Research Article | Open Access
H. Shi, X. R. Luo, G. L. Lei, L. Q. Zhang, L. K. Zhang, Y. H. Lei, "Diagenesis and Fluid Flow Variability of Structural Heterogeneity Units in Tight Sandstone Carrier Beds of Dibei, Eastern Kuqa Depression", Geofluids, vol. 2017, Article ID 6593913, 15 pages, 2017. https://doi.org/10.1155/2017/6593913
Diagenesis and Fluid Flow Variability of Structural Heterogeneity Units in Tight Sandstone Carrier Beds of Dibei, Eastern Kuqa Depression
Tight sand gas plays an important role in the supply of natural gas production. It has significance for predicting sweet spots to recognize the characteristics and forming of heterogeneity in tight sandstone carrier beds. Heterogeneity responsible for spatial structure, such as the combination and distribution of relatively homogeneous rock layers, is basically established by deposition and eodiagenesis that collectively affect the mesogenesis. We have investigated the structural heterogeneity units by petrofacies in tight sandstone carrier beds of Dibei, eastern Kuqa Depression, according to core, logging, and micropetrology. There are four types of main petrofacies, that is, tight compacted, tight carbonate-cemented, gas-bearing, and water-bearing sandstones. The brine-rock-hydrocarbon diagenesis changes of different heterogeneity structural units have been determined according to the pore bitumen, hydrocarbon inclusions, and quantitative grain fluorescence. Ductile grains or eogenetic calcite cements destroy the reservoir quality of tight compacted or tight carbonate-cemented sandstones. Rigid grains can resist mechanical compaction and oil emplacement before gas charging can inhibit diagenesis to preserve reservoir property of other sandstones. We propose that there is an inheritance relationship between the late gas and early oil migration pathways, which implies that the sweet spots develop in the reservoirs that experienced early oil emplacement.
Tight gas sandstone is similar to conventional sandstone gas reservoirs, but with lower permeability, generally less than 0.1 mD, and lower effective porosity, which is historically not economically producible unless the well is stimulated by a large hydraulic fracture treatment or produced by use of horizontal or multilateral wells . Tight sandstones can be developed more easily than shale reservoirs as the rocks generally have more brittle behavior and are easier to complete for production . Although shale gas exploration and development have witnessed a revolutionary breakthrough in the USA recently, more than 50% of natural gas production is from tight sand reservoirs . Since several large tight sand gas fields were found in the Ordos Basin and Sichuan Basin of western China, tight sand gas has also played an important role in the domestic total gas production in China .
Carrier beds, which are a mixture of various permeable bodies under a regional seal, are hydrodynamically connected during oil and gas migration for their macroscopical and physical connectivity which is influenced by the reservoir heterogeneity [5, 6]. Heterogeneity was originally defined as difference or diversity in kind from other things, or consisting of parts or things that are very different from each other . Reservoir heterogeneity can be referred to as spatial distribution of reservoir attributes, such as porosity, permeability, and pore structure, which was mainly caused by depositional process, diagenesis, and tectonism during the buried period . The difference and diversity from rock grain component and spatial fabric of sediments or fractures were proposed to result in reservoir heterogeneity [9, 10]. The grain component includes the mineralogy, shape, size, or sorting. The spatial fabric includes the rhythmicity, continuity, or scale. Fitch et al.  suggested that a formation only with variable mineralogy, grain size, and shape but without spatial organization would appear isotropic in all directions on a large scale. This means that the spatial fabric responding to the combination and distribution of various homogeneous layers is the fundamental origin of heterogeneity in clastic rocks. Therefore, this kind of heterogeneity was termed structural heterogeneity, which ultimately determines the attribute anisotropy of sandstone reservoirs and also causes irregular secondary migration and distribution of hydrocarbon [12, 13]. To comprehend the characteristics and diagenesis history of basic heterogeneity units in the carrier beds is conducive to accurately predict hydrocarbon migration pathways.
In recent years, great progress has been achieved in heterogeneity and its effects on oil migration in conventional and low-permeability sandstone carrier beds [6, 14, 15], but it is still poorly constrained for the structural heterogeneity units and their forming mechanisms in tight gas sand carrier bed. The goals of this study are to (1) recognize the diagenesis variation processes of different structural heterogeneity units, (2) assess migration pathways of the early oil and late gas, and (3) determine the main geological controls on sweet spots which mean sandstone reservoirs with high-yield gas production, taking Dibei gas field in Kuqa Depression as an example.
