Alberta is blessed with an immense resource of 1.6 trillion barrels of bitumen in place, locked mainly in sandstone reservoirs in an area the size of Florida. Of that total, just 4.2 billion barrels had been produced through the end of 2003, leaving 174 billion barrels in remaining proven reserves. Alberta's staggering wealth of crude bitumen propels Canada to second place in world oil reserves, behind only Saudi Arabia. Production of crude bitumen from Alberta's oil sands currently exceeds 1 million barrels a day, a meteoric rise from the 500,000 per day produced in 1997. And that's just the beginning. According to the Canadian Association of Petroleum Producers, close to US$36 billion will be spent during the next five years in new oil-sands projects. If projections hold true, by 2015 Canada will be producing 2.7 million barrels of bitumen a day. Furthermore, oil-sands production could grow for another 40 years to the astonishing level of 11 million barrels per day, according to a Calgary-based FirstEnergy Capital Corp. forecast (on a very provisional basis). The Alberta Energy & Utilities Board (EUB) defines crude bitumen as a type of viscous, heavy oil that will not flow to a well in its natural state. Alberta's oil sands occur in the Athabasca, Cold Lake and Peace River regions in its northeastern quarter. At first glance, oil sands appear very appealing because the resource is gigantic and well defined. However, the developments require lengthy lead times and are extremely capital intensive. The same projects have been on and off the drawing boards of companies for decades, their popularity fluctuating with the price of bitumen, which trades at steep discounts to lighter oils. The extraction processes use copious amounts of energy and water, and production costs range from a low of below US$8 per barrel for cold methods to a high of more than US$15 per barrel for cyclic steam recovery. And, labor shortages and eye-popping cost overruns have unnerved some investors. That said, oil-sands development is barreling forward, thanks both to high commodity prices and rapidly advancing technology. About 65% of Alberta's crude bitumen is produced through open-pit mining, and in-situ methods account for the remaining 35%. (For more on the mining side of oil sands, please see "E&P Momentum" in this issue.) Alberta's in-situ production totaled 323,000 barrels per day in 2003, and about two-thirds of that was recovered by secondary thermal techniques. The other third was produced via cold primary methods. Most of the thermal bitumen is produced by the cyclic steam stimulation (CSS) process, which has been used commercially for two decades. The other popular thermal technology, steam-assisted gravity drainage (SAGD), is the new fair-haired child of the oil patch. It is the promise of SAGD (say "sag-dee") that has opened broad areas of the oil-sands regions to commercial production, and that is responsible for the massive increase that Canada was able to post in its oil reserves in 2003. And it is SAGD that will propel in-situ production to supply some 40% of the 2.7 million barrels of bitumen a day Canada expects to be making in 2015. SAGD Roger Butler, a professor at the University of Calgary, is credited with invention of the SAGD process in the late 1970s. Success didn't come overnight. Many recovery approaches were tried and rejected before SAGD's ascension, including firefloods, steam drives, injections of various additives along with steam, and hydraulic fracturing. Groundbreaking as it was, the SAGD concept was ahead of its time and had to wait for drilling technology to catch up. That's because SAGD requires the ability to drill two horizontal wells up to 1,000 meters in length at shallow depths, and furthermore to stack the wells precisely parallel to each other with about five meters of separation. After the wells are drilled, steam is circulated into both legs for several months to establish conductivity, and then it is injected into the upper leg only. A cone-shaped steam chamber forms in the reservoir as the steam rises, cools and condenses to water. The heated bitumen flows into the lower well along with water; the bitumen is produced and mixed with a diluent-either condensate or synthetic crude oil-so it can be shipped via pipeline. The water, produced along with the bitumen, is separated, cleaned and recycled for new steam generation. Once its initial pilot was in place, SAGD was remarkably successful and operators soon embraced it. SAGD works well with shallow, high-viscosity and low-mobility reservoirs, and was an excellent fit for the vast resources in the Athabasca region's McMurray formation that were too deep to be mined. The process delivers exceptional recovery factors, typically above 50%, and high per-well production rates. It is also continuous, with no production down time. The obvious concern with SAGD-and indeed with all thermal projects-is the cost of the heat. As a broad rule, about 65% of the cost of producing a barrel of bitumen from