More than 560 trillion cubic feet of natural gas permeate the Antrim shale formation of the Michigan Basin, yet less than one percent has been developed as proven reserves. Successful gas production is known to occur in naturally fractured areas of the Antrim and similar formations, but finding these underground fracture systems has been a difficult and elusive process for exploration companies in Michigan and around the world.
      Advanced Resources International (ARI) Inc. of Denver, Colo., has developed techniques to assist in detecting natural reservoir fractures without drilling exploratory wells. Working under NASA's Earth Observations Commercial Applications Program (EOCAP), ARI has correlated satellite imagery with common oilfield geophysical data to locate fracture zones in the Antrim shale.
      ARI, a geological and engineering technical services firm that specializes in developing unconventional hydrocarbon resources such as coalbed methane and gas-bearing shales, received its cost-sharing EOCAP contract in mid-1994. Following a year of technical validation including work in Michigan, ARI is expanding the technique to fractured reservoirs in Africa, New Zealand and other parts of the United States.
      "Naturally fractured reservoirs represent one of the largest undeveloped sources of future gas reserves in the world," said ARI Vice President David Decker. "One of the primary reasons NASA chose to fund this project is the significant worldwide impact it may have on natural gas exploration methods and their successful application."
      Natural fractures occur in many different types of gas-bearing reservoirs, including coal seams, explained Decker. These cracks and crevices are significant in hydrocarbon production because they provide escape routes through which gas can be produced from otherwise impermeable rock. The fractures are usually created in localized portions of the reservoir due to newly created structures or reactivation of pre-existing structures within the underlying basement rock.

A Fractured Reservoir
The Antrim formation is a shallow late Devonian shale that occupies about 33,000 square miles under the northern half of Michigan's Lower Peninsula. At least 50,000 wells penetrate the Antrim, of which only 3,000 actually produce gas from it. Most of the Antrim wells originally targeted oil-bearing reservoirs below the shale. It was not until the 1980s that the Antrim was purposely targeted for production.
      Successful Antrim producers are concentrated in a 1,000 square mile area of the Michigan Basin in Otsego County and average 100,000 cubic feet per well per day of natural gas. A variety of well logs confirm that this producing trend is heavily fractured. Wells drilled outside the fracture zone have been unable to produce from the Antrim, and attempts to induce fractures in the formation have been futile.
      "Basement rock lies below the Antrim at a depth of about 9,000 feet and it has undergone extensive movement throughout the geologic evolution of the basin that probably relates to fracturing in the Antrim," said Alan Klawitter, ARI's remote sensing specialist. "We knew other parts of the Antrim are fractured, and we just had to find them."
      The Antrim made an excellent candidate for ARI's fracture detection process because it is a relatively shallow reservoir. More than two-thirds of the gas producing wells are less than 2,000 feet deep. ARI demonstrated that fractures in a shallow reservoir are much more likely to show surface expressions in the landscape that might be detectable using Landsat imagery.
      Another reason for targeting the Antrim was the volume of available well log information. It is one of the most actively drilled gas plays in the United States, and currently the Gas Research Institute (GRI) is conducting numerous studies of reservoir conditions. GRI made these geological, geochemical, and geophysical data sets available to ARI for the project.
      "Most of the scientific data used in this project are available for most reservoirs or can readily be acquired," said Klawitter. "One of the objectives of EOCAP is to utilize standard data types whenever possible."

Finding the Fractures
The overall goal of ARI's fracture detection technique is to correlate the existence of known fracture zones with signature characteristics in interpretive data such as geologic, aeromagnetics, gravity and Landsat Thematic Mapper imagery. Exploration companies can save a considerable amount of time and money if overlaid signatures in interpretive data can be used to prioritize areas most likely to contain fracture zones instead of randomly drilling wells to find them.
      From log data taken from successful Antrim producing wells, ARI was able to map the location and concentration of some fractures in the Otsego County trend. Circumferential Acoustic Sonic Tool (CAST) logs actually showed a three-dimensional view of cracks in the wellbore. ARI used these CAST logs to create a map of some subsurface fracture trends throughout the Antrim producing zone.
      Hydrologic studies of Antrim formation water conducted at the University of Michigan also helped ARI track the fractures. Through chemical analysis of Antrim well brine, researchers could detect trace elements and compounds from other formations that indicated communication between the shale and surrounding formations.
      With the known fractures mapped, ARI obtained a Landsat TM image of Otsego and surrounding counties.
      "To locate possible surface expressions of basement faulting, we chose a TM band combination that displayed the maximum spectral diversity in that particular area," said Klawitter. "We used a TM 5,4,1 (RGB) combination."
     Klawitter applied an omni-directional edge-enhancement filter to the Landsat scene. The filter highlighted linear features without any preference to direction. ARI technicians then overlaid a mylar sheet on the image and mapped linear features of non-cultural origin. The length, location and orientation of each feature was digitized so that a computer-generated rose diagram could be created to display predominant orientations of the linear features.
      The rose diagram showed three significant peaks in linear feature orientation. One of these was disregarded because it related to known glaciation patterns in Otsego County. The other two orientation peaks clearly depicted northeast and northwest trending surface features in the area. ARI then plotted only features within these trends on the map that displayed the fracture zone locations found in the CAST log data from the successful producing wells.
      "The CAST data had indicated a strong preference of the northeast trending fractures relating to production," said Klawitter. "The overlay map confirmed that good wells encountered northeast trending surface features."

