The Ekofisk Field, Evolution of the Understanding of Reservoir Fluid Flow Controls

James E. Sylte
ConocoPhillips, Houston, TX, USA

The intent of this presentation is to discuss the current understanding of fluid flow control in this complex naturally fractured chalk reservoir, and how this understanding evolved over the past 40 years.

The Ekofisk field is a naturally fractured chalk reservoir located in the Norwegian sector of the North Sea. This volatile oil reservoir is characterized by thickness up to 1000’ in the crest, porosities in the 25-40% range and low initial water saturations. The productive surface area is 12,000 acres. The field was initially over pressured to 7120 psig at 10,400’, with a bubble point pressure of 5545 psig. Two thirds of the oil in place is contained in the Ekofisk formation of Danian age, and the lower third in the Tor formation of Maastrichtian age.

The tight chalk matrix consists of microscopic skeletal fragments. This matrix is typically very high in porosity, but consists of low permeabilities on the order of 1-2 md. The high porosities have been preserved by a combination of early oil migration into the reservoir and overpressures, which effectively transferred overburden stresses from the rock matrix to the reservoir fluids. The productivity of Ekofisk is due to an extensive natural fracture system. The two main fracture types are tectonic fractures and stylolite-associated fractures. Tectonic fractures are common to abundant in both the Ekofisk and Tor formations, while stylolite associated fractures are generally associated with the middle Tor formation.

The Ekofisk field was discovered in December, 1969, and was the first major oil discovery in the North Sea. Initially, there were concerns if the high porosity chalk reservoir could maintain commercial production rates due to closure of the natural fractures as pressures were drawn down with production. To evaluate this, an early production system was put in place consisting of subsea tie-backs of the four exploration wells to a central production facility located on a converted jack-up rig. Production began in July, 1971, only 18 months after discovery. The results were positive with the four wells maintaining nearly constant rates of about 10,000 bopd each for 3½ years.

Permanent production facilities began operation in 1975. All produced gas was reinjected until the completion of the gas export pipeline in 1977. Limited volumes were injected from then until 1997, to accommodate gas availability in excess of sales. Rapid gas breakthrough was observed in a number of wells, and the field was producing with very high GOR’s early in the field life. The GOR was believed to be channeling through the fracture network.

A peak production rate of 350,000 bopd was achieved early in the project, with production declining to below 70,000 bopd by 1987. This natural decline was in line with expectations. Gas breakthrough was limiting the area of viable producers due to gas cycling issues.

Water injection was considered an alternative solution gas drive supplemented with crestal gas injection, but it was recognized that was extensive channeling of water through the fracture system could render the flood ineffective. A series of water injection tests was initiated early in the life of the field in an attempt to determine if a waterflood would be viable. Initially, this consisted of laboratory experiments and simulation studies. This progressed to a pilot injection test in the Tor formation which began in 1981. The results were positive and in line with laboratory experiments. Extensive dual porosity simulations were performed to validate the results. This was the basis for the approval of the full scale water injection in Ekofisk, which was implemented in a series of phases beginning in 1987. Capillary imbibition was believed to be the primary drive mechanism.

The pilot injection was then extended to the Lower Ekofisk formation. Although the laboratory data was less encouraging in the Ekofisk than Tor formation, the large volume of oil in place (and potential waterflood reserves) was the primary motivation for doing this test. The results were more positive than indicated from the lab data alone. This was attributed in part to core preparation techniques, but more importantly to the establishment of a viscous component to the drive mechanism. These results opened the door for further expansion of the waterflood.

The full field waterflood itself was implemented in 3 main phases. The initial phase began in 1987 focused on the northern Tor formation. The second phase began in 1991, extending the waterflood to the southern portion of the field, and adding injection intervals in the Lower Ekofisk formation. The third phase was implemented in 1993, extending the flood vertically for all the Ekofisk intervals. A steady rise in oil rates was seen form the low of 70,000 bopd in 1987, up to a plateau 300,000 bopd in 1998. This plateau was held nearly 10 years. Additional optimization projects have continued to improve recovery and extend the waterflood to the extreme southern extent of the field into the tighter and less fractured intervals.

A key to the success of the waterflood has been a functional understanding of the fracture network in the field. Early on it was recognized that injection/production well pairs located directly along the lines of fracture trends resulted in early water breakthrough and high water cuts. Well patterns which forced water to flow perpendicular to fracture trends resulted in long periods of water free production. Thief zones were also identified early, and due primarily to low vertical permeabilities, excessive water production could be avoided by not perforating these intervals. In cases where these zones were perforated, excessive production of injection water was seen.

Further challenges in the future include EOR projects such as CO2 injection or chemical flooding, and the application of these technologies are heavily dependent on fluid flow through the matrix blocks in a post-waterflooded state without the benefit of capillary imbibition to draw fluids from the fractures into the matrix blocks. Viscous, gravity, and diffusion forces will be the primary transport mechanisms for contact of the injected solvents with the residual oil in the matrix blocks.

A number of technological advances have been important in the understanding of fluid flow at Ekofisk. Advances in reservoir characterization such as time lapse seismic have led to the more accurate mapping of water movement. Well planning in which extensive data can be simultaneously visualized, and aggressive well monitoring, have allowed for identification of the best possible drilling locations. Drilling advancements including horizontal wells, have increased the volume of data available considerably. Integrated geologic, reservoir, drilling, and operations, have resulted in improved understanding of fluid flow, which has contributed significantly to the goal of improved sweep efficiency.

Acknowledgments

The author acknowledges permission to present this poster from ConocoPhillips Skandinavia AS, Total E&P Norge AS, ENI Norge AS, Statoil Petroleum AS and Petoro AS

AAPG Search and Discovery Article #120034©2012 AAPG Hedberg Conference Fundamental Controls on Flow in Carbonates, Saint-Cyr Sur Mer, Provence, France, July 8-13, 2012