TY - JOUR
AU - Rahimpour-Bonab,, H
AB - Abstract Reservoir rock typing is the most important part of all reservoir modelling. For integrated reservoir rock typing, static and dynamic properties need to be combined, but sometimes these two are incompatible. The failure is due to the misunderstanding of the crucial parameters that control the dynamic behaviour of the reservoir rock and thus selecting inappropriate methods for defining static rock types. In this study, rock types were defined by combining the SCAL data with the rock properties, particularly rock fabric and pore types. First, air-displacing-water capillary pressure curues were classified because they are representative of fluid saturation and behaviour under capillary forces. Next the most important rock properties which control the fluid flow and saturation behaviour (rock fabric and pore types) were combined with defined classes. Corresponding petrophysical properties were also attributed to reservoir rock types and eventually, defined rock types were compared with relative permeability curves. This study focused on representing the importance of the pore system, specifically pore types in fluid saturation and entrapment in the reservoir rock. The most common tests in static rock typing, such as electrofacies analysis and porosity–permeability correlation, were carried out and the results indicate that these are not appropriate approaches for reservoir rock typing in carbonate reservoirs with a complicated pore system. rock typing, water saturation, capillary pressure, pore type 1. Introduction Rock typing is a method for classifying reservoir rocks according to their ability of fluid conduction and saturation. In integrated reservoir rock typing, the depositional fabric, diagenetic features, petrophysical properties and dynamic characteristics should be considered. Accordingly, the rock typing process has two parts: the first part, which includes geological and petrophysical studies, is known as static rock typing and the second part, which includes fluid distribution and fluid-rock interaction characteristics in the reservoir, is dynamic rock typing. The results of these classifications must work well together to be applicable in a real reservoir model. The study of depositional fabric, diagenetic features and petrophysical properties characterizes the static properties of the reservoir rock. The most important feature in this category which makes a link between static and dynamic properties is porosity. Depositional and diagenetic processes control the fluid flow and saturation behaviour in the reservoir rock by making and altering the pore system, including pore throats and pore bodies. Thus depositional fabric and diagenetic features should be studied to characterize the pore system in the reservoir rock. For this purpose detailed petrographic analysis needs to be performed on reservoir rock samples. Wireline logs and correlation between porosity and permeability are other common instruments in routine rock typing analysis and reservoir rock characterization, which can be seen to have been applied in many investigations (e.g. Skalinski et al2006, Rushing et al2008, Porras and Campos 2001, Saboorian Jooybari et al2010, Alpin et al2002, Bagheri and Biranvand 2006). Porosity-permeability correlation can only be useful in reservoir rocks with interparticle porosity as this type of porosity is simple and can show an applicable relationship with permeability. However in other types of porosity, particularly dissolution enlarged pores such as molds and vugs, there are many parameters that control the relationship between porosity and permeability (this is the common feature of most carbonate reservoirs). These types of porosity are too complicated to be correlated easily with permeability and also cause fluid entrapment. The classification of reservoir rocks based on wireline logs shows inconsistencies with other types of classification, particularly pore geometry based (Chehrazi et al2011) and SCAL defined rock types. The most common wireline logs in rock typing analysis, such as RHOB, DT, and NPHI, usually classify the reservoir rocks according to the responses they have received from the fluid content or the physical properties of the rock. The wireline logs are unable to distinguish between different pore types or pore connectivity. The porosity–permeability correlation might be a good indicator of the applicability of logs in rock typing analysis. Good correlation between porosity and permeability is representative of the simplicity and the limited variety of pore types in the reservoir rock, thus the porosity percentage is the main controlling parameter in pore system characteristics, as well as the controlling parameter in wireline logs. However, these routine rock typing methods are unable to capture the fluid flow properties in the studied reservoir. Fluid flow and saturation characteristics in a reservoir rock are the crucial constants that control the quality of reservoir rock and need to be considered in reservoir rock typing. The pore geometry of rock and fluid–rock interactions, such as interfacial tension (IFT) and wettability, are the most important parameters that control this fluid flow and saturation behaviour. The reservoir under study is the Permo–terriasic carbonates of Dalan and Kangan Formations in the South Pars Gas field. Depositional environment, diagenetic history and the reservoir quality of this field have been subjected to many investigations (Alsharhan and Narin 1997, Al-Aswad 1997, Moradpour et al2008, Estafili-Dizaji and Rahimpour-Bonab 2009, Rahimpour-Bonab 2007, 2010, Rahimpour-Bonab et al2009, Tavakoli et al2011). In this paper, the rock typing analysis was performed by classifying the drainage capillary pressure data (air-displacing-water capillary pressure curves) and combining this with rock textures and pore types. Petrophysical properties were determined in each facies, and eventually, defined rock types were compared with the relative permeability curves to determine the effect of pore system structure and pore types on fluid entrapment. The purpose of this study was direct classification of the reservoir rocks by SCAL data, detecting the effect of rock properties, particularly pore types, on the reservoir quality and the fluids, behaviour in the pore system. Considered parameters were capillary pressure, irreducible water saturation (in capillary pressure curves) and residual oil saturation (in relative permeability curves). The second purpose was enquiring as to the applicability of some routine methods such as electrofacies analysis and porosity–permeability correlation for an effective classification of reservoir rock, whilst also explaining why static and dynamic rock types are occasionally incompatible. 