GRI's Devonian Shales Research Program: A Progress Report

Extensive data were gathered during the pumping and flowback of a nitrogen foam stimulation treatment on a Devonian Shale gas well in eastern Kentucky. Subsequent analysis of these data revealed that 1) duplicate monitoring equipment, properly calibrated, yielded nearly identical data, 2) the job execution, largely because of extensive quality control, closely followed job design, 3) friction loss in the pumping annulus could be calculated at all stages of the treatment, and 4) both perforation friction loss and the number of holes open could be estimated throughout the job. As part of the Comprehensive Study Well program sponsored by the Gas Research Institute (GRI). this fracture treatment design, planning, execution, and analysis involved cooperation among planning, execution, and analysis involved cooperation among several GRI contractors, the service company, and the operator. On-site monitoring equipment included densiometers, pressure gauges and transducers, and various types of flowmeters. Two control vans monitored real-time data, and permitted adjustment to one of three tentative pumping schedules based on downhole treating pressures during the prepad stage. Direct measurements led to pressures during the prepad stage. Direct measurements led to real-time calculations of foam quality, foam rate, and downhole proppant concentration. proppant concentration. When the data were analyzed, the wealth of data coupled with the redundancy allowed various calculations to be made with confidence. Direct comparisons between the two sets of monitoring equipment showed near-perfect agreement. Similarly, the actual data matched the design estimates. Recently published results dealing with discharge coefficient change and hole enlargement formed the basis of our analysis of perforation friction. This analysis has been performed on one other set of data from an offset well and led to similar conclusions. Comparable data is being acquired from other wells and may eventually provide a valuable database for fracture design in the Devonian Shales. Introduction Beginning in November 1988, Ashland Exploration, Inc. drilled, completed, and tested the E. J. Evans No. 89 (EJE89) in the Lower and Middle Huron intervals of the Devonian Shale. This well was one of approximately 30 AEI wells completed during the late 1980's in the E. J. Evans lease in Breathitt County, KY, on the edge of the Big Sandy Gas Field. The EJE89 was selected to be Comprehensive Study Well No. 4 (CSW4) in a research program sponsored by the Gas Research Institute (GRI). The CSW program in the Devonian Shales was a pan of GRJ's continuing research efforts designed to increase the supply of natural gas. This field oriented program had the following main objectives:to develop a better understanding of the geologic controls on production,to refine recently developed formation evaluation tools for selecting completion intervals, andto improve stimulation practices in the Devonian Shales. The focus of stimulation research in the CSW program was on pumping treatments carrying more sand at higher concentrations, developing an optimal fluid system for the Devonian Shales, and improving quality control. Breathitt County Shale wells are characterized by relatively thin Shale intervals with 250 to 300 ft gross height from the Cleveland to the bottom of the Lower Huron, and relatively low initial reservoir pressures of 450 to 650 psia. Permeability-thickness products of 1 pressures of 450 to 650 psia. Permeability-thickness products of 1 to 5 md-ft are common, translating to gas flow rates of 15 to 25 Mscf/D, even after stimulation. Historically, the most common type of stimulation for the Shales in this area had been pumping straight nitrogen. Within the past few years, operators began using nitrogen foam carrying low sand concentrations (1 to 2 lb/gal). As with other areas of Devonian Shale production, there appeared to be no particular treatment strategy that outperformed others. In this well, particular treatment strategy that outperformed others. In this well, as with other wells in the CSW program, we recommended pumping a higher sand concentration (6 to 7 lb/gal) nitrogen foam treatment. The CSW4 was spudded on November 13, 1988 and air-drilled, reaching a total depth of 2805 ft on November 17. Following the logging runs, which included a borehole television survey, 5 1/2-in, 15.5 lb, J-55 casing was cemented in place to a depth of 2791 ft, with the plug at 2760 ft (RKB). Fig. 1 is a well mechanical diagram for the CSW4. On January 6, the operator placed 17 perforations in the Lower and Middle Huron, from 2477 to 2672 ft. P. 131

