Search

Middle and Upper Devonian gas shales of the Appalachian Basin host a sequence of mode I cracks from early veins to late unmineralized joints. These fractures develop during a cycle of burial and exhumation starting with horizontal veins, then vertical veins, then vertical joints, and ending with pancake joints. Veins are most common in the cross-fold orientation in gas shale of the deep hinterland, whereas joints are common in the shallow foreland fringe. Of particular interest for the recovery of natural gas are joints striking northeast to east-northeast in shallow core and outcrops of Appalachian gas shales, particularly in the underpressured, southwestern part of the basin. Northeast to east-northeast fractures are sparse in the deeper, northeastern part of the basin where the Marcellus gas shale is highly overpressured. The common occurrence of unmineralized joints in shallow gas shales is best explained with a stress–pressure–depth model for the evolution of pore pressure in the Appalachian Basin. Our model shows that by Skempton’s poroelastic relaxation, gas-charged porosity can drive natural hydraulic fractures during exhumation. In the shallow foreland fringe of the Appalachian Basin, exhumation was sufficient to favor the propagation of east-northeast–striking neotectonic joints in the contemporary tectonic stress field. These joints appear mainly at shallow depths where they facilitate leakage to form an underpressured gas column trapped by groundwater (i.e., Big Sandy gas field, Kentucky). Exhumation was insufficient for Skempton’s poroelastic relaxation to favor propagation of east-northeast joints in the deep, central part of the basin where capillary forces maintain overpressured gas shale.

The Middle Paleozoic section of the Appalachian Plateau exhibits a mechanical stratigraphy defined by layers that emit seismic energy with unique signatures in response to a strain energy accumulated on time scales associated with local, regional, and plate-scale processes. The Earth is in a state of frictional equilibrium, which means that even small changes in effective stress cause brittle failure and the concomitant release of ambient seismic energy. Stress changes as low as 0.001 MPa, the level of stress changes during Earth tides or the transmission of a fluid pressure wave, can activate failure on critically oriented fractures. These phenomena lead to a release of ambient seismic energy, which can be mapped using seismic emission tomography (SET) methods to image fracture networks emitting coherent seismic waves. We used a buried array of 54 sondes to identify active fracture networks over a contiguous volume of 3.76 km3 within Middle Paleozoic rocks hosting two Marcellus gas shale wells drilled under the Appalachian Plateau of Lycoming County, Pennsylvania, USA. We sampled ambient seismic emissions before and after two stimulations and found that the pattern was repeatable. The fracture patterns illuminated by ambient seismic emissions defined a mechanical stratigraphy populated by clouds of seismic activity separated by packages of beds emitting relatively less seismic energy. The unique attribute of the beds emitting less seismic energy is a lower least horizontal stress (Shmin) relative to adjacent mechanical units in the section. These low stress beds include the bottom portion of both the Marcellus and Burket/Geneseo black shales. There are three thicker mechanical units carrying clouds of higher energy emissions. These three units include siltstones of the Brallier above the Burket/Geneseo package, silty shale beds of the Mahantango between the Marcellus and Burket/Geneseo packages, and Silurian-Devonian carbonates below the Marcellus package. In map view, emission patterns in the Brallier follow Alleghanian J2 joints. Patterns in the Mahantango are consistent with slip along columnar joint zones like those cutting upward in outcrops of shale on the Appalachian Plateau. In sum, SET reveals a mechanical stratigraphy based on the release of strain energy from three major units of the Middle Paleozoic section.

Thermal cracking of organic matter (OM) can enhance gas-shale porosity. However, thermally matured shales are generally characterized by low porosity with high OM content in tectonic exhumation settings. The Wufeng–Longmaxi shale in the Fuling gas field, China, was investigated via bulk analysis and scanning electron microscopy to reveal the influence of tectonic exhumation on the porosity of OM-rich shale. Thermal maturity and organic and mineralogical compositions are regionally homogeneous in the Fuling gas field. Shale porosity is positively correlated with total organic carbon (TOC) and clay mineral content in the gentle and open anticlines, respectively. The OM-hosted pores were preserved with spherical morphologies because of overpressure-associated overburden stress unloading in the gentle anticlines, whereas OM-hosted pores in the open anticlines are characterized by high aspect ratios and are regularly aligned perpendicular to lateral compression stress. A simple simulation confirmed that tectonic folding and compaction during exhumation caused the loss of porosity in the inner arcs of the open anticlines. Only pressure shadows formed by assemblages of clay mineral platelets and mechanically competent grains inhibited tectonic compaction of some OM-hosted pores in the open anticlines. Tectonic compaction of porosity generally increased the mean value of water saturation in the open anticlines. The extremely high water saturation (>60%) in the high-TOC interval (>2 wt. %) was generally associated with natural fractures, suggesting the occurrence of gas leakage.

More than 1 million wells may have been completed using hydraulic fracturing techniques in the USA alone; however, there have been few case studies exploring the impacts on water resources due to methane migration. This study evaluated the results of three investigations initiated by the US Environmental Protection Agency, that were subsequently described in published studies at Dimock in Pennsylvania, Parker-Hood County in Texas, and Pavillion in Wyoming, as well as another study completed at Sugar Run in northeast Pennsylvania. In addition, earlier investigations at Shaws Corner in Pennsylvania, Jackson County in West Virginia, Garfield County in Colorado, and Bainbridge in Ohio are summarized. The most common cause of incidents was the presence of uncemented sections of production casings in wells that allowed gas migration from intermediate depths to shallow freshwater aquifers. In three cases, an inadequate depth of the primary top of cement (TOC) also contributed to impacts. Sources of methane were best identified through analyses of isotopes on samples from production casings, annular spaces, and water wells. In Dimock, some isotope signatures changed with time, after the completion of remedial actions. In Parker-Hood County, where impacts were not related to gas well activity, noble gas analyses were also needed to determine the source of gas. At Pavillion, where maximum methane concentrations in water wells were

The in-situ stress state of the eastern margin of the Sichuan Basin, where there is a complicated fault zone, is poorly understood. This study offers vital insights into the in-situ stress state based on information from 17 wellbores in the Wufeng-Longmaxi Shale near the eastern boundary fault (i.e., the Qiyueshan Fault) of the Sichuan Basin, China. Drilling-induced tensile fractures and borehole breakouts revealed a decrease in the angle between the SHmax (maximum horizontal stress) orientation and fault strike when the wellbores are close to the damage zone of the Qiyueshan Fault. This stress rotation may be caused by the lateral contrast in the Poisson's ratio between the country rock and damage zone of the Qiyueshan Fault. A high angle between the SHmax orientation and fault strike also appeared in the individual fault blocks of the Qiyueshan Fault as a result of local stress concentration. The higher magnitudes of horizontal stresses indicate stronger lateral compression in the Qiyueshan Fault Zone. Owing to the lateral compression of a “sandwich” consisting of a high-E (Young's modulus) roof and low-E shale intervals, natural fractures mainly appeared in the low-E shale intervals when the wellbore was close to the fault damage zone. The low fracture intensity in the low-E shale intervals of the wellbores located at the boundaries of the stress concentration blocks may be associated with an extensional stepover process.

