Kuwait has a considerable reserve of untapped heavy oil; plans were developed by KOC and embarked on a project to increase its oil production by 2020. Heavy oil production is an ambitious project and will be significant contributer to the overall increase in oil procuction capacity.
As the term "heavy oil?? suggests, it is a very viscous oil. The most common methods of heavy oil recovery are:
- Cold Heavy Oil Production with Sand (CHOPS)
- Cyclic Steam Stimulation
- Steam Flooding
- Steam Assisted Gravity Drainage (SAGD)
With the focus on the second method, Cyclic Steam Stimulation to enhance the recovery of heavy oil, the design of the cement became important in terms of endurance over the life of the well. In this technique the casing / cement would be exposed to steam injection temperatures as high as 500°F.
In such cases, the cement sheath may crack due the extreme forces acting on the thin sheath of the cement. It is therefore important to know the Young's modulus of both the formation and the cement. This will allow the slurry properties to be adjusted by the use of additives to lower the Young's modulus of the cement to less than of that of the formation. This will prevent damage to the cement sheath.
A fit-for-purpose cement slurry was designed accordingly and applied on a South Ratqa well. Well testing during and after 45 days of steam injection demonstrated that the cement maintained its integrity despite the challenging conditions.
Cement Slurry Design
A project was initiated to investigate the mechanical integrity of various cement slurries subjected to 500°F steam-injection cycles. The overall aim was to achieve a flexible cement design that would withstand the induced stress applied in this particular situation. (Figure 1).
Live samples (cement, location water and additives) were air-freighted to the USA (Baker Hughes Pressure Pumping Technology Center in Texas) to avoid any design flaw factors, and maintain reproducible slurries upon actual job execution.
The cement slurry testing was done as per the schedule shown in Figure 1. This schedule was applied to most testing, including the determination of the following parameters:
- Destructive compressive strength
- Ultrasonic compressive strength
- Unconfined tensile strength
- Confined tensile strength
- Ultrasonic Young's modulus
- Ultrasonic Poisson's ratio
- Confined Young's modulus
- Confined Poisson's ratio
The Marrat reservoir in Dharif field is a deep, sour, high pressure oil accumulation of Jurassic age containing light under-saturated oil of 36-380 API. The carbonate reservoir has a porosity range of 10-20% with permeability of 1-10 md. The field was put on production in 1989 through one well. Subsequently, 10 wells were added gradually developing the field. As of date, the field has produced about 12.5% of oil in place, lowering the reservoir pressure from 10,525 to 7,000 psi.
At present, oil production from the field is about 13,500 bbls/day. Due to low permeability, some wells produce with high drawdown approaching asphaltene onset pressure (AOP), estimated at 3,400 psi. This causes Asphaltene deposition in the tubing that requires cleaning to maintain the production level. The major challenges now are to produce the wells above AOP to avoid asphaltene precipitation in the wells or reservoir while sustaining the production level and maximizing recovery.
Hence, Full Field Model (FFM) for simulation studies was constructed and history-matched. Under depletion case, where the wells produce above AOP, field produced about 24% STOIIP. The water injection case shows significant increase in recovery to 40% STOIIP. Since no prior experience of water injection is available for such tight deep carbonate reservoirs in West Kuwait Fields, several key studies such as a) RCAL & SCAL b) Core flood Study c) Water Compatibility & Scale Prediction modeling d) Injectivity test, were carried out to address water injection feasibility.