The Devonian oil shales of the eastern United States constitute one of the nation's major energy resources. The eastern Devonian oil shale resource is estimated to exceed 400 billion bbl of synthetic oil, if all surface and near-surface shales of ore quality were strip or deep mined for above-ground hydroretorting. Work done at the Institute of Gas Technology since 1972 under the sponsorship of the American Gas Association, The Gas Research Institute, and the U.S. Department of Energy has shown that if retorted in hydrogen gas at temperatures of 500 to 730°C and pressures End_Page 1285------------------------------ of 20 to 50 atm, the eastern Devonian shales are a viable source of synthetic liquid or gaseous fuels. Experimental work, in equipment capable of processing up to 1 ton/hour of shale, has confirmed the technical and economic feasibility of above-ground hydroretorting of oil shales. Work done to date on nearly 500 samples from 12 states indicates that the HYTORT process can give organic carbon recoveries from 2 to 2.5 times those of conventional retorting of the Devonian shales, so that the HYTORT process yields 25 to 30 gal/ton on syncrude at many localities, compared with 10 to 15 gal/ton using Fischer Assay retort methods. Criteria for inclusion of shale in estimates of recoverable resources for the HYTORT process are (1) organic carbon of at least 10% by weight, (2) overburden of less than 200 ft (59 m), (3) volumetric stripping ratios of less than 2.5 to 1, and (4) stratigraphic thickness of 10 ft (3 m) or more. Resource estimates include: Kentucky (Ohio, New Albany, and Sunbury Shale), 190 billion bbl; Ohio (Ohio and Sunbury Shale), 140 billion bbl; Tennessee (Chattanooga Shale), 44 billion bbl; Indiana (New Albany Shale), 40 billion bbl; Michigan (Antrim Shale), 5 billion bbl; and Alabama (Chattanooga Shale), 4 billion bbl. Recoverable resources have not been identified in West Virginia, Georgia, Oklahoma, Illinois, Arkansas, or Missouri outcrops. Co-production of uranium and metals is a possibility in the areas favorable for syncrude production. End_of_Article - Last_Page 1286------------

This paper summarizes the results of a cleanout study performed on six Devonian Shale wells in southwestern West Virginia. This study was part of a joint Columbia Gas System Service Corporation (CGSSC), Columbia Natural Resources (CNR), and Gas Research Institute (GRI) project which involved 'cleaning out' shale rock debris from several old, natural, openhole Devonian Shale wells for the purpose of evaluating Devonian Shale production mechanisms. Based on well test analysis and four years of production data, it appears the cleanout procedure is economically attractive for increasing production from these types of wells. Six wells in Mingo, Wayne, and Logan Counties, WV were selected as candidates for study. Using a cable tool rig, Columbia found the Devonian Shale openhole sections in these wells filled up and bridged off with shale rock debris. The debris was removed with a bailer, but some tool sticking problems were encountered. Post-cleanout rates increased as much as six-fold over precleanout rates as a result of these operations (73 to 411 Mscf/D for one well). Pre- and post-cleanout well test analyses performed on three wells indicated the cleanout operations resulted in increases in completion efficiency (more negative skin factors), permeability-thickness products, or both for each well. Analysis of four years of post-cleanout production data has shown the initial post-cleanout flow rates have been sustained, and it appears the cleanout operations have increased the ultimate recoveries for these wells. Based on these initial results, CNR performed a pilot cleanout project in 1989 on four wells using a rotary rig for the cleanout operations. Many problems encountered using the cable tool rigs were avoided with the new procedure. Initial results have been encouraging. Our analysis indicates a cleanout operation, on average, could increase estimated recovery by 210 MMscf over a 20-yr period. Assuming typical operating economic data, this increase in recovery could translate to a net present value of $197,000 per well. If 10% of the over 15,000 Devonian Shale wells in the basin were candidates for cleanout, this workover operation could translate to an increase in reserves of 315 Bscf with an associated net present value of $296 million.

Fracture-enhanced reservoirs are widespread in the middle and upper Devonian sands and shales of the Appalachian Basin. These zones of fracture enhanced permeability and productivity can be delineated with surface and downhole geochemical techniques. Careful use of surface geochemical surveys can greatly improve our ability to target and develop these resources.

Economic gas production from the Devonian Shale requires permeable pathways combined with matrix storage. These pathways may include fractures, bedding planes or silt layers. The Gas Research Institute is sponsoring a research project to evaluate the relationships these geologic features and productive gas flows have with the eventual aim of developing better exploration, stimulation and production strategies. This study will investigate wells throughout the basin, with a wide range of potential productivity in order to determine what variations in geology result in marginal versus good producers. As a first phase of this project, detailed data on the geology and reservoir parameters are being collected on four wells in southern Ohio, which are being recompleted in the Devonian Shale. Two additional wells have been investigated in a pilot study, the remainder are currently under investigation.

