When compared with steam-assisted gravity drainage (SAGD) operations in the McMurray Formation, Athabasca Oil Sands, SAGD projects in the Clearwater Formation at Cold Lake did not perform as expected, likely because of reservoir properties. This paper will use the Orion SAGD case study to: (1) investigate the impacts of reservoir properties on the SAGD thermal efficiency by field evidences; (2) identify key geological parameters influencing each well pad; and (3) summarize major geological challenges for Orion SAGD expansion.
Wireline log data were interpreted to characterize reservoir properties, which were used to build 3D models. 3D visualizations and 2D cross sections of the reservoir revealed spatial distribution and heterogeneity of each property. SAGD production performance was analyzed using: (1) temperature profiles that monitored the growth of the steam chamber; (2) cumulative steam-oil ratios (CSORs); and (3) oil production rates (OPRates), which are direct indicators of thermal efficiency.
Results show that impermeable barriers and low-permeability zones were detrimental to steam injectivity and steam chamber growth, as observation wells in Pilot Pads 1 and 3 did not detect any steam saturation. High-permeability zones favored high steam injectivity and mobility, especially in Pad 105. Steam chambers were irregularly shaped by high shale-content zones, as two sharp spikes displayed on the temperature profile in Pad 103. Low oil-saturation zones and thin net-pays increased the CSORs, as seen in Pads 106 and 104. Impermeable barriers are almost horizontal, making no difference on well pad orientation by their dip angles. Lack of porosity variation made it difficult to identify the impact of porosity on each well pad.
The relatively extensive distribution of impermeable barriers between and above well pairs, as well as the relatively large area of low oil saturation and thin net-pay, were identified as major geological challenges.
Heterogeneity in the Athabasca oil sands can impede the growth of SAGD steam chambers. Here, we show how controlled-source electromagnetic (EM) methods can be used to detect growth-impeded regions and monitor changes in steam chamber growth. Our achievements are two-fold. We first generate a background resistivity model based on well logging at a field site in the Athabasca oil sands and then estimate the resistivity of the steam chambers using an empirical formulation that incorporates the effects of temperature on the surrounding rocks. Using the resulting 3D model, electromagnetic responses for any EM survey can be computed. The second, and more important, achievement illustrates that imaging SAGD chambers, as they grow in time, may be possible with cost-effective surveys. Our example uses a single transmitter loop with receivers in observation wells. In the wells, only the vertical component of the electric field is measured. Even with this limited data set, the images obtained through 3D cascaded time-lapse inversion identifies the location and extent of an impeded steam chamber. The proposed EM survey acquisition time and processing should be relatively fast and cost effective, and are expected to yield sufficient information to help make informed decisions regarding SAGD operations.
Steam Assisted Gravity Drainage (SAGD) is an in-situ recovery process used to extract bitumen from the Athabasca oil sands in northeast Alberta. In SAGD, two horizontal wells are drilled at the bottom of the reservoir (Dembicki, 2001). Steam is injected into the top well and produces a steam chamber that grows upwards and outwards. At the edge of the chamber, the heated, fluid oil and condensed water flow through the formation and are collected by the underlying horizontal production well. The chamber expands further into the bitumen reservoir as the oil drains (Butler, 1994).
The success of this technique is dependent upon steam propagation throughout the bitumen reservoir. However, reservoir heterogeneity, such as clay beds and mudstone laminations, can cause low-permeability zones that can impact the growth of the steam chambers (Strobl et al., 2013; Zhang et al., 2007). This affects the amount of produced oil and exemplifies the importance of monitoring the steam chamber growth. Successful monitoring can aid in optimizing production efforts by increasing understanding of the reservoir, decreasing the steam-to-oil ratio, locating missed pay, identifying thief zones, and more efficiently using resources (Singhai and Card, 1988).
Because the electrical conductivity of a lithologic unit is affected by steaming, electric and electromagnetic methods are promising tools to detect and image SAGD steam chambers. Additionally, these types of surveys can be much more cost-effective than seismic methods (Engelmark, 2007; Unsworth, 2005). Electric and electromagnetic surveys can also be readily installed as permanent installations. Tøndel et al. (2014) used a permanent electrical resistivity tomography (ERT) installation in the Athabasca oil sands to monitor SAGD steam chamber growth over time. From their study, electrodes can stand up to the high-temperature environment in boreholes surrounding the steam chambers while geophones can break down over time. Devriese and Oldenburg (2015) showed how the method can be extended to frequency- and time-domain EM. Permanent installations can also provide multiple data sets per year, without being limited by access to the area in wintertime only.
Connacher's first oil sands project, the Pod One facility at Great Divide, has been operational since 2007. The successful SAGD project has produced approximately 7 million barrels of bitumen. During the past three and a half years, the impacts of certain predicted reservoir challenges and opportunities have become apparent.
While the quality of the oil sands in this first phase of Pod One is generally good, Pad 101 South in particular has geological zones that affect SAGD operation. This includes a bitumen lean zone, and a gas cap overlying the main bitumen channel/s. Early field results matched with detailed simulations have shown positive results in maximizing well pair production. For the purposes of this paper a lean bitumen zone differs from an aquifer in two ways. The lean zone is not charged, and is limited in size. The operation is also complicated by the fact the gas bearing zone has been depleted through earlier production.
Connacher's operating practice at Great Divide attempts to achieve a pressure balance between the 3 zones (rich oil sands, lean zone, gas cap) to reduce steam loss and maximize production rates. Reducing the pressure encourages steam chamber development growth horizontally and ensures that steam contacts the highly saturated bitumen areas. How this is achieved with the highest positive impact on well productivity is illustrated with operational data and analysis including the results of simulations that recommended the optimum operating strategies.