Adding Geophysical Data
Not all northeast trending lineaments coincided with high production areas because surface features are only one indicator of fracturing. Geological and geophysical data were added to the map to complete the puzzle.
      Aeromagnetic and gravity surveys are two interpretive data sets used to create a unique data signature for the productive regions. Magnetic and gravity surveys can tell geologists many things about subsurface features, including variations in basement lithology and structure.
      "A critical assumption in the Michigan Basin is that Antrim fractures are related to basement faulting," said Klawitter. "Vertical and lateral displacements of basement rocks are expressed as abrupt changes or variations in magnetic and gravity readings."
      Klawitter explained that because of the locations and density of geophysical data points, the displacements may appear in the surveys as gently sloping variations over distances of a half mile. To assist in locating possible basement faults, ARI applied second derviative functions to the survey data. Second derivative functions highlight maximum gradient changes in the magnetic and gravity data. Those areas where gradient changes are sharpest may occur at the edge of basement faults. When these results were mapped, contour lines denoted severe gradient changes and indicated the possible presence of a fault.
      Results of the second derivative surveys were mapped on top of the other data layers for visual analysis. ARI examined the map to determine if any correlation could be discerned among the various layers of well log, subsurface structure, imagery, magnetic and gravity data. There searchers found a remarkable correlation between the data signatures.
      "Successful producing wells were located where the maximum curvature of the subsurface structure, magnetic and gravity data intersected the northeast trending surface features found in the Landsat imagery," said Klawitter.

Assessing the Results
ARI conducted the data mapping both inside and outside of the known fracture trend. Project technicians pinpointed several other possible fracture areas to the west of the Otsego trend where subsurface structure, magnetic and gravity anomalies coincided with northeast trending zones of concentrated and aligned linear features. One of these areas was in Manistee County, Mich.
      The technicians confirmed the presence of fractures in Manistee by researching the drilling success of operators there. Without any communication with ARI, one independent operator was in the process of drilling several wells in Manistee at about the same time that ARI was completing its mapping project.
      ARI learned that the operator had drilled four wells in the exact area of the county where the combined data signature indicated a possible fracture zone. Each of these four wells is successfully producing from the Antrim shale. CAST logs have confirmed the presence of a fracture zone in the region.
      For further validation of the detection method, ARI has teamed with another operator in the Michigan Basin who plans to drill several test wells targeting other potential fracture zones that were detected in the project. This cost-sharing project will occur in early 1996.
      "Once those results are confirmed, we will take the technique on the road to target other reservoirs," said Decker.
      ARI believes the Chattanooga, New Albany and Rome shales in the U.S. will make excellent targets because they are so similar to the Antrim. The consulting firm is already using the methodology for coalbed methane reservoirs in Zimbabwe, Australia and New Zealand.

EOCAP Impact
NASA's EOCAP program is sponsored by the Office of Space Access and Technology at the Stennis Space Center in Mississippi. The mission of the program is to collaborate with commercial firms to validate remote sensing applications and develop them for commercial market requirements.
      The fracture detection process is a perfect example of what EOCAP is trying to achieve, according to EOCAP Manager Mark Mick. "Remote sensing was often over sold as a stand-alone technology, but this project introduces satellite imagery as one tool out of many that can solve a problem efficiently when they are used together."
      The fracture detection project will continue for at least another two years. In the meantime, EOCAP is continuing to provide remote sensing technical support to ARI as the firm develops its marketing plan for the technology. The organizations are considering a variety of multi-media presentations that may be used to disseminate information about the technique to oil and gas companies around the world.
      "We may videotape the entire detection process, including fieldwork, data processing and analysis, and then put it on a CD-ROM," said ARI's David Decker. "Instead of sending a brochure out for people to read about it, they can actually watch on screen how the fractures are found."

About the Author:
Kevin Corbley is the principal in Corbley Communications, which provides public relations and marketing services in the fields of remote sensing, GPS, digital mapping, and GIS. He is located in Lakewood, Colo., and may be reached at 303-987-3979.

Back