Fracture Identification and Productivity Predictions in a Carbonate Reef Complex

Abstract

Abstract In order to optimise well completions and production in naturally fractured reservoirs and to predict the effectiveness of enhanced oil recovery schemes, it is important to locate and describe the insitu fractures. A combination of log measurements and core data were employed to locate and orient the fractures and assess their overall contribution to reservoir performance in a Devonian reef complex. These techniques include the imaging of the wellbore with the Formation MicroScanner Log (FMS)* and the Borehole Televiewer (BHTV). The image data were compared to detailed core analysis, where available, and were used to determine the length, orientation and probability of fractures as a function of porosity. The Array-Sonic* tool was also run to help identify fractured zones using full waveform data and to quantify producibility from Stoneley wave data in comparison to flowmeter production logs. From the comparison of the images with core, the utility of the FMS and BHTV Logs were established in identifying fractures and fracture sets as well as determining the orientation and vertical extent of the fractures. In addition, the response of FMS images to different porosity types was confirmed by core data. Comparison of the Array-Sonic data and the flowmeter production log data established the usefulness of the Stoneley wave attenuation as a means of identifying permeable zones. The study results permit better definition of tool response to porosity types and fractures, and reveals the influence of fractures on the overall production of the reef complex. These results have implications on the initial and secondary production techniques of this and other carbonate reservoirs. Introduction In low porosity, low permeability zones, natural fractures are the primary source of permeability which affect both production and injection of fluids. The open fractures do not contribute much to porosity, but they provide an increased drainage network to any porosity. They may also connect the borehole to remote zones of better reservoir characteristics. Whatever the conclusions inferred from fracture location to date, only dynamic measurements of pressure or production results can tell whether or not an open fracture or fractured zone will produce. The productivity of the well will also vary widely with lateral and vertical extent of the fractures, fracture width, and matrix porosity and permeability. The initial impetus for this study came from a requirement to better understand the influence of fractures on production in a carbonate reef complex, in preparation for a miscible flood project. Preliminary core analysis indicated the presence of fractures, but the degree, effectiveness and orientation of fractures was not obvious. The objectives of the logging program were first to identify fractures, second to orient the fractures and third to predict their influence on the production of individual wells and their impact on miscible fluid flow through the reservoir. To accomplish this, several logging measurements were undertaken to find and orient the fractures on five producing wells. These measurements centered on imaging techniques using the FMS Log and BHTV data. In addition to standard resistivity and porosity log data, Array-Sonic Log data were recorded to provide compressional, shear and Stoneley wave data. The wells were selected to provide limited areal coverage and to maximize the available core data for comparison to the log data. The logging was accomplished in two stages, the basic logs for porosity, lithology and water saturation were run when the wells were initially drilled and the imaging and Array-Sonic data were recorded during a workover of the barefoot completed wells. In addition flowmeter data was available for all five wells, which helped to identify producing zones in each well. Production information, including pressure and tracer data from the reservoir suggested that the wells were all in communication, that there was pressure communication between the various layers of the reservoir and that some areas of the reservoir exhibited a preferred flow orientation. In all, five wells were logged with FMS and BHTV tools for the study. P. 579^