Derek A. Reynolds, Ken P. Seymour
Ranger Oil (U.K.) Ltd.
Aberdeen
Excessive water production from hydraulically fractured wells in a poor quality reservoir in the North Sea prompted the drilling of a horizontal well. Gas production from the horizontal well reached six times that of the offset vertical wells, and no water production occurred. This horizontal well proved commercial the western section of the Anglia field.
Horizontal drilling in the North Sea is as an effective technology to enhance hydrocarbon recovery from reservoirs that previously had proven uncommercial with other standard techniques. It is viable for the development of marginal reservoirs, particularly where conditions preclude stimulation from hydraulic fracturing.
FIELD DEVELOPMENT
The Anglia gas field, situated in North Sea Blocks 48/18 and 48/19, was discovered by Well 48/18b-1, spudded in February 1972 (Fig. 1). Despite the relatively early discovery of a significant gas accumulation in the Anglia field, depressed gas prices delayed detailed seismic and appraisal well work until 1984. Appraisal activity in the field has continued because of an extensive gas transmission system now in place and the need for increased U.K. gas supplies. Annex B approval has been granted for the Phase I development of the eastern side of the field.
The Anglia field is a normally pressured (4,000 psia), Rotliegendes, dry gas reservoir, found at a depth of approximately 8,500 ft subsea (SS) true vertical depth (TVD).
Prior to the drilling of horizontal Well 48/18b-9, nine wells had been drilled through the structure finding varying degrees of reservoir quality. Rock permeability is generally poor though variation between sands is significant.
The field consists of two sections, the east and west regions. Wells in the eastern section have attained flow rates up to 36 MMscfd. However, low productivity, often noncommercial, in the western section of the field prompted hydraulic fracturing of those wells.
Unfortunately, difficulties with frac containment incited excessive water cuts, up to 110 bbl/MMscf. This negated the viability of fracture stimulation as a means of effecting commercial gas production.
Horizontal drilling was considered for development of the western area of the field because of the perceived advantages of the horizontal wells in tight reservoirs. These wells could capitalize on the increase in horizontal permeability anisotropy associated with aeolian sandstones. The larger well bore contact area is similar to the theory behind hydraulic fracturing.
WELL PLAN
Ranger Oil (U.K.) Ltd. first considered horizontal drilling in the Anglia field in early 1988. 1 2
A general east-west orientation of the horizontal section, normal to the strike of dune foreset bedding, would maximize permeability anisotropy effects. The two areas selected included Bi, near Well 48/18b-3, and B2, near Well 48/18b-1 (Fig. 2).
The proximity of B1 and B2 to previously drilled and tested wells would enable a full comparison of a horizontal well to a conventional vertical well in the same section of the reservoir.
The initial surface location was selected as a midpoint between B1 and B2. Any future development of the west field could then use this location for the platform position.
WELL DESIGN
Down to the base of the 12-1/4 in. hole, the well was designed like any standard deviated well in the Southern Gas basin. Because of the uncertainty of the length and composition of the evaporate section in the Zechstein group, it was impossible to accurately predict build rates in this section. Ranger Oil therefore decided to build angle in the anhydrite/dolomite sequence at the base of the Zechstein group and in the Rotliegendes where the build rate could be controlled more accurately.
The 9-5/8 in. casing would be set in the Werraanhydrite at a well deviation of about 40, with the final build to the horizontal section achieved in the 8-1/2 in. section. The horizontal section vertical depth was fixed relative to the gas-water contact, and the well profile could be closely defined and followed.
The proposed build up would begin at 6,975 ft TVD and finish at the start of the horizontal section at 8,855 ft. The programmed build rates were up to 4.5/100 ft in the 12-1/4 in. hole and up to 7/100 ft in the 8-1/2 in. hole section.
Throughout the build section of the 8-1/2 in. hole and the horizontal section, the bottom hole assembly would incorporate logging while drilling (LWD) tools. These tools would provide real-time neutron density, resistivity, gamma ray, weight on bit (WOB), torque, direction, inclination, and tool face throughout the build section.