The present paper shares the results of above studies which indicate that water injection is a viable option to maintain the reservoir pressure to produce the wells above AOP as well as to maximize recovery. Pilot water injection is planned through one well for which the area has been optimized using FFM. At present Pilot Water injector and source wells have been drilled and injection will be initiated with commissioning of surface facilities
Dharif field is NNE trending elongated anticlinal structure with faulted western limb. The Marrat reservoir in this field has developed in carbonate aggradational and progradational depositional setting. The field was discovered in 1988, put on production in 1989 and gradually developed with additional producers until 2004 (Fig-1). As of today, total 13 deep wells have been drilled in this field of which eleven are completed in the Marrat reservoir, while two are completed in a shallower Jurassic reservoir. The reservoir porosity ranges between 10-20 % while the average permeability is low, ranging between of 1-10 md with locally higher permeability of about 20 - 30 md in some layers. The average net reservoir thickness is about 200 ft and water saturation is less than 15 %. Initial oil water contact (OWC) was estimated to be 13,360 ft Subsea. The initial reservoir pressure was 10,525 psi at 13,200 ft SS (datum). The oil is under saturated with saturation pressure as 1,959 psi. Oil is light and the density is 36-380 API. The asphaltene onset pressure (AOP) is nearer to 3,400 psi, at a temperature of 2350 F.
The Large Scale Steamflood Pilot (LSP) is a project aimed to determine the feasibility of economically steamflooding the Wafra First Eocene carbonate reservoir. The field is located in the Partitioned Neutral Zone, between Kuwait and Saudi Arabia. The reservoir is a dolomite, with 14-20*API oil. The LSP consists of sixteen inverted 5-spot injection patterns. A fully integrated workflow was matured to maximize the value of information provided by four full cores that were collected when drilling the LSP wells. Core work will support reservoir characterization and dynamic simulation, essential tools for project decision-making. High-level workflow consisted of the following phases: (i) define field, laboratory and office activities, (ii) identify and prioritize stakeholders, (iii) delineate project schedule and assign responsibilities. Coring and core
analysis for heavy oil involves short, mid and long term activities, that may require several years of planning and execution. Planning and frequent communication engaged core experts very early in the work process. Their input was used to shape the project, assuring reliable execution of dependent and independent tasks as work progressed. Synergies between subject matter experts were promoted, and proved to add value to the project. Due to the organizational efforts, the project schedule was not affected by personnel changes. Concerning lab measurements, the operator's heavy oil experts recommend Best Practices for the determination of relative permeability. Long equilibrium times, crude oil instabilities, and viscous fingering are challenges unique to heavy oil systems. Limited capability for such measurements exists in the industry. Heavy oil tests are not routine and should be carefully assessed. We hope that the integrated workflow proposed in this paper provides guidance to similar projects on planning and execution of heavy oil coring programs and analysis.
The Large Scale Steamflood Pilot (LSP) is the third in a series of staged tests conducted to validate the feasibility of applying the enhanced oil recovery technology of steamflooding to unlock the production potential of the heavy oil Eocene reservoir in the onshore Partitioned Neutral Zone (PNZ). Refer to Fig. 1 for the PNZ location. Previous tests included the Small Scale Steamflood Test (SST), which was successfully completed in 2008, and simple steam stimulation testing, conducted in the
late 1990s. The LSP consists of sixteen inverted 5-spot injection patterns. The project is expected to lead to full-field steamflooding of the First Eocene reservoir, marking the first commercial application of a conventional steamflood in a carbonate reservoir anywhere in the world.
The First Eocene is the shallowest reservoir at Wafra field. Average depth to the top of the reservoir is about 1,000 feet. The stratigraphic interval averages 750 feet thick with a gross average porosity of 35% based on well log and core data, and a gross average permeability of 250 md based on core plug measurements. Based on current field practice a porosity cutoff of 35% is used to define net reservoir. The average porosity in the net reservoir is 43% and the net average permeability is about 280 md. The reservoir was discovered in 1954. Full field development and production commenced in 1956. Current oil cumulative from the reservoir is over 300 MMBO. Oil production exceeds 25,000 BOPD of 14-20 ºAPI high-sulfur oil. The First Eocene is a depletion drive reservoir, with partial solution gas drive and limited aquifer support. The aquifer support is not sufficient to maintain the reservoir pressure at current production rate.
A fully integrated workflow was matured to maximize the value of information provided by four LSP cored wells. Core work will support reservoir characterization and dynamic simulation, essential tools for project decision-making. The purpose of this paper is to describe the approach that was followed to maximize the value of heavy oil core analysis and support the LSP development with appropriate petrophysical data.