Summary Extensive study of natural fractures in oriented core, well logs, induced fracture orientation, lineaments, and Seisviewer TM surveys resulted in a log-core correlation of subsurface fracture systems and established a relationship between surface and subsurface directional aspects of Devonian shale. Core and well log analysis techniques discussed have the potential to identify the major producing intervals in Devonian shale. Introduction The primary objectives of the U.S. DOE's Eastern Gas Shales Project (EGSP) are to characterize the nature of the reservoir, determine how gas can be extracted in a cost-effective manner, and reduce the uncertainty of the resource base. The unconventional reservoirs in Devonian shale are considered potentially the most productive underdeveloped source of natural gas in the northeastern U.S.. Field data indicate that gas is produced from Devonian shales mainly as a function of the presence and density of natural fractures. Although the detection of a fracture system does not guarantee commercial production, it increases that probability. In support of this effort, a series of joint industry/U.S. DOE research projects was designed to investigate and study the nature, orientation, and density of natural fractures in Devonian shales. Within several project areas in the Appalachian basin. holes were drilled, cored, and logged (Table 1). This paper deals with the detailed correlation of the natural fractures determined from these cores and geophysical well logs to determine the validity and reliability of logging tools to locate fractured internals in Devonian shales. In the latter part of the paper, correlation of natural and hydraulically induced fracture orientations, along with surface photolineaments traces, are discussed for a specific project area. Background The Morgantown Energy Technology Center (METC) is responsible for the management and technical direction of the U.S. DOE research and development activities related to the unconventional gas recovery (UGR) program. As part of the UGR program, the EGSP was formulated in 1975 to determine the resource potential of Devonian shale that underlies the Appalachian, Illinois, and Michigan basins. Within the EGSP. resource characterization activities including drilling, coring. and logging of Devonian shale wells are complete. This paper addresses primarily the coring and logging activities related to the identification of fractures, Correlation of natural fracture density and orientation as derived from well logs and oriented formation cores may aid in defining and identifying gas reservoirs and improving prospects of commercial gas production from shales. The objective of coring various Devonian shale intervals is to provide chemical and physical characterization of the cored formations. Formation characteristics studied include mechanical rock properties, geochemistry, and natural fracture incidence and orientation. Two types of fractures are detected in Devonian shales: existing natural fractures and fractures that occur as a result of stresses produced by the core extraction process. A classification scheme for identifying and defining the coring-induced and natural fractures has been developed. Kulander et al. defined such terms as hackle, petal, and torsion to classify coring-induced fractures. JPT P. 1371^

SummaryThe in-situ stress contrast between a reservoir rock and the surrounding formations is important to the design and analysis of hydraulic fracture treatments. The stress contrast between layers controls the fracture's vertical (height) growth, which in turn affects the fracture length and width. As part of the Gas Research Inst.'s (GRI's) Comprehensive Study Well (CSW) program in the Devonian shales of the Appalachian basin, we measured in-situ stresses directly and used these values to calibrate acoustic log measurements to develop stress profiles across the Devonian shales. This discusses the measurement and interpretation of in-situ stresses, the use of acoustic logs to determine mechanical properties and a more complete stress profile, and the practical use of the stress profile in fracture-treatment design and analysis.In-situ stresses were measured directly throughout the Devonian shale interval with both openhole and cased-hole stress tests, conducted with nitrogen, a specially designed downhole shut-in tool, and a downhole quartz pressure gauge with surface readout. The paper describes the procedure used and presents interpreted stress-test results from two CSW's. Full-waveform acoustic tools were run to define elastic properties of the Devonian shales. These properties also were determined under both static and dynamic conditions from whole-core plugs taken in the shales. Logs were calibrated to core-measured values, and then elastic properties determined from the logs were used to compute an in-situ-stress profile, which was calibrated against field measurements of in-situ stresses. Finally, we discuss how these data are used to design and analyze hydraulic fracture treatments in CSW's.

This paper presents the results from a preliminary investigation into the reservoir mechanisms controlling gas production from the Antrim Shale of the Michigan Basin. We used a two–phase, dual porosity, finite– difference reservoir simulator incorporating shale– specific reservoir characteristics to match a well's gas and water production history. We also analyzed the sensitivity of Antrim Shale gas production behavior to drainage area and several reservoir properties. The results indicate that long–term gas production from the Antrim Shale is primarily controlled by natural fracture properties, desorption properties, and drainage area. Based on development strategies for coalbed methane reservoirs, there is an optimum well spacing for effectively dewatering the Antrim Shale, thus accelerating gas production and improving ultimate recovery. Finally, we found that shale matrix diffusion properties have little effect on short–term gas productivity, and preliminary results also suggest long–term gas productivity does not depend on matrix diffusion.