2. Geological setting South Pars gas field was formed due to the positive nature of a huge NNE–SSW trending structural feature, which is known as the Qatar Arch (figure 1). The Zagros folded belt bounds the Arch to the north and north east which is located in the interior platform of the Arabian plate. The thinning of the Permian sediments in this field revealed the existence of a syndepositional structural high in the south-eastern part of the Zagros foldbelt. The Persian Gulf basin is one of the most prolific gas and oil accumulations in the world, containing 55–68% of the world's recoverable oil reserves and more than 40% of its gas reserves (Konyuhov and Maleki 2006). In the South Pars field, gas resources are accumulated in P-T stratigraphic units, including upper Dalan (late Permian) and Kangan (Triassic) formations. These two units are composed of dolomite, limestone and an evaporate series, which are representative of deposition in the shallow-marine condition (Alsharhan and Nairn 1997, Ehrenberg et al2007, Esrafili-Dizaji and Rahimpour-Bonab 2009, Rahimpour-Bonab et al2010). These units constitute very extensive natural gas reservoirs in this field and the Persian Gulf area (Kashfi 1992). Figure 1. Open in new tabDownload slide Location map of the South Pars–North Dome super-giant gas fields in the Persian Gulf. Figure 1. Open in new tabDownload slide Location map of the South Pars–North Dome super-giant gas fields in the Persian Gulf. 3. Material and methods Plug samples from the reservoir interval in subject wells were selected for the following special analyses: (1) Capillary pressure analysis at ambient (centrifuge method) conditions. (2) Relative permeability tests at ambient conditions in water–oil, gas–oil and gas–water systems. The results of gas–oil relative permeability test were selected for this study. Capillary pressure curves at ambient conditions: air-displacing-water capillary pressure curves were acquired starting from 0.2 bar up to 70 bar (in laboratory air–water system). In this test, soxhlet cleaned samples were used. The plugs were loaded into a centrifuge and subjected to non-stop centrifugation under air. Sample saturations were determined by weighing the samples at each pressure step, after 12 or 24 h rotational time. Based on the experimental results, capillary pressure curves were determined. After repeated miscible cleaning of all samples, they were dried at 95 °C in a dry oven to a constant weight and had porosity determined in a Boyle's law helium porosimeter to a precision of ±0.1%. All samples then had air permeability determined in a precision permeameter to ±2% using mass flow control to eliminate non-Darcy (visco-inertial) flow. A relative permeability test was performed on preserved samples in ambient conditions using the centrifuge technique and a measurement of Swi, Sor and relative permeability end points was carried out after the following work flow: Mild cleaning to remove the crude oil from the sample by core flooding at ambient temperature. The process has been proved not to alter the wettability characteristics of the rock. Sample drying and saturation under synthetic brine. X-ray CT scanning of the samples to identify and quantify their internal heterogeneity. Several hundreds of thin sections from the pay zones of three wells in the South Pars gas field of the Persian Gulf were selected and studied. Samples were impregnated with blue-dyed epoxy to describe texture and diagenetic features. Samples were also examined with a scanning electron microscope (SEM) to characterize pore types. The reservoir units are composed of dolomite, limestone and evaporate series in this field. Gas is hosted by these Permo–Triassic carbonates of the upper Dalan (late Permian) and Kangan (Triassic) formations in the field. The classification of capillary pressure curves was carried out based on the irreducible water saturation and the capillary pressure at different water saturations on capillary pressure curves, rock properties were then determined in each class. Available relative permeability curves from one well were attributed to corresponding rock samples in reservoir rock types (RRTs) to determine the residual oil saturation in these RRTs. Eventually, electrofacies analysis was performed in reservoir intervals in subject wells. The electrofacies analysis was performed using the multi-resolution graph based clustering (MRGC) method, which was introduced by Ye and Rabiller (2000). 4. Rock typing Traditionally reservoir rock typing is the classification of the reservoir rock into petrophysical units using geological properties, porosity–permeability correlations, wireline logs and mercury injection curves (e.g. Porras 1998, Boada et al2001, Leal et al2001, Skalinski et al2010, Lehmann et al2008). More integrated reservoir rock typing includes incorporating SCAL data into these classes because reservoir rock samples with similar petrophysical properties sometimes show different fluid flow and saturation characteristics. Among different types of capillary pressure tests, mercury injection capillary pressure analysis is the best instrument for the classification of reservoir rock based on the pore geometry. Capillary pressure analysis applying the centrifuge and porous plate tests at different systems (air–water, oil–water, etc) enable one to characterize the ability of reservoir rock to transmit and entrap the reservoir fluids. Of these two tests, the centrifuge test is more common because it is faster and more convenient than the other. In this study rock types were defined by direct classification of the air-displacing-water capillary pressure curves and combining the