an in-situ operation is for energy, and the balance is largely for labor. That's inverse to mining, which has labor costs of some 70% of a produced barrel. About 1,000 cubic feet of natural gas is needed to produce three barrels of steam. The steam/oil ratio (SOR) compares the cold-water barrel equivalent of steam generated per barrel of bitumen produced, and is commonly used to measure the energy intensity of the thermal processes. At present, mature SAGD projects have SORs between 2.5 and 3. In a typical project, steam is injected for five or so years, after which the wells are blown down and production declines. SAGD is an intricate and tricky process, and many variables must properly align to be successful. As use of the method has expanded over a wide area, issues and challenges have continued to crop up. Problems encountered include heterogeneities in the reservoirs that can interfere with steam rise. Thick, clean, high-quality reservoirs are necessary, with 15 or more meters of pay with high bitumen saturations and high permeabilities. The best reservoirs in the McMurray formation are channel sands, and associated shales and mud clasts can wreak havoc on steam distribution. In some areas, the bitumen reservoirs are capped by water or natural gas-filled zones, or underlain by bottom water. In these circumstances, major concerns are the presence of thief zones and production of too much water. Indeed, the competing rights of SAGD and natural gas producers are a very contentious issue in Alberta. Throughout parts of the Athabasca region, gas zones directly overlie the bitumen and the respective licenses are held by different entities. When gas production reduces reservoir pressures, the layers overlying the bitumen can soak up the steam injected for SAGD. At this time, the EUB has shut in many of the gas wells that sit on top of bitumen reservoirs. Additionally, the scalability of the SAGD projects has been problematic. Designs that work well in a small-volume pilot can fail at the much grander sizes needed for commercial production. These projects require substantial surface infrastructure, including steam generators, water recycling plants, storage tanks, pipelines and cogeneration units. Three large-scale SAGD projects-Foster Creek, MacKay River and Firebag-are currently producing bitumen in Alberta, all operated by senior Canadian firms. Several pilots are jumping to commercial scale, and many other proposals are in development by an array of operators ranging from start-ups to established independents to international majors. SAGD production is now just over 90,000 barrels a day; in the unlikely case that all the announced projects are built as planned, SAGD well pairs could be making more than 1.5 million barrels a day in eight to 10 years. Foster Creek EnCana Corp. operates Foster Creek, one of the first and currently the largest SAGD project in operation, which is making 30,000 barrels of crude bitumen a day from 31 producing well pairs. "By the end of the fourth quarter, we will be producing 40,000 barrels per day," says Harbir Chhina, vice president, EnCana oil-recovery business unit. "And by the end of 2006, we will have 50,000 barrels a day." EnCana owns 12 sections of land with a 100% working interest at Foster Creek. The leases contain original bitumen in place of 1.7 billion barrels, and recovery is anticipated to be 70%. The average thickness of the reservoir that is amenable to the SAGD process is 20 to 30 meters, average porosity is 33%, and the average permeability is seven Darcies. The oil sands have 85% bitumen saturation and the gravity is 10 degrees. EnCana produces an average of 1,000 barrels per day per well pair; each well pair currently costs C$2 million to drill and complete. A single well-pair pilot was started at Foster Creek in 1997, and three more well pairs followed. The first commercial project was 20,000 barrels per day, which the company later expanded to 30,000 daily. EnCana blends its bitumen with condensate, raising the gravity to 19 degrees, and markets that product directly without further upgrading. "We think SOR is the most important parameter in SAGD economics-it is more important than the productivity of the wells," says Chhina. During the first quarter of this year, Foster Creek's SOR was 2.3, down from the 2.5 it posted during the past two years. "We have one well on blowdown, but over the next 12 months we will have about 20 wells on blowdown, and we expect the SOR to drop to 2." EnCana is especially interested in the possibilities of low-pressure SAGD production. Historically, gas lift has been the usual method of SAGD production, due to the high bottomhole temperatures. But gas lift requires a substantial level of reservoir pressure to flow the wells. "Pressure and temperature go together with steam-the higher the pressure, the higher the temperature, and the higher the temperature the more heat we lose." To operate at lower pressures, EnCana has installed 20 to 30 electric submersible pumps (ESPs) in