Under a contract with the Gas Research Institute, the West Virginia Geological and Economic Survey has begun a two-year project to determine the relationship between Devonian shale production and certain geologic and technical factors, such as geologic structure. Six counties in western West Virginia were selected for study. The north-south-trending Burning Springs anticline bisects those counties and marks the western edge of the Salina salt basin. Two horizons were selected for structural datums--the base of the Huron Shale Member of the Ohio Shale (Upper Devonian), and the base of the Mississippian Greenbrier Group. Point-to-point contouring methods, rather than generalization, revealed deformation containing differing structural styles (domains) throughout the study a ea, and those domains may have a relationship to the Salina salt basin (Late Silurian). Structural cross sections reveal that the southern end of the Burning Springs anticline is a gentle fold amplified by thrust faults, probably ramping up from a sole fault in the Salina salt beds. Mississippian, Pennsylvanian, and Permian strata are draped over those thrusts. Lineaments, as defined by structure contours, mark the boundaries of subthrust sheets along the Burning Springs anticline. Orientations of those structural lineaments coincide with orientations of mapped surface lineaments. Oil production from the Devonian shales is confined to the eastern flank of the Burning Springs anticline in areas overlying the Salina salt basin, and may be related to fracturing caused by intense deformation, in contrast to less intense deformation of rocks west of the salt basin. End_of_Article - Last_Page 1449------------

A previously watered-out and abandoned gas reservoir in the Sub-Clarksville formation of the Mt. Selman field in North East Texas was reentered and secondary gas recovery initiated by pumping of water. Pumping was done with a sucker rod pump and bottom hole pressures in the producing well and a nearby observation well was measured via gas filled capillary size tube extending into open ended larger tubing downhole. Numerical modeling to match past performance and project possible future production was done before and during the field test to compare theoretical and actual production results. A series of draw-down and build up tests over a three month period were performed to evaluate the reservoir and determine its potential for secondary gas recovery. No residual gas cap of mobile free gas was found. The remaining free gas was apparently trapped as an immobile dispersed phase within the reservoir rock matrix. The amount of gas dissolved in the reservoir brine was about 7 Scf/Stb (1.28 std m3/m3). The produced gas/brine ratio varied from 7 to 25 Scf/Stb (1.28 std m3/m3 - 4.5 std m3/m3) indicating that some immobile gas was mobolized by pumping water to lower the reservoir pressure. Gas analysis has shown the trapped free gas to be richer in heavier hydrocarbons than the dissolved gas.

Gas reserves of the Devonian shales of the Appalachian basin constitute a large, underdeveloped resource producing from fractured reservoirs. As part of ongoing Gas Research Institute research, K and A Energy Consultants, Inc., is identifying geological controls on gas production. Preliminary findings indicate that local gas production is controlled by a combination of structure and stratigraphy. Regional geological review indicates that Devonian sedimentation and structure is influenced by repeated reactivation of basement faults. Site-specific geologic studies indicate that depositional and structural mechanisms vary substantially throughout the basin. Gas production on the eastern margin of the producing area is controlled by an Alleghenian thrust front located by Grenville normal faults. High-capacity wells are associated with tear faults in the thrust sheets. To the southwest, deformation is controlled by both Grenville and Rome trough basement faults. Reactivation of these faults during later orogenic events produced a complex of high-angle reverse and strike-slip faults. Fracturing in the Devonian shales is produced by shearing and flexure associated with these structures. Syndepositional movement of the basement structures influenced the deposition of coarser grained turbidites and tempestites. The combination of fractures and coarser clastic beds provides effective reservoir systems. The shale contains abundant organic material consistingmore » of terrestrial plant debris and marine algal remains. Thermal maturation of this material produced gas which charged the lower reservoir systems. Exploration along reactivated structural trends is an effective strategy for locating Devonian shale gas accumulations. This approach may also apply to other producing strata in the basin.« less

