons; i.e., those with average TOC values in excess of 1.0%. Consequently, although hydrocarbons have been generated, as proven by the mud log gas shows, modelling of the well establishes that this gas was retained in the Lockatong, resulting in an unconventional resource accumulation. If the Laughrey et al. (undated) data set for the Cabot-KBI well is considered to be purely random (i.e., ordered only by depth and a regularly spaced sample interval, with no intent to high-grade the samples) and the Pratt et al. (1986) data set is assumed to be purely biased (i.e., samples intentionally selected towards higher organic content), then the likely hydrocarbon generation potential is a mixture of the two. Nevertheless, the true petroleum potential for the Lockatong Formation as a source rock or unconventional reservoir rock appears to be restricted either geographically, stratigraphically, or both.
Within the Upper Triassic-Lower Jurassic Passaic Formation, sandstones are interbedded with relatively thin organic-rich shales, setting up potential source-reservoir relationships (Olsen et al., 1996). These potential source rocks are moderately rich (ranging from 0.4% to 2.3% TOC, averaging 1.5% TOC; Pratt et al., 1986) and thermally mature (in the peak gas to past peak gas range of 1.4% Ro to 2.8% Ro, averaging 2.3% Ro; Malinconico, 2002, 2010).
Work by Kotra et al. (1985), Malinconico (2002, 2010), and Schamel and Hubbard (1985) suggest that the Newark basin may be at a relatively advanced thermal state throughout, especially in the older, buried section. Malinconico (2010) and Pratt et al. (1986) however, report that the youngest Jurassic shales are still at immature oil levels of thermal maturity.
With respect to the Lockatong, potential conventional reservoirs may exist in the underlying Triassic Stockton Formation and overlying Passaic Formation. Even though very low porosity has been measured from outcrop samples, subsurface measurements indicate porosities from 7 to 31% in arkosic sandstones of the Stockton Formation. Permeability values range from 0.05 to 78.1 mD (Rima et al., 1962; Sloto et al., 1996).
Conventional sealing strata may be present in the shales of the Lockatong Formation that overlie the Stockton sandstone, in the shales interbedded with sandstones of the younger Passaic Formation, and possibly in Jurassic sills overlying the Passaic Formation. The effectiveness of one or more potential seals at the top of Lockatong and within the middle Passaic Formation may have been demonstrated by the initial detection of gas when the Cabot-KBI well drilled into the lower Passaic and the upper Lockatong formations (Ryder et al., 1994). Potential pore throat traps are demonstrated by the presence of small tension fractures filled with bitumen, intergrown vein-filled calcite and bitumen, and bitumen-stained porous sandstones (Pratt and Burruss, 1988).
Malinconico (2002, 2010) examined the thermal maturity of seven Newark Basin Coring Project core holes and the Cabot-KBI deep well to estimate the amount of overburden that had been removed after deposition of the Lower Jurassic. These studies concluded that less than 3,200 ft (975 m) was eroded from the northwestern portion of the basin, where Lower Jurassic strata are exposed, to as much as 20,000 ft (6,096 m) of section removed in the southern portion of the basin. Additional data from the Cabot-KBI well (Laughrey et al., undated) indicated that only about 6,500 ft (1,981 m) of synrift and postrift overburden was removed from the southern basin area (Post and Coleman, 2015, this volume). From fission track analysis, Steckler et al. (1993) estimated that a minimum of approximately 10,000 ft (3,048 m) of section was removed from the Newark basin and surrounding area. Uplift and erosion was geologically fast probably beginning during the early Middle Jurassic (~178 Ma) and continuing until the onset of Cretaceous onlap (possibly as early as 125 Ma) (Huntoon and Furlong, 1992; Malinconico, 2002, 2010).
Surface mapping suggests that four-way and three-way structural closure are present in several places within the basin (Drake et al., 1996 and supporting references). The single publically available 2D seismic line (NB-1, Costain and Çoruh, 1989) shows several areas of two directional structural dip and several areas of updip stratigraphic termination, which if present in the third dimension could provide adequate conventional hydrocarbon traps. Additionally, porethroat traps are evident from the microscopic accumulation of hydrocarbons as described from thin sections (Walters and Kotra, 1990).
Thermal maturity modeling using well and core data indicates that petroleum was generated, expelled, and migrated during the late Early Jurassic (Fig. 4B; Katz et al., 1988). This timing puts the potentially maximum basin-wide expulsion and migration of petroleum approximately at the beginning of uplift and erosion, suggesting that most (or all?) conventional traps may have been ruptured and the contents leaked and migrated. This analysis puts the critical moment at about the end of the Early Jurassic (Fig. 4B).
Preservation time becomes the key petroleum system element in understanding the oil and gas resource potential of the Newark and perhaps other eastern U.S. Mesozoic synrift basins. With the uplift and erosion of the basin beginning at about the time of peak expulsion and migration, the preservation of petroleum in conventional traps becomes highly problematic. Gas shows in the Cabot-KBI well while drilling through the Lockatong Formation indicate that some volume of natural gas remains in the basin, probably as a continuous (or unconventional) accumulation.
In summary, rifting probably ceased by Middle Jurassic, after which uplift began, likely along east-dipping Paleozoic orogenic fault zones reactivated as oblique strike-slip fault zones during the Late Jurassic and Cretaceous (Withjack et al., 2013). Transpressive faulting set up conditions for erosion and exhumation of the basin.
The USGS recognized that the Newark basin was a complex composite rift complex composed of multiple subbasins, compartmentalized by late synrift and postrift oblique slip faulting. The structure and stratigraphic character of the basin indicated that for assessment purposes it should be divided into a north Newark basin and a south Newark basin along a generally trending northwest–southeast line (Milici et al., 2012). The south Newark basin was estimated to contain undiscovered, technically recoverable mean resources of 876 BCFG (24.8 BCM) and 4 MMBNGL (0.6 MMCM) (Table 1; Milici et al., 2012) in a composite continuous gas petroleum system. The north Newark basin was not assessed because of the lack of data at the time of the assessment (Milici et al., 2012).
Gettysburg basin (Pennsylvania and Maryland)
The geology of the Gettysburg basin of Pennsylvania and Maryland has been described by Glaeser (1966) and Root (1988). For this discussion, the Gettysburg basin is limited to the extent of Triassic and Jurassic sedimentary and igneous rocks of the Newark Supergroup in southeastern Pennsylvania and north-central Maryland (Basin 4, Table 1; Fig. 1). It is equivalent to the Gettysburg remnant of Faill (2003).
The Gettysburg basin is a complex half-graben, with the major bounding fault on the northwest side. It is cut by several oblique slip faults generally trending east–west. The basin extends northeastward for 74 mi (119 km) and is 17 mi (27 km) wide at its widest extent (Fig. 1). It is approximately 18,000 ft (5,486 m) deep and contains up to 22,000 ft (6,705 m) of recorded strata (Root, 1988).
Only four petroleum exploration wells have been drilled in the basin, all without success. One well, the Leib No.1, had an unsubstantiated report of oil in 1961 (Harper, 1994, unpublished; IHS Enerdeq, 2014).
The stratigraphy of the Gettysburg basin is similar to that of the Newark basin. The Triassic New Oxford Formation (alluvial-fluvial conglomerate, sandstone, and red mudstone) lies at the base of the synrift section and is overlain by the Gettysburg Formation (alluvial, fluvial, and lacustrine red mudstone, sandstone, and conglomerate) at the top of the sequence (Root, 1988). The formations range in age from Late Triassic (Early Carnian) to Early Jurassic (Hettangian) (Fig. 5; Kozur and Weems, 2010).
Only minimally effective potential source rocks have been found within the Gettysburg basin. Root (1988) reports collecting lean, dark gray to black laminated lacustrine shales with TOC values between 0.1% to 0.9%, averaging 0.5% (n=7). De Wet et al., (1998) report TOC values of 0.5% to 2.4%, averaging 1.2% (n=14) from pedogenic and lacustrine micrites at two localities. The organic-rich mudstone and wackestone contain macroscopic plant fragments. The shales reported by Root (1988) are also highly thermally mature, having Ro values of 1.5 to 3.0%. These values are consistent with the presence of extensive igneous intrusions in the northwestern half of the basin. Coal beds have reportedly been mined in Pennsylvania within the Gettysburg basin in the mid-19th century; however, no evidence of these deposits have been found by Frazer (1880) in his report later in that century.
The only data available to characterize potential reservoir rocks are groundwater data. Approximately 70% of wells sampled by Wood and Johnston (1964) flow at rates of 10 gpm (= 342 oil field barrels per day, 38 l/min) or less. No porosity or permeability values are currently available.
Sealing strata appear to be pervasive throughout the basin, although the basin is highly fractured based on groundwater flow and surface mapping. There are no data to estimate directly the amount of overburden which has been removed.
Surface mapping indicates that structural closures are present, but rare in the basin (Root, 1988). No other evidence of traps is presently available.
Data to estimate the timing of petroleum generation, expulsion, and migration are sparse. However, the limited thermal maturity data indicate that if any generation and expulsion took place, it has occurred during synrift deposition. The lack of any discovery of potentially rich effective source rocks, suggests that no significant volume of thermogenic hydrocarbons was created. An estimate of a critical moment, if one existed, is probably sometime in the Middle Jurassic (Fig. 4C).
Rifting probably ceased by Middle Jurassic, after which uplift began, likely along southeast-dipping Paleozoic orogenic fault zones. Oblique strike-slip faulting developed and likely extended into the Cretaceous. Compression associated with this postrift faulting set up conditions for erosion and exhumation of the basin.
The USGS did not quantitatively assess the oil and gas resource potential of the Gettysburg basin, but did assign it to a composite continuous gas petroleum system (Milici et al., 2012). The lack of evidence for generated and migrated hydrocarbons suggests that this basin has a very low probability for containing any undiscovered commercial thermogenic hydrocarbon accumulations.
Culpeper basin (Maryland and Virginia)
The geology of the Culpeper basin has benn discussed by Faill (2003), Gore, (1988), Gore et al., (1989), Johnson (1999), and Lee and Froelich (1989). For this review, the bounding elements of the Culpeper basin as discussed by Lee and Froelich (1989) are used. It is equivalent to the Culpeper remnant of the Birdsboro basin of Faill (2003). The Culpeper basin is a half-graben having a western bounding fault (Basin 5, Table 1; Fig. 1). It extends north-northeastward for 112 mi (180 km) and is 12 mi (19 km) wide at its widest extent (Gore et al., 1989). Limited 2D seismic coverage suggests that the basin is approximately 6,500 ft (1,981 m) deep (Schorr, 1986).
Eight oil and gas tests are known to have been drilled in the Culpeper basin of Virginia before 1945. Four have been converted to local water wells at completion. No other data are available for these wells (LeVan, 1962). Unconfirmed oil and gas shows have been reported from water wells and asphaltic residues from outcrops along Interstate 66 near Thoroughfair Gap, Prince William County, Virginia (Froelich et al., 1982). The USGS has drilled six core holes and eight drill holes in the basin (Lee and Froelich, 1989). Several 2D seismic lines have been acquired across the basin, including two by Virginia Tech (Costain and Çoruh, 1989; Schorr, 1986).
Lee and Froelich (1989) define the synrift stratigraphy of the Culpeper basin as the Culpeper Group. The lower Culpeper Group is composed of a conglomerate and sandstone base, followed by sandstone and siltstone, which is overlain by another conglomerate and sand interval just below the Triassic–Jurassic boundary. The upper Culpeper Group contains basal basalt flows interbedded with sandstone, siltstone, and mudstone. A sandstone and conglomerate unit overlies and completes the unit. These formations range in age from Late Triassic (Early Carnian) to Early Jurassic (Hettangian) (Fig. 5; Kozur and Weems, 2010).
Source rock analyses show that organic shales with between 0.2% and 2.0% TOC are present within the basin at an immature to peak gas level of maturity (Ro values of 0.4% to 2.0%; Smith and Robison, 1988). Organic matter is dominated by algae and palynomorphs and has lesser amounts of woody material (Smith and Robison, 1988). Coalified logs have been reported from the Culpeper basin in Maryland, but no coal beds have been described (Robbins et al., 1988).
Reservoir porosity and permeability data are not available from the eight oil and gas test wells. Ground water studies indicate that porosity values range from 5% to 10%; indications from flow rates are that these values decrease with increasing depth. Fractures are essential for higher volume water production (Laczniak and Zenone, 1985). The abundance of low permeability sedimentary and igneous rocks indicates that lack of regional and local sealing lithologies is not a likely problem. However, the pervasive fracture network indicated by groundwater studies suggests that the absence of an effective petroleum seal would be of significant concern for conventional petroleum accumulations.
Without well-based thermal maturity data, there is insufficient information to estimate the amount of overburden potentially removed during postrift uplift and erosion. However, thermal content of the basin as determined from surface measurements is consistent with other onshore eastern U.S. Mesozoic synrift basin, indicating between 3,000 ft (914 m) and 7,000 ft (2,134 m) of synrift strata has been removed by basin inversion. The range of thermal maturities as reported by Smith and Robison (1988) suggests that the Culpeper basin may have had a minimal amount of overburden originally deposited on top of the presently preserved synrift strata and/or a lower heat flow than similarly sized Mesozoic synrift basins along trend.
Potential structural and combination structural-stratigraphic traps are evident on the limited 2D seismic lines (Costain and Çoruh, 1989; Schorr, 1986). Porethroat traps in some units are implied by some groundwater flow data; however the fracture network may invalidate the potential for this style of hydrocarbon traps.
The timing of peak petroleum generation and expulsion is implied to be Early to possibly early Middle Jurassic based on geohistory modeling of similar eastern onshore U.S. Mesozoic synrift basins. With this caveat, the critical moment is assumed to be early Middle Jurassic, prior to basin inversion and removal of late synrift strata due to erosion (Fig. 4D).