Jedaan, Nizar Mohamad R. (Qatar Petroleum) | Fraisse, Christian J. (Total E&P USA, Inc.) | Pluchery, Eric (Beicip-Franlab) | De Groen, Vincent Laurens Nico (Beicip-Franlab) | Dessort, Daniel (Total E&P USA, Inc.) | Al Abdulmalik, Abdulmalik (Qatar Petroleum)
Both core description and the log detection have evidenced the presence of bitumen inside the Bul Hanine Field (figure 1), which can be particularly abundant in some wells. This tar mat severely impacts reservoir production behaviour because it acts as a permeability reducer and a barrier to flow. Properly understanding its distribution and its propagation throughout the reservoir is then essential for the prediction of reservoir performance under various development plans, for instance when water flooding the field.
The objectives of this study were to:
Fulfilling these objectives has allowed more accurate volumetric estimations, taking the tar mat into account in the dynamic reservoir modelling as well as in planning further development of the field.
Tar mat occurrence was investigated across more than 5400 ft of cores from 26 wells, 90 well logs and a large number of cuttings samples. Two tar mats were identified in the reservoir. The upper tarmat was formed in the crestal area at early stage of the oil charging (early phase segregation?). The second major one was formed at deeper depth.
The tar mat in the Jurassic reservoirs is composed of asphaltenes. Tar mat formation is explained as follows:
The methods applied in this study include geochemical characterisation of the bitumen of the Bul Hanine Field, a quantification of the tar content in cores using simple techniques (optical observation, Rock-Eval, Iatroscan, image analysis), and extending this quantification through wireline data in non-cored wells and then, subsequently across the field. In the reservoir model, through the relationship between reservoir quality (rock-type) and bitumen content, the distribution of tar mats can be inferred and traced across the entire field.
Tar mat occurs in the Bul Hanine Field, particularly in Jurassic reservoir1. Bitumen occurrence can be a problem due to its effect on oil in place calculation (since bitumen is not movable, it should then be removed from the volumetric calculation) and its impact on reservoir quality. Tar mat impacts on the development plans of an oil field when it behaves as a permeability barrier. Injecting water under the Tar mat might result in inadequate pressure support because of poor communication across the Tar mat2.
For these reasons, it is important to know where Tarmat occurs in the field (both laterally and vertically), and what controlled its distribution. This information, supplemented by a good knowledge of compartmentalization of the field, could then be used to plan the location and design of peripheral field injectors and ensure optimum sweep efficiencies.
Objectives of This Paper
A series of investigations was carried out with the aim of data collection in order to:
The detailed study of the tarmat in Bul Hanine Field was carried out using standard techniques used in Organic Geochemistry.
This paper presents the process, and, results of the analyses of pore/grain morphology of rock fragments, from Mauddud-Burgan reservoir rock in Kuwait, for porosity, permeability and means hydraulic radius (MHR) calculations. The images are captured using SEM and thin-section analyses.
In this study, 2-dimensional images are used to characterize the morphology of the grains and pores, using a two step process. In the first step, the image is captured using a backscattered electron detector (BSE) digital electron microscopy imaging, and thin-section photography. All of the grain/ pore features captured in the image are reported in micrometer units. In the second step, the area of such features is scanned using image analysis software that has the ability to accurately measure several morphological parameters of pore and grain spaces.
A robust technique of visual estimate is used, which has the advantage of speeding the image analysis process. The visual analysis software tool counts different pores and grains and also measures their shapes and sizes that are crucial for porosity, permeability, and MHR calculations.
Several morphological features were been selected for measurement, including: object count, area, perimeter, and roundness. These analyses were conducted twice for each selected image - once for grain morphology and once for pore morphology. Porosity, permeability and MHR features are calculated based on area of the pore/ grain features measured from two-dimensional images.
This paper reports a comprehensive study towards quantitative characterization of the permeability distribution in complex Mauddud-Burgan reservoir in Kuwait. The main objective in this study is to develop a generalized strategy for data-mining a large data-set of rock measurements. This study utilizes measurements of petrophysical and grain/ pore morphology properties in order to correlate permeability. Data-set contain measurements obtained from different length scales, ranging from SEM to wire-line log scale.