Stress memory in the Wufeng-Longmaxi Shale located in the vicinity of the boundary fault of the Sichuan Basin was extracted by Kaiser effect analysis; subsequently, shale-gas retention capacity as determined by stress memory was assessed with a simulation. Due to gas generation and geothermal heating, the Kaiser effect rarely reflects stress states before the maximum burial depth. The strong deformation within and close to the fault damage zone negated the stress memory from the early exhumation episode; instead, the stress state of the recent exhumation episode was recorded. The maximum effective horizontal stress (σH-eff) at 120 Ma was ∼85 MPa at ∼6000 m, with a maximum sustainable pressure of ∼135 MPa. The porothermoelastic mechanism caused a decrease of the pressure magnitude but an increase of the pore-fluid factor during exhumation. After the Wufeng-Longmaxi Shale was uplifted from ∼6000 m to ∼3500 m, the pore-fluid factor almost reached 1 and horizontal hydrofracturing caused shale-gas leakage. The σH-eff at 20 Ma was ∼73 MPa at ∼3500 m and it had a gradient of 38 MPa/km. Reverse faulting at 20 Ma induced a remarkable pressure drop, which indicates shale-gas leakage at that time. Subsequently, a transition from reverse to strike-slip faulting further induced fluid pressure reduction and shale-gas leakage.

The Teton anticline and adjacent structures at the deformation front of the Sawtooth Range fold-thrust belt in Montana are fractured in such a way that some have used these as a model for the fractures that propagate during buckle folding. However, advances in understanding both the process of folding in forelands and the evolution of fracture patterns found within these folds suggests that it is time to reinterpret the nexus between fracturing and folding within this classic structure. Data gathering includes two steps starting with the documentation of the along-strike changes in the geometry of the Teton anticlines, using field measurements, aerial photographs and seismic lines, and then the collection of fracture orientation data and abutting relationships. With the benefit of seismic lines, Teton anticline is best described as a fault-related fold, forming in response to changes in the underlying ramp and flat geometries. The propagation of joints (mode I cracks) initiated with formation of two major sets whose orientation is controlled by pre-folding, regional stresses. Two more joint sets propagated in local stress fields, developed in response to the growth of fault-related folds. Some early joints were reactivated as wrench faults during amplification and tightening of the anticlines. The fracture sets identified in this study are consistent with two unrelated mechanisms starting with propagation in a regional stress field. This first mechanism is poorly understood but might be related to regional stretching in the Sawtooth Range orocline. The second mechanism is associated with tangential longitudinal strain during anticline development. Thus, we suggest that fracture networks across folded hydrocarbon reservoirs should be interpreted and characterized in the light of a more complex tectonic setting and the genetic or geologic history of the system as a whole.

The Flat Creek Member of the Utica Formation along Canajoharie Creek south of Canajoharie, NY, is populated with unmineralized pancake joints, a term for horizontal, bed-parallel partings. Each pancake joint propagated outward from a central initiation point with radial growth in excess of 0.5 m in some instances. Propagation arrests with out-of-plane growth displaying either a tilt characteristic of an arrest line or bifurcation into a complex set of cylindrical splays or hackles. A rough surface morphology relative to vertical fractures (i.e., joints and veins) in the Flat Creek coupled with bifurcation at arrest are the manifestation of an accelerating crack tip reaching critical growth at the shear-wave velocity of the shale. The lack of mineralization and containment between vertical fractures are consistent with propagation during exhumation. One driving mechanism for a late propagation event could be axial splitting which places pancake joints in a class of brittle fracture that includes sheeting joints. The high-speed propagation is a manifestation of the rapid release of fracture energy characteristic of sheeting joints. However, we also consider a second driving mechanism, natural hydraulic fracturing. In the Appalachian Basin, high pore pressure is maintained in gas shale at depths less than 50% of the original overburden. If abnormal pore pressure is maintained for 300 my or more, it is present when exhumation carries the shale upward through the crossover depth where the least stress becomes vertical. If pore pressure exceeds the vertical stress, then pore pressure will lift overburden. A possible mechanism for maintaining pore pressure during exhumation to shallow depths is a poroelastic phenomenon known as Skemption's behavior. We posit that pancake joints differ from common sheeting joints because of the additional contribution from abnormal pore pressure as a driving mechanism that allows for deeper propagation depth.

Extensive development of shale gas has generated some concerns about environmental impacts such as the migration of natural gas into water resources. We studied high gas concentrations in waters at a site near Marcellus Shale gas wells to determine the geological explanations and geochemical implications. The local geology may explain why methane has discharged for 7 years into groundwater, a stream, and the atmosphere. Gas may migrate easily near the gas wells in this location where the Marcellus Shale dips significantly, is shallow (∼1 km), and is more fractured. Methane and ethane concentrations in local water wells increased after gas development compared with predrilling concentrations reported in the region. Noble gas and isotopic evidence are consistent with the upward migration of gas from the Marcellus Formation in a free-gas phase. This upflow results in microbially mediated oxidation near the surface. Iron concentrations also increased following the increase of natural gas concentrations in domestic water wells. After several months, both iron and SO42− concentrations dropped. These observations are attributed to iron and SO42− reduction associated with newly elevated concentrations of methane. These temporal trends, as well as data from other areas with reported leaks, document a way to distinguish newly migrated methane from preexisting sources of gas. This study thus documents both geologically risky areas and geochemical signatures of iron and SO42− that could distinguish newly leaked methane from older methane sources in aquifers.

In order to assess the nature of the Appalachian Basin Quiet Zone (ABQZ), a region of low seismic activity, a minimum horizontal stress profile from sonic logs was calibrated using hydraulic fracture stress measurements in a well from McKean County, PA. The calibrated minimum horizontal stress profile that best fits with the hydraulic fracture stress data was established using an extended Eaton's model that treats rock as a transverse isotropic material with a vertical symmetry axis (TIV). To achieve a best fit, several assumptions are introduced: (1) a constant tectonically-induced elastic strain of eHmax = 3.3 × 10−4, (2) an anisotropic velocity model ANNIE approximation for shales, and (3) a bimodal pore pressure distribution with abnormally high pore pressure in the Marcellus gas shale. Other calibration methods assume an isotropic medium, a constant tectonic strain gradient corresponding to zero stress at ground surface, or constraints from sidewall core data to estimate stiffness constants. By testing four calibration methods, we find that stress profiles are affected by multiple factors, including the ratio of maximum tectonically-induced elastic strain over minimum tectonic strain, pore pressure in the Marcellus, Biot's coefficients, and elastic properties. Sensitivity analysis was conducted to analyze the influence of these parameters. Our stress profile is most influenced by pore pressure and tectonically-induced elastic strain while other factors had much smaller effects. Tectonically-included strain in the ABQZ is insufficient to achieve the critical stress necessary to induce frictional slip on favorably oriented faults in the sedimentary cover which presumably speaks to the lack of earthquakes in the basement of the ABQZ. However, induced earthquakes from deep disposal wells suggest that basement at the edge of the ABQZ may be decoupled from the sedimentary cover, in which case the cover has relaxed through a time-dependent deformation mechanism such as pressure solution creep.