Rifting probably ceased by Middle Jurassic, after which uplift began, probably along southeast-dipping Paleozoic orogenic fault zones. Oblique strike-slip faulting developed and likely extended into the Cretaceous. Compression associated with this postrift faulting facilitated basin inversion, exhumation, and erosion of synrift strata.
The USGS did not quantitatively assess the oil and gas resource potential of the Culpeper basin. However, it did assign it to a composite continuous gas petroleum system, recognizing that potential source rocks are present (Milici et al., 2012). The lack of substantiated evidence for generated and migrated hydrocarbons means that this basin has a very low probability for any undiscovered commercial conventional thermogenic hydrocarbon accumulations. The potential presence of continuous hydrocarbon accumulations cannot be determined without additional drilling and testing data.
Barboursville basin (Virginia)
The Barboursville basin is a small, erosional outlier of the Culpeper basin (Basin 6, Table 1; Fig. 1). The geology of the basin is discussed by Lee and Froelich (1989). The Barboursville basin is approximately 10 mi (16 km) long and 3.5 mi (5.6 km) wide at its maximum width. Its synrift thickness varies from approximately 1,000 ft (305 m) to 2,000 ft (610 m) (Lee and Froelich, 1989).
The stratigraphy of the Barboursville basin consists of just the lower Culpeper Group of Lee and Froelich (1989). In this basin, conglomerate beds are missing, and the section consists of sandstone and siltstone. These formations range in age from Late Triassic (Late Carnian) to Late Triassic (Middle Norian) (Lee and Froelich, 1989).
Data are insufficient for characterizing the various potential petroleum system elements in the Barboursville basin. The Barboursville basin has not been assessed by the USGS because of its relatively small size.
Scottsville basin (Virginia)
The Scottsville basin is a small Mesozoic synrift basin in line with and southwest of the Culpeper basin (Basin 7, Table 1; Fig. 1). The geology of the Scottsville basin has been discussed by Johnson et al. (1985a). The Scottsville basin is approximately 21 mi (34 km) long and 3.5 mi (5.6 km) at its maximum width. It has a northwestern basin bounding fault zone and is approximately 5,250 ft deep (1,600 m) (Johnson et al., 1985a; Costain and Çoruh, 1989).
The stratigraphy of the Scottsville basin is not well described. However, Johnson et al. (1985a) indicate the synrift fill is composed of sandstone, siltstone, shale, and conglomerate along with a few igneous dike intrusions.
Data are insufficient to identify or characterize the various potential petroleum system elements in the Scottsville basin. The Scottsville basin has not been assessed by the USGS because of its relatively small size.
Dan River-Danville basin (Virginia and North Carolina)
The geology of the Dan River-Danville basin is summarized in Johnson et al. (1985a); Meyertons (1963); Olsen et al. (2015); Reid (2015, this volume); and Thayer (1970a). This basin is a large, narrow half-graben that extends northwesterly for 110 mi (177 km) and is approximately 8 mi (13 km) wide at its widest extent (Basin 8, Table 1; Fig. 1; Johnson et al., 1985a). It is defined by the outcrop extent of the Mesozoic synrift strata exposed in the Dan River basin of North Carolina and the adjoining Danville basin of Virginia. It has a northwestern major basin-bounding fault and may be as deep as 5,000 ft (1,524 m) (Reid, 2015, this volume).
“Coal” (or more appropriately, black carbonaceous, sulfide-rich shale) was mined locally from the basin during the American Civil War but was never a major commercial deposit because it was too thin, rather impure, and quite limited in extent (Stone, 1910). The Dan River-Danville basin has not experienced any oil and gas drilling activity. On the other hand, three core holes have been drilled by industry in the North Carolina portion, but no oil and gas test wells have been drilled in Virginia.
From the oldest to the youngest, the synrift stratigraphy of the Dan River-Danville basin is composed of a basal coarse-grained sandstone and mudstone (Pine Hall Formation, 240 ft (73 m) to 6,300 ft (1,920 m) thick), an overlying cyclical sequence of mudstone, sandstone, thin coal beds, and minor conglomerates (Walnut Cove Formation, approximately 400 ft (122 m) to greater than 600 ft (183 m) thick), followed by the approximately several hundred ft (m) of interbedded sandstone and mudstone (Dry Fork Formation), cyclical gray and black microlaminated mudstone, and minor sandstone and conglomerate (Cow Branch Formation, approximately greater than or equal to 1,500 ft (457 m) thick), overlain by a thick interval of predominately red sandstone (Stoneville Formation, approximately greater than or equal to 1,500 ft (457 m) thick) (Olsen et al., 1991; Reid, 2015, this volume; Thayer and Robbins, 1994). Source rocks are present in the Walnut Cove and Cow Branch formations (Olsen et al., 2015; Reid, 2015, this volume). These formations range in age from Late Triassic Early Carnian to Late Norian (Fig. 5; Kozur and Weems, 2010).
Source rocks in the Walnut Cove Formation range in organic carbon content from 0.07% TOC to 47.6% TOC, averaging 3.6% TOC. The Cow Branch Formation has source rocks having TOC content ranging from 0.08% to 3.8%, averaging 1.4% TOC. Thermal maturity determined from vitrinite reflectance (Reid, 2015, this volume; Reid, 2015, written communication) shows an average of 1.9% Ro for the Walnut Cove (min. = 1.23%, max. = 3.02%) and 2.08% Ro for the Cow Branch (min. = 1.79%, max. = 2.27%). Stone (1910) reports the coal rank of the basin is semi-anthracite, which equates to a maximum Ro of about 2.25%. Robbins (1982) reports an Ro of 2.15% from a semianthracite coal. Because there are no depth-thermal maturity values to estimate an amount of synrift overburden that was subsequently eroded, a confident estimate of the original thickness of this overburden is not possible at the time of this writing. On the other hand, the Ro values from surface samples and other observations suggest that the current Dan River-Danville basin synrift section either was originally buried to a depth of several thousand feet or experienced locally high heat flow during and shortly after rifting or perhaps a combination of the two (Reid, 2015, this volume; Robbins, 1982).
Thayer et al. (1982) report porosity values from unspecified sandstone samples ranging from 0.3% to 18.1%, averaging 5.7%. Horizontal permeability ranges from <0.01 mD to 14.1 mD, averaging 0.14 mD. Recent studies on shale beds within the basin by Reid et al. (2015) found the mean porosity of the Walnut Cove Formation is 2.7%, with a range of 1.6% to 4.6%. The calculated average permeability is 4.32 x10-5 mD, and the range is 1.06x10-5 mD to 1.47x10-4 mD. For the Cow Branch Formation., the mean porosity is 1.03%, with a range of 0.5% to 2.3%. The calculated average permeability is 3.63 x10-6 mD, and the range is 1.16x10-6 mD to 2.16x10-6 mD (Reid et al., 2015). The possible contribution of a widespread northwest-trending fracture network to net effective permeability has not been considered in these values.
Lithologic units in the Walnut Cove and Cow Branch formations could act as conventional seals, if conventional hydrocarbon accumulations are present. The possible effect of the regional fracture network on the sealing potential has not been evaluated.
Data are insufficient to determine if conventional or unconventional traps are present within the basin.
Rapid subsidence throughout the Late Triassic likely has caused the source rocks within the basin to reach conditions for hydrocarbon generation and expulsion during the latest Triassic. Based on limited data, the critical moment is estimated to be at approximately the Triassic–Jurassic boundary (Fig. 4E).
Detailed study within the basin illustrates some details of inversion not evident in other east coast Mesozoic synrift basin. From quarry exposures, Ackermann et al., (2003) reports a large (length [L] ≈ 50 ft, [15 m]) normal fault that was reactivated as a reverse fault and smaller normal faults (~ 1 in [2 cm] < L < ~ 4 in [10 cm]) that had anomalously high displacements for their lengths. This compression and resulting shortening may have been localized along décollement surfaces that developed in organic-rich black shales.
Based on the distribution and richness of potential source rocks in the basin, the USGS assessed the Dan River–Danville basin as a continuous (unconventional) tight gas basin having an undiscovered, technically recoverable natural gas resource potential at mean values of 49 BCFG (1.4 billion m3) and no natural gas liquids (Milici et al. 2012; Reid, 2015, this volume).
Davie County basin (North Carolina)
The Davie County basin is a small basin that lies southwest of the Dan River-Danville basin in Davie County, North Carolina (Basin 9, Table 1; Fig. 1; Taylor, 1982; Thayer, 1970b). It extends approximately 7.5 mi (12 km) northwestward and 3.7 mi (5.9 km) in width (Olsen et al., 1991). Based on geologic mapping, there are two main basin-bounding faults, one on the east-southeast side of the basin and the other on the west-northwest side of the basin (Taylor, 1982; Thayer, 1970b). Surface mapping indicates that the synrift fill is composed of typical Newark Supergroup lithologies: polymict conglomerate and arkosic sandstone overlain by siltstone and mudstone of alluvial fan, fluvial, and lacustrine origin (Olsen et al., 1991). A preliminary age of Late Triassic has been assigned based on palynology (Litwin, 1989, oral communication cited in Olsen et al., 1991). Because of its relatively small size, the USGS did not assess its petroleum potential.
Farmville basin (Virginia)
The Farmville basin is one of a series of small rift basins that lie between the Culpeper-Scottsville-Danville-Dan River basin trend and the Richmond-Taylorsville basin trend (Basin 10, Table 1; Fig. 1). The geology of the Farmville basin is discussed by Wilkes (1982, 1986). It has a major western basin-boundary fault. An intrabasin, high-angle reverse fault, characteristic of wrench fault systems, is indicated by Wilkes (1982) in the western third of the basin. The basin is 16.6 mi (26.7 km) long and 5 mi (8 km) wide (Fig. 1). The basin is estimated to be approximately 3,000 ft (914 m) deep (Wilkes, 1982).
Organic-rich shale and thick (3 ft to 4 ft; 0.9 to 1.2 m) coal seams are present along the eastern margins. The coal seams have been mined at various levels of industrial and domestic activity from the 1830s to the 1980s. Two oil and gas exploratory wells were drilled in the basin, one of which had a report of an oil and gas show at a depth of 938 ft (286 m) in 1917 before the well was abandoned at 1,518 ft (463 m). The second well, drilled at the eastern basin margin in 1925, did not have any reported shows (Wilkes, 1982).
Robbins et al. (1988) describe the thermal maturity of the palynomorphs recovered from the coal-bearing strata as having equivalent Ro values of 0.9% to 1.1%. Other than these observations, specific data are insufficient to identify or characterize the various potential petroleum system elements in the Farmville basin. The Farmville basin has not been assessed by the USGS because of its relatively small size.
Other small exposed Virginia Mesozoic synrift basins of central Virginia
Four additional, small exposed Mesozoic synrift basins extend southwest of the Farmville basin (Basins 11-14, Table 1, Fig. 1): Briery Creek, Roanoke Creek, Randolph, and Scottsburg basins. A composite stratigraphic section of synrift strata from the Farmville, Briery Creek, and Scottsburg basins extends in age from Late Triassic Early to Middle Carnian (Fig. 5; Kozur and Weems, 2010). Typical of these small basins, data are insufficient for characterizing the various potential petroleum system elements in these basins. Limited data indicate that these basins have been structurally uplifted since synrift deposition (Ramsey, 1986). These basins were not assessed by the USGS because of their relatively small size.
Taylorsville basin (Maryland and Virginia)
The geology of the Taylorsville basin is summarized in Goodwin et al. (1985), LeTourneau (2003), Malinconico (2003, 2008), Milici and Coleman (2015b, this volume), Milici et al. (1991), and Weems (1980). It is a very large Mesozoic synrift basin that is mostly buried by onlapping Atlantic Coastal Plain sedimentary strata. It is defined by potential field geophysics, limited 2D seismic, and at least 29 wells or core holes drilled into Newark Supergroup rocks (Benson, 1992; Hansen and Edwards, 1986; LeTourneau, 2003; Milici et al., 1991; Wilkes et al., 1989). It is approximately 155 mi (249 km) long and 27 mi (43 km) wide (Basin 15, Table 1; Fig. 1). The basin varies in depth from approximately 8,000 ft (2,438 m) in the north to at least 10,000 ft (3,048 m) in its central area. Data are insufficient to estimate a depth in the southern parts of the basin, where the deepest well penetrated only about 220 ft (67 m) of Newark Supergroup rocks (Jacobeen, 1972; LeTourneau, 2003; LeVan, 1962). Its major basin-bounding fault is on the west side of the basin.
The Taylorsville basin was first explored for gas storage potential in Maryland, where 15 2D seismic profiles were acquired and five wells were drilled into Newark Supergroup rocks. The deepest penetration of this drilling program was 234 ft (71 m) of synrift strata. No oil or gas shows were reported (Hansen and Edwards, 1986; Jacobeen, 1972; Teifke, 1973). Prior to 1985, very few wells had been drilled in the Virginia portion of the basin. However, between 1985 and 1992 the Taylorsville basin was the focus of significant 2D seismic acquisition, exploration coring, and new field wildcat drilling by Texaco and Exxon. This partnership resulted in five deep continuous core holes, followed by three conventional exploratory wells. Six wells or core holes had reports of oil and/or gas shows (Milici and Coleman, 2015b, this volume; Tseng et al., 1996). Based on these efforts, the central Taylorsville basin is one of the most thoroughly explored areas in the East Coast Mesozoic synrift basins (Table 2), yet no commercial production has been established to date.
