TITLE
ADVANCED CHARACTERIZATION OF FRACTURED RESERVOIRS IN CARBONATE ROCKS: THE MICHIGAN BASIN
TYPE OF REPORT: SEMI-ANNUAL
REPORTING PERIOD START DATE: OCT. 1, 1998
REPORTING PERIOD END DATE: MARCH 31, 1999
PRINCIPAL AUTHORS: JAMES R. WOOD, MICHIGAN TECHNOLOGICAL UNIVERSITY, HOUGHTON, MI
WILLIAM B. HARRISON, WESTERN MICHIGAN UNIVERSITY, KALAMAZOO, MI.
DATE REPORT WAS ISSUED: APRIL 1999
DOE AWARD NUMBER:DE-AC26-98BC15100
NAME AND ADDRESS OF SUBMITTING ORGANIZATION: MICHIGAN TECHNOLOGICAL UNIVERSITY, 1400 TOWNSEND DRIVE, HOUGHTON, MI. 49931
DISCLAIMER
This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe on any privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation or favoring by the United States Government nor any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government.
ABSTRACT
TABLE OF CONTENTS
TITLE......
DISCLAIMER......
ABSTRACT......
TABLE OF CONTENTS......
LIST OF GRAPHICAL MATERIALS......
EXECUTIVE SUMMARY......
RESULTS AND DISCUSSION......
Task 1. Project Management......
Subtask 1.1 Technical Aspects......
Subtask 1.2 Financial Reports and Accounting......
Task 2. Basin Analysis......
Subtask 2.1 Geology......
Subtask 2.2 Geophysics......
Subtask 2.3 Hydrology......
Task 3. Quantification and Mapping......
Subtask 3.1 Data Acquisition......
Subtask 3.2 Fracture Analysis, Mapping and Visualization......
Subtask 3.3 Fracture Analysis......
Task 4 Geochemical Studies......
Subtask 4.1 Diagenesis......
Subtask 4.2 Fluid Geochemistry......
Subtask 4.3 Hydrocarbons......
Task 5. Technology Transfer......
Subtask 5.1 Public Outreach......
Subtask 5.2 Workshops......
Subtask 5.3 Meetings......
CONCLUSIONS......
REFERENCES......
BIBLIOGRAPHY......
APPENDICIES......
LIST OF GRAPHICAL MATERIALS
EXECUTIVE SUMMARY
The primary objective of this project is for a university–industry consortium to develop a comprehensive model for fractured carbonate reservoirs based on the “data cube” concept using the Michigan Basin as a prototype. This project will combine traditional historical data with 2D and 3D (?) seismic data as well as data from modern logging tools in a novel way to produce a new methodology for characterizing fractured reservoirs in carbonate rocks.
Advanced visualization software will be used to fuse the data and to image it on a variety of scales, ranging from basin-scale to well-scales. Several new logging tools will be used in this study: (1) formation scanner logs will be used to delineate and quantify the fracture systems, and, (2) hydrocarbon logs, which are similar to traditional well logs except they are obtained by measuring hydrocarbon contents from fluid inclusions cuttings and core samples, will be used to identify and map diagenetic “halos”.
In addition to the logging tools, traditional geologic and engineering data will be used to define and interpret the reservoirs. These include: formation tops and reported hydrocarbon shows obtained from driller’s reports, scout tickets, well logs and seismic data. Engineering data will include porosity, permeability, drill stem tests, initial production, API gravity and cumulative production histories. Selected data measured from outcrops and cores will be gathered to provide a regional framework for the type and geometry of natural fracture sets in the Basin.
INTRODUCTION
In many basins world-wide fractured carbonate zones form important oil and gas reservoirs. Hydrocarbon production in the Michigan Basin is primarily from carbonates, both limestone and dolomite. Fractures are known to have multiple effects on Michigan reservoirs:
- Fractures may represent the dominant permeability pathway for fluid flow from the formation to the borehole. For example, the Antrim Shale (Upper Devonian) has a tight, dense matrix of highly organic shale and quartz silt with almost no permeability, but this formation is a prolific producer of natural gas which desorbs from the matrix into fractures and joints, then into the borehole.
- Fractures may improve connectivity of conventional reservoir compartments that are variably distributed throughout a hydrocarbon accumulation. Facies-controlled or diagenetic heterogeneity may be mitigated by vertical and horizontal fractures that connect the compartments to each other and to the borehole.
- Fractures may have enhanced diagenetic processes that create reservoir quality from non-reservoir precursors, or enhance rock properties to greatly improve the porosity and permeability. The Dundee Formation contains many good examples of this type of reservoir, Crystal Field in Montcalm County and Deep River Field in Arenac County, to name a few. Documenting and understanding the control of fractures on diagenetic enhancement of Michigan carbonate reservoirs is one of the primary objectives of this project.