On January 1, 2009, a concrete slab covering a water-pump vault of a water well 400 m north of a Marcellus gas well in Dimock, Pennsylvania, USA was reported to have split into three pieces while being overturned. It was suggested that the cycling on of a water pump sparked the deflagration of a methane-air mixture causing the slab to overturn. Here, the conditions necessary to generate an explosion consistent with evidence, mainly a split and overturned concrete slab unmarked by soot or other evidence of a flame, are analyzed. Using more than one approach, calculations show that the maximum pressure to lift the concrete slab was roughly 0.3 bar. Considering among others the flammable range of methane, the explosion pressure as a function of equivalence ratio, the presence of methane gradients inside the vault, the absence of soot and possible ignition sources, the analysis did not yield a well-defined, credible gas explosion scenario to explain the observed damage, although the possibility cannot be ruled out with absolute certainty.

The Marcellus Formation, a Devonian gas shale in the Appalachian Basin, is a heterogeneous rock as the result of a complex depositional, diagenetic, and deformational history. Although it is overpressured over a large portion of its economic area, the origin and distribution of pore pressure within the gas shale are not well-understood. We have used the sonic properties of the Marcellus and statistical analyses to tackle this problem. The sonic data come from a suite of 53 wells including a calibration well in the Appalachian Basin. We first analyze the influence of various extrinsic and intrinsic parameters on sonic velocities with univariate regression analyses. The sonic velocities of the Marcellus in the calibration well generally decrease with an increase in gamma-ray american petroleum institute (API) and increase with density and effective stress. Basin-wide median sonic velocities generally decrease with an increase in median gamma-ray API and pore pressure and increase with burial depth (equivalent confining stress), effective stress, and median density. Abnormal pore pressure is verified by a stronger correlation between the median sonic properties and effective stress using an effective stress coefficient of approximately 0.7 relative to the correlation between the median sonic properties and depth. The relatively small effective stress coefficient may be related to the fact that natural gas, a “soft” fluid, is responsible for a basin-wide overpressure of the Marcellus. Following the univariate regression analyses, we adopt a multiple linear regression model to predict the median sonic velocities in the Marcellus based on median gamma-ray intensity, median density, thickness of the Marcellus, confining pressure, and an inferred pore pressure. Finally, we predict the pore pressure in the Marcellus based on median sonic velocities, median gamma-ray intensity, median density, thickness of the Marcellus, and confining pressure.

Moment tensors are calculated by using the P-wave first motion peak amplitudes of 59 microseismic events with high signal-to-noise ratio. These events are from a surface microseismic data set gathered during hydraulic-fracture stimulation of the Marcellus gas shale in Washington County, Pennsylvania, USA. The majority of these 59 events have a horizontal nodal plane ([Formula: see text] a few degrees) characteristic of a dip-slip/horizontal-slip moment tensor. If the horizontal nodal plane is an auxiliary, the vertical nodal plane has a pure dip-slip motion, which is inconsistent with the opening motion for vertical hydraulic fractures that enables proppant loading. This points to slip on horizontal nodal planes with the auxiliary vertical nodal planes aligned with the local maximum horizontal stress orientation as indicated by drilling-induced fractures in nearby vertical wells. These 59 microseismic events are caused by slip on horizontal mechanical discontinuities such as bedding planes during the opening of vertical hydraulic fractures, a model first proposed by research teams headed by Rutledge and Eisner, respectively. During several stimulation stages in the Washington County Marcellus gas shale, a pattern of opposite slip direction develops within “double lineaments” of microseismic clouds. This suggests that fracking fluid is not only able to move in the direction of fracture propagation, but it can also spread sideways into previously unstimulated rock. A secondary microseismic cloud consistently initiates at approximately 133 m (400 ft) from the position opposite the central perforation toward the unstimulated heel of the horizontal wells. From these moment tensors, we have concluded that microseismic focal mechanisms with horizontal nodal planes are direct evidence of the presence of treatment fluid in open hydraulic fractures.

This study presents paleomagnetic results from the Oatka Creek Member of the Marcellus Formation. Paleomagnetic specimens were subsampled from a 50.8-mm-diameter subsurface drill core obtained from the Valley and Ridge physiographic province of Pennsylvania. Alternating field (AF) demagnetization techniques demonstrate that the Oatka Creek specimens retain multiple components of magnetization. The lowest-coercivity component, component A, was isolated by peak fields up to 20 mT and has characteristics of a drilling-induced remanence. Component B was isolated over a range of peak AF fields between 20 and 50 mT in the upper 100 m of the core, and points south with intermediate downward inclinations in geographic coordinates. Scanning electron microscopy, energy dispersive spectroscopy, and rock magnetic experiments yield evidence that suggests component B is a detrital remanent magnetization in magnetite. We corrected these directions for bedding tilt and inclination shallowing and calculated a paleopole at lat. 29.4°, long. 116.9°, A95 = 1.3°, with an age of 388–389 Ma constrained by biostratigraphy and radiometric dating. This paleomagnetic pole could help fill a data gap in the paleomagnetic record for North America that spans the Middle Devonian to Early Mississippian Periods. Component C was isolated throughout the core using AF fields between 50 and 100 mT and points south with shallow downward inclinations in geographic coordinates. Component C directions agree with a chemical remanence in pyrrhotite from a previous paleomagnetic study of the Marcellus Formation. Component C is likely also a chemical remanent magnetization in pyrrhotite acquired during maximum burial temperatures achieved in the Permian, which may have catalyzed pyrrhotite formation. Component D was isolated in specimens from the lower 20 m of the core, using peak AF fields between 20 and 100 mT, and it points north with shallow downward inclinations in geographic coordinates. The lower 20 m of the core shows evidence of significant hydrothermal alteration. The component D directions yield a paleopole in good agreement with the Jurassic reference pole, and a pulse of hydrothermal alteration in the lower portion of the Oatka Creek Member is consistent with other thermal events in the region ca. 180 Ma. Our results indicate that subsurface cores from the Appalachian foreland may provide an opportunity to greatly improve the Paleozoic paleomagnetic record for North America.