its oil-sands business unit, and it expects these to drop its SOR well below 2. Christina Lake is EnCana's newest project, and it is groundbreaking in several ways. The company has a six-well pilot that is producing about 7,000 barrels per day, and it plans to expand that to 18,000 to 20,000 per day. Christina is unusual because it has a top layer of gas, which was owned by Devon Energy Corp. The gas reservoir had an original pressure of 2,200 kPa that was subsequently reduced to 500 kPa after several years of production. Since EnCana wanted to operate its SAGD project at low pressures, it arranged with Devon to take over the wells and repressurize the pool with air back up to 2,300 kPa. "Because we believe in low-pressure SAGD, we have always felt there was no harm in producing the gas, as long as we have ample time to repressurize the pool." Furthermore, the project has demonstrated the successful use of ESPs instead of gas lift, which is necessary for the lower pressure operations required in gas-over-bitumen situations. For both projects, the company plans to add 20,000 barrels of daily production every 12 to 16 months, and it expects the two to produce at a combined rate of 200,000 barrels per day by 2011. "We feel that, given the market conditions in Alberta today with labor and technical support shortages, we are better off to build smaller chunks year after year than to build one big project," says Chhina. At present, EnCana has realized a 15% to 20% rate of return after taxes on its SAGD operations, and it doesn't want to endanger that by overreaching its capital efficiency. "SAGD is a child right now. It still has to grow up, and we will find ways to make it better," says Chhina. MacKay River Production began at Petro-Canada's MacKay River project in 2002 just two kilometers east of the Dover property, where the original SAGD pilot was launched. The Dover project was so encouraging that Petro-Canada moved directly to commercial production at MacKay River, skipping a pilot phase. The firm owns a 100% interest in MacKay River, which comprises 76 sections of land. The 11 sections currently under development contain between 230- and 300 million barrels of recoverable oil. MacKay River is making 24,000 barrels of 10-degree-gravity bitumen per day from 25 well pairs. "For the most part, SAGD technology has been working very well in the reservoir," says Sue MacKenzie, Petro-Canada general manager of oil-sands bitumen. "Our challenges have been predominantly in facilities reliability." Indeed, the run time for Petro-Canada's steam-generation and processing plant was initially disappointing. "Water treating has been a problem, and that's probably the biggest part of this business." Progress is being made, however: Petro-Canada increased plant reliability from 68% in the first half of 2004 to 92% in the second half. At MacKay River, the reservoir occurs at between 125 and 150 meters, and carries net pay of 34 meters with porosities of up to 33% and bitumen saturations of more than 80%. The horizontal wells are drilled to lengths of 750 meters, and the well pairs are spaced about 100 meters apart, with 13 on one pad and 12 on a second pad. The project is currently running at an SOR of 2.5 to 3. The original 25 well pairs have begun to decline, so Petro-Canada is drilling its next phase of well pairs to maintain and grow its production levels. Those will come onstream at the end of this year, says Mackenzie. In 2006, it expects to be producing 27,000 to 30,000 barrels per day. "We're at the front end of learning about declines. It looks as if the SAGD production rates stay higher for longer than rates in conventional wells, but drop off more quickly. The relative well lives are shorter, but our SAGD recovery rate is more than 70%, compared with conventional well recoveries in the 20% to 30% range," she says. Earlier this year, Petro-Canada purchased the Dover test facility and associated leases from Devon Energy. Vapor extraction (Vapex) is a new, nonthermal technology that is in the early stages of testing at the site. The DoVap project will inject solvents, such as propane and butane, into the bitumen reservoirs. If successful, Vapex could replace steam in the production process, and greatly reduce the consumption of both natural gas and water. The key issue with the process is the expense of the solvent, and the amount of solvent that can be recycled. The dozen DoVap participants, from industry and government, expect to complete the first phase of the test next year. "We've always been actively engaged in technology development in the oil sands, and it's an important piece of this business. We still have improvements we can make," says MacKenzie. Petro-Canada is also interested in expanding production onto the new leases. The company estimates the property, which it calls MacKay River 2, contains some 100 million barrels of recoverable bitumen and has potential to produce 30,000 barrels a day. And, it has additional properties that are prospective for