2. Geological Setting
The Kuqa Depression, trending in a NEE-SWW direction, with an area of 2.7 × 104 km2, is situated on the northern margin of the Tarim Basin, neighboring the southern Tianshan fault-fold zones. It is one of the most important natural gas production plays in northwestern China . The depression is dominated by four structural belts and three sags , including the northern monocline belt, Kelasu structural belt, Yiqikelike belt, Qiulitage structural belt, Yangxia sag, and Wushi sag (Figure 1). Dibei gas field (or named Yinan-2 gas field) is in the east of Yiqikelike fault belt and borders the Qiulitage structural belt to the south and the Kelasu structural belt to the west. Influenced by Yanshan and Himalaya tectonic movements, Yiqikelike fault zone has a typical foreland thrust deformation, which is characterized as a series of north dipping thrust faults . The most prominent fault in this belt is the Yiqikelike thrust fault, which began to form in the late Eocene (approximately 34 Ma) and intensified during the late Miocene and Quaternary (from 10 Ma to 1.64 Ma) . The study area, located in the footwall of Yiqikelike fault, has a slope topography inclining southward. The north of the study area is relatively flat with a dip angle of approximately 10° and the south has a steeper angle of about 15–30°. There is a series of secondary fault noses and fault anticlines in the slope (Figure 1).
The Mesozoic and Cenozoic, which develop completely in Dibei, are comprised of terrestrial clastic rocks, coal beds, and salt layers (Figure 2). The coal deposits are in the Upper Triassic Taliqike Formation (T3t), Lower Jurassic Yangxia Formation (J1y), and Middle Jurassic Kezileinuer Formation (J3kz). From Paleocene Kumugeliemu Period (E1k) to Miocene Jidike Period (N1j), salt layers with gypsum, anhydrite, and dolomite were interbedded with thin mudstone and shale, which indicates a lacustrine and evaporative swamp environment [19, 20]. Natural gas is abundant in Lower Jurassic Ahe Formation (J1a) and partly in the lower portion of Yangxia Formation (J1y). Ahe Formation is dominated by gray sandstones or conglomerates of braided delta. The upward Yangxia Formation consists of dark mudstones intercalated with a small number of sandstones and coal seams are common. The Lower Jurassic consists of cycle upward from braided river delta to shore-shallow lake .
The tight sand gas in Lower Jurassic of Dibei was discovered in 1998. Well YN2 obtained 10.9 × 104 m3 gas production per day during well production testing in the depths of 4578.8 to 4783 m in Ahe Formation and Yangxia Formation . The relative density of gas is low, approximately 0.6283–0.6335, and about 88.6104%–89.4456% of the content is methane . The natural gas is sourced coal gas with a heavier carbon isotope . Zhu et al.  regarded the Dibei gas field as a condensate gas reservoir with normal temperature (approximately 116–135°C) and overpressure (pressure coefficient of about 1.43–1.87 reactive to hydrostatic pressure).
According to the previous studies on petrology of Lower Jurassic reservoirs in Dibei [16, 25, 26], the sandstones are medium to coarse-grained, moderately well to well-sorted arkosic arenites or litharenites (). Lithic grains consist of mostly volcanic and plutonic rock fragments, some metamorphic rock fragments, and minor sedimentary rock fragments. Intermediate-basic effusive rocks dominate the volcanic rock fragments and metamorphic rock fragments including phyllite or damouritized phyllite, slate, and epimetamorphism fine-grained detrital rocks. This kind of rock fragment group is inferred to represent a terrigenous detrital source supply from South Tianshan recycling orogenic belt. Tang et al.  indicated that a stronger alkaline diagenetic environment existed during the mesogenesis stage of Lower Jurassic sand reservoirs for the common and obvious dissolution of quartz or analcime and precipitation carbonate cement. So, the cement of Ahe Formation mainly includes carbonate (i.e., calcite, ferroan calcite, and dolomite), authigenic clays (i.e., kaolinite, smectite, and illite), and quartz.