Heavy oil production requires large quantities of freshwater for steamgeneration. Currently these operations utilize surface water and shallowgroundwater for supply. An alternative source of water is the brackishgroundwater from basal aquifers which exist in the lower portion of theMannville Group and below the heavy oil horizons. Hence, the term "basalaquifers" is used to describe these water bearing sediments.
Due to corrosion and precipitation concerns, source water must haveacceptable salinity and Total Dissolved Solid (TDS) levels. Salinity and TDS ofgroundwater can be conveniently estimated from electrical logs using empiricalformulae.
Oil sands are found in three places in Alberta: Athabasca, Cold Lake andPeace River (Figure 1). The Cold Lake region has the largest in-situ bitumenrecovery project, where oil sands are heated by steam injection to extractbitumen to the surface. The area of this study is located between Cold Lake andWolfe Lake, about 30 km northwest of the town of Cold Lake, Alberta (Inset inFigure 1). Currently there are five heavy oil plants in this area.
Cyclic steam stimulation (CSS) and steam-assisted gravity drainage (SAGD)are in-situ recovery methods for bitumen which is buried 400 meters belowsurface. Both methods require steam. The current water supply for steamgeneration in the Cold Lake region mainly comes from surface lakes and aPleistocene shallow groundwater aquifer lying in buried valleys. Because of theincreasing demand of water for steam generation and the limitations associatedwith the current water sources, brackish groundwater from the basal aquifersbecomes an attractive and economical alternative supply for heavy oilproduction.
A potential brackish water supply has to have acceptable salinity and TDSlevels to reduce the risk of corrosion and precipitation, and has to possess abalanced flow equilibrium to sustain long term yield. Studies conducted byTerracon in the Cold Lake area indicate that the Cretaceous Mannville Groupcontains three distinct potential aquifer groups capable of sustaining longterm yields. However, the Salinity and TDS of these aquifer groups varylaterally and vertically. Electrical logs for conveniently estimatinggroundwater salinity and TDS were utilized to supplement limited existing waterquality data.
TDS is an appropriate indicator of groundwater freshness. Chlorideconcentration (Cl ) is a good indicator of both groundwater Salinity and TDS.Because there are recognized petrophysical methods for determining bothChloride concentrations and TDS, and there are hundreds of electrical logsavailable in the study area, it is possible to map the lateral and verticaldistribution of Chloride concentration, and to select appropriate withdrawallocations for brackish water supply.
The Chloride concentration and TDS of formation brackish water werecalculated from formation water resistivity (Rwe, Rwz) using an empiricalformula. The Formation water resistivity was calculated from spontaneouspotential readings (SP) and formation temperature (Tf).
Ground deformation was monitored for nearly ten weeks during the first cycle of steam stimulation in a single-well test using an array of high-resolution borehole tiltmeters. The test was conducted in a section of the Athabasca oil sands having properties similar to the unconsolidated oil sands of California. The properties similar to the unconsolidated oil sands of California. The 310 meter injection depth was also comparable to the depth of thermal stimulation in many California oil fields. Ground response indicated that steam injection was not a continuous process, but rather was characterized by numerous episodic events. During these events wellhead pressure dropped (in one case by 2650 kPa), boiler feed rate increased by a few percent, and the ground surface within the instrument array was lifted percent, and the ground surface within the instrument array was lifted up. Pressures again began to rise and the ground surface subsided within a few hours of the beginning of an event, but subsidence always preceded pressure increase. The magnitudes of the pressure and deformation changes pressure increase. The magnitudes of the pressure and deformation changes varied from event to event, apparently unsystematically.
The events are interpreted to have resulted from breakdown of the oil sands and attendant propagation of hydraulic fractures away from the wellbore in approximately horizontal planes. Larger fractures may have continued to propagate until internal pressures were insufficient to lift the overburden, at which time they collapsed. Fracture growth terminated at higher pressures in events for which deformation changes were small, perhaps because of inelastic blunting of the fracture tips. Modelling suggests that the radii of fractures formed in the larger events may have been about 160 meters, whereas those formed in the smallest events had radii of about 40 meters.
A delay of three weeks between the start of steam injection and the occurrence of the first episodic event suggests that there may have been major modification of the in-situ stress state during this period. Pressure records from cold-water hydraulic fracturing a week before the Pressure records from cold-water hydraulic fracturing a week before the start of steam injection indicate that this fracture was vertical, from which we infer that the most compressive component of in-situ stress was also vertical. Gradual heating of the oil sands during steam injection should have closed the vertical fracture by thermal expansion, and then led to an increase of horizontal compression as further lateral expansion was suppressed. Formation of horizontal fractures after three weeks of steaming is consistent with a modified in-situ stress state in which horizontal exceeded vertical compression.
During the months of July, August and September 1979 Gulf Canada Resources Inc. conducted the initial cycle of steam stimulation and production in a single well approximately 50 km south-southeast of Fort McMurray, in the Athabasca oil sands region. The well is located in land survey 11, section 20, township 83, range 6, west 4th meridian. This project comprised the initial stage of Gulf's first single-well test on the lease, and was followed by additional cycles of steam injection and production. Part of the study of the initial steam stimulation cycle consisted of precision monitoring of ground deformation around the well, produced by reservoir response to steam injection. Deformation monitoring was conducted to obtain direct measurements of the physical responses produced by steam stimulation in order to better characterize reservoir processes and to obtain data for comparison with theoretical models of reservoir behavior. One important objective was to obtain an estimate of the dimensions of the thermally stimulated volume of oil sands. This paper describes the results of the monitoring program and offers interpretations of the behavior that was observed.
GEOLOGY OF THE TEST SITE
The oil sands that were the target of the single-well test are in the Cretaceous McMurray Formation. Rich oil sands occur between the depths of 308m and 317m in the well.