SAGD-Meadow Creek, Chard and Lewis-but at this time Petro-Canada is focusing its in-situ efforts on MacKay River. Firebag Suncor's Firebag project covers 1,200 square kilometers some 40 kilometers northeast of its Fort McMurray mining operations. The company estimates the Firebag properties contain nearly 9 billion barrels of bitumen resources, says Steven Williams, Suncor executive vice president of oil sands. In-situ technologies are an integral part of Suncor's overall bitumen supply strategy, although its production of mined bitumen currently dwarfs its in-situ production. The company is developing the energy-intensive Firebag properties to balance its labor-intensive mining operations. "We've put the two cost drivers in tension with each other, because we want to be robust to a world in which energy and labor prices fluctuate," says Williams. Additionally, Suncor obtains third-party bitumen, which it processes in its upgraders. "The real advantages to us of in-situ are the sheer scale of the available resource, and the fact that, relative to mining, it lends itself to small development steps of C$500- to C$600-million investments," he says. "In-situ also has a significantly reduced environmental impact." To date, Suncor has drilled 40 well pairs on four pads at Firebag, and has 12 well pairs on production. The Firebag property has net oil pay thickness of some 45 meters, a cap rock that contains the steam energy, and no top gas, top water or bottom water. Reservoir qualities are outstanding, with 32% porosity, 84% bitumen saturation, horizontal permeability of seven Darcies and vertical perm of about half that. The bitumen is seven degrees API gravity. Steam injection started in October 2003, and the first bitumen was produced in January 2004. The average Firebag well pair has horizontal legs of 1,000 meters, at a true vertical depth of 320 meters. Each well pair is about 150 meters apart. The original reservoir pressure was around 800 kPa, and at present the project is operating at a pressure of around 2,500 kPa. Reservoir temperatures are 200 to 220 degrees Celsius. With stable steam injection, average bitumen production at Firebag is 1,800 barrels per day per well pair, and each pair can access 11.5 million barrels of original bitumen in place. Suncor's best well pair has already produced 900,000 barrels of bitumen, at a peak rate of 2,900 barrels per day. "Over the life of the wells, we expect a SOR of approximately 2," says Williams. "We believe that there is some way to go to reduce energy use. The two biggest technology drivers we have are to reduce the SOR through injection of light hydrocarbon solvents, and installation of downhole mechanical lifts." During 2004, the company had some shutdowns due to issues with its steam generators and produced-water system, and production levels seesawed. More recently, Suncor had a planned one-month shutdown to tie in second-stage facilities. At press time, Firebag was producing 20,000 barrels per day from its first phase at an operating cost of C$19 per barrel. This phase, which cost C$630 million, will be at its full throughput of 35,000 barrels per day by year-end. The second phase, which is about 70% completed, will raise production capacity another 35,000 barrels per day. This stage requires investments of C$500 million, as it can use some of the infrastructure already constructed. "We believe we can get the in-situ costs approximately in the range of mining costs. In-situ is already broadly competitive with mining, and we believe there is room to improve the technology." Beyond Firebag's first two stages, the EUB has approved two additional stages of 35,000 barrels each, but Suncor has not yet determined its pace for those developments. "The technology and the capability of SAGD are broadly in line with our expectations. In terms of the speed in which we've been able to bring it up and meet our requirements, it's been slightly slower," says Williams. "We're optimistic about SAGD's future, but we're realistic that we need to continue to learn and develop the technology." Jackfish Devon Energy's Jackfish project lies 140 kilometers south of Fort McMurray, immediately south of EnCana's Christina Lake operation. The company has a 100% working interest in Jackfish, which is designed to recover 300 million barrels of bitumen at a target rate of 35,000 barrels per day, says John Pearce, Calgary-based exploitation manager, thermal heavy oil. The company is drilling 25 well pairs with lateral distances of 750 to 800 meters in the reservoir, which occurs at 400 meters. Production will start during the first quarter of 2007, and will reach full volumes in 2008. The wells have a seven-year life, so the company plans a continuous drilling project to maintain volumes at peak rates. "We've found that the main challenge in SAGD is finding a simple, high-quality reservoir," he says. At the seminal SAGD pilot at the Dover facility, which Devon operated for seven years, the 20-meter-thick reservoir was very high-quality and