The deepest well, the Texaco No. 1 Wilkins et ux, reached the top of the synrift section at 1,600 ft (488 m) measured depth and Piedmont Paleozoic(?) crystalline basement at approximately 10,000 ft (3,048 m) before reaching total measured depth of 10,135 ft (3089 m). The Texaco No. 1 Roberts et al. also penetrated basement rocks at a depth of approximately 3,405 ft (1,038 m) near the eastern edge of the basin after penetrating approximately 1,450 ft (1,326 m) of synrift strata. Approximately 15,000 ft (4,572 m) of Newark Supergroup synrift strata are estimated to be present within the basin (LeTourneau, 2003).
The synrift stratigraphy of the Taylorsville basin is composed of the basal South Anna Formation (approximately 325 ft (99 m) of primarily fluvial sandstone) overlain by the Falling Creek Formation (approximately 2,200 ft (671 m) of sandstone, siltstone, and minor amounts of conglomerate, shale, and limestone) and the Newfound Formation (500 to 2,000 ft (152 to 610 m) of fluvial sandstone, siltstone, and conglomerate). These formations are equated to the Doswell Formation from outcrop studies (Weems, 1980). Overlying the Newfound Formation (Doswell Formation) is the Port Royal Formation (lacustrine gray and black mudstone and sandstone; approximately 1,300 to 2,600 ft (396 to 793 m) thick) and the Leed-stown Formation (red and gray fluvial mudstone and sandstone; approximately 6,500 ft (1,981 m) thick). Potential source rocks are in the shale and coal of the Falling Creek Formation (LeTourneau, 2003; Milici and Coleman, 2015b, this volume). The synrift section ranges in age from Late Triassic Early Carnian to Middle(?) Norian (Kozur and Weems, 2010).
Public domain organic richness data for potential source rocks in the Taylorsville basin are sparse. Pratt et al. (1985) suggests that the Falling Creek Formation may be an effective source interval, having kerogen characteristics interpreted to be potentially similar to those of younger Culpeper and Newark basin source rocks. LeTourneau (2003) discusses the presence of “dark lacustrine shale” and coal beds in the Port Royal and Falling Creek formations from several wells and core holes without comment as to their potential source rock effectiveness. Potential source rocks are hypothesized from high gamma ray wireline log response in dark gray and black shales in the Newfound and Falling Creek formations (Milici and Coleman, 2015b, this volume).
Robbins et al. (1988) found that the coals in the exposed southwestern portion of the basin had equivalent Ro values of 1.1% to 1.3%. Malinconico (2003) found that Ro values from the central part of the basin ranged from 0.4% (at 2,260 ft (689 m)) to 2.9% (at 10,010 ft (3,051 m)) or a range in thermal maturity from immature oil to past peak gas. In the northern portion of the basin in Maryland, one of the gas storage test wells penetrated coal and coal streaks within the synrift fill; however, no organic richness or thermal maturity values were available (Robbins et al., 1988).
Potential conventional reservoirs have very low wireline uncorrected log-based porosity (near zero) to isolated high values up to 30% in sandstone. The average porosity for the No. 1 Thorn Hill well is approximately 5% (Milici and Coleman, 2015b, this volume; Onstott et al., 1998).
The presence of low porosity and permeability rocks throughout the synrift section suggests that local sealing horizons are present. However, the lack of an upper level low permeability shale, impermeable igneous body, or evaporite suggests that a regional basin-wide seal may be lacking.
Limited 2D seismic profiles or interpretations thereof (e.g., Tseng and Onstott, 1997) suggest that four-way and three-way structural closures are present within the basin creating potential conventional structural and combination traps. The presence of both oil and gas shows recorded while drilling, in conjunction with the overall low porosity and permeability indicate that pore-throat traps are also likely.
With the exception of a small area of exposure in the southwest corner of the basin, the Taylorsville basin is overlain by approximately 1,600 to 1,900 ft (488 to 579 m) of Cretaceous and Tertiary strata of the Atlantic Coastal Plain (Milici and Coleman, 2015b, this volume). Lower(?) Cretaceous sands, silts, and clays of the Patuxent Formation lie at the synrift–postrift unconformity. From vitrinite analysis of the Mesozoic synrift section, Malinconico (2003) concludes that erosion of synrift strata is highly variable over the central and deepest part of the basin. Calculated erosion values range from approximately 680 ft (207 m) (lowest minimum value) to 10,900 ft (3,322 m) (highest maximum value), with an average of approximately 4,900 ft (1,493 m). Malinconico (2003) uses a value of 8,200 ft (2,499 m) for a basin maximum.
Rapid subsidence during the Triassic and possibly Early Jurassic likely caused the source rocks within the basin to reach conditions for hydrocarbon generation and expulsion at about the time of the Triassic–Jurassic boundary. Tseng et al. (1996, 1999) concluded that thermal maximum and by extension maximum burial occurred at about 200 Ma (in the Jurassic Hettangian). Malinconico (2003) suggested this event occurred slightly earlier at the Triassic–Jurassic boundary. Hence, the critical moment was estimated to be in the Early Jurassic (Fig. 4F). Tseng et al. (1999) interpreted uplift to extend from the Early Jurassic into the Cretaceous Campanian, when renewed subsidence began. Early Miocene uplift reversed this trend.
Throughout much of the Atlantic Coastal Plain of Maryland and Virginia, the Cretaceous and Tertiary strata are reverse-faulted in zones which appear to be a continuation of the basin-bounding faults of the Richmond and Taylorsville basins (Mixon and Newell, 1977). During the Jurassic to Late Cretaceous uplift, the Taylorsville basin has undergone differential uplift and topographically driven fluid flow (Tseng et al., 1999). As Jurassic uplift and deformation progressed, petroleum and other formation fluids are interpreted to have migrated through fracture zones in the basin, leading to water washing of any potentially previously trapped oil (Tseng et al., 1999).
If one assumes that the key exploration wells drilled in the central portion of this basin were drilled in seismically defined locations conducive to the discovery of hydrocarbons, then the failure to have a commercial discovery would be a significant factor in downgrading the petroleum potential of the basin. However, all of the previously drilled wells were drilled without consideration of the potential for developing continuous resources in the organically rich shales and possible coal beds using drilling and completion technology that has been significantly advanced since the mid-1980s. Based on the size of the basin, the apparent lack of a major igneous intrusion event, and hints that all of the components of a continuous (unconventional) petroleum system exists within the basin, the USGS assessed the undiscovered, technically recoverable oil and natural gas resource potential at mean values of 1,064 BCFG (30.1 BCM) and 37 MMBNGL (5.9 MMCM) (Milici and Coleman, 2015b, this volume; Milici et al., 2012).
Richmond basin (Virginia)
The geology of the Richmond basin is summarized in Gore and Olsen (1989), Milici and Coleman (2015a, this volume), Milici and Wilkes, (2003), and Ressetar and Taylor (1988). It is defined by outcrop, potential field geophysics, limited 2D seismic profiles, and numerous wells (Benson, 1992; Johnson et al., 1985b; Wilkes et al., 1989). It is interpreted to be a small rhomb-shaped graben extending north to south approximately 38 mi (61 km) and 8 mi (13 km) west to east at its widest point (Basin 15, Table 1; Fig. 1). It may be as deep as 8,200 ft (2,499 m), and the major bounding fault is on the west side of the basin (Dicken et al., 2008; Gore and Olsen, 1989; Mickus et al., 1988).
It has been the site of at least 32 oil and gas exploration wells since 1897, as well as a site of active coal mining from the early 1700s to the 1950s (Milici and Coleman, 2015a, this volume; Wilkes, 1988). The coals are thick and gassy, and many miners have been killed by explosions during the early years of mining. Of the 32 wells drilled, 27 have reached depths of 1,000 ft (305 m) or more (Milici and Coleman, 2015a, this volume). Four of these wells have reached prerift Piedmont Paleozoic(?) crystalline basement (Gore and Olsen, 1989). Of the 27 “deep” wells, 15 (including the four basement wells) have had shows of oil and/or gas and/or coal (Gore and Olsen, 1989; Milici and Coleman, 2015a, this volume). To-date, none of these wells has resulted in established oil and/or gas production. Efforts to establish coal-bed methane production also have not succeeded.
The deepest well, the Cornell Oil No. 1 Bailey, was drilled to a total depth of 7,438 ft (2,267 m) near the center of the basin, reaching Piedmont crystalline basement at 7,110 ft (2,167 m) (Table 2; Gore and Olsen, 1989), suggesting along with other data that the total present-day synrift fill may be approximately 8,600 ft (2,621 m) thick (Gore and Olsen, 1989).
The stratigraphy of the Richmond basin is composed of basal and basin-margin conglomerates and breccias (Boscabel boulder beds; variable thickness), overlain by fine- to coarse-grained fluvial clastics (Lower barren beds; approximately 600 ft (183 m) thick). The overlying coal-bearing strata (Productive coal measures; approximately 600 ft (183 m) thick) extends across most of the basin. Deep-water lacustrine clastics (Vinita Beds; approximately 2,100 ft (640 m) thick) succeed the coal section. The preserved synrift basin fill ends with a thick interval of lacustrine and fluvial-deltaic sandstones (Turkey Branch Formation and Otterdale Sandstone; approximately 5,300 ft (1,615 m) thick; Gore and Olsen, 1989; Milici and Coleman, 2015a, this volume). These formations range in age from basal Late Triassic (Early to Middle Carnian; Kozur and Weems, 2010). Source rocks are present in the Productive coal measures and the Vinita beds.
Current TOC values for source rocks in the Richmond basin are up to 10% in the Vinita beds (Whiteside et al., 2011). Coals in the Productive coal measures are generally highly gassy, having an average gas content of 350 to 530 cubic feet per ton (10.5 to 16.5 m3 per metric ton) (Knox et al., 1992). Thermal maturity analysis shows potential source rocks range from pre-peak oil to peak gas (Malinconico, 2008; Robbins et al., 1988). Potential reservoirs are present in all of the lithologies present within the basin (coal beds, shales, siltstones, sandstones, and conglomerates). Uncorrected density-neutron wireline log porosity values range from near 0% to 14% in sandstones within the basin (Milici and Coleman, 2015a, this volume). The absence of any clear indicators of conventional accumulations together with oil and gas shows in tight sandstones, coal beds, and shale intervals has led the USGS assessment team to assess the Richmond basin as a continuous (unconventional) gas accumulation (Milici et al., 2012).
The presence of hydrocarbon shows throughout the stratigraphic sequences in the synrift formations of the basin clearly indicate that conventional sealing intervals are no longer present. The presence of late stage fractures throughout the basin apparently disrupted any seal integrity that may have existed at the time of basin filling. However, this apparently pervasive fracture network may improve production flow from continuous reservoirs such as coal beds, tight sandstones, and shales (Milici and Coleman, 2015a, this volume).
The Richmond basin is covered today only by a thin veneer of undifferentiated Quaternary sediments. Applying the Dow (1977) method of estimating basin exhumation to Malinconico’s data (2008), approximately 1,500 to 4,000 ft (457 to 1,219 m) of section has been removed. Based on regional stratigraphy, it is very likely that this missing section is Late Triassic to Early Jurassic in age and has been removed by basin inversion during the Middle to Late Jurassic (and possibly extending into the Cretaceous).
Surface mapping and 2D seismic profiles have indicated several three-way and potentially four-way structures. In addition, synsedimentary structural adjustments and late, synrift dike intrusions divided the basin into many compartments (Venkatakrishnan and Lutz, 1988).
Rapid subsidence during the Triassic and likely Early Jurassic caused the source rocks within the basin to reach conditions for hydrocarbon generation and expulsion at about the time of the Triassic–Jurassic boundary (Fig. 4G). The critical moment is estimated to be about that time for primary accumulation of any generated petroleum into the primary traps (Milici and Coleman, 2015a, this volume). The time from the Early Jurassic to the present-day without long-term upper level seals may be a significant factor in the preservation potential for any accumulated and likely remigrated conventional hydrocarbons. The widespread detection of oil and gas shows indicates that continuous (unconventional) hydrocarbons are present and may be concentrated in the source rock intervals rather than conventional reservoir rock intervals.
Based on the large number of drilling hydrocarbon shows and the presence of key petroleum system elements, the USGS assessed the undiscovered, technically recoverable oil and natural gas resource potential at mean values of 211 BCFG (6.0 BCM) and 11 MMBNGL (1.7 MMCM) (Milici et al, 2012; Milici and Coleman, 2015a, this volume). The Deep Run and Flat Branch basins, two small, shallow outlier basins to the northeast of the Richmond basin, were not included in the Richmond basin assessment.
Deep River basin (North Carolina and South Carolina)
The geology of the Deep River basin is summarized in Bain and Brown (1980), Campbell and Kimbell (1923), Reid (2015, this volume), and Reinemund (1955). It is a large, narrow, half-graben that extends northwesterly for 150 mi (241 km) and is approximately 14 mi (23 km) wide in the center of the basin (Basin 17, Table 1; Fig. 1). As discussed here, the Deep River basin consists of the northern Durham subbasin, the central Sanford subbasin, and the southern Wadesboro subbasin. The Crowburg basin of Steele and Colquhoun (1985) and the Ellerbe basin of Bell et al. (1974) are included within the Wadesboro subbasin.
The Deep River basin is mostly exposed, with only a small portion of the Wadesboro subbasin covered by Atlantic Coastal Plain sediments. For this study, the Deep River basin is defined by the outcrop limits of Mesozoic synrift sedimentary and igneous rocks; where it is buried by postrift sediments, it is defined by water wells and potential field geophysical data. Nine oil and gas exploration wells and several 2D seismic lines also help delineate and characterize the basin. It has a southeastern major basin-bounding fault, and the basin may be as deep as 7,100 ft (2,164 m) (Reid, 2009; Reid et al., 2010, 2011). Gravity and magnetic data modeling by Lai et al., (1985) suggests depths as great as 13,000 ft (3,962 m).