Analysis of two 3D surveys, available well data, published outcrop data and subsurface information, as well as production data available from the state of Pennsylvania, demonstrates that wide-azimuth seismic is sensitive to variations in fracturing at the scale of individual pads or even individual wells. These variations in fracturing begin to explain why production varies significantly, even locally, within the Marcellus Shale gas play. Rose diagrams from quantitative fracture analysis using azimuthal seismic velocity volumes were compared with published data from Appalachian black shale outcrops and subsurface fracture models proposed in various papers to validate the results from subsurface data. These analyses provided insight into the rock fabric and the presence of systematic joints that likely affect production. There was a strong correlation between the low anisotropy and low heterogeneity of anisotropy and high estimated ultimate recovery (EUR). Additionally, interpreted fracture trend azimuths differed between areas of larger gas EUR and areas of smaller gas EUR as defined by decline curve analysis. Some perforations were likely to perform much better than others along the borehole, based on observed heterogeneity in the seismic profiles and map view.

Debate continues over paleoenvironmental conditions that prevail during deposition of organic-carbon (C)-rich marine source rocks in foreland basins and epicontinental seas. The focus of disagreement centers largely on paleowater depth and the prevalence of anoxia/euxinia. The issues of paleodepth and water column conditions are important for prediction of lateral variations in source quality within a basin because the viability of a hydrocarbon play depends on a thorough understanding of the distribution of source rock quality and depositional environments. We used inorganic geochemical data from the Middle Devonian Marcellus Shale in the Appalachian Basin to illustrate interpretive strategies that provided constraints on conditions during deposition. Source evaluation typically relies on the analysis and interpretation of organic geochemical indicators, potentially also providing evidence of the degree of thermal maturity and conditions of the preservation of the organic matter. The Marcellus Formation is thermally mature, making the evaluation of the organic-carbon fraction for geologic interpretation inadequate. Because most labile organic matter has largely been destroyed in the Marcellus Formation, analysis of inorganic elements may be used as an alternative interpretative technique. Several inorganic elements have been correlated to varying depositional settings, allowing for their use as proxies for understanding the paleodepositional environments of formations. A high-resolution geochemical data set has been constructed for the Union Springs Member along a transect of cores from proximal to distal in the Appalachian Basin in central Pennsylvania using major, minor, and trace elemental data. Our results suggested that during deposition, the sediment-water interface, and a portion of the water column, was anoxic to euxinic. As deposition continued, euxinia was periodically interrupted by dysoxia and even oxic conditions, and a greater influx of clastic material occurred. Such variations were likely related to fluctuations in water depth and progradation of deltaic complexes from the eastern margin of the Appalachian Basin.

Between 2005 and 2014 in Pennsylvania, about 4000 Marcellus wells were drilled horizontally and hydraulically fractured for natural gas. During the flowback period after hydrofracturing, 2 to (7 to ) of brine returned to the surface from each horizontal well. This Na-Ca-Cl brine also contains minor radioactive elements, organic compounds, and metals such as Ba and Sr, and cannot by law be discharged untreated into surface waters. The salts increase in concentration to () in later flowback. To develop economic methods of brine disposal, the provenance of brine salts must be understood. Flowback volume generally corresponds to ∼10% to 20% of the injected water. Apparently, the remaining water imbibes into the shale. A mass balance calculation can explain all the salt in the flowback if 2% by volume of the shale initially contains water as capillary-bound or free Appalachian brine. In that case, only 0.1%–0.2% of the brine salt in the shale accessed by one well need be mobilized. Changing salt concentration in flowback can be explained using a model that describes diffusion of salt from brine into millimeter-wide hydrofractures spaced 1 per m (0.3 per ft) that are initially filled by dilute injection water. Although the production lifetimes of Marcellus wells remain unknown, the model predicts that brines will be produced and reach 80% of concentration of initial brines after ∼1 yr. Better understanding of this diffusion could (1) provide better long-term planning for brine disposal; and (2) constrain how the hydrofractures interact with the low-permeability shale matrix.

Both bedding-parallel slickensides and cleavage duplexes are forms of mesoscopic-scale detachment faulting populating black (Marcellus and Geneseo/Burket) and intervening gray (Mahantango) shales of the Middle Devonian, a section known for abnormal pore pressure below the Appalachian Plateau. The abundance and the orientation of slickensides and cleavage duplexes in the more organic-rich black shale relative to gray shale suggests that maturation-related abnormal pore pressure facilitates detachment, a mesoscopic manifestation of the Hubbert–Rubey pore pressure model for overthrust faulting. The former are discrete slip surfaces whereas the latter consists of nested, anastomosing slip surfaces, either cutting through bedding or on disrupted bedding surfaces stacked as mesoscopic versions of thrust duplexes. Cleavage duplexes are between a few cm and over 1 m thick with their hanging walls commonly transported toward the Appalachian foreland, regardless of local limb dip. Cleavage duplexes are most common near the stratigraphic maximum flooding surface, the organic-rich section most prone to develop maturation-related pore pressure in the Middle Devonian gas shales. Bedding-parallel slickensides are somewhat more evenly distributed in the black shale but also found in overlying gray shale. In both black and gray shales, slickensides are more abundant on the limbs of folds, an indication of pore-pressure-related flexural-slip folding. On the macroscopic scale, the Pine Mountain Block of the Southern Appalachian Mountains was enabled by a basal detachment cutting along the Upper Devonian Chattanooga black shale which has a thermal maturity sufficient for the generation of abnormal pore pressure. The Pine Mountain block is a large-scale overthrust showing little evidence of collapse of the hinterland side, a credible example of a pore-pressure-aided overthrust fault block of the type envisioned by the Hubbert–Rubey model.

Two of the major joint-driving mechanisms are joint-normal stretching and poroelastic shrinkage, and these lead to joint sets commonly associated with structural bending and natural hydraulic fracturing, respectively. Regardless of joint-driving mechanism, joint infilling is a nonhomogeneous Poisson process in the presence of stress shadows. Through probability modeling, we show that in all cases joint spacing is best fit with gamma distributions. The shape parameter of the best-fit gamma distribution to joint-spacing data is a quantitative means to assess the extent of saturation, which is represented in a new parameter, the joint-saturation ratio (JSR). To test the utility of JSR, we call upon published structural bending joint data (Elk Basin, Lilstock, and Rives plate-bending experiment). The shape parameters for these well-developed structural bending joints are equal to around three, corresponding to a JSR of approximately 30p. Using the same analysis on the spacing of natural hydraulic fractures collected from outcrops in the gas-prone Devonian sections of the Appalachian Basin, we find that natural hydraulic fractures differ in two aspects from structural bending joints. First, the joint spacing is proportional to bed thickness in bedded rocks but not in gas shale sections. Second, the joint saturation of natural hydraulic fractures is generally lower than in well-developed structural bending joints. Thus, the JSR is a means to distinguish the joint-driving mechanism and to represent joint-saturation level independent of bed thickness effects. It can be used to distinguish natural fractures from drilling-induced fractures and to improve the fracture-network modeling.

