Oil Sand Reservoirs Research Articles

Diagenesis of tight oil sand reservoirs: Upper Triassic tight sandstones of Yanchang Formation in Ordos Basin, China

Upper Triassic sandstones in the Ordos Basin, northern‐central China, comprise tight oil reservoirs. Using a combination of thin sections, SEM, BSE, EDS, XRD, and fluid inclusion analyses, 24 core samples from 13 wells were collected to study the petrology, paragenesis, and diagenetic processes and implications for reservoir quality. Quartz cement usually occurs as overgrowths or euhedral quartz. Extensive dissolution and albitization of K‐feldspar can be observed. Five types of carbonate cements, ferrocalcite, ankerite, dolomite, calcite, and siderite, occur during different diagenetic stages. Two main types of illite and 5 main habits of chlorite are observed in this study. Kaolinite mainly occurs as booklets and vermicular aggregates. Diagenetic illite, chlorite, biotite, mixed‐layer illite/smectite (I/S), and other minor minerals are also observed. The diagenetic processes include compaction, alteration of volcanic materials and mica, clay mineral transformation, cementation (silica, aluminosilicate, and carbonate), and dissolution of feldspars and rock fragments. Compaction was a significant porosity‐reducing agent, and the presence of carbonate cement exerts a dominant impact on the reduction of porosity. Quartz cement and authigenic clays are less important; however, it is worth mentioning that pore‐lining clays are conducive to porosity preservation. In this study, most of the porosity variation is caused by a combination of compaction, carbonate cements, quartz cement, and authigenic clays. This study gives insights into diagenetic alterations within tight sandstones and has implications for reservoir quality prediction in similar settings.

Clay minerals in an oil-sand deposit and their effect on oil-bearing properties

ABSTRACTThe current research on the characteristics of clay minerals in oil-sand reservoirs of China is still in its infancy. The burial depth of oil sands in the western slope of the Songliao basin is relatively shallow, and the reservoir contains clay minerals. Scanning electron microscope observations and X-ray diffraction analysis were used to test the chemical composition and characteristics of clay minerals. Comprehensive analysis and further research found that the clay mineral content of the oil-sand reservoir was relatively low and the clay minerals belonged to the illite–montmorillonite mixed-layer, composed of montmorillonite and illite, with occasional kaolinite, but no chlorite. The oil-bearing property of the oil-sand reservoir was consistent with the type of clay mineral found. The reservoir containing a mixed layer of illite–montmorillonite has a relatively good oil-bearing potential, and the oil content ranged from 3.80% to 16.17%.

Insights on Heat Transfer at the Top of Steam Chambers in SAGD

Steam-assisted gravity drainage (SAGD) is the method of choice for producing oil from oil sands reservoirs. In this method, steam is injected into the formation and the oil, upon heating, is mobilized and driven under gravity to a production well. The accumulation of steam within the reservoir is referred to as the steam chamber. One of the critical issues confronting SAGD operators is the thermal efficiency, measured by the steam-to-oil ratio, of their operations since it directly ties to process costs. Using thermocouple profiles from observation wells on three SAGD fields in Alberta, we use error function fits to estimate the thermal conductivity of the shale above the oil formation (found to be from 0.33 to 3.81 W/mK), heat flux at the top of the steam chamber, vertical height of the steam/gas zone above the steam chamber, and accumulated gas volume present. A gas material balance is then derived to estimate the volume of gas that might be generated through in situ chemical processes. The results of the heat transfer analysis performed on the thermocouple data reveal that the gas co-injection during SAGD operations studied did not directly affect the heat transfer rate at the top of the steam chamber since the gas volume added was small. The results also show that a sufficiently large accumulation of gas at the top of the chamber lowers the heat flux at the edge of the chamber.

Screening of Major Chemical Reactions in In-Situ Combustion Process for Bitumen Production from Oil Sands Reservoirs

Screening of Major Chemical Reactions in In-Situ Combustion Process for Bitumen Production from Oil Sands Reservoirs

Optimisation of well constraints based on wellpad system to accomplish a successive thermal process in a heterogeneous oil-sands reservoir

Optimisation of well constraints based on wellpad system to accomplish a successive thermal process in a heterogeneous oil-sands reservoir

Optimisation of well constraints based on wellpad system to accomplish a successive thermal process in a heterogeneous oil-sands reservoir