Spectacular reservoirs, consisting of porous and permeable dolomite pods within tight, non-reservoir quality limestone, represent one of the most prolific, but enigmatic hydrocarbon accumulations in Michigan. Termed “dolomite chimneys”, in the Michigan Basin, they have long been among the most prolific producers of hydrocarbons in the world. Classic fields, such as, Albion-Scipio (Trenton/Black River Fm.) and Deep River (Dundee Fm.) represent these types of reservoirs and have the greatest cumulative (Albion-Scipio, 125 million bbls.) and per acre (Deep River, 28,000 bbls/acre) recoveries in the Michigan basin. However, key aspects of their origin(s), distribution and architecture have been enigmatic. They have been difficult to find and once found, many have proven difficult to produce efficiently.
The Michigan Basin is well suited to serve as a model for understanding the role of fractures in creating these types of reservoirs. It is a mature basin that contains almost 50,000 gas and oil wells with extensive data and rock samples. Over 150 million barrels oil have been produced from this type of fractured carbonate reservoirs in Michigan and adjacent states. The Dundee Formation alone has produced over 350 million barrels, approximately 40-50 million from fractured, dolomitized reservoirs. It has been estimated that nearly this amount of hydrocarbons remains to be recovered.
The data collected and the data cube itself will be on the Internet in digital form together with the software packages required to display and manipulate the data. The software will permit visualization and interpretation on both large and small scales. The main deliverable will be a data cube for the Michigan Basin that will include:
- A library of formation tops picks (300,000+)
- digitized well locations (latitude & longitude; 50,000+)
- scanned images of well header records
- digitized and interpreted logs of key wells
- hydrocarbon logs,
- engineering data, and
- key horizons picked from 2D & 3D seismic data, if possible.
- Core and outcrop measurements of fractures
The basin model and data will be integrated and displayed using a standard GIS format such as that employed by ArcView and ArcInfo. Detailed case histories and tutorials will be provided. Sponsoring organizations include: Marathon Oil Company, Chartwell Properties, L.. L. C., Advanced Hydrocarbon Stratigraphy, Newstar Energy, Dart Energy and Western Atlas Wireline Services.
RESULTS AND DISCUSSION
Task 1. Project Management
Subtask 1.1 Technical Aspects
Project management has gone smoothly. All tasks are on schedule and no major revisions in the SOW or budget have been submitted.
Subtask 1.2 Financial Reports and Accounting
All required reports have been submitted. Budget expenditures are on track. No revisions have been submitted.
Task 2. Basin Analysis
Subtask 2.1 Geology
Subsurface data have been collected for 2 key fractured reservoirs in the Eastern Michigan Basin. These reservoirs are the Deep River Field in Arenac County and the North Adams Field in Arenac Co. Additional data will be collected for similar fields, such as Pinconning Field in Bay Co. The data consist of formation top picks, “top porosity” picks and production data for 106___ wells in Deep River and 52 in North Adams. These data have been loaded into Geographix software and the Atlas software package being developed as part of this project (see __ below for description of “Atlas”).
In addition, data on paper records (scout tickets, driller’s reports and well logs) has been collected on nearly every field in Michigan. Work is in progress to reduce this data to digital form for preparation of basin-scale geologic and structure maps.
Subtask 2.2 Geophysics
2.2.1 Seismic
Three 2D seismic data lines have been obtained and loaded into GeoQuest software. These lines cover Crystal Field in Montcalm County and will be further processed to elucidate the structure at the Dundee level (~3000 ft. subsea) over the field. Another 2D line has been obtained over Deep River. This line is new and was shot by Bay geophysical to demonstrate their ability to “see” through the glacial drift and image the reservoir more accurately. This is a new development not included in the original proposal, but a direction that we will pursue because it is relevant to the study and if verified, will greatly enhance our ability to use 2D seismic in Michigan fields.
2.2.2 Borehole
Six image logs (Baker-Western Atlas CBIL logs) have been acquired from our industry partners and work is beginning on analysis and interpretation. At present the main problem is obtaining software that will read these files. Western Atlas has indicated that they will provide us with a copy of their software package. We also think that GeoQuest will handle these files, but that has not yet been tested.
Subtask 2.3 Hydrology
2.3.1 Fluid Pathways
Work has begun on the main problems associated with development of a basin-scale flow model for fluid (hydrocarbon) migration. The top priority is to develop a geologically realistic model for the “geocontainer”, that is the shape of the basin boundaries and the rock surfaces and volumes that fill the container. We have decided to use existing public domain gravity data to help define the basement . The lithologic fill will be obtained from the data obtained in subtask 2.1
2.3.2 Flow Model
A buoyancy-driven flow model will be developed at a later stage in this project.
2.3.3 Gas and Oil Trapping
Maps showing hydrocarbon trapping in the Michigan Basin will be determined using data obtained in subtask 2.1 as well. We will map the distribution of all known major fields in the Michigan Basin on several scales (field and basin). We will also attempt to locate data for gas and oil shows and map these as well. These data are more difficult to come by as they are not usually reported on scout tickets and only sporadically on the driller’s reports. We will try to obtain and analyze mud logger’s logs for this data.
Task 3. Quantification and Mapping (WBH)
Subtask 3.1 Data Acquisition
3.1.1 Data Cleanup and Digitization
Data acquisition has begun with the collection and organization of the necessary paper records (see subtask 2.1 above). Images of 2000 – 3000 scout tickets have been obtained so far, with about 10-000 – 20,000 (est.) to go. These are TIF (Tagged Image Format) files and work is in progress to write a software package that will retrieve and display these images.
