Horizontal well taps bypassed Dundee oil in Crystal field, Mich.

James R. Wood
J. R. Allan
Jacqueline E. Huntoon
Wayne D. Pennington
Michigan Technological University
Houghton, Mich.

William B. Harrison III
Western Michigan University
Kalamazoo, Mich.

Eric Taylor
Craig J. Tester
Cronus Development Corp.
Traverse City, Mich.

Dundee Horizontal Well as Planned and Executed [26729 bytes]

The Dundee formation (Middle Devonian) has yielded more oil than any other producing interval in Michigan. The Dundee trend, which forms an east-west band across the central Michigan basin, consists of 137 fields which together have yielded more than 350 million bbl of oil (Fig. 1 [18838 bytes]).

The first commercial Dundee production was established at Mt. Pleasant field in 1928, and most Dundee fields were discovered and brought on production during the 1930s-40s.¹ Wells in many of the fields had very high initial production (IP) rates. IPs in excess of 1,000 b/d of oil were common, with values as high as 9,000 b/d reported.

These high flow rates, combined with a thin (10-30 ft) oil column and a strong water drive, resulted in water coning that left significant volumes of oil unrecovered in some fields. One such field, Crystal field in Montcalm County, is the focus of a U. S. Department of Energy (DOE) Class 2 Reservoir Demonstration Project designed to demonstrate that horizontal drilling can recover significant volumes of this bypassed oil.

Demonstration project

In 1993, Michigan Technological University, Western Michigan University, and Cronus Development Corp., (formerly a subsidiary of Terra Energy, Ltd., an independent oil and gas producing company in Traverse City, Mich.) signed a cooperative agreement with DOE under the Class 2 (shallow-shelf carbonate) Reservoir Program and began a 3 year project aimed at evaluating and demonstrating the application of horizontal drilling technology to the recovery of bypassed oil in the Dundee formation in the central Michigan basin. Terra Energy and the two universities provided matching funds.

The project had two goals:

To test the viability of using horizontal wells to recover bypassed oil from the Dundee reservoir in Crystal field, and
To adequately characterize the Dundee reservoir in Crystal field and in 29 additional Dundee fields in the central Michigan basin with similar reservoir lithologies. It has been estimated that as much as 2 million bbl of additional oil might be recovered from Crystal field through horizontal drilling. With 137 Dundee fields in the trend, it is apparent that the amount of incremental oil recoverable through horizontal drilling could be very significant.

Regional setting

The Dundee formation is composed of open-shelf, biohermal, and, locally, sabkha carbonate deposits.^2-3 Stratigraphic nomenclature for the Dundee (Fig. 2 [11142 bytes]) has been inconsistently applied over the years. Some of the confusion resulted from regional facies changes across the basin. Based on observation of a limited number of cores (about 50) and many more wireline logs (about 400) the following generalities can be made:

In the western Michigan basin, the Dundee formation is divided into three members (Fig. 3 [158299 bytes]), the Reed City dolomite, the Reed City anhydrite, and the Rogers City limestone.⁴ In western Michigan, Dundee reservoirs occur in the Reed City dolomite as porous, sucrosic intervals which underlie the Reed City anhydrite, a sabkha to shallow lagoonal evaporite deposit. The Dundee reservoir formed when brines refluxed downward penecontemporaneously with Reed City anhydrite deposition, dolomitizing underlying carbonates. Reed City field is the most productive example of this type of Dundee reservoir.⁵

In the eastern Michigan basin, the Rogers City limestone directly overlies the Dundee limestone (Fig. 3). Because they are lithologically similar, the Dundee and Rogers City limestones are usually undifferentiated in the subsurface. The section in the central and eastern basin from the base of the Bell shale to the top of the Detroit River anhydrite is commonly referred to on drillers' logs simply as the "Dundee."

Throughout much of eastern Michigan, which is near the basin depocenter, the Dundee limestone exceeds 150 ft in thickness and is composed of thin (2-20 ft) coarsening-upward parasequences with grainstone beds at their tops. The grainstones are the productive reservoir lithology. The Dundee-Rogers City contact has been interpreted as a sequence boundary by Curran and Hurley6 and probably correlates with the regressive Reed City anhydrite in western Michigan. West Branch field provides a good example of this type of reservoir.⁶ Stromatoporoid/rugose-coral patch reefs are widespread in the Dundee limestone in eastern Michigan, and in some areas production comes from reef core boundstones and reef flank skeletal sands. South Buckeye field provides an example of Dundee reef production.⁷

In the central Michigan basin, where our seven-county study area is located, core coverage is limited, but those cores available for examination suggest depositional environments similar to eastern Michigan. Throughout a large area in central Michigan, the Dundee formation has been almost entirely converted to dolomite (Fig. 3). Only a thin (<15 ft), tight interval at the top of the rogers city limestone, identified in drillers' logs as the "cap limestone," remains undolomitized. hydrocarbon production in the central basin comes from dolomitized intervals below the "cap limestone."

Initial production values in central Michigan fields group into two categories. Most wells have IPs of several hundred barrels of oil per day, while wells concentrated in "sweet spots" within each field have IPs of several thousand barrels of oil per day.