The Lower Jurassic has ultralow to low porosities (average less than 10%) and low permeabilities ranging from 0.02 mD to 24 mD with an average of less than 1 mD . Previous studies  indicated a number of natural fractures developed in sandstone reservoirs although the tight background and proposed fractures led to some sweet spots. But the fracture intensity of Lower Jurassic sandstones near wells YN4 and YS4 was the highest by the research of tectonic stress field analysis. Nevertheless, those two wells have no gas production but produce a large amount of water instead  (Figure 3). It is obvious that the natural fractures are not adequately reasonable to account for the forming mechanism of sweet spots in Lower Jurassic tight sand reservoirs of Dibei.
3. Sampling and Methods
Two wells (DB102 and YN4, locations in Figures 1 and 3) were selected for petrographic analyses to investigate the heterogeneity characteristics of tight sand carrier beds. The well DB102 has two conventional loggings (gamma-ray and spontaneous potential), a Formation Micro Scanner Image (FMI) logging, and an elemental capture spectroscopy (ECS) logging. The well YN4 only has two conventional loggings (GR and SP). These log data were provided by the Tarim Oil Field Company of PetroChina. Two hundred ninety-four core plugs from the Ahe Formation of the two wells were tested for gas porosity and permeability. Then, these plugs were cut and point counted with 200 counts per thin section to identify the grain and cement components. Basic heterogeneity units can be determined according to the logging, physical property parameters, and petrology data. Three sand samples from different heterogeneity units in well DB102 were selected for mercury intrusion capillary pressure (MICP) to compare the threshold displacement pressure and pore-throat size.
It is difficult to completely investigate all brine-rock interactions in sandstones, because the diagenetic minerals experienced repetitive precipitations and dissolutions with the changes of time, concentration, temperature, pressure, PH, redox, and so on. Compared with diagenetic minerals, it seems relatively easy to determine the activities of organic fluids, because they occurred relatively more shortly in time and less frequently [29, 30]. To identify the phase and age of every charged hydrocarbon is extremely important for clarifying the diagenetic sequences in different heterogeneity units of sandstone carrier beds. So, twenty cores from the Ahe Formation in the study area were cut to fluorescence thin sections to investigate the pore migrabitumen and hydrocarbon inclusions. Fluid microthermometry from Ahe reservoirs in Dibei gas field was measured by Li et al. . The data of homogenization temperatures, which give the minimum temperature of entrapment of a fluid in a mineral, can be used to determine the age of fluid activity integrating with the burial history. Liu et al.  measured quantitative grain fluorescence (QGF) parameters for eighty-six core samples from wells YN2, YN5, YN4, and YS4 (locations in Figure 1) to determine the paleo oil-water contacts. QGF is a technique to test ancient and current hydrocarbon quantity by detecting traces of hydrocarbon compounds trapped as inclusions (QGF) and adsorbed onto grains (QGF on extract, or QGF-E) [33, 34]. QGF index is defined as the average spectral intensity between 375 nm and 475 nm normalized to the spectral intensity at 300 nm and indicates hydrocarbon inclusion abundance. QGF-E intensity corresponds to the maximum spectral intensity normalized to 1 g of quartz sand in 20 mL of DCM solvent, which is used to estimate the concentration of residual hydrocarbon. According to Liu and Eadington , one sand sample with QGF index less than 4 normally is considered to have not been charged with early oil because there are no inclusion hydrocarbons to be detected by fluorescence. Otherwise, the sample is regarded to have experienced early oil filling. Meanwhile, a sandstone sample with low QGF-E intensity (<40 pc) is classified as part of current water leg; otherwise, it is the sandstone absorbed with hydrocarbon onto the grain surface. We used these data to further illuminate the hydrocarbon charge histories of different heterogeneity units.
Furthermore, twelve samples from different heterogeneity units were selected for the scanning electron microscopy (SEM), which was used to clarify diagenesis changes of different heterogeneity units integrated with fluorescence and cast thin sections.
4.1. The Heterogeneity in the Gas Leg and Water Leg
Two types of layers, that is, the gas leg and water leg, could be determined by the formation testing while drilling (FTWD) in the Lower Jurassic of the study area. We have investigated the heterogeneity characteristics in those two types of layers and clarified the reasons for reservoir quality changes according to core, logging, mercury intrusion capillary pressure (MICP), and thin section.
Figure 3 shows that the gas of the Lower Jurassic is distributed in plays near wells YN2 and DB102. The interval of well DB102 in depth from 4931 to 4941 m is determined as a gas leg by FTWD, which comprises two distributary channels from the bottom to the top, respectively, responding to 4936.8–4941 m and 4930–4936.8 m. The bottom of every channel deposited with lag conglomerate and usually eroded the top of the previous channel. The lithology of every distributary channel evolved upward into coarse, medium sandstone and argillaceous siltstone in the end, which represented a typical fining-up cycle (Figure 4).