This paper presents the optimum operations of a wellpad system in a heterogeneous oil-sands reservoir where the flux interference occurs between parallel adjacent steam chambers. The objective values are the minimum steam to oil ratio for energy efficiency and the maximum recovery for profitability. The constraints influencing both objective values are the bottom hole pressure at the injectors, the surface liquid rate, and the steam rate at the producers. The temperature distribution shows that the unbalanced development of steam chambers plays a man-made thief-zone. The energy efficiency can be achieved when the operational conditions make a small amount of steam to release into the reservoir, reduce the interference, and increase steam retention within the reservoir. For maximum recovery, the highest steam injection is necessary to generate more vertical growth notwithstanding that some steam losses occur and the well-developed chamber assists other chambers by providing the steam laterally. [Received: December, 7, 2015 ; Accepted: March 17, 2016]

Performance of multiple thermal fluids assisted gravity drainage process in post SAGD reservoirs

Multiple thermal fluids injection process is a new heat-carrier process proposed in recent years as an EOR process for heavy oil reservoirs. Compared with the conventional saturated-steam injection process, it combines the multiple advantages of immiscible and/or miscible gas injection and thermal recovery processes. In this paper, based on the ideas of the multiple thermal fluids injection process and the conventional steam assisted gravity drainage (SAGD) process, a new recovery process, multiple thermal fluids assisted gravity drainage (MFAGD) process is proposed to enhance heavy oil recovery for post SAGD reservoirs. Two 3D gravity drainage experiments (SAGD and SAGD-MFAGD) are first conducted to explore EOR mechanisms of the multiple thermal fluids in heavy oil reservoirs. Subsequently, numerical simulations are performed to match the experimental measurements. Then the EOR mechanisms and the remaining oil saturation distribution are analyzed. Furthermore, the differences between the SAGD, steam-and-gas-push (SAGP) and MFAGD processes are discussed. From a dimensionless scaling analysis for gravity drainage process, lab scale parameters are converted to field scale ones, and a field scale simulation model is developed. Through the analysis and simulation, the reservoir adaptability of the MFAGD process is investigated, and its operation parameters are numerically optimized. Experimental results indicate that a strategic combination of the SAGD and MFAGD processes can tremendously improve the development of heavy oil reservoirs. The MFAGD process can be adopted as a follow-up recovery technique after the SAGD process for heavy oil reservoirs. As its EOR mechanisms, in addition to the conventional thermal recovery mechanisms of steam injection, the results show that heat insulation, energy recovery, gas dissolution, foamy oil and auxiliary well cleanup are all its important mechanisms. In comparison with SAGD, this new process has a higher recovery rate, a lower steam injecting volume and a lower cumulative steam-oil ratio (cSOR). Numerical results also show that the steam chamber from MFAGD is shaped like a liquid drop instead of a conventional inverted triangle from SAGD. The net-to-gross value and formation thickness have a great influence on the performance of MFAGD. It is found that for MFAGD, the optimal steam-gas ratio in the standard conditions is 1:1, the steam injection rate is 200m3/d, and the operation pressure is 3.0MPa for the study reservoir. The MFAGD process will be a potentially important EOR method for post SAGD heavy oil and oil sands reservoirs.

Research on Influence Factors and Screening Method of Tight Sandstone Oil Reservoir CO 2 Huff and Puff Based on Multi-index

CO2 huff and puff has been continuously considered as an import EOR method for tight oil reservoirs. However, the large number of evaluation parameters for reservoir screening has limited its application. This paper proposed a multi-index evaluation function for CO2 huff and puff method to facilitate the reservoir screening. By the means of orthogonal experimental design, it concludes the main controlling factors for screening a tight sand oil reservoir with multi-stage fractured horizontal wells. In addition, we built a calculation model for this multi-index evaluation function with Box-Benhken experimental design method. And this model has been successfully applied to screen wells for CO2 huff and puff of Y oilfield in Ordos.

A criterion for evaluating the efficiency of water injection in oil sand reservoirs