Since the American Revolutionary War, there has been a long history of coal mining in the Sanford subbasin of the Deep River basin. The coal seams have a high gas content, resulting in a series of major explosions in the coal mines and loss of life. Exploration and resource appraisal of these coal beds indicates that retorting of oil and gas from the bituminous shale in the coal measures would produce between 4 and 13 gal (15 to 49 l) of oil and 700 and 3,200 cubic feet of natural gas per ton of rock (21.8 and 99.9 m/mt) (Reinemund, 1955). Sixteen of 43 coal assessment core holes or petroleum test wells have had oil or gas shows (Hoffman and Nickerson, 1988 (revised 1998); Reid, 2015, written communications).
Chevron drilled the first deep oil and gas exploratory well in the basin in 1974 after acquiring a loose grid of 2D seismic lines. The Chevron No. 1 Groce drilled to a total depth of 6,000 ft (1,829 m) into Piedmont Paleozic(?) basement although wireline logs only reached a depth of 5,342 ft (1,628 m). This well remains the deepest penetration of the Deep River basin, though other wells have reached basement at shallower depths (Table 2). Eight other oil and gas wells have been drilled in the basin. SEPCO (operating as Seaboard) drilled two test wells (No. 1 Butler and No. 1 Hall) on either side of the No. 1 Groce. They found substantial oil and gas shows, but were unable to complete either well.
As with all of the onshore east coast Mesozoic synrift basins, the synrift stratigraphy of the Deep River basin varies across the basin. The Deep River basin section ranges in age from Late Triassic (Middle(?) Carnian to Middle(?) Norian; Fig. 5; Kozur and Weems, 2010). It is composed of a basal fluvial sandstone and conglomerate (Pekin Formation, 1,800 to 4,000 ft (547 to 1,219 m thick) succeeded by fluvial-lacustrine coal, shale, and sandstone (Cumnock Formation; 300 to 800 ft (91 to 244 m thick), that is in turn overlain across most of the basin by the Sanford Formation (fluvial-lacustrine siltstone, shale, and local conglomerates 1,000 to 5,000 ft (305 to 1,524 m) thick) (Olsen et al., 1991; Reinemund, 1955; Thayer, 2006, unpublished report). Approximately 7,100 to 10,000 ft (2,164 to 3,048 m) of Newark Supergroup strata are preserved in the basin (Olsen et al., 1991; Reid, 2015, this volume; Reinemund, 1955). A small area of postrift Upper Cretaceous sands, silts, and clays overlies a portion of the Wadesboro subbasin (Conley, 1962).
Organic richness data for the Cumnock Formation Late Triassic (Carnian} organic-rich shale and coaly source rock interval in the Sanford subbasin indicate an average value of 2.9% TOC, a minimum of 0.01%, and a maximum of 33.6% TOC. Thermal maturity ranges from an Ro value of 0.8% to an average of 1.4% and maximum of 4.8% (Reid, 2015, this volume). The organic matter is a mixture of terrestrial plants and lacustrine algae (Reid, 2015, this volume). The presence of Cumnock Formation source rocks in the Durham subbasin has not been determined, and the Cumnock Formation equivalent in the Wadesboro subbasin has TOC values between 0.04% and 0.3%, averaging of 0.2% (Reid, 2015, written communications).
Organic richness data from the Sanford Formation in the Sanford subbasin ranges from 0.2% TOC to 4.4% TOC, averaging 1.4%. Only one Ro value (1.3%) is available from these analyses, which has a value consistent with the average Ro value of the Cumnock Formation within the Sanford basin (Reid, 2015, written communications).
Limited core data for potential conventional reservoir rocks in the Sanford and Pekin formations indicate low porosity and permeability: porosity values range from 0.1% to 8.1%, an average of 4.2%, and permeability values range from <0.01 mD to 2.2 mD, averaging 0.4 mD (Reid and Milici, 2008). Recently acquired porosity and permeability data for the Cumnock Formation, which is considered a potential continuous (unconventional) reservoir, range from 0.2% to 6.4% porosity, averaging of 1.9%, and 0.05x10-5 mD to 7.11x10-5 mD permeability, averaging 1.68x10-5 mD (Reid et al., 2014a, b). The limited data are currently insufficient to fully characterize the potential continuous (unconventional) reservoir rock characteristics of fractured diabase bodies, which have been demonstrated to host as yet non-economic accumulations of natural gas (Hoffman and Buetel, 1991). Brown (1988) reports two samples of diabase from the Sanford subbasin, which have porosity values of 3.4% and 26.4% and permeability values of 0.04 mD and 2.1 mD, respectively.
Potential seals are evident throughout most of the synrift section in low permeability mudstone, siltstone, sandstone, and diabase. A regionally continuous zone capable of sealing potential continuous accumulations occurs only in the Cumnock Formation. However, the presence of a pervasive fracture network suggests that any accumulation may be heavily compartmentalized.
Surface mapping in conjunction with regional 2D seismic profiles indicated that conventional structural and combination traps were present within the basin. All but one of the oil and gas exploratory wells were drilled along the same 2D seismic line, suggesting that all conventional traps and play types illustrated on that line were sufficiently tested without establishing commercial production. However, attempts to establish unconventional production were apparently not made using modern drilling and completion technology. The presence of pore-throat traps were clearly indicated by the widespread occurrence of oil and gas shows while drilling.
Rapid subsidence during the Triassic likely caused the source rocks within the basin to reach thermal conditions for hydrocarbon generation and expulsion during the latest Triassic. Examination of the thermal maturity of the synrift section suggests that at least 3,000 ft (914 m) of synrift section was removed from the Deep River basin after subsidence ceased, probably in the Late Triassic (Reid, 2015, this volume). The critical moment is estimated to be near the Triassic–Jurassic boundary (Fig. 4H).
Based on the large number of drilling shows and the presence of key petroleum system elements, the USGS assessed the undiscovered, technically recoverable oil and natural gas resource potential at mean values of 1,660 BCFG (47 BCM) and 83 MMBNGL (13 MMCM) in a continuous (or unconventional) accumulation (Milici et al. 2012; Reid, 2015a, this volume).
Buried Onshore and State Waters East Coast Mesozoic synrift Basin Review and Evaluation
Buena basin (New Jersey)
The Buena basin is the northernmost of several buried onshore Mesozoic synrift basins identified or postulated beneath the Atlantic Coastal Plain sediments in New Jersey, Delaware, and Maryland (Basin 18, Table 1; Fig. 1). It is defined by a single 2D reflection seismic profile in support of gravity anomaly mapping (Sheridan et al., 1991). Benson (1992) sets the dimensions to be 36 mi (58 km) in length in a northwesterly direction and 10 mi (16 km) in width and illustrates a northwestern basin-bounding fault. Based on seismic velocity analysis, Sheridan et al. (1991) estimate the basin to contain at least 3,300 ft (1,006 m) of Mesozoic synrift sedimentary rocks underlying approximately 3,500 ft (1,067 m) of postrift Mesozoic and Cenozoic coastal plain strata (Trapp, 1992). Sheridan et al. (1991) also suggest that the southeastern fault zone may be as large as the northwestern fault zone, such that the Buena basin is a full graben. No well has confirmed the identity or character of the synrift fill. This basin was not assessed by the USGS because of the lack of data.
Delaware Bay basin (New Jersey-Delaware)
The Delaware Bay basin is represented by a low aeromagnetic anomaly between Delaware and New Jersey (Basin 19, Table 1; Fig. 1). Benson (1992) illustrates this basin as approximately 41 mi (66 km) long in a northwesterly direction and approximately 7 mi (11 km) wide, with a northwestern basin-bounding fault. No well or seismic data help illuminate the identity or character of this basin. This basin was not assessed by the USGS because of the lack of data.
Queen Anne basin (New Jersey-Delaware-Maryland)
The Queen Anne basin has been identified by several 2D seismic profiles, gravity data, and at least one well (Basin 20, Table 1; Fig. 1; Benson, 1992; Hansen, 1988). It has a main border fault on its western side and minor down-to-the-west faults along the eastern margin (Benson, 1992; Hansen, 1988). It is estimated to be approximately 140 mi (225 km) long and 27 mi (43 km) wide at its widest point (Benson, 1992). As with several exposed basins of similar size (e.g., Newark, Taylorsville, and Deep River basins; Table 1), it is likely a composite basin of several subbasins. Based on sparse 2D seismic and well data, it is at least 7,300 ft (2,225 m) deep, with as much as 4,100 ft (1,250 m) of Mesozoic synrift fill (Hansen, 1988).
Data were insufficient to identify or characterize the various potential petroleum system elements in the Queen Anne basin. Consequently, it was not assessed by the USGS because of lack of data. However, it was included in a collective tight gas continuous assessment unit with three other basins as the Delmarva basins (Milici et al., 2012).
Greenwood basin (New Jersey-Delaware-Maryland)
The Greenwood basin is a buried basin whose identification is based on a single 2D seismic profile and the extent of the aeromagnetic “signature” where that line crosses the signature (Benson, 1992). The basin is interpreted to extend in a northeasterly direction for 38 mi (61 km) and is approximately 7 mi (11 km) wide (Basin 21, Table 1; Fig. 1). Benson (1992) interprets the basin to have a northwest and a southeast basin-bounding fault from a single seismic line. No wells have been drilled to confirm this interpretation or to identify any potential petroleum system elements. The USGS did not assess this basin; rather, it was identified as potentially containing a composite continuous tight gas petroleum system and was included within the Delmarva basins assessment unit (Milici et al., 2012).
Bridgeville basin (New Jersey-Delaware-Maryland)
The Bridgeville basin is a buried Mesozoic synrift basin that has been identified from the intersection of a single 2D seismic profile and the extent of the aeromagnetic “signature” of the basin (Benson, 1992). It is 45 mi (72 km) long and 9.5 mi (15 km) wide (Basin 22, Table 1; Fig. 1). Benson (1992) interprets the basin to have a northwest and a southeast basin-bounding fault from a single seismic line. No wells have been drilled to confirm this interpretation. As with other basins in this area, it was not individually assessed by the USGS. Instead, it was identified as potentially containing a composite continuous tight gas petroleum system and was included within the Delmarva basins assessment unit (Milici et al., 2012).
Worcester basin (New Jersey-Delaware-Maryland)
The Worcester basin is a buried basin that has been identified using one seismic profile, four wells, and a regional aeromagnetic anomaly map (Fig. 2; Benson, 1992; Edwards, 1970; Maher and Applin, 1971; Richards, 1945a). These data suggest that the basin is at least 50 mi (80 km) long and its greatest width is 30 mi (48 km) (Basin 23, Table 1; Fig. 1). Data are insufficient to estimate depth of the basin; however, at least 600 ft (183 m) of synrift section were penetrated by the deepest well in the basin (Anderson et al. 1948).
The stratigraphy of the upper portion of the Worcester basin synrift fill was described by Johnson (1945) and Johnson and Richards (1945) from conventional cores and cuttings from two deep oil and gas exploration wells. No stratigraphic subdivision was presented; the entire section was identified as “Triassic”, “Triassic (?)” or “Probably Triassic”. The sedimentary package was described generally as interbedded red and gray sandstone (occasionally arkosic) and red and green sandy shale. Both wells reached total depth in crystalline rock identified as possibly Precambrian.
Black or dark gray shales are reported only in minor amounts from the synrift section in these two wells (Anderson et al., 1948). Reports on the examination of the cored synrift section give no indication of visual porosity or permeability in sandstone, siltstone, or shale (Johnson, 1945; Johnson and Richards, 1945). No oil or natural gas shows are reported while drilling from any of the wells in the basin.
Data are insufficient to characterize the various potential petroleum system elements in the Worchester basin. The USGS did not assess this basin, and instead considered it a part of the Delmarva basins assessment unit (Milici et al., 2012).
King and Queen basin (Virginia)
King and Queen basin of eastern Virginia is a small buried basin (Basin 24, Table 1; Fig. 1), defined by the presence of a low magnetic anomaly and a single deep water well (Fig. 2; Wilkes et al., 1989). A 2D seismic line acquired along Interstate 64 that crosses through the area of the basin does not image it clearly (Harris et al., 1986; Virginia Tech Geothermal Data Regional Geophysics Laboratory, undated). The basin is interpreted to be approximately 32 mi (51 km) long north–south, 7 mi (11 km) wide east–west, and at least 370 ft (112 m) deep based on Water Well 515, which reached total depth in Mesozoic synrift sediments at 1,689 ft (515 m) (Wilkes et al., 1989). This basin was not assessed by the USGS for oil and natural gas potential because of lack of data.
Toano basin (Virginia)
The Toano basin is another small basin beneath the Atlantic coastal plain, which was identified by low gravity and low magnetic anomalies, a single deep water well, and a single 2D reflection seismic line (Basin 25, Table 1; Figs. 1, 2; Costain and Çoruh, 1989; Schorr, 1986; Wilkes et al., 1989). It is 34 mi (55 km) long and approximately 5 mi (8 km) wide (Fig. 2). Seismic data tentatively suggest that it may have one of the thickest synrift sections in the East Coast Mesozoic synrift basin trend, being approximately 11,000 to 13,000 ft (3,353 to 3,962 m) thick (Costain and Çoruh, 1989; Schorr, 1986). This basin has not been assessed by the USGS for oil and natural gas potential because of lack of data.
Suffolk County basin (Virginia-North Carolina)
The Suffolk County basin is another small basin defined by a low magnetic anomaly (Fig. 2) and a deep observation well, which penetrated 97 ft (30 m) of Mesozoic synrift strata without reaching basement rock (Basin 26, Table 1; Fig. 1; Wilkes et al., 1989). This basin also has not been assessed by the USGS for oil and natural gas potential because of lack of data.
