A project funded by the National Energy Technology Laboratory (NETL) to increase heavy oil reserves in slope and basin clastic (SBC) reservoirs through the application of advanced reservoir characterization and thermal production technologies is expected to eventually add 13 million bbl of incremental oil production in a small portion of the Wilmington oil field in Long Beach, Calif.

The project involves improving thermal recovery techniques in the Tar Zone of Fault Blocks

Figure 1. Heavy oil produced from Wilmington oil field, Long Beach, Calif.
II-A and V of Wilmington field. A primary objective is to transfer technology applicable to other heavy oil formations of Wilmington field and other SBC reservoirs, including those under waterflood. Participants in this project are the City of Long Beach, Tidelands Oil Production Co., the University of Southern California and Stanford University.

The Wilmington field, which runs roughly southeast to northwest through the Los Angeles Basin, is the third largest oil field in the contiguous United States. If new technologies and techniques developed under the project are applied field-wide, it could boost Wilmington’s ultimate oil recovery by 525 million bbl. That jump, in a single oil field, is a 2.5% increase in total US proved oil reserves. An aggressive effort to transfer this technology could boost reserves in similar fields along the California coast by 1.4 billion bbl.

Tidelands Oil Production Co. operates the western portion of the field as a subcontractor to the field owner, the City of Long Beach. Since 1932, more than 3,400 land wells have been drilled in the western portion of Wilmington oil field. By the 1950s, that portion had been completely developed under primary recovery, and waterflooding was started to increase recovery and control subsidence.

Project phases
The first phase addressed several producibility problems in the Tar IIA and Tar V thermal recovery operations that are common in SBC reservoirs. A few of the advanced technologies developed include a 3-D deterministic geologic model, a 3-D deterministic thermal reservoir simulation model to aid in reservoir management and subsequent post-steamflood development work and a detailed study on the geochemical interactions between the steam and the formation rocks and fluids. State-of-the-art operational work included drilling and performing a pilot steam injection and production project via four new horizontal wells (two producers and two injectors); implementing a hot water-alternating-steam (WAS) drive pilot in the existing steamflood area to improve thermal efficiency; installing a 2,400-ft (732-m) insulated, subsurface harbor channel crossing to supply steam to an island location; testing a novel alkaline steam completion technique to control well sanding problems; and starting work on an advanced reservoir management system through computer-aided access to production and geologic data to integrate reservoir characterization, engineering, monitoring and evaluation.

The second phase implemented state-of-the-art operational work to optimize thermal recovery processes, improve well drilling and completion practices, evaluate the geomechanical characteristics of the producing formations and update the 3-D geologic and reservoir simulation models. The objectives were to further improve the characterization of the heterogeneous turbidite sands, identify the high-permeability thief zones to reduce water breakthrough and cycling and analyze the non-uniform distribution of the remaining oil-in-place. This work has resulted in the redevelopment of the Tar II-A and Tar V post-steamflood projects by drilling a few new wells and converting idle wells to more effectively drain the remaining oil reserves by improving sweep efficiency and reducing water cuts while minimizing further thermal-related formation compaction. With no steam currently available to inject, efforts are being made to test cold heavy oil production techniques. The project will use all the tools and knowledge gained throughout the Department of Energy (DOE) project to maximize recovery of the oil-in-place.

Project summary
The 3-D reservoir simulation model was used to drill horizontal producing well UP-957 in March 2004 to the best remaining oil-saturated sands in the D1 sands. The well reached peak oil production in April 2004 at 259 b/d of oil, over 100% better than projected.

The Tar II-A post-steamflood project accommodated the Port of Long Beach (POLB) container terminal expansion by plugging and abandoning four of the best producing wells totaling 345 b/d of oil from January to March 2005. The POLB paid for three replacement producing wells, which were drilled and activated from October 2004 to February 2005. The three replacement wells were initially proposed as the first three DOE BP2 wells to be drilled. Well UP-959 was drilled as an updip directional delineation well that showed significant oil depletion, ranging from 20% to 75% recovery of the original-oil-in-place. Two new horizontal wells were completed in the updip oil-depleted steam chest areas to accelerate the upward migration of hot oil. Well UP-958 was completed in the T2 sands and reached a peak rate of 226 b/d of oil/1,749 b/d gross (87% water cut) in November 2005. Well W-900 was completed in the D1, D1b and D1d sands and averaged 155 b/d of oil at 93% water cut in November 2005.

