CC BY
49ГЕОЛОГО-МИНЕРАЛОГИЧЕСКИЕ НАУКИ
RESERVOIR CHARACTERIZATION OF CARBONATE RESERVOIRS KARACHAGANAK, TENGIZ, KASHAGAN
Altayeva N.
Altayeva Nargiza - Master, SPECIALTY: OIL AND GAS, FACULTY OF OIL AND GAS ENGINEERING, KAZAKH-BRITISH TECHNICAL UNIVERSITY, ALMATY, REPUBLIC OF KAZAKHSTAN
Abstract: this article presents a study which compares carbonate fields in Central Asia and Southeast Asia, their reservoir properties and petroleum system. The Central Asian carbonate reservoirs in the Pricaspian Basin selected for comparison include Kashagan, Karachaganak and Tengiz.
Keywords: carbonate reservoir, porosity, permeability, oil and gas reserves, lithology.
The Central Asian carbonate reservoirs in the Pricaspian Basin selected for comparison include Kashagan, Karachaganak and Tengiz. Average porosities are low and strongly influenced by the complex diagenetic history. Permeability is aided by the presence of fracture networks. Burial depths are significant: over 4 km at least. Probably the biggest challenge, however, in the development of these fields is the large concentrations of H2S and CO-.
Many of the reservoirs in fields are in carbonate successions, and more than 50% of the estimated remaining proven and undiscovered petroleum resources are in carbonate reservoir rocks.
Carbonate reservoirs typically contain highly variable distributions of pore sizes and pore throat geometry. These distributions are a primary control on fluid movement in the reservoir. Knowledge of the pore geometry is critical to optimize the number of wells in a carbonate field, to optimize facilities, and to estimate recoverable reserves.
Rock Types can be synonymous with either depositional or diagenetic texture or combinations of both. In carbonates they are often overlapping with pore types, especially in reservoirs with a high contribution of secondary porosity and associated variety of pore types (i.e. vugs, fractures and modified "micro" pores). Reservoir rock types are conventionally defined by identifying geological lithofacies or rock texture from core and then predicting those geological parameters from logs. Rock type predictions can use a simple correlation, or can be predicted by more complex methods such as neural networks and discriminant analysis. Since log responses are related to different properties of the formation from those commonly used to extract geological attributes, such approaches can be valid only when a clear link between geological facies and petrophysical rock types is established. Carbonate reservoirs are often modified more extensively through diagenetic processes compared to siliclastic reservoirs. The high level of reactivity and chemical instability of carbonate minerals leads to a much more significant impact on key reservoir properties such as porosity through diagenetic modification. The highest impact on the reservoir performance is generally attributed to diagenetic processes altering pore geometry or pore types and associated pore throat distributions. As a consequence, pore typing in carbonates is an important element of rock type definition. Rock type determination is a critical
factor in reservoir characterization which drives the accuracy of the reservoir simulation model prediction. The identification of flow units and petrophysical rock types is especially difficult due to highly variable pore geometry, the lack of correlation between pore size and pore throat size, the lack of correlation between porosity and permeability, and the difficulty in establishing any link between small scale pore properties and large-scale reservoir properties. The pore type classification is based primarily on pore throat size which directly controls flowing dynamic properties. Most of Kazakhstan's oil and gas reserves have not been developed. Many areas remain under- or unexplored. It is expected that a considerable portion of Kazakhstan's potential oil and gas reserves will be located offshore in the Caspian Sea.
It is necessary to mention that most of the samples taken from shallow-water facies of the eastern shelf are represented by clastic carbonate sediments of varying degree of dispersion, from shelly-oolitic sand to carbonate mud. Finer pelitomorphic carbonate muds are found in the area of the outer shelf. The content of carbonate matter in these sediments varies from 40% to 95%. This area adjoins that part of the sedimentation basin where the physiographic conditions of depositional environment are characterized by sharply pronounced aridity without fluviatile supply.
Reservoir characterization is an important component to reservoir management. The characterization process integrates information from two main sources: static data (such as geological interpretation based on out crop, seismic, welllog, core, drilling fluid loss) and dynamic data (such as well test, production logging, and production history). Although the use of dynamic data in the reservoir characterization process has been attempted in the past, its application is still limited and lack in garobust workflow. In this study, numerical well test in techniques (Kamaletal.2005) were applied to understand and characterize the fracture- matrix properties of Naturally Fractured Reservoirs (NFRs), which is especially challenging due to a high degree of uncertainty. The effective utilization of pressure transient data that contain rich information about well sand reservoir should narrow the uncertainty, improve the characterization, and help optimize field development.