Water-filled fractures stimulated by hydraulic fracturing will severely attenuate S-waves. By comparing the P- and S-wave amplitudes generated from perforation shots in the Marcellus gas shale, one can infer the existence and constrain the location of stimulated hydraulic fractures. The waveforms of the perforation shots in unstimulated gas shale are first analyzed to define the prestimulation radiation pattern. Then the change of perforation-shot S-wave waveform is examined for information about the development of hydraulic fractures. S-wave attenuation validates the existence of offstage hydraulic fractures and the invasion of treatment fluid from another neighbor stage. This method might have potential use in designing an active seismic survey to more effectively monitor hydraulic-fracture stimulation beyond the current microseismic mapping techniques.

More than 2 × 10 4 m 3 of water containing additives is commonly injected into a typical horizontal well in gas shale to open fractures and allow gas recovery. Less than half of this treatment water is recovered as flowback or later production brine, and in many cases recovery is

Organic-carbon–rich shales of the lower Marcellus Formation were deposited at the toe and basinward of a prograding clinothem associated with a Mahantango Formation delta complex centered near Harrisburg, Pennsylvania. Distribution of these organic-carbon–rich shales was influenced by shifts in the delta complex driven by changes in rates of accommodation creation and by a topographically high carbonate bank that formed along the Findlay-Algonquin arch during deposition of the Onondaga Formation. Specifically, we interpret the Union Springs member (Shamokin Member of the Marcellus Formation) and the Onondaga Formation as comprising a single third-order depositional sequence. The Onondaga Formation was deposited in the lowstand to transgressive systems tract, and the Union Springs member was deposited in the transgressive, highstand, and falling-stage systems tract. The regional extent of parasequences, systems tracts, and the interpreted depositional sequence suggest that base-level fluctuations were primarily caused by allogenic forcing—eustasy, climate, or regional thermal uplift or subsidence—instead of basement fault reactivation as argued by previous workers. Paleowater depths in the region of Marcellus Formation black mudrock accumulation were at least 330 ft (100 m) as estimated by differences in strata thickness between the northwestern carbonate bank and basinal facies to the southeast. Geochemical analysis indicates anoxic to euxinic bottom-water conditions. These conditions were supported by a deep, stratified basin with a lack of circulation.

As industry evolves in the 21st Century in Africa, there are fundamental questions that are in the best interest of industry to solve in exploration for and production of new prospects and in the revitalization of partially depleted fields. Many of these questions may be addressed using technologies that have evolved during the gas shale revolution of the past decade. This talk will review some of these leading technologies in briefly discussing the following ten questions. What is the state of stress in basins? What information comes from a microseismic survey? How does knowledge of a depositional environment guide the placement of a lateral? How do natural fractures evolve in shales? What information is available when sonic logging is coupled with conventional logs? Where does water go when injected into shale? What are the major subsurface structures that can be detected in multichannel seismic? How does a project deal with the smaller scale (sub-seismic) structures? How is organic carbon distributed in reservoir rock? What are the mechanisms for the development of overpressure in basins?

Extracting gas from shale increases the availability of this resource, but the health and environmental risks may be too high.

Analysis of more than 900 wireline logs indicates that the Middle Devonian Marcellus Formation encompasses two third-order transgressive-regressive (T-R) sequences, MSS1 and MSS2, in ascending order. Compositional elements of the Marcellus Formation crucial to the successful development of this emerging shale gas play, including quartz, clay, carbonate, pyrite, and organic carbon, vary predictably within the proposed sequence-stratigraphic framework. Thickness trends of Marcellus T-R sequences and lithostratigraphic units reflect the interplay of Acadian thrust-load-induced subsidence, short-term base-level fluctuations, and recurrent basement structures. Rapid thickening of both T-R sequences, especially MSS2, toward the northeastern region of the basin preserves a record of greater accommodation space and proximity to clastic sources early in the Acadian orogeny. However, local variations in T-R sequence thickness in the western, more distal, area of the basin may reflect the reactivation of inherited Eocambrian basement structures, including the Rome trough and northwest-striking cross-structural discontinuities, induced by Acadian plate convergence. Episodes of block displacement locally warped the basin into northeast-southwest–trending regions of starved sedimentation and/or erosion adjacent to depocenters in which regressive systems tract deposits were ponded. Block movement appears to have initiated in late Early Devonian time, resulting first in thinning and local erosion of the Oriskany sandstone in northwest Pennsylvania. This study, in addition to providing the basis for a predictive sequence-stratigraphic model that can be used to further Marcellus exploration, tells of a foreland basin more tectonically complex than accounted for by simple flexural models.

The marine Middle and Upper Devonian section of the Appalachian Basin includes several black shale units that carry two regional joint sets (J1 and J2 sets) as observed in outcrop, core, and borehole images. These joints formed close to or at peak burial depth as natural hydraulic fractures induced by abnormal fluid pressures generated during thermal maturation of organic matter. When present together, earlier J1 joints are crosscut by later J2 joints. In outcrops of black shale on the foreland (northwest) side of the Appalachian Basin, the east-northeast–trending J1 set is more closely spaced than the northwest-striking J2 set. However, J2 joints are far more pervasive throughout the exposed Devonian marine clastic section on both sides of the basin. By geological coincidence, the J1 set is nearly parallel the maximum compressive normal stress of the contemporary tectonic stress field (SHmax). Because the contemporary tectonic stress field favors the propagation of hydraulic fracture completions to the east-northeast, fracture stimulation from vertical wells intersects and drains J2 joints. Horizontal drilling and subsequent stimulation benefit from both joint sets. By drilling in the north-northwest–south-southeast directions, horizontal wells cross and drain J1 joints, whenever present. Then, staged hydraulic fracture stimulations, if necessary, run east-northeast (i.e., parallel to the J1 set) under the influence of the contemporary tectonic stress field thereby crosscutting and draining J2 joints.