Characterizing the permeability for the Mauddud-Burgan reservoir is a challenge because of the complexity of this reservoir. The process is dependant on the type of data available for the reservoir. This study strives toward comprehensive data mining to understand the permeability of this complex reservoir.
A Multiple-Layer Feed Forward, MLFF, with back propagation neural network is developed to calculate the permeability at each desired vertical depth in the reservoir. This tool can assist in determining permeability at any vertical depth of the reservoir, within the boundaries of the reservoir model. Knowledge of other petrophysical properties, such as, porosity, pore type, and pore size distribution, as available, are integrated to estimate the permeability.
Injection of seawater through pipelines subject to corrosive conditions may lead to the introduction of dissolved and particulate iron which may cause formation damage and loss in injectivity.This study addresses the estimation of damage potential by a combination of modeling and laboratory testing. The specific goal of the proposed program is to create a database of acceptable iron contamination in injection water, and, injection rates for targeted reservoirs in Kuwait.
Past studies show that iron concentration in injection water fluctuates due to seasonal temperature variations which control the corrosion rate inside pipelines.Conventional methods for quantifying formation damage due to seawater injection do not accurately reflect the processes occurring in the field, such as the variation of iron concentration with temperature.
Traditional "rules-of-thumb" for predicting the lifespan and performance of injection wells ignore the contribution of temperature fluctuations on iron concentration in the injection water.This study proposes a procedure to estimate the potential and tolerance for formation damage in injection well systems due to uncontrolled introduction of iron and other particulate in the injection water.The proposed experiments include:study of the dry rock samples; study of reservoir brine and injected water; and, study of the fluid-rock interactions.
Study of reservoir-rock utilizes routine core analyses to obtain the petrophysical parameters.SEM/ EDS are proposed as tools to characterize mineralogy and pore-morphology.Porosimitery is proposed for obtaining the pore size distributions and the mean hydraulic radius, MHR.Fluid Characterization involves the analysis and interaction of formation brine and the injection fluids.Knowledge of rock mineralogy and pore throat size distribution is expected to help in estimating the acceptable solid and iron concentration in the injected water.Characterization of rock-fluid interactions involves the injection of water with a range of solids and iron concentration in the reservoir cores to identify potential formation damage.
Comparison of petrophysical properties between Eastern U.S. Silurian-age sandstones and Western U.S. Cretaceous-age sandstones illustrates the universality of certain low-permeability reservoir properties. In the Medina insitu porosities are generally within 97% of routine porosities and in Mesaverde-Frontier are approximately 0.8 porosity percent less than routine porosities. In both groups insitu Klinkenberg permeability (ki) exhibits an increasing difference from routine air permeability with decreasing permeability. Both groups of low-permeability sandstones exhibit very similar decrease in "irreducible" water saturation (Siw) with increasing ki. However, lithologic variations that are related primarily to grain size and shaliness cause variance in Siw at any given permeability. Both eastern and western low-permeability sandstones exhibit sharp decrease in gas relative permeability ( krg,Siw) with increasing Siw. The krg,Siw values are generally less than 5% at Siw greater than 60%. Although average krg,Siw is low and average Siw is high, these sandstones typically produce water-free gas. Analysis of cumulative flow capacity in wells in both regions indicates that a significant fraction of total flow capacity often comes from a few, thin, higher permeability intervals within generally low-permeability sandstone. Although storage capacity declines with decreasing permeability due to increasing Siw, low-permeability intervals still represent a significant fraction of total storage.
Lower Silurian-age low-permeability Medina Group (Medina) sandstones of the Appalachian Basin and Upper Cretaceous-age Mesaverde Group (Mesaverde) and Frontier Formation (Frontier) sandstones in several Western U.S. basins are important targets for natural gas exploration and production. Recoverable reserves are estimated to be approximately 30 trillion cubic feet (TCF) in the Medina1 and approximately 10 TCF in the low-permeability Mesaverde and Frontier.2 The abundant gas reserves of the Medina, which drillers locally call the "Clinton," are regionally extensive and without a recognized downdip water contact. Mesaverde and Frontier reservoirs exhibit similar undefinable gas-water contacts. In both areas water can occur updip of gas production.