The gamma-ray (GR) and spontaneous potential (SP) of the underneath distributary channel are widely lower than of the upper distributary channel, suggesting that the reservoirs of the underneath one are more sandy than of the upper one. According to Formation Micro Scanner Image (FMI) and elemental capture spectroscopy (ECS) loggings, the top of every channel is composed of fine-grained deposits, such as silt or mudstone, such as 4930.0–4930.9 m and 4936.0–4936.8 m. Moreover, the sandstones from burial depths between 4938.6 m and 4940.2 m have a high content of carbonate cement (approximately 3.47%–34.92%) and the reservoirs from 4930.9 m to 4935.1 m are rich in muddy content (2.03%–5.41%) compared with others (Figure 4).
The porosity and permeability of fifty-four sand samples in this gas leg, respectively, vary from 0.77% to 6.42% (mean 2.68%) and from 0.016 to 5.107 mD (mean 0.656 mD) (Figures 4 and 5(a)). The cemented and argillaceous sandstone samples all have lower porosities (resp., 0.44%–1.73% and 1.2%–2.91%) and permeabilities (resp., 0.046–1.110 mD and 0.016–0.130 mD) than sandy reservoirs (resp., 1.73%%–6.49% and 0.480–5.110 mD) (Figure 5(a)).
MICP results show that the threshold displacement pressure (approximately 0.2 Mpa) of the gas sandstone (4937.52 m) is obviously lower than of the cemented sandstone (approximately 2.0 Mpa, 4939.52 m) and the argillaceous sample (approximately 8.0 Mpa, 4931.08 m) (Figure 5(b)). Furthermore, the pore-throat size of gas sand sample varies from 4 μm to 8 μm in diameter, but the pore-throat diameters of cemented and argillaceous sandstone mainly range from 0.01 μm to 0.60 μm and 0.005 μm to 0.02 μm (Figure 5(c)).
The sandy sandstones (or gas sandstones) are mainly composed of more rigid grains (i.e., quartz, feldspar, or quartzite fragments) and less cement (Figure 4). The rigid grains can retard compaction and preserve porosity during the buried period . Those reveal that rigid grains (sandy constituents) and carbonate cement are important factors notably influencing the reservoir quality.
The Lower Jurassic water legs in Dibei gas field mainly occur in the north of slope near well YN4, showing that the water legs develop above the gas legs (Figures 1 and 3). This gas/water inversion phenomenon is a typical geologic characteristic of deep basin gas, and Liu et al.  regarded the Dibei gas field as a deep basin gas accumulation. The section in depth from 4450 to 4480 m of well YN4 is identified as a water leg by FTWD and the quantitative grain fluorescence (QGF) results indicated that there was no absorbed hydrocarbon responding to the late charged hydrocarbon on the surface of grains in the section of well YN4 . Therefore, we have chosen this water leg to investigate the reservoir heterogeneity.
Several distributary channels are stacked in the water leg of well YN4 according to the cores and conventional loggings (Figure 6). The lithology of every channel changes from conglomerate in the bottom to silt or mudstone in the top in a fining-up cycle, which is often eroded by the next lag deposits.
Sandy layers saturated with water at those well sites (4451.06–4454.42 m, 4467.72–4474.24 m, and 4476.32–4480 m) with relatively higher reservoir qualities can be determined by core analysis, which indicates that their porosities and permeabilities range from 4.7% to 11.7% and 0.101 mD to 58.53 mD for relatively rich quartz grains and low igneous rock fragments or carbonate cement (Figure 6). Igneous rock fragments, the content of which ranges from 10% to 25% with an average of approximately 15.7%, are abundant in the zones (4454.42–4458.36 m and 4460.80–4467.72 m) that have a poor reservoir quality with 3.1%–8.9% in porosity and 0.02–0.886 mD in permeability. Meanwhile, the reservoirs in 4458.36–4460.80 m and 4474.24–4476.32 m are rich in carbonate cement, with porosities from 3.1% to 8.2% and permeabilities from 0.04 mD to 1.61 mD (Figure 6). Same as in the gas leg, the sandstones with high content of igneous rock fragments or carbonate cement were more impermeable than the ones with abundant rigid framework grains or with little carbonate cement in the water leg.