Well UP-960 was drilled as a vertical infill pattern well to delineate the remaining oil saturations in the steamflood area downdip of UP-959. The well showed pre-waterflood oil saturations in the T sand and steam depleted D sands. Well UP-961 was drilled for the DOE project in place of well W-900 as a horizontal D1 sand well, only instead of drilling into the updip depleted steam chest, the well was completed in the downdip cold tar sands in a highly oil saturated area where vertical wells produced at 98+% water cuts.

The 3-D reservoir simulation model showed that continued operations through the year 2013 would not recover oil from the highly oil saturated D1 sands in the cold, structurally downdip areas south of the steamflood patterns. Recent drilling of replacement Tar II-A downdip water injectors and other new, deeper-zone wells in this area confirmed the very high oil saturations at the top of the D1 sands. When Union Pacific Resources developed its Tar II-A waterflood in 1960, the firm avoided this area because initial vertical waterflood wells had 98 to 99% water cuts; instead, they concentrated their efforts on the up-structure sands. New horizontal well UP-961 was drilled in November 2005 structurally downdip along the top of the cold D1 sands — again a very counterintuitive decision based on past well performance.

This well peaked at 185 b/d of oil and 635 barrels of gross fluid per day (BGFPD) in November 2005. A vertical steamflood pattern infill well, UP-960, was drilled in November 2005 and found the T sands oil saturated at pre-waterflood levels, even though waterflood and steamflood injectors surround the well, whereas the D1 sands appear oil depleted. The T sands do not appear resaturated because the temperature survey showed the sands as relatively cold, as if they never were steamed. These patterns were steamflooded for 6 to 7 years and then hot waterflooded; therefore, it is difficult to understand why the sands are not hot from conductive heat transfer alone. The resistivity log from an offset 1953 well that was completed deeper in the Lower Terminal zone shows similar T sand resistivities to UP-960. The D1 and D3 sands were very hot and exhibited the same characteristics as in well UP-959, only the resistivities were even lower. Based on the oil saturation versus resistivity chart developed for UP-959 and assuming reservoir temperature of 300°F (148.75°C) and 20,000 ppm salinity, the D1 sands (0.7 - 2.0 ohms) had less than 15% So at the top and 40% So at the middle to bottom of the sands. The well was cased to total depth, selectively perforated at the tops of the various T sands and the top of the D1 sands, assuming that the remaining D1 oil will migrate and resaturate the top of the sands, and an inner wire-wrapped screen was gravel-packed inside the casing for sand control. The well reached a peak rate of 70 b/d of oil and 1,328 BGFPD in March 2006.

Seven idle wells were activated and converted for use as Tar IIA producers (wells AT-43, AT-42, AT-63, UP-927) or water injectors (2AT-21, 2AT-22, 2AT-23) during 2005. All of the wells except UP-927 are completed in a single sub-zone, T or D1 sands, to provide better reservoir management control in an effort to increase oil production rates and reduce water cuts. The four producers peaked at 221 b/d of oil and 6,092 BGFPD (96% water cut). Two of the three injection wells were successfully placed on water injection while the third well had casing damage.

Tar II-A oil production from October 2005 to March 2006 averaged 1,228 b/d of oil at a 3.31% oil cut (29.1 WOR), substantially better than in November 2003, when it averaged 902 b/d of oil at a 3.3% oil cut (28.9 WOR). The Tar V post-steamflood pilot project experienced an increase in oil production from two new horizontal wells, wells A-603 and A-115, at the top of the S4 sands outside the steamflood pattern in cold oil. The two wells peaked at 633 b/d of oil and 1,746 BGFPD and were producing 351 b/d of oil and 3,187 BGFPD in March 2006. The offset operator, Thums Long Beach Co., intends to drill similar Tar S sand horizontal wells in fault block V later.

New tools
New tools developed for the project include:
• An advanced computer model to simulate the Wilmington reservoir, which it used to optimize steam, hot water, and water injection without causing surface subsidence, a perennial problem in the field;
• New horizontal well-based steamflooding designed with new three-dimensional computer models;
• A novel alkaline-steam well completion technique that controls excessive production of sand in the well bore, cutting capital costs by 25%;
• A new, commercial technology to scrub out H2S created in the steamflood at a 50% cost reduction; and
• A new steam generator that can burn a variety of low-quality waste gases created by the thermal enhanced oil recovery operations.

NETL project manager Jim Barnes noted that two companies are now marketing DOE-supported technologies as a result of the project: Dynamic Graphics, Inc. (DGI), Alameda, Calif., and Geomechanics International Inc. (GMI), Houston.

“Tidelands teamed with Stanford and the University of Southern California during many of their investigative efforts,” Barnes added. “GMI was started by Stanford researchers, who developed novel well logs calibrated to accurately measure porosity and oil saturation through sound-wave technology.”

Comments