The oldest joint set in Devonian rocks of western New York state has an atypical NS strike and predates regionally more abundant NW-striking and ENE-striking joints driven by hydrocarbon-related fluid decompression. The NS joints originated in higher modulus diagenetic carbonate and were driven initially by a different mechanism, either joint-normal stretching and/or thermoelastic contraction. The origin of these joints in higher modulus carbonate concretions indicates the presence of a tensile stress produced by uniform regional extensional strain. Upper Devonian shale hosting the NS joints crops out in that area of the Appalachian Basin where a Morrowan erosional unconformity marks the region of maximum upward lithospheric flexure of an early Alleghanian forebulge. The NS strike of these early joints points to a forebulge stretching axis oriented approximately east-west and associated in time and space with crustal loading that drove both the Northfieldian Orogeny and the underplating of the Bronson Hill Anticlinorium in New England. Ultimately, subsidence of the Morrowan forebulge buried the Upper Devonian shale succession to the oil window during the latter part of the Alleghanian tectonic cycle resulting in the propagation of fluid driven NW- and ENE-trending joints in black shale.

Two stages of regional cross-fold jointing refl ect the rearrangement of a lithosphericplate-scale stress fi eld during the Ouachita (i.e., Alleghanian) orogeny. These joints overprint northward-verging thrust-related folds in the Ouachita foreland. Stage-one jointing during the Ouachita orogeny is associated with a stress fi eld recorded by an Atokan and earlier regional joint set, which has a NNE strike. Stage-one jointing is separated into two phases in the central Ouachita salient, where a secondorder recess is marked by a prominent bend in the Windingstair fault below the forelandmost sheet of the central thrust belt. The Big Cedar recess developed when the Windingstair sheet overrode the frontal imbricate zone, thereby causing the Big Cedar pinch zone, a structure refl ecting fold-parallel shortening proximal to localized dextral transpression in the Choctaw thrust sheet. A later regional joint set, representing the second stage of jointing during the Ouachita orogeny, affects rocks in the Arkoma foreland as well as in the fold belt. Stage-two jointing records a stress fi eld associated with NNW-directed closure during the fi nal stage of plate convergence; this stage persisted into the Desmoinesian.

Horizontal bitumen-filled microcracks are common within clay laminae of the finely laminated organic carbon-rich shale in the lower half of the heavily jointed Upper Devonian Dunkirk Shale, western New York state. Such cracks are not found higher in the Dunkirk Shale, where moderate bioturbation resulted in a relatively porous and permeable microfabric. Horizontal microcracks in a hydrocarbon source rock that carries regional vertical joints indicating a horizontal least principal stress owe their presence to material properties of the fractured shale and the magnitude and orientation of the crack-driving stress during kerogen maturation. Three material properties favored the horizontal initiation of microcracks in the Dunkirk Shale: (1) the abundance of flat kerogen grains oriented parallel to layering; (2) a marked strength anisotropy in large part caused by the laminated nature of the rock; and (3) the tight, strongly oriented planar clay-grain fabric produced by gravitational compaction of flocculated clay at shallow-burial depth. The latter was especially important to sustaining elevated pore pressure, the crack-driving stress, which was generated by the conversion of kerogen to bitumen. Poroelastic deformation of the low-permeability laminated shale pressurized by catagenesis, perhaps enhanced by compaction disequilibrium prior to kerogen conversion, elevated the in-situ horizontal stress in excess of the vertical stress, which remained constant during pore-pressure buildup, thereby favoring the propagation of microcracks in the horizontal plane.

Linear elastic fracture mechanics predicts that joint orientation is controlled by the stress field in which the joints propagate. Thus joint sets are effective proxies for stress trajectories during joint growth. Complexity in joint orientation indicates stress trajectory variability, a phenomenon quantified using an eigenvalue method that measures dispersion of joint normal vectors (i.e. poles) around the mean vector. Ratios between the eigenvalues of a joint orientation tensor give the clustering strength (ζ) and the shape factors (γ) of the distribution. A joint set that forms in a relatively isotropic rock subject to a rectilinear stress field should exhibit strong clustering and small random orientation variation that can be described by the Fisher statistical model. However, most joint orientation distributions in bedded rocks have non-random variation, greater in strike than in dip. This relative stability of the vertical stress orientation is strongest when joints are bounded by bedding interfaces, reflecting the tendency for deflection in the local stress field arising from the growth of side cracks, joint segments and adjacent joints in joint zones. Even when joint growth across bedding interfaces indicates negligible strength anisotropy, joint orientation distributions reflect less joint–joint interaction during vertical growth than during horizontal growth. As strike variation grows due to the presence of a non-rectilinear stress field, the orientation data better fit a Kent statistical model. Joint sets formed during fold development or in rocks with irregular bedding boundaries are more weakly clustered with Fisher-like orientation distributions. Orientation distributions for joint sets formed throughout a stress rotation have Kent-like shapes that indicate the magnitude of stress trajectory variation and clustering strength that depends on the joint density at each increment of the stress rotation.

Within the Appalachian plateau detachment sheet of southwestern Pennsylvania tectonic thickening is a central element to the growth of anticlines in a Silurian-Devonian section containing a three-tiered mechanical stratigraphy (that is, a basal detachment zone, a lower imbrication zone, and an upper wedge zone). The detachment zone is predominantly within disturbed shale of the Silurian Vernon Formation which sits above a disturbed surface on the Lockport dolomite. Large-scale anticline growth is, in part, a consequence of small-scale blind imbrication of the more competent mechanical layers in the lower portion of the detachment sheet (the imbrication zone). Here, salt within the Silurian Syracuse Formation hosts secondary detachment responsible for imbrication and the development of triangle zones in the core of the anticlines. Some fold amplification is also accomplished by extensive, smaller-scale thrust wedging and concomitant tectonic thickening of the less competent Devonian section in the upper portion of the detachment sheet (the wedge zone). These spatially periodic anticlines are separated by horizontally bedded synclines characterized by the absence of thrust wedging and fault imbrication across all three zones. Tectonic thickening continued behind the foreland propagation of the detachment-tip line within the Vernon shale as indicated by a systematic hinterland increase in the cross-sectional width and amplitude of the anticlines. On the basis of seismic images, each detachment sheet anticline is situated above prominent, periodically-spaced, pre-Alleghanian structures in the footwall of the detachment sheet. These footwall structures arise from a combination of thrust imbrication at the depth of the Ordovician Trenton Group and high-angle, basement-involved faulting at the depth of the Cambrian Gatesburg Formation. Some of the high-angle fault displacement is a consequence of Alleghanian inversion on faults associated with the extensional Rome Trough and other basement structures developed during the Late Proterozoic rifting of the eastern margin of Laurentia. Evidence for tectonic inversion is present in seismic reflection images that show buttress anticlines in the Cambro-Ordovician carbonate section. The superposition of detachment-sheet anticlines above footwall structures strongly suggests that the foreland transport was disrupted by topographic bumps on the base of the detachment zone. Bending of the detachment sheet over these irregularities may have promoted the spatially periodic collapse and concomitant tectonic thickening in the over-riding sheet as it was pushed laterally toward the foreland. A strain hardening of the detachment zone part of the detachment sheet at the topographic irregularities permitted the growth of double-sided tapered wedges (that is, each anticline). The overall structural profile of the detachment sheet is consistent with the growth of a Coulomb wedge whose low basal friction is interrupted by patches of strain hardening.