Successive vertical wells combined with 3D seismic evaluation helped build an accurate map of the reservoir. Ranger Oil divided the reservoir into six zones, with Zones 3 and 5 having the higher permeability sands. The highest permeability exists at the base of Zone 3 and at the top of Zone 5.
The final proposed horizontal section was planned to commence as close as practicable to the surface location and extend for approximately 2,500 ft to maximize the amount of Zones 3 and 5 in the well bore. The optimum vertical depth of the horizontal section was 70 ft above the gas-water contact at 8,790 ft SS with a 15 ft tolerance window.
DRILLING OPERATION
The 36-in., 26-in., and 17-1/2 in. hole sections were drilled vertically using conventional rotary assemblies.
The 17-1/2 in. hole section was drilled using a salt-saturated/polymer water-based system. The length of the 17-1/2 in. section and the lithologies encountered caused a few problems. Unplanned angle building occurred in the 17-1/2 in. hole section (maximum 4), but this was accommodated in the directional program.
Achieving high build rates in the Zechstein group became a problem. The Stassfurt Halite, thought to be 150 ft thick, actually was almost 600 ft thick.
Considerable reactive torque was encountered in this formation causing a reduction in WOB to 15,000 lb to maintain high-side tool face. Because of these drilling parameters and the extended formation, it was not possible to achieve the programmed build rate. The average build through this section was only 1.64/100 ft.
Below this formation, WOB was increased to 30,000 lb which increased build rate to 3.26/100 ft; but this was insufficient to achieve the plan. With certain changes to the BHA it was possible to increase build rate to a maximum of 4/100 ft. As a result, the 9-5/8 in. casing was run higher in the Werraanhydrite than originally planned--this gave more 8-1/2 in. hole to make up hole angle.
BUILDING ANGLE
During the 8-1/2 in. hole, it was essential to achieve build rates in excess of 7/100 ft to reach horizontal at the required TVD. Other operators' experiences indicated that a BHA with a full suite of real-time LWD tools may encounter difficulty in achieving these build rates. However, it was considered necessary to attempt to gain as much information from the well bore as possible.
Drilling proceeded in the Werraanhydrite, at an average rate of penetration (ROP) of 30 ft/hr and with an average build rate of 7.42/100 ft. The bit reached the reservoir at 8,640 ft measured depth (MD), and the ROP increased to 60 ft/hr.
After drilling approximately 75 ft of the reservoir and with the hole angle at approximately 43, it became increasingly difficult to deliver weight to the bit. Several changes to the LWD configuration were required until ultimately all the LWD sensors except gamma ray were removed. This maximized flow rate through the drillstring and provided more flexibility to the BHA to facilitate sliding of the drillstring in the oriented mode. Average build rates dropped to less than 5/100 ft. It became clear that reaching horizontal by the desired TVD was no longer likely with a conventional bent housing motor. A more aggressive building assembly with an eccentric housing motor was run in the hole. The eccentric housing motor could build angle at 13/100 ft, necessary to reach horizontal at the targeted TVD.
Initial surveys recorded average build rates of 9.7/100 ft, rising at the end of build section to 11/100 ft. A survey at 9,160 ft MD recorded 84.2 inclination, extrapolated to an inclination of 91 at the bit at a TVD of 8,837 ft. Horizontal was reached within 3 ft of the targeted TVD. Because of the higher radius at the end of the build section, the horizontal section had been reached 200 ft closer to the surface location than planned.
One bit and BHA drilled the first 1,048 ft of horizontal hole at an average ROP of 25 ft/hr. The next 814 ft of hole required six bits because the very abrasive sands with intermittent claystone sections caused extreme wear to the cutters. One bit drilled the last 1,036 ft, but this section needed a BHA change to stop a drop tendency and build angle back to required TVD.
The end of the horizontal section was extended from the programmed TD until it was evident that as much of the permeable sands as possible had been drilled. The total length of the horizontal hole was 2,940 ft (Fig. 3).