4.2. Activities of Hydrocarbon Fluids
The fluorescence micropetrology results suggest three types of organic fluids with different occurrences and fluorescent colors in the gas sandstone sample (4935.01 m, well DB102). The first type is solid pyrobitumen with dark-brown to black fluorescence under UV excitation that usually appears in intragranular dissolution pores of feldspar and fragmented grains (Figure 7(a)). The second type is liquid bitumen or inclusion with yellow to orange fluorescence in some intergranular pores, mineral lattices, or quartz microfractures (Figures 7(a)–7(d)). The third type is gas-liquid (or light) bitumen or inclusion with blue to white fluorescence that mainly occurs in quartz microfractures (Figures 7(c) and 7(d)). There are no pore bitumen and hydrocarbon inclusions in the argillaceous (Figure 8(a)) and calcite-cemented (Figure 8(b)) sandstones.
Generally, the fluorescence of oil changes with increasing maturity, regardless of the source rocks, from red to blue due to an increase in the ratio of saturated hydrocarbons to aromatics and API°, which is called blue shift . Based only on the fluorescent color, the maturity of hydrocarbons increased from pyrobitumen to light bitumen. Orange bitumen was often observed coating the black pyrobitumen (Figure 7(a)), and the microfractures filled with orange inclusions were usually cut by the microfractures full with light inclusions (Figure 7(b)). This implied that low mature oil first migrated into some of the intergranular pores and intragranular dissolution pores and converted to solid pyrobitumen and liquid migrabitumen. Some liquid bitumen moved into intergranular pores, mineral lattices, and quartz microfractures as bitumen in pores or inclusions in cement. In a final stage, highly mature oil and gas occupied some of the healed microfractures in quartz grains and the residual pores, where the oil largely transformed into light bitumen or inclusions (Figure 7(c)).
According to previous work [27, 31, 38], two types of hydrocarbon inclusions occur in the Ahe sandstones of Dibei. Some hydrocarbon inclusions, mainly in quartz overgrowths or healed microfractures, show orange-yellow fluorescence under UV light, and the homogenization temperatures of their coeval aqueous inclusions are 85–105°C. This suggests that early oil invaded into the Ahe Formation from the late Suweiyi (approximately 25 Ma) to the Jidike Period (approximately 20 Ma). The other type of hydrocarbon inclusions consists of oil-gas two-phase mixed inclusions with blue-white fluorescence, which mainly appears in microfractures. Their coeval brine inclusions have homogenization temperatures ranging from 115°C to 150°C, which implies a late gas charge between the Kangcun Period (approximately 15 Ma) and Kuqa Period (approximately 5 Ma) (Figure 9).
The QGF index values of hydrocarbon inclusions in the Lower Jurassic sandstones (86 samples) are, respectively, 1.5–142.6 and the QGF-E intensities of the absorbed hydrocarbons are 12.2–29538.5 pc. Figure 10 reveals that 82.6% (71 samples) of all samples were charged by late gas (QGF-E intensity > 40 pc) and approximately 57.7% (41 samples) of those sandstone samples had experienced early oil flowing (QGF index > 4). Eighty-two percent (about 50) of the samples that were saturated by the early oil have QGF-E intensity values more than 40 pc. Furthermore, the sand samples with gas accumulations (outlined by the dashed oval in Figure 10), which have higher QGF extraction intensities (i.e., more than 300 pc), only have medium QGF indices (i.e., 6–30) (Figure 10).
4.3. Diagenesis Difference
Thin sections from the argillaceous layer (4930.90–4935.10 m) of well DB102 suggest that the sandstones generally are rich in ductile grains, such as igneous, phyllite, slate, or fine-grained sedimentary fragments, which are soft to be deformed and result in a large porosity loss on mechanical compaction (Figure 11(a)). The corresponding SEM shows that authigene fibrous illite and siderite are around framework grains (Figure 11(b)). The tight carbonate-cemented sandstones of the Lower Jurassic in the study area have low porosity and most of the framework grains appear point-contacted under the microscope (Figure 11(c)), which illustrates that mechanical compaction is relatively weak and calcite should be cemented in the eodiagenesis stage. Although some grains are coated with a little of illite smectite mixed layer in sheet shape, the pores are almost filled with calcite crystals (Figure 11(d)). Compared with the argillaceous sandstones, sandy reservoirs with gas (Figure 11(e)) or water (Figure 11(f)) are characterized with adequate quartz and feldspar grains and a small quantity of ductile grains, obviously appearing more porous and permeable with residual intergranular and intragranular pores.