Elisabeth City basin (Virginia-North Carolina)
The Elisabeth City basin lies astride the Virginia-North Carolina border (Basin 27, Table 1; Fig. 1; Olsen et al., 1991). It is defined primarily by a low aeromagnetic anomaly and three oil and gas tests (Daniels and Zietz, 1978). One of these wells, the deGrandlee Exploration No. 1 Foreman, found shows of gas in the Cretaceous above at least 2,900 ft (884 m) of Triassic red and carbonaceous shale, and plant fossil fragments and some sandstone interbedded with fresh diabase. A second show of gas was detected in what may be the upper synrift Mesozoic section (Richards, 1954). Two other wells encountered Mesozoic synrift sediments within the area of the low aeromagnetic anomaly. No further information is available on this basin. Consequently, it was not assessed by the USGS for oil and natural gas potential because of lack of data.
Cumberland-Marlboro basin (North Carolina and South Carolina)
Of the areas that have low magnetic anomalies (Fig. 2) and no clearly condemning well data and are inferred to be buried Mesozoic synrift basins, one of the largest of these is termed here the Cumberland-Marlboro basin. This interpretation is based on an extensive, low aeromagnetic anomaly that trends subparallel to the Deep River basin 18.5 mi (30 km) to the northwest. The interpreted basin (Basin 30, Table 1; Fig. 1) is 117 mi (188 km) long and 12 mi (19 km) wide at its widest extent. Water well data are insufficient to estimate basement depth or synrift thickness. The current outline of this inferred/proposed basin is drawn to exclude obvious non-Newark Supergroup type rocks, such as highgrade igneous and metamorphic rocks encountered by water wells in the area, but to include those rocks whose identity could be consistent with Mesozoic synrift strata. The outline also includes an area in northeast South Carolina that has at least two water wells that found “unmetamorphosed presumed lower Mesozoic (Upper Triassic-Lower Jurassic) synrift sedimentary rock sand[stone] or diabase” (Benson, 1992).
The few North Carolina water wells within the area of Cumberland-Marlboro low magnetic anomaly encountered rocks potentially consistent with Mesozoic synrift sedimentary rocks (i.e., HO-P-1-70, phyllite and quartzite; CD-P-1-67, metagraywacke; CD-T-1-86, phyllite and metasandstone; SA-T-1-XX, metamudstone; Lawrence and Hoffman, 1993). This basin was originally was interpreted by Bonini (1964) and Bonini and Woollard (1960) as a “buried Triassic Basin” and given the name “Fayetteville basin”. Its identification was based on a lower than expected basement velocity of 12,500 ft (3,810 m) per second and an interpretation that two wells with bottom cuttings in sericitic phyllite and volcanic material were, in fact, “Triassic sediments (?)”.
Schipf (1964) disagreed with Bonini (1964) citing his (Schipf, 1961) groundwater study data as evidence that the basin was a geophysical anomaly (velocity and magnetic low) associated with metavolcanic-epiclastic rocks of the Eastern Slate Belt (Brown and others, 1985). Schipf (1964) cited records from 25 wells in the Fayetteville area which bottom in “slate” according to drillers' and Schipf's logs. Schipf (1964), however, recognized the possibility that the gray slates in these wells might be gray shales of Triassic age, such as those of the Cumnock Formation in the Durham–Sanford Basin, but considered it “unlikely.” Because of the size of this geophysical anomaly, the true nature of these non-specific bottom-hole samples should be reinvestigated.
As is often the case with these buried basins, data were insufficient to identify any potential petroleum system elements in the Cumberland-Marlboro basin at the time of the USGS resource assessment. The USGS did not assess this basin specifically. However, based on its size and proximity to the Deep River basin, it was identified as potentially containing a composite continuous tight gas petroleum system (Milici et al., 2012).
In mid-2015, the North Carolina Geological Survey conducted drilling and coring operations at three locations within the outline of the Cumberland-Marlboro basin (Fig. 1; Reid, 2015). All three core holes recovered metamorphic basement rock beneath Cretaceous sedimentary rock of the Atlantic coastal plain, indicating that the large magnetic low and refraction seismic velocity low used to postulate the outline of the Cumberland-Marlboro basin are not the geophysical responses to a large, strike-parallel Mesozoic synrift basin seaward of the Deep River basin (Reid, 2015; Reid et al., in progress).
Other Buried Mesozoic synrift Basins of North Carolina
A narrow Mesozoic synrift basin, the Bertie County basin, centered in Bertie County, North Caorlina, was identified based on a single core hole drilled in 2004 (Basin 28, Table 1; Fig. 1; Weems et al., 2007). This basin did not have a pronounced negative aeromagnetic anomaly, nor was it confirmed by reflection seismic profiling. Sixty-eight and one half feet (21 m) of strata characteristic of Mesozoic synrift basin fill beneath a Cretaceous section were core-drilled. Weems et al. (2007) described these strata as clayey, sandy conglomerate (with approximately 75% clasts of ferruginous Triassic sandstone and 25% Piedmont metamorphic rocks), clayey, silty, quartz sandstone, and sandy siltstone, most of which are dark red-brown in color. No black or dark gray shales were reported. No information was available to estimate depth or synrift fill thickness. This basin was not assessed by the USGS due to its size and lack of data.
Another possible basin, approximately 63 mi (101 km) long by 12 mi (19 km) wide (Basin 29, Table 1; Figs. 1, 2), the Craven-Jones basin, was identified in Jones and Craven counties, North Carolina (Daniels and Zietz, 1978; Won et al., 1979). Further investigation by Sampair (1979), who tentatively named this feature the Graingers basin, resulted in the conclusion that the strata beneath the Atlantic coastal plain within the area of the low magnetic anomaly was not Mesozoic synrift material, but rather phyllitic basement rock of undetermined, but possibly Paleozoic age, based on recovery from three core holes. Based on vertical electrical soundings at the core hole locations, crystalline rock was estimated to be approximately 4,500 ft (1,372 m) below the Lower Cretaceous-phyllite unconformity (Sampair, 1979). This basin was not assessed by the USGS due to lack of data.
Two additional buried basins have been postulated in southeast North Carolina and northeast South Carolina: the Columbus and Columbus South basins (Basins 31 and 32, Table 1; Figs. 1, 2). Delineation of these basins is based solely on the low aeromagnetic anomalies in the area (Fig. 2). No 2D seismic profiles have confirmed the existence of these basins. Benson (1992) shows the location of a single deep well that penetrated presumed Upper Triassic-Lower Jurassic synrift strata in the Columbus South basin. In 1939, Palmetto Oil drilled two wells to “bedrock” at 1,150 ft (350 m) in Horry County, South Carolina, near a “considerable show of gas in the soil” and “some seepage of oil to the top of the earth” (Richards, 1945b, p. 918). These wells are located directly adjacent to the proposed Columbus South basin (Figs. 1, 2) and very near a suggested border fault. These basins were not assessed by the USGS due to lack of data.
Florence basin (South Carolina)
The Florence basin is a small Mesozoic synrift basin buried beneath the Atlantic Coastal Plain in northeastern South Carolina (Basin 34, Table 1; Fig. 1). It is defined by several wells, magnetic data, and seismic refraction studies (Bonini and Woollard, 1960; Darton, 1896; Richards, 1945b; Siple, 1958; Zietz et al., 1982). The geology of the Florence basin has been described by Steele and Colquhoun (1985). The basin is approximately 40 mi (64 km) long and 13 mi (21 km) wide at its widest point (Bledsoe and Marine, 1980). There are no published estimates of synrift thickness or basin depth; however, it lies beneath 600 to 700 ft (183 to 213 m) of Upper Cretaceous Atlantic Coastal Plain sediments.
The stratigraphy of the Florence basin is not formally defined. Only a single, old, deep water well penetrating slightly more than 700 ft (213 m) of brown and gray sandstone tentatively identifies those rocks as being related to the Triassic Newark Supergroup (Darton, 1896). Drilling of the well was stopped upon reaching a “hard black rock,” likely a Mesozoic diabase intrusion. Other water wells in the basin have yielded “red sandy clay,” “hard red clay,” and “red clay with sand and gravel” (Steele and Colquhoun, 1985).
There were insufficient data with which to determine the various potential petroleum system elements in the Florence basin. The Florence basin was not assessed by the USGS because of its relatively small size. However, it was recognized as potentially having a composite continuous gas petroleum system (Milici et al., 2012).
(South) Florence basin (South Carolina)
A possible southern component of the Florence basin is recognized in this paper as the South Florence basin (Basin 35, Table 1; Fig. 1). It is defined by a single well and aeromagnetic data (Fig. 2). It is approximately 30 mi (48 km) long and 9 mi (14 km) wide at its widest extent (Fig. 1). In 1987, a Texaco test well was continuously cored from 1,275 to 5,889 ft (389 to 1,795 m) and recovered 3,314 ft (1.010 m) of red sand and shale and 1,293 ft (394 m) of gabbro; no shows were reported (IHS Enerdeq, 2014). The plotted well location lies on the edge of the aeromagnetic low anomaly used to define the basin. No further information is available on this area. As with the Florence basin, this basin has not been assessed because of its relatively small size and lack of data.
Marion-Dillon basin (South Carolina)
A possible small, buried basin, termed the Marion-Dillon basin in this paper, is located near the South Carolina-North Carolina line in Marion and Dillon counties, South Carolina. This basin is postulated based on a characteristic low aeromagnetic anomaly, similar to the response for others confirmed as Late Triassic–Early Jurassic rift basins (Basin 33, Table 1; Figs. 1, 2). No wells or seismic data are available to confirm its location and structural characteristics. As such, this potential basin has not been assessed by the USGS.
Lexington basin (South Carolina)
A second possible small, buried basin is located near the center of Lexington County, South Carolina. This basin, termed here the Lexington basin, is postulated based on the character of a low aeromagnetic anomaly in the area (Basin 36, Table 1; Figs. 1, 2). No wells or seismic data are available to confirm its location and structural characteristics; consequently, this potential basin was not assessed by the USGS.
Calhoun basin (South Carolina)
A third possible small, buried basin is located near the center of Calhoun County, South Carolina. This basin, the Calhoun basin, is postulated based on a low aeromagnetic anomaly (Basin 37, Table 1; Figs. 1, 2). No wells or seismic data are available to confirm its extent and structural characteristics. The lack of data precludes assessing this potential basin.
South Georgia basin (Georgia and South Carolina)
A large area of southeast Georgia and southern South Carolina is characterized by early Mesozoic synrift strata overlying Paleozoic rock. This area, which appears to contain a number of Mesozoic synrift basins, is collectively termed in this study the South Georgia basin (Basin 38, Table 1; Fig. 1; McBride et al., 1987). It is defined by several public-domain 2D seismic profiles, regional potential field geophysical data, and a number of wells (Blount et al., 2011; Brantley et al., 2012; Chowns and Williams, 1983; Daniels et al., 1983; Domoracki, 1995; Heffner, 2013; McBride and Nelson, 1988; McBride et al., 1987; Sartain and See, 1997; and Waddell, 2013). However, even with this data set, a consensus outline delimiting this basin complex is not evident in the literature (see discussion in Heffner, 2013). For the purposes of the USGS resource assessment (Milici et al., 2012), the outline for the South Georgia basin assessment unit was determined by basin outlines from Chowns and Williams (1983), Daniels et al. (1983), and Horton et al. (1991) in addition to a regional aeromagnetic anomaly map derived from U. S. Geological Survey and National Geophysical Data Center (2002). The southern extent is arbitrarily placed where the trend of the Brunswick aeromagnetic anomaly (McBride and Nelson, 1988) cuts across the basin, separating the South Georgia basin (as herein defined) from the North Florida basin (as defined below) (Fig. 1).
Within the South Georgia basin complex of this report are the Jedburg, Dunbarton, Colleton-Dorchester, and Riddleville subbasins (Basins 38a to 38d respectively, Table 1; Fig. 1; discussed below), which have been previously identified and their designation/identification has been in common usage prior to the 2012 USGS resource assessment. The Kibbee, Unadilla, and Albany subbasins as defined by Heffner (2013) are not discussed independently in this paper. However, aspects of those areas are described in the general discussion of the South Georgia basin complex. The Summerville basin of Steele and Colquhoun (1985) is not examined separately, but is included within the analysis of the greater composite South Georgia basin.
Much of the South Georgia basin appears to be overlain by a regional seismic event termed the “J-reflector” (McBride et al., 1989). This event originally has been tied to the near Triassic–Jurassic boundary event associated with the widespread eruption of basalt flows and diabase dikes and sills of the Central Atlantic Magmatic Province (CAMP) (Marzoli et al., 2011). In the area near Charleston, South Carolina, where the “J-reflector” was first described, the USGS has drilled three core holes, one of which penetrated Mesozoic synrift sedimentary strata below a thick basalt sequence (Gohn, 1983; Rankin, 1977). Other wells and core holes have also penetrated a thick basalt interval in southern South Carolina, reaching total depth in what is likely Mesozoic synrift sedimentary rocks (Heffner, 2013).
The presence of the J-reflector has been used to portray the limit of the South Georgia basin (McBride et al., 1989). However, detailed reexamination of the available 2D reflection seismic and well data indicate that the “J-reflector” is not a consistently continuous seismic reflector and is not always associated with basalt or other mafic rock types (Heffner et al., 2012). Instead, it seems best associated with the onlap surface of the Atlantic Coastal Plain sediments, whether that surface is developed on Piedmont crystalline rocks or early Mesozoic synrift basin strata. As such, the J-reflector is not considered a reliable delimiter of the areal extent of the South Georgia basin complex (Heffner et al., 2012).