rock properties (rock texture and pore types). The classification is proposed to avoid the inconsistency between routinely defined and SCAL defined rock types and to detect the relationship between the fluid saturation and the geological properties in the reservoir rock samples. Capillary pressure curves obtained from capillary pressure analysis in air–water, air–oil and oil–water systems do not represent the pore geometry of reservoir rock samples properly. This is due to the lack of necessary pressure for the fluid displacement, and also the influence of the reservoir fluids–rock interactions such as wettability and interfacial tension, on fluid distribution in the porous media. However, these curves are strongly affected by the pore geometry (according to the result of the study) and they are better representatives of the fluid distribution in the reservoir conditions. Six different classes were obtained from the classification of the air-displacing-water capillary pressure curves. Capillary pressure curves were classified according to the similarity in irreducible water saturation (Swir) and displacement pressures at 85%, 60%, 35% and 16% water saturation. The most similar curves were placed in a class. The result of this classification is presented in table 1 by averaging the capillary pressure and Swir in each class. The capillary pressure was determined at several points to consider the entire curve for the classification. A mean curve was determined by averaging all curves in each class (figures 2 and 3). Average values of permeability and porosity were also determined for these classes (table 1). Finally, rock types were defined by incorporating the geological properties (depositional textures and pore types) into the classes. Figure 2. Open in new tabDownload slide Classified air–water capillary pressure curves and the average curve in each class. Figure 2. Open in new tabDownload slide Classified air–water capillary pressure curves and the average curve in each class. Figure 3. Open in new tabDownload slide The average capillary pressure curves for each rock type. Figure 3. Open in new tabDownload slide The average capillary pressure curves for each rock type. Table 1. The average values of saturation properties and petrophysical parameters in each rock type. Rock type . Swir (%) . P (85%) . P (60%) . P (35%) . P (16%) . K (MD) . PHI (%) . Dominant rock fabric and pore types . 1 9.56 1.67  4.76  15.14  64.26 406.48 20.48 Dolo-grainstone with preserved interparticle porosity 2 16.57 1.72  5.25  38.09 122.44  53.67 27.70 Grainstone with interparticle porosity and intergranular cement 3 10.60 4.86 13.81  31.92 131.72   8.67 21.48 Well sorted dolo-grainstone with moldic porosity 4 27.55 12.47 54.78 209.64 490.59   3.40 20.18 Poorly sorted grainstone with moldic porosity 5 19.53 2.64 20.74 120.80 529.15   0.85 18.50 Grainstone with unsorted moldic and vuggy porosities 6 71.10 106.28 >2000 >2000 >2000 <0.1  6.55 Grainstone with totally plugged molds Rock type . Swir (%) . P (85%) . P (60%) . P (35%) . P (16%) . K (MD) . PHI (%) . Dominant rock fabric and pore types . 1 9.56 1.67  4.76  15.14  64.26 406.48 20.48 Dolo-grainstone with preserved interparticle porosity 2 16.57 1.72  5.25  38.09 122.44  53.67 27.70 Grainstone with interparticle porosity and intergranular cement 3 10.60 4.86 13.81  31.92 131.72   8.67 21.48 Well sorted dolo-grainstone with moldic porosity 4 27.55 12.47 54.78 209.64 490.59   3.40 20.18 Poorly sorted grainstone with moldic porosity 5 19.53 2.64 20.74 120.80 529.15   0.85 18.50 Grainstone with unsorted moldic and vuggy porosities 6 71.10 106.28 >2000 >2000 >2000 <0.1  6.55 Grainstone with totally plugged molds Open in new tab Table 1. The average values of saturation properties and petrophysical parameters in each rock type. Rock type . Swir (%) . P (85%) . P (60%) . P (35%) . P (16%) . K (MD) . PHI (%) . Dominant rock fabric and pore types . 1 9.56 1.67  4.76  15.14  64.26 406.48 20.48 Dolo-grainstone with preserved interparticle porosity 2 16.57 1.72  5.25  38.09 122.44  53.67 27.70 Grainstone with interparticle porosity and intergranular cement 3 10.60 4.86 13.81  31.92 131.72   8.67 21.48 Well sorted dolo-grainstone with moldic porosity 4 27.55 12.47 54.78 209.64 490.59   3.40 20.18 Poorly sorted grainstone with moldic porosity 5 19.53 2.64 20.74 120.80 529.15   0.85 18.50 Grainstone with unsorted moldic and vuggy porosities 6 71.10 106.28 >2000 >2000 >2000 <0.1  6.55 Grainstone with totally plugged molds Rock type . Swir (%) . P (85%) . P (60%) . P (35%) . P (16%) . K (MD) . PHI (%) . Dominant rock fabric and pore types . 1 9.56 1.67  4.76  15.14  64.26 406.48 20.48 Dolo-grainstone with preserved interparticle porosity 2 16.57 1.72  5.25  38.09 122.44  53.67 27.70 Grainstone with interparticle porosity and intergranular cement 3 10.60 4.86 13.81  31.92 131.72   8.67 21.48 Well sorted dolo-grainstone with moldic porosity 4 27.55 12.47 54.78 209.64 490.59   3.40 20.18 Poorly sorted grainstone with moldic porosity 5 19.53 2.64 20.74 120.80 529.15   0.85 18.50 Grainstone with unsorted moldic and vuggy porosities 6 71.10 106.28 >2000 >2000 >2000 <0.1  6.55 Grainstone with totally plugged molds Open in new tab RRT 1, having the lowest Swir and displacement pressure at different water saturations, shows the best reservoir quality. Dolo-grainstone with well-preserved interparticle porosity is the dominant rock fabric and pore type in this class. RRT 1 to 6 show an increasing trend in Swir and capillary pressure, and a decrease in permeability and reservoir quality. The capillary behaviour and permeability values are not compatible in RRT 2. Although the average permeability value in RRT 2 is high, Swir is higher than less permeable rock types and also the capillary pressure shows dual behaviour at low (16% and 35%) and high (60% and 85%) water saturations. This behaviour is related to the complexity of the pore system, which is not uniformly distributed in the reservoir rock. This phenomenon shows that permeability and capillary behaviour are not always compatible, because the permeability of rock samples increases as a result of large and interconnected pores or even fractures. But the capillary pressure increases due to the effect of small pore throats. RRT 5 is segregated from RRT 4 because of the differences in petrographic features and the presence of large molds and vugs in RRT 5. RRT 5 includes some rock samples displaying fluid flow characteristics and petrophysical properties with an approximately lower reservoir quality in comparison to RRT 4. RRT 6 is the non-reservoir rock type in the classification. Its irreducible water saturation is very high and displacement pressure at high wetting phase saturation surpasses 2000 psi, which is more than the attainable pressure in centrifuge tests. 