The nature of carrier beds is the migration of hydrocarbons or water. We can use the petrofacies that include fluids information as the basic structural units to destruct sand carrier beds. Petrofacies are the reservoir genetic units with a number of petrophysical and percolation properties that are the comprehensive reflection to the deposition, diagenesis, and tectonism . According to the heterogeneity in the gas leg and water leg, we proposed four basic types of petrofacies, respectively, tight compacted (argillaceous), carbonate-cemented, gas-bearing, and water-bearing sandstones in the Lower Jurassic reservoirs of the study area (Figures 4 and 6), which correspond to the four heterogeneity units.
The tight compacted sandstones, usually varying from 2.1% to 5.8% in porosity and from 0.001 mD to 0.130 mD in permeability (Figures 4 and 5), generally deposited in a relatively low-energy sedimentary environment and more than 30% of the framework grains were igneous and sedimentology rock fragments (Figure 6). During the eodiagenesis, mechanical compaction makes these ductile grains rearranged, deformed, and finally be pseudomatrix. Even negligible porosities may be found in ductile-rich sandstones at less than 2000 m [35, 40–43]. This leads to a subtotal loss of primary porosity in tight compacted sandstones and no hydrocarbon fluids flow into those sandstones (Figure 8(a)). Furthermore, igneous rock fragments are easy to evolve into smectite [43, 44], which can further transform to authigenic fibrous illite under the conditions of abundant potassium and aluminum [45, 46] (Figure 11(b)). The illite has wide surface area, pore-bridging texture, and significant associated microporosity and may significantly increase flow-path tortuosity. It can have a severely detrimental effect on reservoir permeability and dramatically increase the irreducible water saturation and capillary pressure [47, 48]. Therefore, the mechanical compaction in eodiagenesis and the illitization of smectite in mesodiagenesis are the main diageneses to completely damage the quality of tight compacted sandstones.
The tight carbonate-cemented sandstones also have extremely low porosity and permeability dominated from 1.0% to 1.9% and from 0.03 mD to 1 mD (Figures 4 and 5). Previous work (i.e., [43, 49, 50]) suggested that the carbonate cement commonly appears at sequence boundaries, flooding surfaces, or sedimentary transformation surfaces in response to the calcite nucleation sites. In eodiagenesis, calcite cement in continental deposits usually might be elongated in the direction of regional groundwater flow as concretions and preferentially cement in the channels and distributary channels with abundant nucleation sites, such as calcrete and mud intraclasts. In addition, the zones rich in mud barriers and interlayers or bioclastics, under or above permeable beds, might produce a large amount of carbonate-cemented layer or calcretes [51–53]. Luo et al.  proposed a hive-like framework model for the Donghe Sandstone reservoir in the Hadexun area of the Tarim Basin, which was compartmentalized by a large number of calcite concretions serving as effective fluid flow barriers. In the Ahe Formation, we also found that the carbonate-cemented sandstones usually locate near the lag deposits of some channels (Figures 4 and 6). In view of the absence of pore bitumen and hydrocarbon inclusions and the point-contacted framework grains in the carbonate-cemented sandstones (Figure 8(b)), we believe that the calcite precipitated in the eodiagenesis stage although the calcite may be cracked by the microfracture during the mesodiagenesis (Figure 11(d)).
The gas-bearing sandstones mainly are composed of rigid, coarse, and fine-sorting grains and muddy matrix is scarce in the pores, which represents the deposits in strong energy environment, such as the lag of distributary channel (Figure 4). The porosity and permeability of gas-bearing sandstones dominated with 3.0%~8.0% and 0.5~10 mD. Although the mechanical compaction created a porosity loss, the presence of rigid framework grains had resistance to the compaction. The dissolution of feldspar and lithic rock fragments contributed a small amount of pores (Figure 11(e)). Tang et al.  proposed that the secondary pores should originate from the fresh water leaching in the eodiagenesis. In light of Figure 10, 82.6% of all samples were charged by the late gas (QGF-E intensity > 40 pc), 57.7% among the gas-bearing sandstone samples had experienced early oil charge (QGF index > 4), and the other 42.3% (QGF index < 4) should be influenced by the microfracture to act as the gas migration pathways. Approximately 82% of the samples that were saturated by the early oil (QGF index < 4) then were charged by the late gas. These suggest that the percolation media for the early oil successively act as pathways for the late gas in the Ahe Formation. Furthermore, the high QGF-E intensities (>300 pc) only occur in the samples charged with the early oil instead of the sandstones that were affected by microfractures, which implies that the microfractures are not the main migration pathways or accumulation media. The early oil emplacement in the sandstone should be proposed as the primary reason for the late gas percolation.