Transected joints (i.e. systematic joints that strike at an angle to the present fold axis trend) occur on the flanks of Split Mountain, a Laramide anticline near the eastern end of the Uinta Mountains, Utah. The common orientation on both flanks for these WNW-striking joints is inconsistent with joints driven by a syn-folding stretch normal to the direction of highest curvature. A smaller dispersion of the poles to these transected joints occurs when they are rotated with bedding to their ‘pre-fold’ orientation. This dispersion of poles is inconsistent with a post-fold genesis in a regional stress field but permits the possibility that these WNW joints propagated as a systematic set prior to Laramide folding. A pre-fold interpretation is substantiated by a regional WNW-striking joint set within Cretaceous and older rocks in the surrounding Piceance, Uinta, and southeastern Sand Wash basins. During tilting accompanying the upfolding of Split Mountain, most joints of this WNW-striking regional set remain locked without slipping under a shear stress. Fracture toughness and frictional strength are two rock properties that serve to lock a joint until a critical resolved shear stress is achieved. A gravity load caused down-dip slip on some joints that were tilted to a dip of about 62°. This suggests that a local principal stress remained roughly vertical during bedding rotation. Assuming fracture strength and friction prevented slip on most joints during tilting, the ratio of least horizontal, S h , to vertical stress, S v , at the critical tilt angle was approximately 0.55.

Alpine inversion in the Bristol Channel Basin includes reverse-reactivated normal faults with hanging wall buttress anticlines. At Lilstock Beach, joint sets in Lower Jurassic limestone beds cluster about the trend of the hinge of the Lilstock buttress anticline. In horizontal and gently north-dipping beds, J3 joints ( 295‐2858 strike) are rare, while other joint sets indicate an anticlockwise sequence of development. In the steeper south-dipping beds, J3 joints are the most frequent in the vicinity of the reverse-reactivated normal fault responsible for the anticline. The J3 joints strike parallel to the fold hinge, and their poles tilt to the south when bedding is restored to horizontal. This southward tilt aims at the direction of s1 for Alpine inversion. Finite-element analysis is used to explain the southward tilt of J3 joints that propagate under a local s3 in the direction of s1 for Alpine inversion. Tilted principal stresses are characteristic of limestone‐shale sequences that are sheared during parallel (flexural-flow) folding. Shear tractions on the dipping beds generate a tensile stress in the stiffer limestone beds even when remote principal stresses are compressive. This situation favors the paradoxical opening of joints in the direction of the regional maximum horizontal stress. We conclude that J3 joints propagated during the Alpine compression caused the growth of the Lilstock buttress anticline. q 2001 Published by Elsevier Science Ltd.

Two general types of sedimentary structures serve as initiation points (IP) for cross-fold joints in siltstone beds of the Ithaca Formation, Appalachian Plateau, New York: planar bed boundaries and geometrically complex features such as bedforms, trace fossils, and soft-sediment deformation structures. The relative abundance of each IP type varies with the stratigraphic position of the jointed body. When they are mechanically isolated by thicker shales, siltstone bodies have a higher percentage of IP associated with complex features. This association supports the hypothesis that sedimentary structures serve to concentrate stress, thus becoming preferred sites for primary joint initiation. Secondary joint initiation occurs at planar bed boundaries through interaction with pre-existing joints in adjacent siltstone beds that are mechanically coupled to form composite joints. The propagation path of joints passing through planar bed boundaries of composite joints can be traced backward to primary IP at complex sedimentary structures. Modeling suggests that when beds are mechanically coupled, the modest joint-tip stress concentration across a shale layer (e.g.

Four trends of joint sets (WNW‐ESE, NW‐SE, NNW‐SSE and NE‐SW) are found in upper Turonian carbonate rocks within the Neqarot syncline of south-central Israel. The two most predominant sets strike parallel to the trend of maximum compressive stress directions (SH) associated with the plate-related Syrian Arc stress field (SAS; WNW‐ESE) active during the Cretaceous to present and the perturbed regional stress field (NNW‐SSE) related to stress accumulation on the Dead Sea Transform during the Miocene to the present. Eighty-two percent of the beds in this study contain joints parallel with the latter trend, whereas 42% contain joints parallel to the former trend. All beds with layer thickness to spacing ratio (FSR) . 1.5 have NNW‐SSE joint sets compatible with the Dead Sea Transform stress field (DSS), whereas all joints sets that are not compatible with the DSS stress field fall beneath this value for FSR. Considering lithology, joints in five of six chalky limestone beds and all marly limestone beds are compatible with the DSS, whereas joints compatible with the SAS do not develop in these marly and chalky limestone beds. In the study area, the joint sets lack a consistent formation sequence where more than one set is found in a single bed. We use these observations to conclude that all studied joints are Miocene or younger, that the regional stress field from the Miocene to the present fluctuated, between DSS and SAS states, and that the higher FSRs correspond to a greater amount of joint-normal strain in response to the DSS. q 2001 Elsevier Science Ltd. All rights reserved.

Structural geometries and kinematics based on two sets of joints, pinnate joints and fault striations, reveal that some mesoscale faults at Split Mountain, Utah, originated as joints. Unlike many other types of faults, displacements (D) across faulted joints do not scale with lengths (L) and therefore do not adhere to published fault scaling laws. Rather, fault size corresponds initially to original joint length, which in turn is controlled by bed thickness for bed-confined joints. Although faulted joints will grow in length with increasing slip, the total change in length is negligible compared to the original length, leading to an independence of D from L during early stages of joint reactivation. Therefore, attempts to predict fault length, gouge thickness, or hydrologic properties based solely upon D‐Lscaling laws could yield misleading results for faulted joints. Pinnate joints, distinguishable from wing cracks, developed within the dilational quadrants along faulted joints and help to constrain the kinematics of joint reactivation. q 2001 Elsevier Science Ltd. All rights reserved.