5. Depositional textures and pore types RRTs do not necessarily show the same depositional textures and diagenetic histories, but the determination of depositional texture and diagenetic processes is necessary for understanding the variations in the pore system and the controlling parameters in reservoir quality. The complexity in carbonate reservoirs is the result of variety in depositional textures and the intensity in the diagenetic processes. Cementation, dissolution and dolomitization are the most important diagenetic features that extensively change the initial rock fabric and pore system in the studied reservoir. Different types of porosity could be inherited from the primary fabric or created by the diagenetic process. The relationship between porosity and the dynamic behaviour has been considered in many rock typing studies. Although the fluid flow and saturation behaviour are strongly affected by the rock–fluid interactions and wetting phase condition, the effect of the pore system as the conductor of the fluids is undeniable. Comparison between SCAL defined rock types and the rock properties reveals that there is not a remarkable correlation between porosity percentage and the fluid behaviour or permeability in the studied reservoir. The crucial parameter that controls the reservoir quality, fluid flow and saturation characteristics is the pore system structure such as the pore types, pore throat connectivity and pore to pore throat size. The cement content is very important in fluid behaviour in the reservoir rocks. Pore throat size and connectivity decrease extensively with increasing cement content because the pore system is interconnected by the pore throats which are very sensitive even to a small amount of cement. The flow direction, speed and saturation would change by pore throat occlusion, thus two samples with different cement content show different fluid flow characteristics and could be classified as different rock types. Dunham (1962) and Lønøy (2006) classifications were used for rock texture and pore type classification, respectively in this study. According to the classification, grainstone (dolo-grainstone) and well-preserved interparticle porosity are the main depositional texture and pore type in RRT 1 (figure 4(a)). This RRT shows the best reservoir quality, thus the interparticle porosity is the best pore type which makes an interconnected system in the reservoir rock samples and helps the fluid movements. RRT 2 includes grainstone with two different types of porosity. The first type is the grainstone with interparticle and moldic porosities (figure 4(b)). Moldic porosity causes fluid entrapment, but the connectivity of these molds by the interparticle porosity compensates this effect. The second type is grainstone with interparticle porosity which is plugged partially by intergranular cements. This is displayed in figure 4(c), and also shows partially plugged pore throats. Dominant rock fabric and pore type in RRT 3 are grainstone and moldic porosity (figure 4(d)). These grainstone samples are mainly ooidal-peloidal, thus the moldic porosity shows good sorting. The main rock fabric in RRT 4 is grainstone with moldic porosity, but it is progressively filled by different types of cement. The interparticle porosity is also present in some samples but it is also plugged progressively by the cement. The porosity sorting is poor due to the irregular distribution of cements in the pore system (figures 2(e) and (f)) and wide range of particle size. Rock samples in RRT 5 are grainstone with unsorted moldic and vuggy porosities (figure 4(g)). RRT 6 includes non-reservoir rock samples. Mudstone with microporosity that has not been enhanced by dolomitization and dissolution or dolo-mudstone with high cement content, and grainstone with moldic porosity that is mainly plugged, are the rock fabrics and pore types in this RRT (figures 4(h) and (i)). Figure 4. Open in new tabDownload slide The main rock fabrics and pore types in each rock type. RRT 1: (a) dolo-grainstone with interparticle porosity; RRT 2: (b) dolo-grainstone with moldic and interparticle porosity, (c) well sorted dolo-grainstone with interparticle porosity which is partially plugged by dolomite cement; RRT 3: (d) well sorted ooidal grainstone with moldic porosity; RRT 4: (e) dolo-grainstone with interparticle porosity which is mainly plugged by dolomite cement, (f) grainstone with moldic porosity which is progressively filled by calcite and anhydrite cements; RR T5: (g) unsorted grainstone with unsorted moldic porosity; RRT 6: (h) grainstone sample in which interparticle and moldic porosities are totally plugged by calcite and anhydrite cements, (i) dolo-mudstone with patchy anhydrite cement. Figure 4. Open in new tabDownload slide The main rock fabrics and pore types in each rock type. RRT 1: (a) dolo-grainstone with interparticle porosity; RRT 2: (b) dolo-grainstone with moldic and interparticle porosity, (c) well sorted dolo-grainstone with interparticle porosity which is partially plugged by dolomite cement; RRT 3: (d) well sorted ooidal grainstone with moldic porosity; RRT 4: (e) dolo-grainstone with interparticle porosity which is mainly plugged by dolomite cement, (f) grainstone with moldic porosity which is progressively filled by calcite and anhydrite cements; RR T5: (g) unsorted grainstone with unsorted moldic porosity; RRT 6: (h) grainstone sample in which interparticle and moldic porosities are totally plugged by calcite and anhydrite cements, (i) dolo-mudstone with patchy anhydrite cement. This classification shows that the depositional texture and total porosity are not the determinant parameters in the reservoir quality and fluid movements. Diagenetic processes mainly modify the reservoir quality in rock samples with similar depositional textures. The porosity percentage in rock samples with moldic porosity and low reservoir quality (e.g. RRT 3 or RRT 4) and samples with the best reservoir quality (RRT 1) are similar. 