had no bottom water, top water or top gas. "It was almost too easy. But as commercial projects have begun developing, we have found that levels of risk are added for all the different complexities we encounter in the reservoirs." At Jackfish, the challenge is bottom water. "If we run our SAGD operating pressure lower than the bottom-water pressure, we could conceivably start producing that water. If we run at an operating pressure much in excess of the bottom-water pressure, potentially the steam chamber could flow down into the water and take off on us." Consequently, Devon has designed its process at Jackfish to be in balance with the bottom-water pressure. "We're focused on the overall performance of the project-execution, start-up and production. Following that, we will grind away on the operating cost on all fronts," says Pearce. "You can do all sorts of calculations on paper, but it's all theory until we get the oil flowing down the pipe." Other SAGD projects Hangingstone. Several other hefty SAGD projects are slated to start producing significant volumes of bitumen in the next few years. Japan Canada Oilsands operates a long-running pilot at Hangingstone, which is currently producing 9,000 barrels per day. The three-stage demonstration project began in 1999; two phases of commercial production are planned that will bring production up to 60,000 barrels per day by 2012. Surmont. On the cusp of commercial production are Surmont, Long Lake and Joslyn. Surmont is a nine-township project operated by ConocoPhillips; Total is its partner. Next year, Surmont will begin producing 25,000 barrels a day and could eventually recover between 5- and 10 billion barrels. Peak production is forecast at 100,000 barrels per day, which will be reached in 2012. Long Lake. Nexen and Opti Canada's 60,000-barrel-a-day Long Lake project is being followed with high interest. The Calgary-based partners are building a combined SAGD facility and upgrader that will be fueled with synthetic gas made from asphaltene residue, a process that will drastically cut the use of natural gas. As of this spring, 71 horizontal wells had been drilled at Long Lake and a SAGD pilot was operating. Commercial in-situ production is scheduled to begin in late 2006, and upgrader operation should begin the following year. Joslyn. And, Deer Creek Energy's Joslyn project has both mining and in-situ components. A SAGD pilot was launched in 2004, and some 270 wells have already been drilled. Deer Creek, a Calgary firm, expects to start steaming a 10,000-barrel-a-day commercial phase early next year. It could be producing 40,000 barrels per day by 2010. Whitesands. An intriguing new in-situ approach is being launched by Calgary-based Petrobank Energy at its Whitesands project. The company will use toe-to-heel air injection (THAI), a process in which air is injected into a vertical well drilled above the distal end of a horizontal well, and the air/bitumen mixture is set on fire. The THAI process could open a wide slice of noncommercial resource to exploitation, as it has the potential to operate in lower pressure, lower quality, thinner and deeper reservoirs than current methods. At press time, the company was drilling its pilot wells and expects to report initial results early next year. Sunrise. Further out, Husky Energy applied for regulatory approval for its 200,000-barrel-a-day Sunrise project last year. This property contains possible recoverable reserves of 3.2 billion barrels, and an initial phase of 50,000 barrels a day could start in 2008. Great Divide. Another project is being spearheaded by Connacher Oil & Gas Ltd. The company has applied for approval to produce 10,000 barrels a day at its Great Divide project. If it receives timely approvals, operations may begin near year-end 2006. Cyclic steam stimulation Although SAGD has been grabbing the headlines, the bulk of Alberta's thermal bitumen is produced by the tried-and-true CSS method from the Cold Lake region. Imperial Oil was the pioneer of in-situ recovery methods, and Cold Lake is the granddaddy of all in-situ projects. Cold Lake produces 140,000 barrels of 10-degree bitumen per day from 3,800 wells and it contains 760 million barrels of reserves. During 2004, the 45 million barrels that Imperial produced at Cold Lake comprised more than 5% of Canada's crude oil output. Imperial operated CSS field pilots in the Cold Lake region between 1964-75 and started commercial production in 1985. The bitumen occurs in the Lower Cretaceous Clearwater formation, an unconsolidated reservoir that attains thickness of 50 meters and is covered by 400 to 750 meters of overburden. In the CSS process, more familiarly called huff-and-puff, pressurized steam is injected into the target reservoirs. The steam softens and dilutes the bitumen, and the high pressure also causes fracturing, which improves permeabilities. The wells are shut in after the steam is injected. After a "soak" period, the same wells are put on production for several months. The process is then