The geology of the South Georgia basin complex is discussed by Chowns and Williams (1983), Blount et al. (2011), and Heffner (2013). As defined here, the South Georgia basin complex extends approximately 300 mi (483 km) in a northeast direction and 140 mi (225 km) in a northwest direction. Because it consists of several subbasins, the depth to basement for the individual rift segments varies. The maximum depth appears to approach 17,000 ft (5,182 m) (Sartain and See, 1997). Heffner (2013) presented a shallower maximum depth of approximately 13,000 ft (3,962 m) based on observed conflicts between well data and potential field calculations. Where illustrated by 2D seismic profiles, the subbasins all appear to be fault-bounded, with one side having a more pronounced major fault (McBride et al., 1987, 1989; Waddell, 2013)
No source rocks have been geochemically identified from Mesozoic synrift strata penetrated in wells in Georgia. Dark gray shales and coals are present in the Hunt No. 1 Stalvey (Table 2); however, their source potential was not measured. Relatively high gamma ray excursions on the wireline logs of the Surface Exploration Co. No. 1 McNair well (Table 2) suggest that potentially high organic content layers may be present in that area.
Unlike other individual Mesozoic synrift basins, a general stratigraphic succession has not been established for the South Georgia basin complex. However, with the possible notable exception of relatively thick coal or lacustrine organic shale formations, all of the other stratigraphic components typical of the other Newark Supergroup synrift stratigraphic successions are present.
Significant among the wells which penetrated the thickest synrift section are the Chevron No. 1 Snipes (11,456 ft; 3,492 m), the Leighton No. 1 Dana (6,030 ft; 1,838 m), the Surface Exploration Co. No. 1 McNair (17,815 ft; 5,430 m), the SEPCO (Essex) No. 1 Light-sey (12,750 ft; 3,886 m), and the SEPCO No. 1 McCoy (9,400 ft; 2,865 m) (Table 2).
The No. 1 Snipes (Jeff Davis County, Georgia; drilled in 1981) penetrated at least 3,770 ft (1149 m) of Newark Supergroup strata overlying 3,567 ft (1,087 m) of Paleozoic sedimentary rocks before possibly encountering basement at 11,406 ft (3,477 m) (Chowns, 2009). Prior to the drilling of the Surface Exploration No. 1 McNair et al. well (Turner County, Georgia), this was the deepest well in Georgia. Waddell (2013) reported approximately 465 ft (142 m) of section having greater than 10% porosity in synrift sandstones, interbedded with siltstone, mudstone, and diabase in the No. 1 Snipes well.
The Leighton No. 1 Dana well (Table 2; Pulaski County, Georgia; drilled in 1957) drilled through 3,685 ft (1123 m) of Newark Supergroup synrift strata. Gas shows were reported from pink fluvial sandstones having 6 to 9% porosity near 3,000 ft (914 m) measured depth. Oil shows were also reported from sands in the “3,000-foot” interval and at 1,700 ft (518 m) and 4,870 ft (1,484 m). The operator had extreme difficulty testing in the “3,000-foot” zone and attributed this to drilling mud damaging the borehole and establishing an oil-water emulsion in the vicinity of the reservoirs and the well bore. The No. 1 Dana well was plugged after acidizing and swabbing established only a non-commercial gas flow (Leighton, 1957, unpublished report). Oil and gas shows were also reported from the Leighton No. 1 Tripp well (Pulaski County, Georgia; drilled in 1957) near the top of the synrift section (IHS Enerdeq, 2014). Near the No. 1 Dana well, the Atlanta Gas Co. No. 1 Griffith well (Pulaski County, Georgia; drilled in 1974) penetrated 3,983 ft (1,214 m) of Newark Supergroup redbeds and diabases, with gas shows reported near a depth of 3,000 ft (914 m) (Anonymous, undated).
In 1996, the Surface Exploration Partners No 1 McNair et al. well (Table 2; Turner County, Georgia; drilled in 1997) penetrated at least 10,108 ft (3,081 m) of Newark Supergroup strata before encountering coarse-grained igneous rocks (diorite or gabbro) at 15,258 ft (4,820 m), which might be Paleozoic-age basement. Drilling continued to a total depth of 17,815 ft (5,430 m), thus making this currently the deepest well drilled in Georgia. Low-level gas shows were recorded scattered throughout the well, the greatest concentrations of shows between 12,880 ft (3,926 m) and 13,315 ft (4,058 m). This interval of mostly sandstone had several thin layers of high gamma ray response, which may be organic-rich shales or coals. However, neither was described on the mud log through this section. A second gas show (methane with a trace of butane) was reported at approximately 15,500 ft (4,724 m) in igneous or meta-igneous rock. The well was plugged and abandoned without testing because of low porosity and permeability in potential conventional reservoirs.
Natural gas shows, and a few oil shows, have been reported in wells scattered throughout the South Georgia basin complex. Several of these wells have had follow-up drilling in close proximity to the original hole, but no successful flow testing has been reported.
Akintunde et al. (2011) presented data from wells in the South Georgia basin complex showing porosity and permeability data consistent with values from other East Coast Mesozoic synrift basins: i.e., porosity values ranging from 2.1% to 13.3% and permeability values ranging from 0.0023 mD to 0.16 mD (excluding the No. 1 Lightsey core data discussed below).
Publically available 2D reflection seismic data suggest that several types of traps, both structural and combination structural-stratigraphic, are present within the South Georgia rift basin complex (McBride et al., 1987, 1989; Waddell, 2013). Scattered evidence indicates that the major fault zones within the basin complex were active into the late Eocene (Prowell and O’Connor, 1978).
Basin modeling by Waddell (2014) indicates that the South Georgia basin subsided rapidly from approximately 230 Ma to 200 Ma, after which uplift and inversion began. This exhumation ended at approximately 140 Ma.
Approximately 9,000 ft (2,743 m) of Mesozoic synrift fill are estimated to have been eroded during this episode. Adequate biostratigraphic data are lacking, such that a reasonably confident burial history profile cannot be developed. However, it appears that based on what is known, the South Georgia basin had a similar history to other east coast Mesozoic synrift basins, and a critical moment for any possible petroleum generation and entrapment occurred near the time of the Jurassic-Triassic boundary (Fig. 4I).
Dunbarton basin (Dunbarton basin of Heffner, 2013)(South Carolina)
With the exception of the southwestern-most end of the Deep River basin (the Wadesboro subbasin), all of the occurrences of Newark Supergroup rocks in South Carolina are beneath the Atlantic Coastal Plain. The Dunbarton basin (Basin 38a, Table 1; Fig. 1), located in the southwestern part of the state, has been drilled and studied extensively for waste disposal purposes and for interpreting the structural history of the U. S. Department of Energy’s Savannah River Site. This basin is discussed in Domoracki (1995), Marine (1974), Marine and Siple (1974), Petersen et al. (1984), Steele and Colquhoun (1985), and Siple (1967). The Dunbarton basin is approximately 30 mi (48 km) long and 9 mi (14 km) wide, underlies approximately 1,200 ft (366 m) of coastal plain cover, and has a synrift fill greater than 3,000 ft (914 m) as determined from drilling and possibly as much as 12,000 ft (3,656 m) based on geophysical studies (Cumbest et al., 1992; Domoracki, 1995; Marine, 1974; Snipes et al., 1993b; Stieve and Stephenson, 1995). Marine (1974) suggested that a synrift thickness of 5,300 ft (1,615 m) best explains the observed level of formation pressure.
The Dunbarton basin is a half-graben having a basin-bounding fault on the northwest side. Richers (2000, pers. comm. in Dennis et al., 2000) theorized that a pervasive flushing event (the “greening” event) was the result of the flow of reducing water accompanying migrating hydrocarbons along the basin-bounding border fault. This “greening” event probably occurred at approximately 86.3 Ma, based on its presence in Upper Cretaceous Coniacian strata and absence in Santonian rocks (Dennis et al., 2000). Petrographic study indicated that the Triassic synrift interval has also undergone hydrothermal alteration and subsequent thermal metamorphism (Snipes et al., 1993a).
Surface-based soil gas geochemical sampling has indicated the possible presence of two different suites of source rocks in the Dunbarton and Riddleville basins. One is a normal, oil-prone marine source, and the second is a more gasprone terrigenous source (Richers et al., 2000). However, no direct observation of a source-rock quality interval has been reported.
Effervescing dissolved gas from two test wells in the Dunbarton basin contain the same ratio of hydrogen to nitrogen (approximately 67% hydrogen to 33% nitrogen) as analyzed from coal in the Deep River Basin in North Carolina (Marine, 1974; Reinemund, 1955). Minor amounts of methane were also detected from these two wells (0.021 to 0.098 mole %; Marine, 1974). Other than these gas analyses, no hydrocarbon shows have been reported in the various test wells in the Savannah River Plant area.
Potential conventional reservoir rocks in the Dunbarton basin as observed in slim-hole core and drill cuttings, are mostly fine to very coarse-grained arkosic sandstone and conglomerate. Analyses of slim-hole core in these lithologies indicate total porosity values ranging between 2.0% and 12.4% and mean effective porosity >1 μm, varying between 1.7% and 6.5% (Thayer, 1993). Permeability ranges from 0.000066 mD to 0.016 mD (Marine, 1974). Low matrix permeability is the result of high clay content, tight grain packing, compressive infiltration of clay and ductile rock fragments into primary pore space, and cementation (Thayer and Summer, 1996).
Sealing potential within the synrift section is potentially good based on the permeability measurements noted above. Data from two early test wells indicate that the Dunbarton basin is slightly overpressured (0.453 psi/ft (3.12 kPa) and 0.473 psi/ft (3.26 kPa) as compared with 0.433 psi/ft (2.99 kPa) for freshwater wells in the coastal plain aquifer; Marine, 1974). These data suggest that the Mesozoic synrift section is sealed from the overlying coastal plain cover.
The Dunbarton basin is currently covered by approximately 1,150 ft (350 m) of Cretaceous and Cenozoic strata (Domoracki, 1995). Reactivated Paleozoic shear fault zones appear to have controlled initial rift development and postrift sedimentation of the Cretaceous through Tertiary coastal plain (Cumbest et al., 1993). Faults bounding the Dunbarton basin appear to have offset strata as young as Late Eocene and to have a reverse slip component (Snipes et al., 1993b; Stieve and Stephenson, 1995); the present-day reverse offset of the basement-coastal plain unconformity is approximately 100 ft (30 m) (Cumbest et al., 2000), and there is as much as 325 ft (99 m) of postrift uplift and at least 200 ft (61 m) of erosion of Upper Cretaceous strata (Faye and Prowell, 1982).
Marine (1974) suggests from Gulf Coast analogs that between 7,000 to 13,000 ft (2,133 to 3,962 m) of late synrift and postrift material was deposited during the Jurassic and Cretaceous over the present-day Triassic section and then removed by subsequent erosion. Following the cessation of Triassic-Jurassic rifting, from the early Late Cretaceous to the end of the Paleogene, several episodes of oblique-slip, reverse faulting occurred (Aadland and Thayer, 2000). These faults have not been active since the Miocene (Wheeler, 2002).
Two-dimensional seismic profiles suggest several types of structural traps may be present within the synrift and early postrift section. Pore-throat traps are also likely, based on sparse petrographic data.
The Dunbarton basin appears to have evolved in a similar manner to other East Coast Mesozoic synrift basins: rapid subsidence from Middle–Late Triassic through Early Jurassic likely driving any potential source rocks through the oil generation window and into the gas generation window. With ensuing uplift, beginning possibly in the Middle Jurassic, maturation and thermogenic hydrocarbon generation eventually ceased. As in the case in the other Mesozoic synrift basins, the critical moment for the Dunbarton basin is in the Early Jurassic.
Riddleville basin (Riddleville basin of Heffner, 2013) (South Carolina)
The Riddleville basin (the Riddlesville basin in some reports) appears to be a southwestern extension of the Dunbarton basin (Basin 38b, Table 1; Fig. 1). Several wells have been drilled in and around its margins. In 1978, the SEPCO and GEO78 No. 1 Taylor well drilled to 3,808 ft (1,161 m) and penetrated 2,632 ft (802 m) of mostly diabase associated with minor sandstone and siltstone (Table 2). Minor gas shows in several intervals were recorded when the well drilled out of diabase and into interbedded sandstone and siltstone. SEPCO continued their drilling program in 1980 with the No. 1A Malpasse well, which was drilled approximately 0.8 mi (1.3 km) northwest of the No. 1 Taylor well and penetrated approximately 4,550 ft (1,387 m) of basin margin fanglomerate on the northwest basin margin. No oil or gas shows were reported. The No. 1 McCoy well was subsequently drilled as an offset approximately 1.8 mi (2.9 km) southwest to the No. 1 Taylor. The SEPCO No. 1 McCoy drilled through 8,275 ft (2,522 m) of Newark Supergroup strata before reaching total depth of 9,386 ft (2,861 m) in Piedmont schist (Chowns, 2009). Multicolored to red sandstone and mudstone were encountered before drilling through 1,530 ft (466 m) of varicolored sandy conglomerate above the contact with Piedmont rock. No oil or gas shows were reported. All three wells were plugged and abandoned.
Based on COCORP seismic profiles, Petersen et al. (1984) characterize the Riddleville basin as an asymmetric graben having the main basin-bounding fault on the northwest side. The basin has been identified originally from magnetic data and a deep water well (Daniels et al., 1983). It trends northeastward for approximately 78 mi (125 km), has a width of approximately 30 mi (48 km) (Heffner, 2013), and is approximately 8,200 ft (2,499 m) deep (Petersen et al., 1984).