3.1.2 Gridding
Progress on the data gridding front has been made in several directions. One, an economical commercial shareware package has been located (QuickGrid, Perspective Edge Software, . QUIKGRID is a program written for MS WINDOWS which will read in a set of scattered data points (x, y, z) which represents a surface. The program will generate a grid from this data and then display the surface as a contour map, QUIKGRID runs under Windows 95/98 or Windows NT. The generated grid may be output to a file as a series of XYZ triplets, in the DXF 3DFACE format or as an ER Mapper Raster Dataset.
Second, a new gridding algorithm has been under development as part of this project. The goal was to grid tops data in such a way that any field-scale fracture information would be preserved and then to be able to view the gridded data in 3 dimensions. So far an algorithm has been developed and applied to field-scale data for several fields. It is not clear whether this algorithm will enable field-scale fracture detection, but it does appear to provide an unbiased picture of the data. The 3D code for the visualization is described in 3.2.2 below.
3.1.3 Database Management
All data associated with this project to date has been placed into a MS Access database as promised. In addition, all documents related to the project (reports, software, etc.) have also been placed in a digital database that consists of the MS Windows normal file structure.
Subtask 3.2 Mapping and Visualization
2D Mapping
Maps of Deep River Field and Adams\N. Adams fields have been prepared using Geographix. These maps show the “top porosity” (Figure __). A common practice in Michigan when developing prospects is to map a parameter termed “Top of Porosity”, essentially the first encounter of the drill bit with diagenetic dolomite. This parameter is recorded in scout tickets and driller’s reports for several of the carbonate units in the Michigan Basin, including the Dundee and Trenton Formations. In many Dundee Reservoirs, the main pay zone is altered limestone that occurs off structure. Most of these traps are classified as stratigraphic.
The Deep River Field in Arenac County, MI. is a good example. Here the main reservoir is a diagenetic dolomite lying roughly NW – SE in a trend that is clearly off the structural highs mapped on the surface of the Dundee Formation. In the accompanying figure, the dolomite trend is superimposed on the structure contours of the Dundee Formation (lines) and show the initial production (IP) for each well as a white circle. The diameter of each circle is proportional to the IP. Tight, unaltered, limestone (“Tombstone”) forms an excellent seal where present. Where the dolomitization has penetrated through to top of the limestone, the overlying Bell shale forms an effective seal. Two prominent dolomite lineations are apparent, but only one, the main NW – SE, trend was developed.
3.2.2 3D Mapping
A 3D code for displaying the gridded data described in 3.1.2 above is under development. A working version of the code has been written in Visual Basic (VB) is available in a pre-release form.. The program is based on a commercial software tool sold as an ActiveX object by KL Group named “Olectra Chart”. This tool provides all the graphics capability and allows us to bind our databases to a chart and then display the data
3.2.3 Reports and Maps
Michigan Atlas – An electronic atlas of the major fractured reservoirs in the Michigan Basin is being prepared as part of this project. Thirty earlier fields were placed in the atlas as part of a previous DOE-sponsored study. These fields will be re-mapped looking for evidence of fractures, similar to the Deep River and N. Adams Fields. In addition, as many new fields as we can find and treat, perhaps 12-15, will be added to the atlas. The final format for the atlas is still under development, but the plan is to make it comprehensive (e.g. treat more than just the fracture characteristics) and also tailor it for the World Wide Web.
This atlas will be one of the primary deliverables from this project and will be one of the main means of technology transfer. As such, we plan to present it at various meeting over the course of the project, beginning with the MOGA (Michigan Oil and Gas Association) meeting in May at Traverse City, Michigan. We hope to obtain input and additional data from the companies and geologists actively working the Michigan Basin. So far, this task is moving along nicely and should be ready for a first release on CD ROM at the end of the 1st project year this October.
Subtask 3.3 Fracture Analysis
An extensive bibliography relating to Michigan carbonates and fractured reservoirs is being compiled. It should be completed for the annual report. Several unpublished master's theses from Michigan universities, which researched geology and fractures in the basin, have been obtained. These references contain regional and field-scale data on fracture frequency and orientation collected from cores and outcrops. These data, usually in the form of "rose diagrams", in being compiled into a digital format for integration into the data cube.
Cores and core analyses from Dundee wells are undergoing preliminary examination for fracture occurrence. A search of log archives is also beginning to locate log suites that contain borehole imaging or fracture identification logs. Data from these logs will be integrated into the data cube as it is compiled.
Task 4 Geochemical Studies
Subtask 4.1 Diagenesis
Dolomitization is the primary diagenetic process operating in the carbonate rocks of Michigan. It is also responsible for much of the reservoir capacity in formations such as the Dundee, and as such, has been more carefully documented in driller’s reports and mud logs. As a result, the field-scale diagenesis can be mapped in fair detail in the Dundee and Trenton formations just by picking the data out of driller’s report. (It is often denoted as “Top Porosity”.)