The sandstones saturated with only water are the most common petrofacies after deposition. In eodiagenesis, the sandstones with a large number of ductile grains became tight owing to mechanical compaction and some other sand reservoirs were adequately cemented with calcite displacing formation water. After the early oil emplacement, part of these sandstones became saturated by oil and the others still kept saturated with water. Most of the oil-charged reservoirs were occupied with the late gas and part of water-saturated sandstones were slightly filled by the late gas then (Figure 10). The current water-bearing sandstones that we proposed usually have relatively higher porosity and permeability ranging from 3% to 15% and from 0.2 mD to 10 mD (Figure 6). Tang et al.  observed the dissolution of feldspar and litholic fragment and the kaolinite derived from feldspar in the weakly acidic environment of eodiagenesis stage. Mesodiagenesis calcite, dolomite, albite, and illite were also the common authigene minerals in the water-bearing sandstones. There are more abundant diagenetic reactions in these sandstones in the presence of pore water in theory compared with the other heterogeneity units, but they are not crucial for recognizing the hydrocarbon migration and accumulation in the sandstone carrier beds .
Based on these analyses, we proposed a simple diagenesis sequence model (Figure 12) to illuminate the diagenesis changes of the four basic structural heterogeneity units in the sandstone carrier beds of our study area. The tight units including the compacted and carbonate-cemented sandstones all became impermeable for the early oil and late gas in eodiagenesis stage, respectively, owing to the strong compaction on ductile grains and abundant calcite cement. Early oil emplacement primarily created the gas-bearing sandstones for the late gas migration and accumulation, following with the microfractures for the other gas migration pathways. The water-bearing sandstones were never charged by any hydrocarbon and there were abundant brine-rock reactions.
There are four basic structural heterogeneity units, respectively, the tight compacted, tight carbonated cemented, gas-bearing, and water-bearing sandstones in the sandstone carrier beds of Dibei, eastern Kuqa Depression, due to the investigation of petrofacies by logging, core, and mercury intrusion capillary pressure. Two stages of hydrocarbon charges have been identified according to fluorescence thin section, microthermometry of fluid inclusion, and quantitative grain fluorescence. The early oil emplacement occurred from 25 Ma to 20 Ma, which corresponds to early mesodiagenesis. The late gas charge into the reservoirs began at 15 Ma and stopped at 5 Ma. Early oil emplacement is proposed to be the primary reason for the sand reservoirs to be charged by the late gas. Microfracture also has created about 42.3% migration pathways for gas but the gas saturation in these sandstones is generally low.
The eodiagenetic compaction of rich ductile grains causes the tight compacted sandstones to be impermeable for the early oil and late gas. The tight carbonate-cemented sandstones usually rich in nucleation sites are completely cemented by the eodiagenetic calcite before hydrocarbon charging. Early oil never flowed into the water-bearing sandstones and they are extremely difficult to act as the gas migration pathways. Therefore, the sweet spots for gas production in the Lower Jurassic of Dibei should develop in the sandstones that experienced early petroleum emplacement.
The authors declare that they have no competing interests.
This study was supported by the National Natural Science Foundation of China (41372151, 41102078) and CAS-CNPC Strategic Cooperation Project (RIPED-2015-JS-272).
- S. A. Holditch, “Tight gas sands,” Journal of Petroleum Technology, vol. 58, pp. 86–93, 2006.
- Y. Z. Ma, W. R. Moore, E. Gomez, W. J. Clark, and Y. Zhang, “Tight gas sandstone reservoirs, part 1 overview and lithofacies,” in Unconventional Oil and Gas Resources Handbook, chapter 14, pp. 405–427, Boston Gulf Professional, 2016.
- EIA (Energy Information Administration), Annual Energy Outlook 2012 with projections to 2035 DOEEIA 0383, 2012, http://www.eia.gov/forecasts/aeo/.