Steam Assisted Gravity Drainage (SAGD) has been implemented at a large scale in the development of ultra-heavy oil reservoirs in Karamay city, the Xinjiang province of China. Before circulating steam, byproduct water from previous SAGD projects is injected into a SAGD wellpair in order to create a dilative and highly permeable zone in the inter-well domain. A criterion is desired to evaluate how efficient this injection/stimulation has been performed. The criterion was based on pore pressure distribution in the domain encompassing the dual wells, which was the outcome from finite element analysis using the Drucker-Prager model to describe oil sand. The analysis used constitutive parameters calibrated by laboratory results and the real-time injection data input as loading conditions. Based on the simulated pressure profile, communication parameter (CP) was proposed to represent the efficiency of water injection, with a magnitude varying from 0 to 1. It was found that CP gradually and continuously increased with the injection process, demonstrating that the efficiency increases with time and pressure. Moreover, by monitoring the time point at which the injection rate dramatically increased in the field, a group of threshold CPs were established based on the previously completed stimulation projects on a variety of oil sand reservoirs. The criterion was then fulfilled by comparing the calculated CP to the threshold CP of a specific reservoir type. By applying the criterion in the stimulation projects of two wellpairs in the field, it was observed that immediately after the CP exceeds the threshold value of the corresponding reservoir type, the pressure of the dual wells was quickly responsive to each other and the injection rates sharply increased. The preheating periods of the two wellpairs decreased from conventionally 180–300 days to 60 and 90 days, implying the success of reservoir stimulation. The proposed criterion can not only be used to quantitatively describe how efficient a water injection process has achieved, but also to determine whether the stimulation of an oil sand reservoir has been successfully accomplished. When the criterion is satisfied, the field engineers can confidently move on to the steam circulation stage.

On the physics of cyclic steam stimulation

Steam fracturing as done during cyclic steam stimulation is an effective thermal process for initiating recovery from viscous oil reservoirs such as oil sands reservoirs found in Alberta, Canada and Liaohe, China. A key component of these processes is the ability to inject high temperature steam into the formation to fracture it which in turn raises its permeability and mobilizes the oil by lowering its viscosity. The dynamics of steam fracturing are not fully resolved especially how steam fingers into the reservoir and how its state changes as heat losses occur from the injected steam. The results of this study reveal that steam condensate, pressurized by the steam vapour upstream, fractures the formation. The results also show that dilation of the reservoir during steam injection relieves the pressure which in turn lowers the steam injection pressure below that of the case where no dilation occurs.

Controls on reservoir heterogeneity of tight sand oil reservoirs in Upper Triassic Yanchang Formation in Longdong Area, southwest Ordos Basin, China: Implications for reservoir quality prediction and oil accumulation

Compared to conventional reservoirs, pore structure and diagenetic alterations of unconventional tight sand oil reservoirs are highly heterogeneous. The Upper Triassic Yanchang Formation is a major tight-oil-bearing formation in the Ordos Basin, providing an opportunity to study the factors that control reservoir heterogeneity and the heterogeneity of oil accumulation in tight oil sandstones.The Chang 8 tight oil sandstone in the study area is comprised of fine-to medium-grained, moderately to well-sorted lithic arkose and feldspathic litharenite. The reservoir quality is extremely heterogeneous due to large heterogeneities in the depositional facies, pore structures and diagenetic alterations. Small throat size is believed to be responsible for the ultra-low permeability in tight oil reservoirs. Most reservoirs with good reservoir quality, larger pore-throat size, lower pore-throat radius ratio and well pore connectivity were deposited in high-energy environments, such as distributary channels and mouth bars. For a given depositional facies, reservoir quality varies with the bedding structures. Massive- or parallel-bedded sandstones are more favorable for the development of porosity and permeability sweet zones for oil charging and accumulation than cross-bedded sandstones.Authigenic chlorite rim cementation and dissolution of unstable detrital grains are two major diagenetic processes that preserve porosity and permeability sweet zones in oil-bearing intervals. Nevertheless, chlorite rims cannot effectively preserve porosity-permeability when the chlorite content is greater than a threshold value of 7%, and compaction played a minor role in porosity destruction in the situation. Intensive cementation of pore-lining chlorites significantly reduces reservoir permeability by obstructing the pore-throats and reducing their connectivity. Stratigraphically, sandstones within 1 m from adjacent sandstone-mudstone contacts are usually tightly cemented (carbonate cement > 10%) with low porosity and permeability (lower than 10% and 0.1 mD, respectively). The carbonate cement most likely originates from external sources, probably derived from the surrounding mudstone. Most late carbonate cements filled the previously dissolved intra-feldspar pores and the residual intergranular pores, and finally formed the tight reservoirs.The petrophysical properties significantly control the fluid flow capability and the oil charging/accumulation capability of the Chang 8 tight sandstones. Oil layers usually have oil saturation greater than 40%. A pore-throat radius of less than 0.4 μm is not effective for producible oil to flow, and the cut off of porosity and permeability for the net pay are 7% and 0.1 mD, respectively.