6. Relative permeability Relative permeability is an accurate measurement for understanding the dynamic behaviour of the reservoir units which can be used for the rock typing analysis. Saturation history, rock property, wettability, fluid saturation, IFT, temperature, overburden pressure, viscosity of the fluids, the process of handling samples and laboratory conditions all have strong effects on relative permeability and the result of this test. Regardless of other factors impacting relative permeability, the effect of pore geometry, particularly pore types, on hydrocarbon entrapment is considered in this study. The relationship between porosity–permeability and residual oil–gas saturation in sandstone and carbonate reservoirs has been investigated by many authors. Most of these studies considered this subject to find a relationship between residual oil–gas saturation and grain sorting, microporosity and particularly porosity percentage (e.g. Wardlaw and Cassan 1978, Delclaud 1991, Jerauld 1997, Karine et al2001). Obviously finding a relationship between porosity and fluid entrapment in carbonate reservoirs is more complicated than for sandstone reservoirs. The result of the relative permeability tests in the water–oil system was applied in one of the subject wells. Reservoir rock samples in this well are distributed among RRT 1, 3 and 4. The variations in studied curves among the three RRTs show a mainly similar trend. The tendency in relative permeability curves is in agreement with the classification of rock types and the variations in pore structure (figure 5). The residual oil saturation increases from RRT 1 to 4 with a decrease in the pore system and reservoir rock quality and an increase in irregularity of pore types (table 2). According to figure 3, the enormous variation in relative permeability curves cannot be considered only as a result of the variation in the wettability of rock samples, as they are all carbonates and were cleaned before the tests. Figure 5. Open in new tabDownload slide Oil–water capillary pressure curves which are related to RRT 1, 3 and 4. Figure 5. Open in new tabDownload slide Oil–water capillary pressure curves which are related to RRT 1, 3 and 4. Table 2. Air permeability, the permeability of oil at Swir and residual oil saturation of samples A, B and C in RRT 1, 3 and 4. Sample . RRT . Kair (md) . Koil at Swi . Sorw% . A 1 660.5 257.0 30 B 3 88.0 16.8 45.7 C 4 5.5 1.7 50.7 Sample . RRT . Kair (md) . Koil at Swi . Sorw% . A 1 660.5 257.0 30 B 3 88.0 16.8 45.7 C 4 5.5 1.7 50.7 Open in new tab Table 2. Air permeability, the permeability of oil at Swir and residual oil saturation of samples A, B and C in RRT 1, 3 and 4. Sample . RRT . Kair (md) . Koil at Swi . Sorw% . A 1 660.5 257.0 30 B 3 88.0 16.8 45.7 C 4 5.5 1.7 50.7 Sample . RRT . Kair (md) . Koil at Swi . Sorw% . A 1 660.5 257.0 30 B 3 88.0 16.8 45.7 C 4 5.5 1.7 50.7 Open in new tab 7. Discussion In this study, rock typing analysis was performed using different methods to find the crucial parameters controlling the reservoir quality and introduce the best way for classification of reservoir rocks. First, rock typing was carried out by classifying the reservoir rock samples according to their capillary pressure and water saturation using the air-displacing-water capillary pressure curves. This method was used for the purpose of classifying the reservoir rock samples based on their capillary behaviour and fluid saturation. Rock samples were classified into six classes, based on the similarity in swir and the capillary pressure, generally increasing from class 1 to 6. Accordingly, the irreducible water saturation and capillary pressure at different saturations increase with decreasing reservoir quality. Petrophysical properties such as porosity and permeability were compared to the result of the classification, and then RRTs were determined by incorporating the geological properties into defined classes. The variations in the rock properties, such as depositional texture, pore types and the cement content, among classes were studied to determine the effect of these properties on the fluid behaviour of the reservoir rock. Correlation between properties such as porosity and permeability with SCAL defined rock types or fluid behaviour in the reservoir has been considered by many authors (e.g. Hamon 2003, Gomes et al2008, Chekani and Kharrat 2009), but applying pore types in rock typing methods is not common in investigations. According to figure 6, porosity–permeability correlation is very weak in the reservoir. Total porosity has been increased due to the large volume of the moldic and vuggy porosities. However, these types of porosity cause fluid entrapment and reduce the oil and gas recovery. The results show that the initial rock texture could change to different types of reservoir rock with different qualities as a result of diagenetic overprints. For example, grainstone is present in all RRTs; however, the reservoir quality is not similar due to the difference in pore types and the cement content. Each rock type shows distinct fluid saturation and a narrow range of rock properties. Figure 6. Open in new tabDownload slide Porosity–permeability correlation in the studied reservoir. Figure 6. Open in new tabDownload slide Porosity–permeability correlation in the studied reservoir. According to the result of the classification, RRT 1 with the best reservoir quality includes samples with well-preserved interconnected interparticle porosity. From petrographic analysis, an increase in swir and displacement pressure in capillary pressure-defined classes is compatible with decreases in interparticle to moldic and vuggy porosities ratio, porosity sorting, and an increase in the intergranular and pore-filling cements from class 1 to 6. The water saturation and permeability show dual behaviour in RRT 2. The irregular and patchy distribution of the intergranular