The first approach (Jennings-Lucia) links rock-types petrophysical properties, i.e. porosity, permeability and capillary properties, to a rock-fabric classification that groups carbonate rocks in three classes:
1 Grainstones, dolograinstones and large crystalline dolostones;
2 Grain-dominated packstones, fine and medium crystalline, grain-dominated dolopackstones, and medium crystalline, mud-dominated dolostones;
3 Mud-dominated limestones and fine crystalline, mud-dominated dolostones.
Each class should occupy distinct areas in the porosity-permeability bi-dimensional space with permeability increasing according to the increasing inter-grain porosity, grain size and sorting. It is assumed that pore throats, the factor which is really controlling permeability, are directly elated to particle size, sorting and interparticle porosity. On this basis generalized power laws porosity-permeability relationships were estimated for the three classes using worldwide data. The relevant ranges of existence were estimated according to the following equations:
K = ea(X) V(X)
Where 9 is the interparticle porosity given in fraction of bulk volume, K is the permeability in mD, and a(X) and b(X) are parameters dependent on the rock fabric number X:
1 XCarbonate rock Class 1 :> 2.5, < 4
2 XCarbonate rock Class 2 :> 1.5, < 2.5
3 XCarbonate rock Class 3 :> 0.5, < 1.5
and
1 a(X)=a0 - a1 ln(X) (where a0 = 22.56, a1 = 12.08 )
2 b(X)=b0 - b1 ln(X) (where b0 = 8.671, b1 = 3.603 ).
These general power law relationships, to be adjusted according to the specific data of different reservoirs in which they are applied, are characterized by the rock-fabric numbers Xwhich have a geological significance. This link between petrophysic and geology is the main characteristic of this approach which should allow to drive the permeability reservoir
10
modeling according to geological concepts. The weak point on the contrary appears to be the diagenetic overprint, considered by the authors as generally negligible.
In Eni the most commonly used technique to perform this "Integrated Petrophysical Characterization" is the process named as "Cluster Analysis". This methodology is based on a statistical algorithm that cluster logs data according to their similarity, i.e. Average Euclidean Distance in the multi-NLog-dimensional space. It is so obtained a zonation along the wells which, tuned according to core data when available, is characterized by distinct lithological, petrophysical and geological characteristics. The so called "Cluster Analysis" process is in fact an "Unsupervised Classification" of log data, i.e. without any "a priori" guidance of classes to be defined. This "Classification" is therefore optimized, tuned and characterized "a posteriori" with core data. This procedure has been largely and successfully used by Eni in the last 20 years for silici-clastic reservoirs but it results to be less effective for carbonate rocks. The Silici-clastic rocks are in fact strongly characterized by its original texture, i.e. grain size, shape, sorting and packing, and generally they are not strongly affected by the diagenetic overprint. The relevant permeability is thus well correlated to porosity. Silici-clastic rocks have therefore flow characteristics mainly related to factors like porosity, shalyness and lithology which are well recognizable by conventional logs.
Rock contains abundant microscopic flaws. Microcracks, voids, pore space or grain boundaries are here considered as microscopic fractures. For simplicity, Griffith modeled such flaws as strongly elliptical microfractures, now known as Griffith (micro)cracks. He considered the stress concentrations associated with these microfractures and the energy that it takes for them to grow and connect. He then obtained much more realistic (although not perfect) estimates of tensile strength.