Reservoir Simulation of Hydrogen Sulfide Production During a Steam-Assisted-Gravity-Drainage Process by Use of a New Sulfur-Based Compositional Kinetic Model

Summary Steam injection is commonly used as a thermal enhanced-oil-recovery (EOR) method because of its efficiency for recovering hydrocarbons, especially from heavy-oil and bitumen reservoirs. Reservoir models simulating this process describe the thermal effect of the steam injection, but generally neglect the chemical reactions induced by the steam injection and occurring in the reservoir. In particular, these reactions can lead to the generation and production of the highly toxic and corrosive acid gas hydrogen sulfide (H2S). The overall objective of this paper is to quantitatively describe the chemical aquathermolysis reactions that occur in oil-sands reservoirs undergoing steam injections and to provide oil companies with a numerical model for reservoir simulators to forecast the H2S-production risks. For that purpose, a new sulfur-based compositional kinetic model has been developed to reproduce the aquathermolysis reactions in the context of reservoir modeling. It is derived from results gathered on an Athabasca oil sand from previous laboratory aquathermolysis experiments. In particular, the proposed reactions model accounts for the formation of H2S issued from sulfur-rich heavy oils or bitumen, and predicts the modification of the resulting oil saturate, aromatic, resin, and asphaltene (SARA) composition vs. time. One strength of this model is that it is easily calibrated against laboratory-scale experiments conducted on an oil-sand sample. Another strength is that its calibration is performed while respecting the constraints imposed by the experimental data and the theoretical principles. In addition, in this study no calibration was needed at reservoir scale against field-production data. In the paper, the model is first validated with laboratory-scale simulations. The thermokinetic modeling is then coupled with a 2D reservoir simulation of a generic steam-assisted gravity drainage (SAGD) process applied on a generic Athabasca oil-sand reservoir. This formulation allows investigating the H2S generation at reservoir scale and quantifying its production. The H2S- to bitumen-production ratio against time computed by the reservoir simulation is found to be consistent with production data from SAGD operations in Athabasca, endorsing the proposed methodology.

Improving Steam-Assisted Gravity Drainage performance in oil sands with a top water zone using polymer injection and the fishbone well pattern

In western Canada, a significant number of oil sands reserves have little or no cap rock with a top water zone. Due to huge heat loss to the top water zone, the conventional Steam-Assisted Gravity Drainage (SAGD) process is uneconomical when applied directly in this type of reservoir.In this study, it is proposed that a high temperature polymer solution can be injected into the bottom of the top water zone to establish a stable high viscosity layer that will prevent steam from leaking into the top water zone. In order to select a suitable polymer that has stable viscosity under high temperature, the viscosities of different polymer solutions at different temperatures were measured and the concentration of the selected polymer solution was optimized. Furthermore, in order to extend the connection area between the oil sands and the steam chamber, the fishbone well pattern was applied instead of the single well pair pattern.Numerical simulations were performed to evaluate the feasibility of using the selected polymer in the fishbone well pattern to improve SAGD performance in oil sands with a top water zone. The numerical simulation model was based on a typical Athabasca oil sands reservoir. In this study, the effects of steam injection pressure, polymer solution injection time, steam injection rate, and different fishbone well patterns on the performance of the SAGD process were studied and optimized.The numerical simulation results suggest that the fishbone well pattern could extend the steam distribution and that polymer injection is able to prevent heat from leaking into the top water zone. Compared to the conventional SAGD process in an oil sands reservoir with a top water zone, the optimal case using a one-fishbone well pattern and polymer injection could enhance the oil production significantly. Under these conditions, the oil recovery factor in this study increased from 10.58% to 59.02%; the cumulative steam oil ratio decreased from 10.44m3/m3 to 3.85m3/m3; and the cumulative injected energy oil ratio decreased from 24.00GJ/m3 to 8.87GJ/m3. This indicates that the SAGD process with the one-fishbone well pattern and polymer injection is able to improve SAGD performance in oil sands with a top water zone.