Porosity and permeability values for potential conventional reservoir rocks in the Riddleville basin have not been reported. They appear to be uniformly low based on examination of select wireline porosity logs (Williams et al., 2013). Potential source rocks have not been identified in well cuttings or cores taken from the basin or suggested by high gamma ray curve excursion on wireline logs. Consequently, the potential source rocks for the reported oil and gas shows in the No. 1 Taylor well are either suggested to be below the total depth of the well or located deeper in the basin, towards the southeastward. Well data are currently insufficient to determine the thermal maturity of the basin and to estimate the critical moment and the amount of overburden originally present but now removed.
Colleton-Dorchester basin (Ehrhardt basin of Heffner, 2013) (South Carolina)
The Colleton-Dorchester basin has been the site of significant seismic acquisition and the drilling of two significant wells in the Mesozoic synrift basins of the eastern U.S (Basin 38c, Table 1, Fig. 1). In 1984, the SEPCO (operating as Essex in partnership with Texaco) No. 1 Lightsey well (Colleton County, South Carolina) drilled over 10,760 ft (3,280 m) of Newark Supergroup strata before being plugged and abandoned following mechanical failure while drilling to a total depth of 12,750 ft (3,886 m) (Table 2). No shows were reported in this test of the 14,300 ft (4,359 m) deep basin (Smith and Foley, 1988). The well encountered a full suite of Newark Supergroup sedimentary rocks and several diabase intervals. Traverse (1985, 1987) identified fossil pollen species indicative of Late Triassic (Late Carnian) age sedimentary rocks. These data appear to provide the only public biostratigraphic information on the Mesozoic synrift section of the South Georgia basin complex.
Presumably, other efforts to define this interval were barren of diagnostic flora or fauna (e.g., Gohn et al., 1983). Smith and Foley (1988) stated that the entire synrift sedimentary section is Triassic, whereas three of the eleven diabase intrusions were dated as Early to Middle Jurassic. No specific potential source rocks were identified; however, several thin zones of gray, carbonaceous shale were reported on the mud log. Occasional bits of coal or possibly solid hydrocarbon were also observed while drilling. Several thin (less than ~2 ft (0.6 m) thick) wireline log gamma ray spikes suggestive of organic shale or coal may be present between 4,950 ft (1,509 m) and 7,450 ft (2271 m). Whole core analysis of two cores showed porosity ranged from 1.6% to 6.2%, and permeability values were consistently less than 0.01 mD. Sidewall cores, however, had substantially higher porosity and permeability values: 22.6% to 32.5% porosity and 1.5 mD to 8.9 mD permeability (determined empirically). These values were likely elevated by the percussion impact of the sidewall-coring device (Core Laboratories, 1984, unpublished core analysis).
More recently, the Rizer No. 1 Test Boring was made approximately 2 mi southwest of the Lightsey well in order to test the potential for CO2 sequestration (Brantley et al., 2012; Waddell, 2013). Preliminary results from this effort confirmed the low porosity and permeability values reported from the Lightsey well (Waddell, 2013): average porosity values of 2.6% (vertical) and 3.1% (horizontal) and permeability values of 0.0032 mD (air, vertical) and 0.0049 mD (air, horizontal). Drilling operations from near the bottom of the well recovered a core consisting of finely laminated dark gray and light gray strata suggestive of lacustrine microturbidites (Brantley et al., 2012). The laminated interval in the Rizer well does not appear to correlate to the Lightsey well, suggesting that potential source rocks may not be continuous across the basin/subbasin possibly due to penecontemporaneous fault segmentation of synrift depositional environments. Reprocessing of several 2D seismic lines in the area, support an interpretation of rift basin inversion along transpressive fault zones that appear to compartmentalize the subbasin as well as the larger South Georgia rift basin complex (Brantley et al., 2012; Waddell, 2013).
Jedburg basin (Jedberg basin of Heffner, 2013) (South Carolina)
The Jedburg basin, apparently an extension of the Colleton-Dorchester basin, was originally described by Behrendt (1985, 1986) from 2D reflection seismic profiling associated with USGS research into the 1886 Charleston, South Carolina, earthquake (Basin 38d, Table 1; Fig. 1). Within the Jedburg basin, an early (1920 or 1921) oil and gas exploratory well was drilled to a depth of at least 2,570 ft (783 m) (Table 2). The well penetrated the Late Cretaceous-Late Triassic(?) unconformity at 1,580 ft (482 m) and continued through 990 ft (302 m) of Newark Supergroup strata of mixed coarse-grained sandstone and gravel, variably colored clay beds and diabase (Cooke, 1936).
The basin is interpreted to be approximately 76 mi (122 km) long and 27 mi (43 km) wide (Heffner, 2013), and may be as deep as 9,800 ft (2,987 m) (Costain and Çoruh, 1989). The basin is capped by the “J” seismic reflector. Other than the description supplied by Cooke (1936) and data from water wells within the basin (Blount et al., 2011), there is little information to characterize potential source, reservoir, sealing rocks, or thermal maturity of the strata within the basin. No oil or gas shows have been reported. The presence of a thick overlying basalt implies that the area may be more thermally mature than a potentially similar area without the igneous influence, depending on whether or not the igneous activity was accompanied by sustained increased heat flow. Using various geological, geophysical, and topographical data, Rhea (1989) recognizes that this general area has undergone repeated uplift, although the study did not estimate a net cumulative uplift at the level of the “J” reflector.
The South Georgia basin complex and its individual subbasins were not assessed for oil and gas resources by the USGS. However, a composite, continuous gas petroleum system was recognized as present (Milici et al., 2012).
North Florida basin (Georgia and Florida)
The composite South Georgia basin continues southward into north Florida, where it is bifurcated by the Brunswick magnetic anomaly (Fig. 2; Daniels, et al., 1983; Nelson et al., 1985b), which is used to demark the northern boundary of the North Florida basin (Basin 39, Table 1; Fig. 1). This basin complex is defined by several wells, potential field geophysical data, and 2D reflection seismic profiles (Ball et al., 1988; Blount, et al., 2011; Dutch, 2013; Nelson et al., 1985a; SEI Seismic Exchange, 2014). The outline for the USGS resource assessment examination has been significantly influenced by the work of Sartain and See (1997). The North Florida basin complex as used in this paper is approximately 160 mi (257 km) long and 100 mi (160 km) wide and includes the Tallahassee Graben of Smith (1983) and at least two Mesozoic grabens identified by Dobson and Buffler (1991) on the Middle Ground Arch of offshore and coastal Florida. The depth to prerift basement varies from approximately 6,000 ft (1,829 m) in the northern part of the basin complex (Mitchell County, Georgia) to over 13,000 ft (3,962 m) in the southern part (offshore Franklin County, Florida) (Williams et al., 2013).
The Mesozoic synrift basins of north Florida continue westward into the Florida Panhandle and southern Alabama as the eastern extent of the Eagle Mills rift basin trend of the Gulf Coast basin (Basin 40, Table 1; Fig. 1) and southeastward along the trend of the Florida Escarpment (Bartok, 1993). The western boundary of the North Florida basin in this paper as set here is an arbitrary line dividing the more northeast–southwest trend of the east coast Mesozoic synrift basins from the more east-west trend of the Gulf Coast Eagle Mills rift basin trend.
The presence of early Mesozoic rocks in this basin was established by the Stanolind No. 1 Pullen well (Mitchell County, Georgia; Applin, 1951). Wells that have reached the Mesozoic synrift sedimentary rocks in the North Florida basin complex usually did not penetrate more than a few tens to a hundred feet into synrift strata, because the synrift strata was considered economic basement.
Lithologies of potential source rocks have not been reported. The synrift section consists of interbedded low porosity and permeability sandstone and mudstone and a significant thickness of mafic igneous intrusions (Blount et al., 2011; Mitchell-Tapping, 1982). Where the Jurassic salt overlies the synrift section, seismic imaging of potential structural and combination traps is difficult to nearly impossible.
The North Florida basin complex and its individual subbasins were not assessed for oil and gas resources; rather, a composite, continuous gas petroleum system was recognized as potentially present (Milici et al., 2012).
Offshore synrift basins (Federal Outer Continental Shelf areas)
Offshore, on the U.S. Outer Continental Shelf, 2D reflection seismic, gravity, and magnetic data can be used to confirm at least seven Late Triassic–Early Jurassic synrift basins that are often composed of multiple subbasins (Fig. 1). These include the Norfolk, New York Bight, Long Island, Nantucket, Atlantis, Franklin, and Yarmouth basins. In addition, other small, shallow, named, and unnamed basins occur on the Atlantic continental shelf, primarily in the Gulf of Maine (Ballard and Uchupi, 1972; Benson, 1992; Post and Coleman, 2015, this volume; Uchupi and Ballard, 1975; Uchupi and Bolmer, 2008), Many of these basins have undergone recent reexamination and their petroleum potential is discussed in Post and Coleman (2015, this volume).
The East Coast Mesozoic synrift basins of the eastern U.S. formed immediately prior to the birth of the Atlantic Ocean in the earliest Jurassic, between North America and West Africa. Preexisting weak zones in Alleghenian thrust faults and cross-strike structural discontinuities failed during extensional and translational faulting as Pangea broke apart, localizing and delimiting the main extensional fault zones. These rift basins extend from offshore Maine to western Florida and from the Appalachian Piedmont and Blue Ridge provinces to the Atlantic continental shelf. Either because of differential initial basin formation, postrift uplift, or both, the outcropping strata of the basins vary from Late Triassic (Carnian–Norian) to Early Jurassic (Hettangian–Pliensbachian) in age (Fig. 5).
In well-developed basins, cycles of lacustrine and fluvial-fan delta sedimentation are preserved. Organicrich source rocks are common in all of the exposed basins; presumably, similar strata are present in many of the buried basins, as well, even though only limited data suggest their possible widespread occurrence. Sandstone reservoirs capable of flowing economically attractive rates are known only from water well data in the Newark basin (Rima et al., 1962). Whereas, the Richmond, Taylorsville, Deep River, and Newark basins have been tested by “explorationally-smart” wells, all of the buried basins are inadequately explored to determine their true hydrocarbon potential.
Oil and gas shows are common in many of the basins from Georgia to New England. The oil recovered in a few tests has been typed with extract from Newark Supergroup source rocks in that basin (Cornet, 1990, unpublished). Four-way closed structures are not common in the East Coast Mesozoic synrift basins. The most common conventional accumulation traps are likely to be fault-related three-way structural and combination traps, stratigraphic traps on a faulted ramp or nose, or pore-throat traps in a variety of reservoir settings. Sufficiently impermeable beds within the basins are present for lateral and possible vertical conventional seals. The uncertainty of an upper-level, basin-wide seal, in place prior to oil generation and migration and still preserved within the basin, may be the single, most high risk factor in conventional hydrocarbon exploration.
Jurassic-age mafic igneous activity has affected basin prospects throughout the trend. However, the metamorphic aureoles are usually only a few 10's to 100's of feet (3 to 30 m) thick and do not adversely affect most of the basin. Nevertheless, possible hydrothermal fluids associated with igneous emplacement may have diagenetically altered potential reservoir rocks such that they are damaged when exposed to conventional oil and gas field drilling and completion fluids.
Points of concern
Two areas of concern remain to be addressed: (1) the wide-spread presence of possible drilling and completion fluid-reactive minerals in potential reservoir sandstones, such as laumonite and/or chlorite, and (2) the consequences of postrift uplift and erosion prior to the deposition of a potential long-term, effective, upper-level seal before the migration of oil.
The possible presence of fluid-reactive minerals in potential reservoir sandstones: This review notes that several wells having significantly large oil or gas shows could not be successfully tested or completed following acidizing and, in some instances, fracture stimulation. The data are insufficient to draw substantiated conclusions; however, a condition may exist whereby common oil field acidizing processes may fatally damage these synrift sandstone reservoirs.
In 1985, SEPCO (operating as Shore) drilled the No. 1 Hicks well, a 4,580 ft (1396 m) basement test in the Richmond basin (Table 2). The well was drilled on the updip edge of a rotated fault block having dubious structural closure (Gore and Olsen, 1989). Gearheart wireline cores were taken in prospective reservoir rocks. Three sands had calculated low permeability pay between 2,890 ft (881 m) and 3,060 ft (933 m). Porosities ranged from 7 to 12%. Permeabilities ranged from 0.08 to 18 mD, but most were below 9 mD. The core data confirmed the wireline log analysis.
Special core analysis performed by Core Laboratories, Dallas (unpublished, 1985) states that the potential reservoir rocks at 2,912 ft (888 m) and 3,052 ft (930 m) have reduced porosity and permeability as a result of pore-lining and pore-filling laumontite (a low temperature calcium zeolite). In these two samples, laumontite was the most abundant cementing agent even though it occurred in minor amounts. In this setting, it probably formed after oil migration, sealing off oil-filled pores. Laumontite was assumed to have resulted from alteration of feldspar grains within the rock and is a widespread accessory mineral in Late Triassic–Early Jurassic synrift rocks from Georgia to Nova Scotia (Aumento, 1966; Chowns and Williams, 1983; Van Houten, 1962).
In other hydrocarbon basins, laumontite appears to be related to a late, hydrothermal event. However, in the East Coast Mesozoic synrift basins, laumontite occurs not only in the basin sandstones but also in surrounding rock, indicating that the entire area has undergone a regional diagenetic event (Chowns and Williams, 1983). Laumontite commonly replaces feldspar grains and matrix and fills voids or fractures. It may, in fact, form an upper level, diagenetic seal. Production beneath laumontite-bearing rocks is present at Kettleman North Dome field (San Joaquin Valley, California) and McAllen Ranch Field (Vicksburg Formation, Hidalgo County, Texas) (Crossey et al., 1984). When exposed to acid, laumontite will alter to kaolinite, silica, water, and free calcium ions.