Joints are open fractures in which the fracture walls move away from each other. Joints may form due to several causes such as:
• burial (compaction)
• heating (dilation)
• cooling (contraction)
• tectonic stress
Table 1. Comparative characteristics of carbonate reservoirs Karachaganak, Kashagan, Tengiz to certain parameters
Karachaganak Tengiz Kashagan
Location of the reservoir It is located about 150 kilometres (93 mi) east of Oral (Uralsk) in the northwest of Kazakhstan It is located on the south side of the Pri-Caspian basin on the northeastern edge of the present-day Caspian Sea. It is located in the northern part of the Caspian Sea close to Atyrau
Type of reservoir gas condensate field oil and gas field oil field
Depth approximately 5,000 metres from 13,000 feet to over 15,000 feet. 4,500 metres (14,800 ft)
Lithology The reservoir is a carbonate massif that consists of heterogeneous reef (Permian) and platform carbonates (Carboniferous, Devonian) - primarily limestone and dolomite, with overall low porosity and permeability. The North Caspian is a pericratonic depression of Late Proterozoic-Early Paleozoic age. It is a massive Carboniferous-Devonian carbonate reservoir. The Tengiz platform is a cyclic alternation of grainstone and packstone with thin volcanic ash layers and later strongly modified by diagenesis. The Tengiz flank areas are dominated by microbial boundstone deposition in an upper slope environment and detrital sediment in the lower slope. Carbonate facies in the Tengiz flanks are generally tight with bitumen cement and are dissected by fracture systems which act as fluid conduits. The Paleozoic reservoir at Kashagan field is characterized by a relatively porous carbonate platform interior surrounded by a highly-cemented margin called the rim. Most of the faults and fractures are concentrated near the rim.
Karachaganak Tengiz Kashagan
Reservoir porosity range from 9.67% to 11.70% porosity of the reservoir interval is 15% range from 0.1 to 10.8%
Reservoir temperature 70 to 95 °C 343-368 °F nearly 200° F 80° C
Crude Characteristics 45-50° API Gravity 3.5-5.0 mol % H2S 48.2° API Gravity 12.5 mol % H2S 42-45° API Gravity 18-20 mol % H2S
Seismic survey Each of the seismic facies is in turn defined by reflector characteristics, such as continuity, intensity, and geometry. Seismic mapping of the abovementioned elements with corroboration from other subsurface data identifies heterogeneity at well, flow unit, and field scales, including 1)bedset-scale variations internal to clinoforms and wedges; 2) along strike and temporal variations in clinoform or wedge style; 3)platform-scale asymmetry. Comprehensive logging and coring programs are required for newly drilled wells and for existing wells that can be deepened through the reservoir. Recent applications of NMR and array induction logging technology, in conjunction with more standard logging tools, have improved understanding of the reservoir. Unfortunately, difficulties associated with the calibration of old openhole logs, sparse core coverage, and a major diagenetic overprint of solid bitumen combine to limit the identification of effective reservoir at Tengiz based on openhole log data alone. A limitation of production logs is that they only measure fluid entering the wellbore and are not necessarily indicative of flow in the reservoir away from the well. Faults commonly strike parallel to the rim boundary. At the borehole-scale, image logs also show a fracture set striking parallel to the rim, with secondary sets at high angles to the rim. Rim parallel features are consistent with their formation during early syndepositional margin collapse and differential compaction of the platform and rim.
Joint is a name used by geologists to dilational fractures. Mechanicists and engineers, differenly, named these fratures as cracks. Joints are hence open fractures, along which the wall rocks move away from each other.
Mechanical stratigraphy represents a subdivision of rock into discrete fracture intervals according to the mechanical properties of these intervals.
Joints develop within mechanical units (competent layers) and terminate against mechanical interfaces (traction-free surfaces), which may be represented by both stratigraphic contacts and weak rock layers that resist to fracture propagation.
Examples
Karachaganak field
Fig. 1. Strustural cross sections through the Karachaganak field showing the buildup associated with
the Permo-Carboniferous reef complex. The flanks of the structure are overlain by salt so imaging them seismically is dificult. Consequently, the new 3D seismic acquisition program being planned for 1999 may lead to a different interpretation of the flanks
Conclusion
The combined study of rock fabrics and physical properties of reservoir rocks is an important aspect of reservoir characterization, because it directly links the geophysical exploration tools to the reservoir properties. Results of such a study provide a fundamental understanding of the petrophysical responses to geological features and their dynamic changes. The aim is to directly convert the attributes of the physical properties to the geological reservoir parameters.
References
1. Amyx J.V., Bass, D.M., Whiting R.L. "Petroleum Reservoir Engineering". McGraw-Hill Book Company. New York, I960.
2. Sieve Elliott H.H., Hsu Terry OHeam Ian, Vercesi F. Sylvester Ricardo. The Giant Karachaganak Field, Unlocking Its Potential. Karachaganak Integrated Organization London, England. UK autumn, 1998.