Optimal Fracturing Parameters Condition of Integrated Model Development for Tight Oil Sands Reservoir with 2D Fracture Geometry Using Response Surface Methodology

In this study, oil production in tight oil reservoir was declined due to high heterogeneity, complicated and complexity of reservoir of the low permeability reservoir ranges from 0.1 md to 5 md and reservoir porosity ranges from 10 to 18% lead to low fracture conductivity among the fractures. The challenge deal with this problem is to stimulate the reservoir of hydraulic fracturing for maximum oil production is necessary for the study. The application of the central composite design and response surface are the best tool in order to optimize the operating parameters based on the maximum net present value and through the series calculation, the optimal operating parameters of hydraulic fracturing have been determined of 46 bpm of injection rate, 88.5 min of the injection time and 0.002 ft/min0.5 of the leak-off coefficient and the maximum net present value of 46.5 $mm. Finally, the integrated model development of hydraulic fracturing includes of the normal faulting stress regime, fracturing fluid selection and fluid model, proppant selection, fracture geometry, pressure model, material balance, fracture conductivity and simulation production of fractured well and unstimulated well that have been presented in this study. With the using two dimensional Perkins-Kern-Nordgren fracture geometry models coupled with carter leak-off as the 2D PKN-C has been used to account for leak-off coefficient, spurt loss in term of power law parameters to propagate the fracture half-length, fracture width. The result of the fracture conductivity of fractured well at the fracture half-length of 1,940 ft and average fracture width of 0.32 in by series calculated propped fracture concentration of 1.63 lb/ft2 and fracture closure pressure of 4,842.59 psi to fracture conductivity of 6,200 md-ft, which value is measured by the laboratory experiment. The post-fracture production has been shown the fold of increase oil of 18.7 and the oil production rate of fractured well demonstrated much rising compared to oil production rate of unstimulated case.

Assessment of reservoir heterogeneity by using produced water chemistry in SAGD

The key challenge of extracting oil from oil sand reservoirs is the viscosity of the oil which is typically between 100 000 and several million cP. To reduce the viscosity of the oil, high pressure, high temperature steam, typically between about 185 and 250 °C, is injected into the reservoir by using recovery processes such as the steam-assisted gravity drainage (SAGD) process. In this process, steam heats the bitumen and as a consequence, its viscosity drops to about 5 cP and it readily flows under gravity within the reservoir. One key issue that has not gained much attention with respect to SAGD process evolution are steam–rock reactions, water geochemistry, and how the produced water composition varies as the process evolves. Here, we examine how the produced water composition varies in SAGD operations. For the first time, we show that the produced water composition can be used to detect shale barriers and contact of the steam chamber with the overburden. As yet, the produced water composition is not used to understand in situ process development but as we show here, this could be a rich data source for understanding process dynamics. Copyright © 2016 John Wiley & Sons, Ltd.

Application of Grid-Free Geostatistical Simulation to a Large Oil-Sands Reservoir

Summary Geostatistical simulation is performed for reservoir characterization to depict local variability in the modeled properties. The conventional simulation methods are implemented in a grid-dependent manner that makes regridding of realizations, refinement of existing grids, and the simulation on irregular grids challenging. The grid-free-simulation (GFS) method has been recently developed for flexible reservoir characterization. The geostatistical realizations of a reservoir are expressed as an analytical function of the coordinates of the simulation locations and, thus, are infinitely resolvable. The resulting model is conditioned to primary scattered point-scale hard data and secondary exhaustively sampled block-scale soft data. The former data are sampled along wells, whereas the latter data are from seismic surveys. The GFS methodology is applied to the Firebag oil-sands thermal project operated in northern Alberta, Canada. The conditioning data are point-scale core measurements, log observations, and block-scale acoustic impedance (AI). The models of correlated porosity, permeability, and water saturation attributes are constructed on three different grids by facies and are consistent with each other. These models are intended for resource estimation, reserves estimation, and subsequent-flow simulation, respectively. The modeling results of the grid-independent simulation method are promising for industrial application to petroleum reservoir characterization.

Integrated Model Development for Tight Oil Sands Reservoir with 2D Fracture Geometry and Reviewed Sensitivity Analysis of Hydraulic Fracturing

In this study, oil production rate in tight oil reservoir is much declined due to low permeability reservoir of 0.5 mD, low porosity of 0.15%, high reservoir depth range from 9,962.5 ft to 10,037.5 ft, high closure pressure up to 4,842.5.9 psi among fractures that lead to poor fracture conductivity in the fractures of the reservoir. Usually, high closure pressure is easy to reduce conductivity very fast during oil production. Need to stimulate the reservoir in order to enhance oil production is important to successful project by the application of integrating model development for tight oil reservoir based on the two dimensional Perkins and Kern Nordgren Carter fracture geometry in term of the power law parameters of hydraulic fracturing by systematic of model asNormal faulting stress regime, fracturing fluid model, fracture geometry model, pressure model, material balance, proppant selection, fracture conductivity that have been presented in detail of the research. By series calculation and laboratory experimental for fracture conductivity under closure pressure and proppant fracture concentration of 1.5 lb/ft2 that fracture conductivity of 5,700 mD.ft of Carbo-Lite ceramics proppant size of 20/40, proppant density of 169 lb/ft3.In order to estimate the effects of the operating fracturing parameters condition of hydraulic fracturing as the injection rate, injection time, leak-off coefficient on the fracture half-length, fluid efficiency, the net fracture pressure, productivity, dimensionless fracture conductivity, the sensitivity analysis has been proposed for the analysis. The research is provided the new tool for hydraulic fracturing stimulation in order to advance in knowledge for engineer in the field. The post fracture production has been shown about 7.4 fold of oil production increment.