cement in grainstone samples causes the pore system to be unsorted in RRT 2. The pore throat diameters decrease partially in the rock samples due to the increase in cement content in the interparticle spaces, thus reducing the accessibility to some parts of the pore system. The smaller pore throats cause fluid entrapment and increase the swir and capillary pressure at low water saturations, but the capillary pressure in high water saturations is low and permeability is high due to the effect of the large pore throats. Thus the result of the capillary pressure analysis is not always compatible with the permeability values in complicated carbonate reservoirs. In relative permeability curves, the residual oil saturation increases from RRT 1 to 4 with a decrease in sorting and connectivity of porosity and an increase in irregularity of pores. The trend in relative permeability curves is compatible with the defined rock types. In RRT 1, the interparticle porosity constructs an excellent pathway for the fluid movements. The irreducible water saturation in capillary pressure analysis and the residual oil saturation in relative permeability curves are very low in this rock type. RRT 4, having an unsorted pore system, shows the highest residual oil saturation. RRT 3 and 4, including samples with similar depositional textures (grainstone) and lithology, are different in cement content, porosity sorting, water saturation and dynamic properties. The moldic porosity in RRT 3 is well sorted but RRT 4 includes samples with poorly sorted and progressively plugged moldic porosity. Generally, unsorted and separate molds and vugs (RRT 5) cause high pore to pore throat ratio and make a good shelter for the entrapped fluids and thus reduce the reservoir quality (figure 7). Figure 7. Open in new tabDownload slide Comparison between the hydrocarbon displacement in the excellent interparticle porosity pathways and entrapment in moldic and vuggy porosities. Figure 7. Open in new tabDownload slide Comparison between the hydrocarbon displacement in the excellent interparticle porosity pathways and entrapment in moldic and vuggy porosities. Using wireline logs for the classification of reservoir rock in rock typing analysis is a common performance in reservoir studies. More common logs such as RHOB, NPHI, DT, etc are unable to differentiate between pore types. As different pore types control the quality of the pore system in the rock samples, the wireline logs cannot classify the rock samples properly for rock typing analysis in complicated carbonate reservoirs with various pore types. The result of electrofacies analysis in sandstone reservoirs or a reservoir with interconnected interparticle porosity might be in agreement with the reservoir quality and fluids behaviour. The porous media are simple in these reservoirs, with porosity and permeability showing good correlation. Therefore, parameters that reduce the porosity volume such as matrix and cement are inversely related to the reservoir quality and the porosity percentage directly controls the permeability. In carbonate reservoirs, such as that studied, this compatibility does not exist. In carbonate reservoirs with various types of porosity there are several parameters that control the reservoir quality. This complication in reservoir rock and the lack of correlation between porosity and permeability are related to the complicated pore geometry of rock and this is the result of intense diagenetic processes. Wireline logs differentiate the reservoir rock based on the physical properties and fluid content. The pore system quality and the connectivity of pore throats cannot be studied using this instrument. Thus, electrofacies analysis, which is unable to differentiate the complex pore systems that control the dynamic behaviour of the reservoir rock, is not an appropriate approach for rock typing analysis in such reservoirs. Electrofacies analysis was performed in the reservoir interval in studied wells to examine the applicability of this method in classification of the reservoir rock. The results show that there is no agreement between the electrofacies analysis result and defined rock types. These electrofacies are not compatible with SCAL defined classes or classified pore types. Table 3 shows the result of the classification of rock samples based on the electrofacies analysis. This table indicates the saturation properties and permeability of rock samples in one electrofacies. It is obvious from the table that rock samples with totally different saturation properties are classified in similar electrofacies. According to figure 8, the air–water capillary pressure curves which are representative of the accessibility of the pore system for fluids are distributed in electrofacies illogically. Figure 8 and table 3 indicate that the rock samples with the best fluid flow characteristics and the poorest one are in similar electrofacies. Thus the electrofacies analysis was not successful in classifying the reservoir rock according to the pore system quality. Each electrofacies also includes samples with approximately similar porosity percentage but different pore types. Therefore, this method is not appropriate for rock typing analysis in the studied’ reservoir. Figure 8. Open in new tabDownload slide The air-displacing water capillary pressure curves in defined electrofacies. Figure 8. Open in new tabDownload slide The air-displacing water capillary pressure curves in defined electrofacies. Table 3. The result of electrofacies analysis on studied samples. This table shows the saturation and petrophysical properties of rock samples in electrofacies A. Swir . P (85% saturation) . P (60% saturation) . P (35% saturation) . P (16% saturation) . K (md) . 5.0 4.4    9.8    23.9    55.1    2.3 25.9 1.2    9.6   237.5     735    2.0 27.8 15.7   80.8   279.4   750.7    0.9 84.3 189.3 >2000 >2000 >2000 <0.01 12.1 5.5   39.2   244.7   621.3    0.4 29.9 16.6   41.8   133.0   179.2   11.5 Swir . P (85% saturation) . P (60% saturation) . P (35% saturation) . P (16% saturation) . K (md) . 