Prediction of gas producibility from low-permeability sandstones is complicated because conventional log-analysis interpretation and formation-evaluation parameters are commonly not applicable. Furthermore, conventional core-analysis petrophysical values can differ significantly from reservoir values. Early work on petrophysical properties of low-permeability sandstones demonstrated that pore-volume compressibility is small and that increasing confining stress and water saturation cause permeability to decrease.3,4,5 Jones and Owens6 quantified these effects and concluded that the presence of a thin, sheet-like, tabular pore structure could explain the response to confining stress. Ostensen7 provided a comprehensive theoretical analysis of the relationship between grain boundary micro-crack dimensions and permeability. Dutton et al.2 summarized the properties of low-permeability sandstones in the U.S. and provided extensive references to previous studies. Recently, Byrnes 8 summarized insitu rock properties for low-permeability sandstones in Rocky Mountain basins and Castle and Byrnes9 presented insitu rock properties for the Medina in northwestern Pennsylvania.
Laboratory displacement measurements and compute simulations were performed to support field evaluations of residual oil saturation for a high permeability sandstone reservoir exhibiting moderate permeability sandstone reservoir exhibiting moderate to weakly water-wet characteristics. Logging a sponge core data had indicated low oil saturations in zones depleted by natural bottom water drive. The laboratory displacement tests were designed to help further verify these results and to provide data most applicable for the prediction of future field performance. Reservoir conditions displacement test data were in reasonable agreement with the field results, but laboratory conditions tests predicted higher residual oil saturations. Water predicted higher residual oil saturations. Water flood and centrifuge measurements were in agree under appropriate experimental conditions.
Simple simulations were conducted to evaluate factors affecting the gravity drive process during vertical displacement of oil by water. Particular attention was paid to sensitivity analyses of the effects of the oil relative permeability and viscosity characteristics. The results obtained are useful for reservoir management questions relating to production rates, coning phenomena, and time-lapse monitoring of saturation changes.
Effective reservoir management requires the integration of information from a variety of sources. Geological and engineering models are ideally used as dynamic tools, to be revised and upgraded as needed to account for the most recent field performance. Likewise, laboratory-derived data on performance. Likewise, laboratory-derived data on rock, fluid, and fluid flow properties are subject to review and verification based on observed reservoir production characteristics.
This paper describes a laboratory test program in support of field evaluations of residual oil saturation for a high permeability sandstone reservoir in the Middle East that produces by natural water drive. The laboratory work was initiated to resolve previously-observed differences in oil displacement efficiency from centrifuge and waterflood tests on the reservoir rock. The present work also provided a better definition of present work also provided a better definition of core sample handling and testing methods most applicable for use in the prediction of field performance. A second phase of the study involved performance. A second phase of the study involved some simple simulations to evaluate factors affecting the efficiency of the gravity drive mechanism during vertical displacement of oil by water.
FIELD DETERMINATIONS OF RESIDUAL OIL SATURATION
Sponge coring and extensive logging in one well indicated that the residual oil saturation averaged only about 14% in a continuous sand section depleted by bottom water drive. The naturally depleted water-swept interval in this well extended 30 feet above the original oil-water contact.
The well was drilled using a water-base bland mud formulation having an API filtrate loss of less than 8 cc. The friable, poorly-consolidated nature of the reservoir rock caused low core recovery over the swept zone (about 7 ft. out of 28 ft. cut), in contrast to good overall recoveries achieved in other wells using more conventional plastic sleeve core barrels. It was nevertheless possible to process ten, 3 1/4" diameter whole core samples and process ten, 3 1/4" diameter whole core samples and the associated sponge from the swept zone. Table 1 gives a summary of the results. The average oil saturation was determined to be 14% at reservoir conditions, in good agreement with a range of 12-17% measured by different logs and a single well tracer test.