On the Use of Particle-Size-Distribution Data for Permeability Prediction

SummaryParticle-size distribution (PSD) is a list of values that defines the relative amount of particles present according to the size in a sample. The PSD of the McMurray Formation sediments characterizes rock granulometry and is a fundamental indicator of the nature of the sediment. The size distribution of the component solid particles in the McMurray Formation sediments relates to their porosity; volume of water and bitumen contained within the pore space; and the depositional environment, including lithological association, stratigraphy, areal distribution, and associated physical processes.PSD is known to be a significant factor for evaluating bitumen recovery from an oil-sand mine. This is because presence of fines (evaluated by PSD analysis) affects the hot-water-separation process and processing-plant recovery prediction and provides grade control. Presence of more fines translates into lower recovery from commercial oil-sand processing. In this study, we investigate whether the PSD should be also considered a critical parameter for evaluation and estimation of permeability of an oil-sand reservoir.We show, by use of the data from Cenovus Energy's Telephone Lake lease, that there is a strong relationship between permeability and PSD data. We also show that the information provided by the PSDs for permeability prediction is more significant than that inferred from a simple porosity/permeability relationship. Subsequently, we comment on permeability modeling by use of PSD data and list the techniques available for cleaning and modeling of multivariate PSDs.We document a methodology for modeling of PSDs and provide a work flow for incorporating these data in improved understanding and modeling of permeability and its distribution.

On fingering of steam chambers in steam‐assisted heavy oil recovery

Western Canadian oil sands reservoirs are among the largest petroleum accumulations in the world. Given original oil viscosity up to 5,000,000 mPa‐s, these oils are currently recovered from these reservoirs using steam which heats the oil to ∼250°C with reduced viscosities <10 mPa‐s. A key issue faced by thermal recovery processes is the uniformity of the steam chamber within the reservoir. Nonuniformities of the chamber arise from multiphase flow instabilities in the porous media where fingering has been explained by penetration of low viscosity steam into viscous oil. Here, fine‐grid thermal reservoir simulation reveals that fingering takes place in the gas phase beyond the chamber edge in a zone created by gas exsolution due to elevated temperature beyond the edge of the steam chamber. The results suggest that nonuniform chambers will occur in perfectly homogeneous reservoirs which implies that uniform chambers along wells may be impossible to achieve. © 2015 American Institute of Chemical Engineers AIChE J, 62: 1364–1381, 2016

Prediction of steam-assisted gravity drainage steam to oil ratio from reservoir characteristics

Forecasts suggest that production of bitumen from oil sands reservoirs will increase by a factor of at least 2.5 times over roughly the next 15 years. Although a significant economic benefactor to the Canadian economy, there are challenges faced by oil sands operators with respect to greenhouse gas emissions and water consumption. For the Athabasca deposit, the current oil recovery process of choice is the SAGD (steam-assisted gravity drainage) method where high temperature and high pressure steam is injected into the oil sands formation. At present, there are more than ten SAGD operators in Alberta, Canada and results to date reveal that the geology of the reservoir impacts SAGD performance. Given the growth of the SAGD industry in Alberta, forecasting tools are required that can predict performance versus reservoir characteristics. Here, we present a neural network-based model to predict the SOR (steam-to-oil ratio) in oil sands reservoirs by using log and core data to characterize the reservoir porosity, permeability, oil saturation, depth and thickness. Our analysis confirms that the lower the porosity, permeability, and oil saturation of the reservoir, the worse the performance of the operation. In other words, the lower the quality of the reservoir, the lower the oil rate, and the higher the SOR. Our analysis also shows that well performance (i.e., SOR), is predictable with a relatively high degree of accuracy (R2∼0.80) using log and core data via a neural network model. These results imply that the depth of the reservoir, gamma ray readings, and permeability are the most important determinants of the variation in SOR.