repeated. For CSS to work, the bitumen in the reservoir must have at least some mobility, and the reservoir seal must be able to contain the high-pressure steam. Recovery factors for the process are generally in the high-20% range, and SORs typically fall between 3 and 4. At present, Imperial has a steam capacity at Cold Lake of 600,000 barrels per day, and it uses 220,000 cubic feet of gas daily. The steam is pumped into the reservoir at pressures averaging 1,600 psi (11,000 kPa) and temperatures of around 300 degrees Celsius. During the past 12 years, production at Cold Lake has grown at an average annual rate of 4%. The company is presently producing from phases 11-13, and during 2004 it drilled more than 200 development wells. Indeed, Imperial expects to drill up to 200 wells annually for the next 15 years. Its 2005 program includes horizontal CSS wells and first development work in the northern extension of its main area. Last year, Imperial received regulatory approval for two future expansion areas, and it is performing technical work on those. Another senior Canadian firm, Canadian Natural Resources Ltd., runs two additional projects in the Cold Lake region that date back to the 1980s. It acquired the Wolf Lake and Primrose properties in 1999 as part of the purchase of BP's oil assets in Alberta. The projects combine CSS and SAGD technologies, and are currently producing 45,000 barrels per day. Canadian Natural is expanding the projects to 60,000 barrels per day in 2006, and plans to reach production of more than 120,000 barrels per day by 2009. The company uses CSS in the Clearwater formation sands that have higher clay content, and the SAGD process in the overlying Grand Rapids zone, which is cleaner. In addition to its Cold Lake assets, Canadian Natural owns the Kirby property in the Athabasca region that is prospective for SAGD production from the McMurray. Likewise, it has noted SAGD potential on portions of its Horizon properties, where it is developing a world-scale open-pit mining operation. The Cold Lake area will also host two projects that will use only SAGD. Construction is under way at Husky's Tucker Lake venture, which the company says will be commissioned in third-quarter 2006. Tucker Lake will make about 30,000 barrels per day from the Clearwater formation; probable and possible reserves are more than 350 million barrels. And, BlackRock Ventures is contemplating a commercial SAGD development at its Orion project, adjacent to Imperial's Cold Lake operations. The Calgary company has been operating a pilot in the Clearwater formation for six years; this year it will decide whether to move forward with a 20,000-barrel-per-day commercial project. Peace River region During the late 1970s, Shell launched a cyclic steam project in the Peace River deposit, Alberta's third major oil-sands region. During the past 25 years, the company has produced 50 million barrels of bitumen from Lower Cretaceous Bluesky-Gething sandstones that occur at about 550 meters. Shell picked the Peace River area for its pilot in its pressure-cycle steam drive technique because it wanted reservoirs with several meters of bottom water. It injected steam into the water interval and the steam rose and softened the bitumen, which then flowed down into the water zone. The approach was successful in the pilot, but when Shell expanded it in 1985 to more than 200 wells, results were not so good. The steam tended to leak into the producing wells, and the SOR jumped to unacceptable levels. The company also experimented with SAGD at Peace River, drilling two pilots in the 1990s. Again, it experienced problems: in its first pilot, the bottom water zone was so thick that it cooled the injected steam too much. In its second test, permeability variations in the reservoir didn't allow the steam to rise uniformly. Shell then went back to a variation of its cyclic steam process. The soak radial technique used a vertical well with several horizontal laterals. The company experimented with haybob, tuning fork and multilateral J layouts for the horizontal legs before settling on a simpler design that it uses today. Now it drills horizontal J wells that are shaped like their eponymous letter. These deliver more heat to the toe regions of the laterals, and allow the bitumen to settle into the heels of the wells. The method is still a high-pressure cyclic process, but it incorporates a gravity component. This year, Shell has brought Peace River production close to the license capacity of 12,000 barrels per day, and it is evaluating its Carmon Creek expansion project, which could grow production to about 30,000 per day. If it decides to move forward, construction on the expansion could start in 2007. Clearly, the industry faces a considerable challenge in moving from the present production levels in thermal in-situ projects to the lofty expectations for the future. Companies will have to bridge this gap with much hard work, considerable capital and relentless technical innovation.

Comments