The Gas Research Institute (GRI) has been sponsoring a cooperative well program with Ashland Exploration, Inc., (AEI) during the past two years targeting the Devonian Shale and Berea sandstone formations in Pike County of eastern Kentucky. Operators typically complete both the shales and Berea in one well bore in this area. This presentation summarizes the research results of the Berea cooperative well, the COOP 2 (Ashland FMC 80). The specific objectives of the Berea evaluation in the COOP 2 were to develop an integrated reservoir description for stimulation design and predicting long-term well performance, identify geologic production controls, determine the in-situ stress profile, and develop Berea log interpretation models for gas porosity and stress. To satisfy these objectives, data were collected and analyzed from 146 ft of whole core, open-hole geophysical logs, including formation microscanner and digital sonic, in-situ stress measurements, and prefracture production and pressure transient tests. In addition, data from a minifracture, a fracture stimulation treatment, and postfracture performance tests were analyzed. The authors determined the integrated reservoir/hydraulic fracture descriptions from analyzing the data collected in the open- and cased-hole, in addition to the log interpretation models developed to accurately predict gas porosity and stress profiles. Results canmore » be applied by operators to better understand the Berea reservoir in the study area, predict well performance, and design completion procedures and stimulation treatments. The methodology can also be applied to other tight-gas sand formations.« less

Analysis of Devonian shale production histories without careful data screening can easily mislead even the most exacting investigator. Trends established from cumulative production data may not accurately represent the true production pattern. The difference between a good well and a poorer producer may be the result of fracture depletion or the length of actual producing time (among other factors) rather than a change in reservoir quality. The use of matrix controlled production information, derived from decline curves, appears to resolve this problem.

Determination of the total organic carbon (TOC) based on conventional well logs using artificial neural network

Introduction Until the past few years, the U.S. supply of natural gas has always exceeded demand. Enough natural gas was available from conventional reservoirs that it was neither necessary nor economic to develop low productivity reservoirs. The current situation in the U.S., however, necessitates that additional supplies of natural gas be developed to meet the nation's demand. Since 1974, advance in technology and improved economic incentives have resulted in a dramatic increase in the exploration and exploitation of tight gas basins. Tight gas reservoirs are often termed as "unconventional" reservoirs. The exact definition of conventional and unconventional reservoirs is vague. In economic terms, a conventional reservoir is one in which a reasonable profit can be made at low gas prices and without requiring large volume stimulation prices and without requiring large volume stimulation treatments. Likewise, an unconventional reservoir can be described as one which requires the higher gas prices and large volume fracture treatments before a prices and large volume fracture treatments before a reasonable profit can be made. In engineering terms, the best way to classify reservoirs is by permeability. Of course other factors are important, such as porosity, net gas pay and reservoir pressure; however, permeability is normally the dominant parameter when evaluating tight gas reservoirs. Considering the current gas price and the level of stimulation technology, it is the opinion of these authors that the "permeability cut-off" for conventional reservoirs is approximately 0.1 md. Normally, if the permeability is greater than 0.1 md, an undamaged well will recover 70–80 percent of the gas in place. When the permeability is about 0.01 md, large fracture treatments are required to recover 40–60 percent of the gas in place. In reservoirs where the permeability ranges from 1 to 5 microdarcies, large fracture treatments are required to recover about 10 to 30 percent of the gas in place. As mentioned above, other parameters such as reservoir pressure, well spacing porosity, etc will affect the pressure, well spacing porosity, etc will affect the recovery efficiency; however, in general terms, the recovery efficiencies given above are realistic. Substantial quantities of natural gas exist in these unconventional, low permeability reservoirs. The location of the basins which contain a majority of these tight gas reservoirs have been known for many years. Most of these basins are found in the western portion of the U.S. Figure 1 illustrates the location of the most important basins. Figure 1, was reproduced from a Department of Energy report which was published earlier this year. That report presented a comprehensive study concerning the "enhanced recovery of unconventional gas". Four sources of unconventional gas were considered:tight gas basinsThe Devonian shalegeopressured aquifersmethane from coal seams To determine the potential of each of these unconventional sources of gas, geologic data, engineering analyses and economics were combined to estimate the recoverable gas from each source. A portion of that study - the tight gas basins and the Devonian shale - required the use of numerical reservoir simulation to predict well performance. A single phase, two dimensional, finite difference reservoir model was used for this purpose. The model has been described in an earlier publication. The objectives of this paper are to present the techniques used to model each of these reservoirs and to demonstrate how computer history matching can be coupled with geologic information to develop a realistic interpretation of insitu reservoir conditions. The first use of the numerical model in this study was to perform a sensitivity analysis using "typical" tight gas reservoir parameters. Computer runs were made to isolate the effects of formation permeability and fracture length upon recovery permeability and fracture length upon recovery efficiency for this average data set.