The interaction between propagating joints and embedded concretions in a Devonian black shale near Seneca Lake, NY, permits identification of the loading configurations responsible for two joint sets of different ages striking at nearly the same orientation. The earlier set consists of systematic joints cut by later Alleghanian joints of the Appalachian Plateau. The later set consists of non-systematic curving cross joints that abut these same Alleghanian joints. Field evidence shows that concretions functioned as stiff inclusions in a compliant black shale. As a consequence of this elastic contrast, local perturbations in the remote stress field persisted around the concretions during burial, tectonic deformation, and exhumation. These stress perturbations influenced joint propagation paths of both joint sets. Our conclusions about loading configurations are based on finite-element modeling of the effect of the local stress perturbation on concretion-modified joint propagation. Modeling shows that the local stress perturbation from a thermoelastic loading was responsible for deflecting cross joints away from concretions in a curved propagation path near the concretion. This load configuration also led to arrest of cross joints before they penetrated the shale–concretion interface. At greater distances from the concretion, the propagation path of cross joints was controlled by the contemporary tectonic stress field. The interface between concretions and the surrounding shale was strongly bonded, as indicated by the crossing of the interface by some of the systematic ENE joints. Higher compressive stress levels within the concretions relative to the shale suppressed joint development in the concretion, causing the arrest of those joints once they had driven across the interface and a short distance into the concretion. Numerical modeling shows that interface penetration by the systematic ENE joints is consistent with a fluid load, the same loading configuration postulated for the subsequent Alleghanian joints. The traces of the systematic ENE joints align on opposite sides of concretions, rather than curving toward the concretion as predicted by two-dimensional models of the fluid load. Co-planar traces are indicative of large, planar joints propagating in-plane around the concretion, making it energetically inefficient for the crack front to curve as it enters the local stress perturbation near the concretion. A fluid load for the systematic ENE joints came from high pore pressure during the pre-Alleghanian stages of burial of the Devonian Catskill delta complex.

One model for the development of hybrid shear fractures is transitional–tensile fracture propagation, a process described as the in-plane propagation of a crack subject to a shear traction while held open by a tensile normal stress. Presumably, such propagation leads to a brittle structure that is the hybrid of a joint and a shear fracture. Crack–seal veins with oblique fibers are possible candidates. While these veins clearly show shear offset, this is not conclusive evidence that a shear traction was present at the time of initial crack propagation. Many recent structural geology textbooks use a parabolic Coulomb–Mohr failure envelope to explain the mechanics of transitional–tensile fracturing. However, the laboratory experiments cited as demonstrating transitional–tensile behavior fail to produce the fracture orientation predicted by a parabolic failure envelope. Additional attempts at verification include field examples of conjugate joint sets with small acute angles, but these conjugate joints form neither simultaneously nor in the stress field required by the transitional–tensile model. Finally, linear elastic fracture mechanics provides strong theoretical grounds for rejecting the notion that individual cracks propagate in their own plane when subject to a shear traction. These observations suggest that transitional–tensile fracture propagation is unlikely to occur in homogeneous, isotropic rock, and that it is not explained by a parabolic Coulomb–Mohr failure envelope as several recent structural geology textbooks have suggested.

Vertical joints in Devonian clastic sedimentary rocks of the Finger Lakes area of New York State are ornamented with arrays of fringe cracks that reveal the complex deformational history of the Appalachian plateau detachment sheet during the Alleghanian orogeny. Three types of fringe cracks were mapped: gradual twist hackles, abrupt twist hackles, and kinks. Gradual twist hackles are curviplanar en echelon fringe cracks that propagate with an overall vertical direction within the bed hosting the parent crack and are found in all clastic lithologies of the detachment sheet. Abrupt twist hackles propagate as planar features in thick shale beds above or below the siltstone beds hosting parent joints. Kinks propagate horizontally as planar surfaces from the tips of parent joints in siltstone beds. The breakdown of the parent joint into either gradual or abrupt twist hackles depends on the orientation and magnitude of the remote stress field, internal fluid pressure, and the elastic properties of the bed. The twist angle of gradual twist hackles is larger in coarser clastic beds, indicating that stress and internal pressure are more important parameters than elastic properties in controlling breakdown. Assuming that the vertical stress axis (S v ) equals 78 MPa at 3 km burial depth, the difference in twist angle between sandstone and shale beds is used to estimate the maximum horizontal stress difference in the shale beds as S H ‐ S h ∪ 2.5 MPa when S H ‐ S h ∪ 12 MPa in sandstone beds. The twist angle of the fringe cracks and the abutting relationships of parent joints give an indication of the overall change in stress field orientation within the detachment sheet during Alleghanian tectonics. These parent joints indicate a regional clockwise stress rotation of Alleghanian age concordant with the twist angle of fringe cracks throughout the western part of the study area. A counterclockwise twist angle in the eastern portion indicates a local stress attributed to drag where no salt was available to detach the eastern edge of the plateau sheet. The clockwise change in stress orientation is consistent with the rotation in stress orientation found in the anthracite belt of the Pennsylvania Valley and Ridge, but is opposite to the sense of rotation in the southwestern portion of the detachment sheet (western Pennsylvania and West Virginia). The two regional rotation domains are separated by the Juniata culmination.

In rocks without systematic mechanical discontinuities (e.g., granite), joint spacing follows an approximately log-normal frequency distribution (i.e., the distribution has a kurtosis near zero). Joint spacing in rocks with systematic mechanical boundaries differs from the spacing in isotropic rocks, exhibiting consistently positive values of kurtosis (i.e., the distribution is more clustered around the mode than a perfectly log-normal distribution). We attribute this difference in joint-spacing distribution to mechanical boundaries such as bed partings in sedimentary rocks that constrain joint height and control joint spacing. Existing systematic joints can also act as mechanical boundaries during the development of later ‘cross’ joints. Two parallel mechanical boundaries determine the mechanical-layer thickness that influences joint spacing. In this paper, we investigate the effect of sampling geometry and mechanical discontinuities on joint-spacing statistics. In many situations, neither pavement surfaces nor properly oriented boreholes are available for measuring the spacing of cross joints that develop between existing systematic joints. When joint-spacing data come from scanlines that are oblique to a systematic joint set (e.g., crossing many systematic joints on a sub-vertical outcrop face), we consider whether the median spacing of the systematic joint set is statistically equivalent to the mechanical-layer thickness thought to control cross-joint spacing between individual pairs of systematic joints. The basis of our analysis is the fracture spacing index (FSI), which is the slope of a line fitted to a plot of mechanical-layer thicknesses vs. median joint spacing. We collected joint-spacing data along oblique scanlines from a large outcrop of the Devonian Brallier Formation, a distal turbidite sequence, near Huntingdon, PA. Our analysis indicates that the spacing of cross joints correlates better with a mechanical-layer thickness defined by the median systematic (‘strike’) joint spacing (r 2 D 0:78, FSID 1.02) than with a mechanical-layer thickness defined by the stratigraphic bed thickness (r 2 D 0:69, FSID 0.97). This is consistent with the conclusions of previous workers. We also note that the spacing data from the cross joints exhibit a higher (i.e., more positive) value of kurtosis than the data from the earlier strike joints (1.66 vs. 0.98). This is consistent with the idea that mechanical discontinuities alter joint spacing in a systematic manner. In this case, the cross joints may have been influenced by two sets of mechanical boundaries (bedding and existing joints), whereas the earlier strike joints were constrained by only one set (bedding). © 1998 Elsevier Science B.V. All rights reserved.