5.0 4.4    9.8    23.9    55.1    2.3 25.9 1.2    9.6   237.5     735    2.0 27.8 15.7   80.8   279.4   750.7    0.9 84.3 189.3 >2000 >2000 >2000 <0.01 12.1 5.5   39.2   244.7   621.3    0.4 29.9 16.6   41.8   133.0   179.2   11.5 Open in new tab Table 3. The result of electrofacies analysis on studied samples. This table shows the saturation and petrophysical properties of rock samples in electrofacies A. Swir . P (85% saturation) . P (60% saturation) . P (35% saturation) . P (16% saturation) . K (md) . 5.0 4.4    9.8    23.9    55.1    2.3 25.9 1.2    9.6   237.5     735    2.0 27.8 15.7   80.8   279.4   750.7    0.9 84.3 189.3 >2000 >2000 >2000 <0.01 12.1 5.5   39.2   244.7   621.3    0.4 29.9 16.6   41.8   133.0   179.2   11.5 Swir . P (85% saturation) . P (60% saturation) . P (35% saturation) . P (16% saturation) . K (md) . 5.0 4.4    9.8    23.9    55.1    2.3 25.9 1.2    9.6   237.5     735    2.0 27.8 15.7   80.8   279.4   750.7    0.9 84.3 189.3 >2000 >2000 >2000 <0.01 12.1 5.5   39.2   244.7   621.3    0.4 29.9 16.6   41.8   133.0   179.2   11.5 Open in new tab As a result, before rock typing analysis, the controlling parameters in fluid flow and saturation characteristics in the reservoir need to be identified, for example, an integrated petrographic analysis should be performed and the effect of each feature on the reservoir quality needs to be identified. 8. Conclusions Integrated rock typing was carried out using SCAL data (air-displacing-water capillary pressure and oil–water relative permeability curves), geological properties (depositional texture, pore type and cement content) and some petrophysical parameters (porosity and permeability) of the reservoir rock and the main conclusions are the followings: The pore system as the main controlling parameter in static and dynamic properties of reservoir rock needs to be considered more precisely. Accordingly, characterizing the geological properties in reservoir rock is fundamental to reservoir studies because the pore system is the result of interactions between depositional texture and diagenetic overprints. The studied carbonate reservoir is too complicated to be classified easily by routine methods such as porosity–permeability correlation or electrofacies analysis. The porosity–permeability correlations are mostly weak in such reservoirs due to the intense effect of diagenetic processes. Although moldic and vuggy porosities increase the porosity percentage, they do not improve the reservoir quality and also cause fluid entrapment. Pore geometry and pore types are the most appropriate parameters for the classification of reservoir rocks in static rock typing because the pore system conducts fluids by controlling the size and connectivity of the pore throats. Thus the result of the applied methods for reservoir rock characterization needs to be corrected based on the pore system structure and characteristics. Otherwise, porosity percentage or porosity–permeability correlations are not an appropriate instrument for rock typing or reservoir quality control. There is an obvious relationship between pore types and SCAL defined rock types. This relationship shows the importance of the rock properties in fluid flow and saturation behaviour in reservoir rock. Elecrofacies analysis was carried out as a common instrument in rock typing and the conclusion is that electrofacies analysis is not compatible with SCAL defined rock types or variations in classified pore types and thus wireline logs are not appropriate instruments for rock typing in carbonate reservoirs with such complicated pore systems because variations in pore systems do not have an effect on log defined classes. As a result, for utilizing each method in reservoir study and to avoid inconsistency of the results, the most important approach is to determine the main controlling parameters in reservoir quality and dynamic behaviour of reservoir rock because each method is appropriate for some types of reservoir rock with specific characteristics. Blind rock typing without knowing the crucial parameters in reservoir quality leads to defining rock types with a lack of necessary correspondence with production behaviour, which results in the creation of an unrealistic static/dynamic model for a reservoir. References Al-Aswad A A . , 1997 Stratigraphy, sedimentary environment and depositional evolution of the Khuff Formation in south-central Saudi Arabia , J. Phys. G: Nucl. Part. Phys. , vol. 20 (pg. 1 - 20 ) OpenURL Placeholder Text WorldCat Alpin G F , Dawans J M L , Sapru A K . , 2002 New insights from old data: identification of rock types and permeability prediction within a heterogeneous carbonate reservoir using diplog and openhole log data Abu Dhabi Int. Petroleum Exhibition and Conf. Abu Dhabi, United Arab Emirates 13–16 Oct. (Society of Petroleum Engineers) 78501-MS Alsharhan A S , Nairn A E M . , 1997 , Sedimentary Basins and Petroleum Geology of the Middle East The Netherlands Elsevier Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bagheri A M , Biranvand B . , 2006 Characterization of reservoir rock types in a heterogeneous clastic and carbonate reservoir , J. Sci. Technol. , vol. 32 (pg. 29 - 38 ) OpenURL Placeholder Text WorldCat Boada E , Barbato R , Porras J C , Quaglia A . , 2001 Rock typing: key approach for maximizing use of old well log data in mature fields, Santa Rose field, case study SPE Latin American and Caribbean Petroleum Engineering Conf. 25–28 March Buenos Aires, Argentina (Society of Petroleum Engineers) 69459-MS Chehrazi A , Rezaee R , Rahimpour H . , 2011 Pore-facies as a tool for incorporation of small-scale dynamic information in integrated reservoir studies , J. Geophys. Eng. , vol. 8 (pg. 202 - 224 ) 10.1088/1742-2132/8/2/008 Google Scholar Crossref Search ADS WorldCat Crossref Chekani M , Kharat R . , 2009 Reservoir rock typing in a carbonate reservoir SPE/EAGE Reservoir Characterization and Simulation Conf. 19–21 Oct. Abu Dhabi, United Arab Emirates (Society of Petroleum Engineers) 123703-MS Delclaud J . , 1991 Laboratory measurements of the residual gas saturation 2nd European Core Analysis Symp. 20–22 May London (pg. 431 - 451 ) Dunham R J . , 1962 Classification of carbonate rocks according to their depositional texture , Classification of Carbonate Rocks Ham W E . Am. Assoc. Petrol. Geol. , vol. 1 (pg. 108 - 121 ) Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ehrenberg