The 8-1/2 in. hole was drilled with a standard 70/30 (oil/water ratio) oil-based mud. This type of mud was chosen because of experience from previous wells that had no apparent formation damage from oil-based drilling fluid or from barite used as a weighting material.
No hole cleaning, torque, or drag problems were encountered during drilling or tripping throughout the horizontal section. A policy of backreaming and washing while pulling out of the hole minimized any hole fill or cuttings build up problems.
TESTING
The well was tested through a 5-1/2 in. predrilled, uncemented liner. This allowed flow contribution from the whole Rotliegendes section. The 5-1/2 in. liner crossed over to a 7-in. liner, with a tieback packer and polished bore receptacle (PBR) seal assembly set in the 9-5/8 in. casing (Fig. 4). The only problem in running the completion was the external casing packers which failed to inflate in the 8-1/2 in. hole.
The flow rate on this horizontal well was the highest recorded of all the Anglia wells. After being brought on with nitrogen, it had an initial flow rate of 39.5 MMscfd at a tubing head pressure of 2,665 psi with a 52/64-in. choke. The well flowed for 7 days with the flow rate decreasing slightly to 35.2 MMscfd at tubing head pressure of 2,398 psi. The condensate production averaged 1-1.5 bbl/MMscf.
At the end of the 7-day flow, the well was shut-in downhole for a 2-day pressure buildup survey. Following the pressure buildup, a spinner tool was run on coiled tubing, but the tool failed. No further attempts were made to obtain these data because the main objective had been achieved, namely proving commercial rates from Anglia West.
SUSPENSION
On completion of testing operations, the well was suspended pending production from the western part of the Anglia field in 1992 or 1993.
The 48/18b-9 appraisal well was suspended on a long-term basis. The method had to equate with a permanent abandonment (that is, the well could not be suspended with gas across the horizontal section). Brine was used to minimize formation damage. This resulted in losses to the formation which were cured with calcium carbonate. The well was subsequently suspended with three cement plugs: one plug set above the 7-in./5-1/2 in. crossover, another plug set across the liner lap, and the last plug set in the 9-5/8 in. casing at the 13-3/8 in. casing shoe depth. No mechanical plugs were used. The well was suspended with 10-ppg sodium chloride inhibited brine placed above the top plug and with a series of corrosion caps.
In October 1990, approval was given for the development of the eastern part of the Anglia field, with a requirement to deliver first gas by Dec. 1, 1991. The project consists of five wells from a normally unmanned platform, Anglia Alpha. One of the five wells was completed with a horizontal section of 3,568 ft.
Plans for the western part of Anglia include the use of subsea completions. Well 48/18b-9 will be tied back to Anglia Alpha as the first step.
RESULTS
- The 48/18b-9 well was drilled, tested, and suspended in a total of 87 days.
- The length of horizontal section drilled was 2,940 ft, a U.K. continental shelf record.
- The stabilized test flow rate from the well was 36 MMscfd, approximately a sixfold increase in absolute open flow potential from the offset vertical well.
- No major operation difficulties were encountered with regard to hole cleaning, torque and drag directional control, surveying, logging, running the liner, or testing.
- For long term suspension, research is required into how best to kill and suspend this type of well while leaving the horizontal full of gas.
ACKNOWLEDGMENT
The authors thank their colleagues for help in this article. The authors would also like to thank the Chairman of Ranger Oil (U.K.) Ltd. and the Anglia development project partners, Conoco Ltd., Elf Aquitaine, Amerada Hess Ltd., and Triton, for their cooperation and permission to write this article.
REFERENCES
- K.J. Weber, "Computation of Initial Well Productivities in Aeolian Sandstone on the Basis of a Geological Model, Leman Gas Field, U.K.," The Society of Economic Paleontologists and Mineralogists, 1987.
- Diane K. Neil Mills, "The Feasibility of Horizontal Wells in the Development of a Southern North Sea Gas Field," MSC Thesis, 1988.
Copyright 1991 Oil & Gas Journal. All Rights Reserved.