While the presence of laumontite is widespread in the Newark Supergroup rift basins, it is not uniformly distributed. Sandstones derived from medium- to highgrade metamorphic rocks appear to be more susceptible to laumontite crystallization than sandstones from felsic igneous rocks (Chowns and Williams, 1983). All that is needed, apparently, is feldspar, calcite, and clay minerals, and high pH-low pCO2 fluids (Chowns and Williams, 1983; Crossey et al., 1984). Laumontite is also a frequent alteration product of the zeolite analcite, a very common constituent of the mudstones of the Lockatong Formation of the Newark basin (Blatt et al., 1972). Because analcite forms in response to high rates of evaporation in the lacustrine basins, it may have been a significant constituent of many Late Triassic-Early Jurassic synrift lacustrine-deposited units.
Chlorite is present as pore-lining films in some Richmond basin samples. Chloriterim coating, in the past, has been shown to be the primary agent for porosity retention in otherwise low porosity rocks. If such is the case in the normally tight Newark Supergroup sandstones, chlorite, while not widespread, may be present only in porous rocks, i.e., those which one would try to complete as a reservoir. Iron-rich chlorite will tend to produce an iron hydroxide gel upon reaction with hydrochloric acid, inhibiting reservoir performance. Other minerals may also be present and potentially affect reservoir performance, but these two apparently are the most critical.
The other point of concern is represented by the consequences of postrift uplift and erosion prior to the deposition of a potential long-term, effective, upper-level seal before the migration of oil. Data are insufficient to accurately model the thermal history of each basin. However, those models which have been developed indicate that most of the basins evolved in a similar fashion. All basins appear to have developed along prerift Paleozoic or Proterozoic zones of weakness, with the southern basins potentially developing and ceasing subsidence earlier, i.e., Triassic Early Carnian to Middle Norian for the southern basins ranging to Early Jurassic Hettangian to Sinemurian in the northern basins (Fig. 5).
Source rocks, where identified, developed at slightly different times in most of the basins. However, all likely went through peak oil level of thermal maturity at approximately the the time of the Triassic–Jurassic boundary. Data from several basins (e.g., Newark, Taylorsville) suggest one or more periods of structural readjustment and possible inversion during the primary period of basin subsidence. However, the main episode of uplift and exhumation probably began during the latest Triassic to early Jurassic, and this primary phase of structural inversion and exhumation was probably complete by Middle Jurassic. This event essentially stopped all thermal maturation of synrift source rocks through a reduction of temperature and pressure.
Additional episodes of uplift also occurred in the Early Cretaceous and Tertiary (possibly Eocene and Miocene). With the possible exception of the postrift succession in the offshore New England rift basins (Klitgord and Hutchinson, 1985) and the Apalachicola Embayment of panhandle Florida (Mitchell-Tapping, 1982), the oldest postrift sedimentary succession of any significant thickness over the rift basins in the onshore and state waters is likely the Lower Cretaceous lower Potomac Group and its equivalents (Dorf, 1952; Hansen, 1969; Trapp, 1992). This interval is typically a basal coarse-grained sandstone with minor siltstone and mudstone beds: a poor potential upper-level seal. However, Owens and Gohn (1985, their Figure 2-10) illustrate several thick (100 ft, 30 m) prodelta clay units of the Lower Cretaceous Potomac Group that may be effective, local seals. In the offshore New England basins and the Apalachicola Embayment, Middle to Late Jurassic sequences onlap the synrift strata. Internal, but not basin wide, seals are obviously present in most of the thick synrift sequences.
With the clear evidence that most (if not all) rift basins of the onshore and state waters areas were uplifted and exhumed during the first half of the Jurassic, it is equally clear that any hydrocarbons that were generated and migrated into available traps were also eroded with the 0.6 to 1.9 mi (1 to 3 km) of latest synrift–earliest postrift or sag strata. If hydrocarbons were not eroded, then they were affected by the decrease in overburden, temperature, and pressure caused by overburden removal. Where those basins came into contact with the local or regional aquifer system, hydrologic and biochemical factors were also likely at work reducing the quantity and quality of any accessible liquid hydrocarbon accumulations.
Even though the data are limited, the following characteristics of multiple East Coast Mesozoic synrift basins are consistent with those of basins with petroleum accumulations and host reservoirs substantially modified by uplift and induced topographically-drive fluid flow (Corcoran and Doré, 2002; Cramer et al., 2002; Doré et al., 2002; Parnell, 2002):
The apparent lack of conventional structurally trapped petroleum accumulations (or conversely, the potential widespread presence of unconventional [continuous] petroleum accumulations; e.g., the Newark, Richmond, Taylorsville, and Deep River basins).
Relatively small conventional petroleum accumulations and a disproportionate volume of gas over oil, probably because of the devolatilization and differential loss of light weight hydrocarbons due to diffusion; e.g., the Richmond and Deep River basins.
Variably pressured basins, ranging from slightly overpressured, normally pressured, to probably underpressured; e.g., the Newark, Richmond, Taylorsville, Deep River, and Dunbarton basins.
Development of natural hydraulic fractures as the uplifted synrift section is differentially depressurized; e.g., the Richmond and Deep River basins.
Development of a topographically driven flow system and likely accompaniment of exsolution of gas from low-permeability reservoirs potentially producing a gas-rich system; e.g., the Taylorsville basin.
Suite of postdepositional diagenetic mineralization and mineral alteration; e.g., the Hartford and South Georgia basins.
Fracture-dependent fluid flow; e.g., the Hartford and Gettysburg basins.
Thermally mature and compacted deep basin rocks at or near surface; e.g., the Hartford, Newark, Gettysburg, Culpeper, Richmond, Taylorsville, and Deep River basins.
“Unfulfilled expectations”; e.g., the Newark, Richmond, Taylorsville, Deep River, and South Georgia basins.
Conventional seal failure; e.g., the Hartford, Taylorsville, Deep River and Richmond basins.
With no production statistics to examine from which to project potential future discoveries, the USGS assessment team reviewed potential productive rift analogs (Charpentier et al., 2007) and potential continuous resource (tight gas accumulation) analogs (Fig. 6; U. S. Geological Survey Oil and Gas Assessment Team, 2012). Of the 2,677 non-US assessment units (AUs) catalogued in Charpentier et al. (2007), 94 are in rift settings and 63 of these are in Mesozoic synrift basins. Of these 63, 32 have Cretaceous source rocks, 28 have Jurassic source rocks, and three have Triassic source rocks. The three with Triassic source rocks are the South Mangyshlak AU (Middle Caspian Basin); Locker-Mungaroo/Barrow AU (Northwest Shelf Australia); and Zala-Drava-Sava Basins AU (Pannonian Basin). These three have marine source rocks; only the Zala-Drava-Sava Basins have some lacustrine source contribution. The time of peak maturation for these basins is in the Tertiary. None of these rift basin AUs having Triassic source rocks have been used as analogs, because peak source rock maturation is in the Tertiary and only one has some lacustrine source contribution mixed with a marine component.
In 1981, production was established from Triassic reservoirs in the Essaouira basin of Morocco (Table 2; Anonymous, 1986). This Triassic rift basin lies on the conjugate Atlantic margin opposite the Fundy basin of Nova Scotia, Canada (Figs. 1, 6). This discovery supported decisions by several companies to conduct exploration drilling the United States during the 1980s. However, the petroleum system setting for the Triassic reservoirs of the Essaouira basin is significantly different from that of the U.S. Atlantic Mesozoic synrift basins. In the Essaouira basin, the probable source rocks are either prerift Silurian shales or perhaps Carboniferous coals, and the seal is Upper Triassic (“Keuper”)–Lower Lias salt (Hafid, 2000). This potential analog also has not been used, because its petroleum system elements of source and seal are so different from those demonstrated by outcrop data and drilling in the eastern U.S. Mesozoic synrift basins.
A suite of data from USGS assessments of tight gas accumulations in the United States (variable reservoir rock lithologies with shale and coal source rocks) had previously been collected and merged, so that a full spectrum of such accumulations could be compared with one another (U.S. Geological Survey Oil and Gas Assessment Team, 2012). By comparing the petroleum system characteristics of the east coast onshore and state waters Mesozoic synrift basins with those of previously assessed 46 tight gas AUs from this compilation, the USGS assessment team estimated undiscovered, technically recoverable oil and natural gas resource potential of the Newark Supergroup basins. The results of the USGS resource assessment show that the resource potential of the East Coast Mesozoic onshore and state waters rift basins fall within the range of other previously assessed tight gas AUs (Fig. 7).
The USGS examined 13 basins and assessed five of these: the Taylorsville, Richmond, south Newark, Deep River, and Dan River-Danville basins (Table 1, Milici et al., 2012). During the review, it was clear that conventional accumulations (i.e., those with a downdip water level and gravity segregated pay) were probably not present in volumes greater than the minimum level for the assessment. The presence of unconventional accumulations was clearly apparent in those basins with sufficient data in the form of coal-bed gas, shale oil and gas, and tight gas sandstones. The review team concluded that mean values of 3.86 TCFG (109 BCM) and 135 MMBNGL (21.5 MMCM) were likely present in undiscovered, technically recoverable volumes in the five assessed basins. The total range of potential resources extended from 1.774 TCFG (50 BCM) (at the 95% fractile) to 7.0565 TCFG (200 BCM) (at the 5% fractile) and 56 MMBNGL (8.9 MMCM) (at the 95% fractile) to 260 MMBNGL (41.3 MMCM) (at the 5% fractile). No volumes for free oil were thought to be present in quantities larger than the minimum level for national assessments, so no oil volumes were estimated (Milici et al., 2012).
At least 39 Mesozoic synrift basins (as individual or composite basins) occur in the eastern United States (onshore and state waters) from New England to north Florida. At least five of these basins have identified or postulated petroleum systems with all of the critical elements for generation, migration, and accumulation. The USGS has estimated that these five basins have cumulative mean estimated undiscovered, technically recoverable resources of 3.86 TCFG (109 BCM) and 135 MMBNGL (21.5 MMCM) (Milici et al., 2012). Eight other basins have been examined but not assessed because of the lack of sufficient petroleum system data. An additional 26 other basins also have been examined for this study to provide further insight into the overall history of the petroleum systems of east coast Mesozoic synrift basins.
Within the studied exposed basins, rifting probably began in the Late Triassic Carnian and continued with minimal disruption until probably the Late Triassic Norian in the Richmond basin, the middle of the Norian in the other southern exposed basins (Deep River, Dan River-Danville, and Taylorsville basins), and Early Jurassic in the northern exposed basins (Culpeper, Gettysburg, Newark, and Hartford basins). The major basin-bounding faults appear to be rejuvenated Paleozoic Alleghenian or Taconic thrust faults or fault zones, and occur on the west or northwest side of all basins except the Hartford, Pomperaug, and Deep River basins. Five basins (Buena, Davie County, Greenwood, Bridgeville, and North Florida basins) appear to be more symmetrical grabens having major faults on both sides of the basin. Available data suggest that most if not all of the onshore and state waters rift basins have undergone uplift and exhumation. Insufficient thermal maturation data limit the accuracy of estimation of the amount of removed overburden from these basins, but amounts probably range from about 3,200 to 11,000 ft (975 to 3,353 m). Two of those basins, the Richmond and Deep River basins, have the least amount of overburden removal and the best record of detected oil and gas.
For those basins with sufficient data to make the estimate, the critical moment ranges from near the end of the Triassic to the Middle Jurassic. With the cessation of subsidence and the beginning of basin inversion and uplift (and coincidently the end of petroleum generation), the potential for conventional hydrocarbon seal and trap rupture and charge leakage became high. Limited data suggests that many of the onshore east coast Mesozoic synrift basins have undergone some degree of topographically driven fluid flow and water washing of conventional hydrocarbon accumulations. Natural gas enrichment likely occurred throughout the basins as a result of devolatilization of oils and gas exsolution from deep basin pore waters.
Enhanced fracture networks developed during uplift, along with the development of new fractures and/or the dilation of old (Alleghenian) fracture systems. All of these processes occurred during the Middle and Late Jurassic and prior to the Early Cretaceous. Continental shelf subsidence, during which major coastal plain fluvial sediment dispersal systems developed and were preserved by subsequent marine inundation and deposition continued from the Jurassic to the Holocene. For most of the east coast onshore and state waters Mesozoic synrift basins, this period of exposure and erosion lasted for at least 70 million years in the southern basins and 50 million years in the northern basins. Hence, petroleum preservation is the main risk for conventional play development within these basins. Until modern drilling and completion technology is attempted in these basins, the full potential for converting unconventional (continuous) resources to reserves remains to be evaluated.
The authors wish to thank Tina Roberts-Asby and David Brown for their thoughtful reviews and comments. We thank Robert Ryder (USGS, retired) for generously sharing decades of informal information collected on the east coast Mesozoic basins from a variety of sources. Steve Snyder created the original aeromagnetic anomaly map from which Figure 2 was developed. We also thank the attendees and participants at the USGS sponsored 2008, 2009, and 2010 U.S. Mesozoic Basins Energy Resources Potential Workshops with the state geological surveys of the eastern U.S. and our friends and colleagues David Brown, MaryAnn Malinconico, Jeff Reid, and Martha Withjack.
Figures & Tables
Petroleum Systems in “Rift” Basins
- Atlantic Ocean
- basin analysis
- case studies
- Eastern Canada
- Eastern U.S.
- Maritime Provinces
- natural gas
- North Atlantic
- Northwest Atlantic
- Nova Scotia
- petroleum exploration
- potential deposits
- sedimentary basins
- sedimentary rocks
- structural controls
- thermal maturity
- United States