S N , Nadeau P H , Aqrawi A A M . , 2007 A comparison of Khuff and Arab reservoir potential throughout the Middle East , Am. Assoc. Pet. Geol. Bull. , vol. 86 (pg. 1709 - 1732 ) OpenURL Placeholder Text WorldCat Esrafili-Dizaji B , Rahimpour-Bonab H . , 2009 Effects of depositional and diagenetic characteristics on carbonate reservoir quality: a case study from the South Pars gas field in the Persian Gulf , Pet. Geosci. , vol. 15 (pg. 325 - 344 ) 10.1144/1354-079309-817 Google Scholar Crossref Search ADS WorldCat Crossref Gomes J S , Riberio M T , Strohmenger C J , Neghaban S , Kalam M Z . , 2008 Carbonate reservoir rock typing—the link between geology and SCAL Int. Petroleum Exhibition and Conf. 3–6 Nov. Abu Dhabi, United Arab Emirates (Society of Petroleum Engineers) 118284-MS Hamon G . , 2003 Two phase flow rock typing: another perspective SPE Annu. Technical Conf. and Exhibition 5–8 Oct. Denver, CO (Society of Petroleum Engineers) 84035-MS Jerauld G R . , 1997 Prudhoe Bay gas–oil relative permeability Soc. Petrol. Eng. , vol. 12 (pg. 66 - 73 ) 10.2118/35718-PA Karine S , Gérald H , Joël B , Vincent T . , 2001 Residual gas saturation of sample originally at residual water saturation in heterogeneous sandstone reservoirs Int. Symp. Society of Core Analysts Pau, France 21–24 Sept., 2003 , vol. vol 14 Kashfi M S . , 1992 Geology of the Permian ‘supergiant’ gas reservoirs in the greater Persian Gulf area , J. Pet. Geol. , vol. 15 (pg. 465 - 480 ) OpenURL Placeholder Text WorldCat Konyuhov A I , Maleki B . , 2006 The Persian Gulf basin: geological history, sedimentary formations, and petroleum potential , Lithol. Miner. Resources , vol. 41 (pg. 344 - 361 ) 10.1134/S0024490206040055 Google Scholar Crossref Search ADS WorldCat Crossref Leal L , Barbato R , Quaglia A , Porras J C , Lazarde H . , 2001 Bimodal behaviour of mercury injection capillary pressure curve and its relationship to pore geometry, rock quality and production performance in a laminated and heterogeneous reservoir SPE Latin American and Caribbean Petroleum Engineering Conf. 25–28 March Buenos Aires, Argentina (Society of Petroleum Engineers) 69457-MS Lehmann C T , Al-Hosany K , Cobb D O , Al-Hend A . , 2008 Rock typing of upper Jurassic (Arab) Carbonates, offshore Abu 13th Abu Dhabi Int. Petroleum Exhibition and Conf. 3–6 Nov. Abu Dhabi, United Arab Emirates (Society of Petroleum Engineers) 117889-MS Lønøy A . , 2006 Making sense of carbonate pore systems , Am. Assoc. Pet. Geol. , vol. 90 (pg. 1381 - 1405 ) OpenURL Placeholder Text WorldCat Moradpour M , Zamani Z , Moallemi S A . , 2008 Controls on reservoir quality in the lower Triassic Kangan formation, southern Persian Gulf , J. Pet. Geol. , vol. 31 (pg. 367 - 386 ) 10.1111/j.1747-5457.2008.00427.x Google Scholar Crossref Search ADS WorldCat Crossref Porras J C . , 1998 Determination of rock fabrics and pore types from pore throat radius and bulk volume water, and their relations to lithofacies, Carito North field, eastern Venezuela basin SPWLA 39th Annu. Logging Symp. paper OO Porras J C , Campos O . , 2001 Rock typing: a key approach for petrophysical characterization and definition of flow units, Santa Barbara field, Eastern Venezuela basin SPE Latin American and Caribbean Petroleum Engineering Conf. Buenos Aires, Argentina 25–28 March Society of Petroleum Engineers pg. 69458-MS 10.2118/69458-MS Rahimpour-Bonab H . , 2007 A procedure for appraisal of a hydrocarbon reservoir continuity and quantification of its heterogeneity , J. Pet. Sci. Eng. , vol. 58 (pg. 1 - 12 ) 10.1016/j.petrol.2006.11.004 Google Scholar Crossref Search ADS WorldCat Crossref Rahimpour-Bonab H , Asadi-Eskandar A , Sonei R . , 2009 Effects of the Permian–Triassic boundary on reservoir characteristics of the South Pars gas field, Persian Gulf , Geol. J. , vol. 44 (pg. 341 - 364 ) OpenURL Placeholder Text WorldCat Rahimpour-Bonab H , Esrafili-Dizaji B , Tavakoli V . , 2010 Dolomitization and anhydrite precipitation in Permo-Triassic carbonates at the South Pars gasfield, offshore Iran: controls on reservoir quality , J. Pet. Geol. , vol. 33 (pg. 43 - 66 ) 10.1111/j.1747-5457.2010.00463.x Google Scholar Crossref Search ADS WorldCat Crossref Rushing J A , Newsham K E , Blasingame T A . , 2008 Rock typing keys to understanding productivity in tight gas sands SPE Unconventional Reservoirs Conf. 10–12 Feb. Keystone, CO, USA Society of Petroleum Engineers 114164-MS Saboorian Jooybari H , Mowazi G H , Jaberi S R . , 2010 New approach for rock typing in one of Iranian carbonate reservoir (a case study) Int. Oil and Gas Conf. and Exhibition Beijing, China 8–10 June Society of Petroleum Engineers 131915-MS Skalinski M , Gottlib-Zeh S , Moss B . , 2006 Defining and predicting rock types in Carbonates-preliminary results from an integrated approach using core and log data from the Tengiz field , Petrophysics , vol. 47 (pg. 37 - 52 ) OpenURL Placeholder Text WorldCat Skalinski M , Kenter J , Jenkins S , Tankersley T . , 2010 Updated rock type definition and pore type classification of a carbonate buildup, Tengiz field, Republic of Kazakhestan SPE Caspian Carbonates Technology Conf. 8–10 Nov. Atyrau, Kazakhstan Society of Petroleum Engineers 139986-MS Tavakoli V , Rahimpour-Bonab H , Esrafili-Dizaji B . , 2011 Diagenetic controlled reservoir quality of South Pars gas field, an integrated approach , Geoscience , vol. 343 (pg. 55 - 71 ) 10.1016/j.crte.2010.10.004 Google Scholar Crossref Search ADS WorldCat Crossref Wardlaw N C , Cassan J P . , 1978 Estimation of recovery efficiency by visual observation of pore systems in reservoir rocks , Bull. Can. Pet. Geol. , vol. 26 (pg. 572 - 585 ) OpenURL Placeholder Text WorldCat Ye S J , Rabiller P . , 2000 A new way for elecro-facies analysis: multi-resolution graph-based clustering SPWL: Society of Petrophysicists and Well-Log Analysts 41st Annu. Logging Symp. 4–7 June © 2013 Sinopec Geophysical Research Institute
TI - Integration of rock typing methods for carbonate reservoir characterization
JF - Journal of Geophysics and Engineering
DO - 10.1088/1742-2132/10/5/055004
DA - 2013-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/integration-of-rock-typing-methods-for-carbonate-reservoir-g3qoadHUr5
VL - 10
IS - 5
DP - DeepDyve
ER -