Initial production rates are related to reservoir quality, which may be controlled by structural position. Presence of fractures and vugs in cores suggest that higher production rates are associated with zones which contain fracture and solution-enhanced porosity.

The productive zone in central Michigan normally occurs immediately beneath the cap limestone in the upper part of the dolomitized Dundee formation and is known in this region as the Dundee porosity zone. It is underlain by a thick water leg with an active water drive. Winterfield field⁸ and Crystal field (this article) provide good examples of central Michigan Dundee production from dolomite reservoirs enhanced by fracturing and solution. It is in this central Michigan productive belt that Crystal field, and the other 29 Dundee fields being characterized in this project, are located.

Crystal field history

Crystal field (Fig. 4 [19804 bytes]) was discovered on May 29, 1935, with the drilling of the Daily Crude, J. Tow No. 1 well.⁹

The discovery well was drilled on an anticlinal structure defined by subsurface mapping using data from nearby wells. Development was rapid; 193 producing wells were drilled, most before 1939 (Fig. 5a [11928 bytes]).

Decline was also rapid (Fig. 5b [9071 bytes]). Most wells had a productive life of less than 5 years.

The oil leg in the Dundee reservoir is thin (typically <30 ft) and is underlain by a thick water leg with an active hydrodynamic drive. as a result, wells that were produced at excessive rates coned water and rapidly watered out. by 1940, 95% of crystal field's present cumulative production of nearly 8 million bbl of oil had been recovered. by 1995, when this project's demonstration well was drilled, only seven conventional wells remained on production in the field, the best one producing 5 b/d of oil and 600 b/d of water.

Two productive lithologies can be inferred from a contour map of IP values (Fig. 6 [16104 bytes]). Most wells had IPs of several hundred barrels of oil per day. However, in several "sweet spots" in the field, IPs of more than 1,000 b/d were common. Nine wells had IPs greater than 5,000 b/d, and the maximum IP for the field exceeded 9,000 b/d.

Core cut in a test well for this project in Crystal field, and core from a wildcat well located 8 miles to the northwest contain fractures and vugs, suggesting that fracture and solution-enhanced porosity may be responsible for the high-IP "sweet spots" in Crystal field.

Next: Drilling of and production from the demonstration well and conclusions.

The Authors

J. R. Allan operates Allan Geological Analysis, which specializes in core, reservoir, and regional studies; carbonate geology; and inorganic geochemistry. He holds a PhD from Brown University. He is a research associate professor with the Department of Geological Engineering and Sciences at Michigan Technological University. He was previously a senior research associate with Chevron Oil Field Research Co., where he studied carbonate reservoirs in the Arabian basin, Western Canada sedimentary basin, and most major U.S. carbonate provinces.
William B. Harrison III is a professor of geology at Western Michigan University and director of the Michigan Basin Core Research Laboratory. He has a PhD from the University of Cincinnati in paleontology and sedimentology. His areas of specialization include stratigraphy and sedimentology of the Michigan basin, reservoir characterization of oi- and gas bearing strata in Michigan, and resource assessment of Michigan oil and gas fields.
Jacqueline E. Huntoon is an associate professor in the Department of Geological Engineering and Sciences at Michigan Technological University. She has a BS degree from the University of California at Santa Cruz and an MS degree from the University of Utah. She received a PhD in 1990 from Pennsylvania State University. Her research projects use both field data and computer modeling to analyze the development of sedimentary basins in a variety of geologic settings.
Wayne D. Pennington is a professor of geophysics in the Department of Geological Engineering and Sciences at Michigan Technological University. He specializes in petrophysics, acoustic properties of rocks, and formation stresses and strength for wellbore stability, fracture design, and induced seismicity. He has previously held positions at the University of Texas at Austin's Department of Geological Sciences and Marathon Oil Co.'s Petroleum Technology Center and holds degrees in geology and geophysics from Princeton University, Cornell University, and the University of Wisconsin-Madison.
Eric T. Taylor is an independent geologist in Traverse City, Mich. After graduating from Adrian College, Adrian, Mich., in 1979, he worked as a geologist for Wiser Oil Co. in Dallas, and then for H.E. Tope Inc. from 1979-84. Since 1984 he has been an independent geologist and has worked extensively with Terra Energy Ltd. and Cronus Development Corp.
Craig J. Tester is vice-president of Cronus Development Corp. in Traverse City, Mich. He has a BS in mechanical engineering from Michigan Technological University. After holding various positions in petroleum engineering and formation evaluation with Amoco Production Co. in Texas and Oklahoma, he served as vice-president of production and exploration with Terra Energy Ltd. He has been actively involved in Michigan basin oil and gas exploration and production since 1982.
James R. Wood is a professor of geology at Michigan Technological University. He spent 11 years at Chevron Oil Field Research Co. as a senior research associate and 3 years at the University of Wyoming as an assistant professor. He specializes in geochemistry of oilfield fluids and visualization, interpretation, and management of large geological data sets. He received a PhD from Johns Hopkins University.