- C. N. Zou, Z. Yang, S. Z. Tao et al., “Continuous hydrocarbon accumulation over a large area as a distinguishing characteristic of unconventional petroleum: the Ordos Basin, North-Central China,” Earth-Science Reviews, vol. 126, pp. 358–369, 2013.
- X. Luo, Y. Lei, L. Zhang et al., “Characterization of carrier formation for hydrocarbon migration: concepts and approaches,” Acta Petrolei Sinica, vol. 33, no. 3, pp. 428–436, 2012.
- Y. H. Lei, X. R. Luo, G. Q. Song et al., “Quantitative characterization of connectivity and conductivity of sandstone carriers during secondary petroleum migration, applied to the Third Member of Eocene Shahejie Formation, Dongying Depression, Eastern China,” Marine and Petroleum Geology, vol. 51, pp. 268–285, 2014.
- J. A. Simpson and E. S. Weiner, The Oxford English Dictionary, Clarendon Press Oxford, Oxford, UK, 1989.
- K. J. Weber, “How heterogeneity affects oil recovery,” Reservoir Characterization, vol. 487, p. 544, 1986.
- L. W. Lake and J. L. Jensen, “Review of heterogeneity measures used in reservoir characterization,” Society of Petroleum Engineers of AIME, vol. 3, pp. 409–439, 1989.
- R. Nurmi, M. Charara, M. Waterhouse, and R. Park, “Heterogeneities in carbonate reservoirs: detection and analysis using borehole electrical imagery,” The Geological Society, vol. 48, pp. 95–111, 1990.
- P. J. R. Fitch, M. A. Lovell, S. J. Davies, T. Pritchard, and P. K. Harvey, “An integrated and quantitative approach to petrophysical heterogeneity,” Marine and Petroleum Geology, vol. 63, pp. 82–96, 2015.
- X. Luo, “Simulation and characterization of pathway heterogeneity of secondary hydrocarbon migration,” AAPG Bulletin, vol. 95, no. 6, pp. 881–898, 2011.
- X. Luo, C. Hu, Z. Xiao et al., “Effects of carrier bed heterogeneity on hydrocarbon migration,” Marine and Petroleum Geology, vol. 68, pp. 120–131, 2015.
- J. Zhao, X. Luo, B. Zhang, F. Zhao, and Y. Lei, “A quantitative characterization and effectiveness evaluation on sandstone carrier beds of the silurian kalpintag formation in the tazhong area, tarim basin,” Acta Petrolei Sinica, vol. 32, pp. 949–958, 2011.
- H. Zhao, X.-R. Luo, Z.-Y. Xiao et al., “Interlayer features and petroleum geology meaning of Donghe sandstones in the Hadson oilfield,” Natural Gas Geoscience, vol. 25, no. 6, pp. 824–833, 2014.
- C. Zou, J. Jia, S. Tao, and X. Tao, “Analysis of reservoir forming conditions and prediction of continuous tight gas reservoirs for the deep Jurassic in the eastern Kuqa depression, Tarim basin,” Acta Geologica Sinica (English Edition), vol. 85, no. 5, pp. 1173–1186, 2011.
- J. Chengzao and L. Qiming, “Petroleum geology of Kela-2, the most productive gas field in China,” Marine and Petroleum Geology, vol. 25, no. 4-5, pp. 335–343, 2008.
- J. Cai and X. Lü, “Substratum transverse faults in Kuqa Foreland Basin, northwest China and their significance in petroleum geology,” Journal of Asian Earth Sciences, vol. 107, pp. 72–82, 2015.
- T. Zhang, X. M. Fang, C. H. Song, E. Appel, and Y. D. Wang, “Cenozoic tectonic deformation and uplift of the South Tian Shan: implications from magnetostratigraphy and balanced cross-section restoration of the Kuqa depression,” Tectonophysics, vol. 628, pp. 172–187, 2014.
- Z. Wu, H. Yin, X. Wang, B. Zhao, and D. Jia, “Characteristics and deformation mechanism of salt-related structures in the western Kuqa depression, Tarim basin: insights from scaled sandbox modeling,” Tectonophysics, vol. 612-613, pp. 81–96, 2014.
- X. Tang, S. Yang, and S. Hu, “Thermal and maturation history of jurassic source rocks in the kuqa foreland depression of tarim basin, NW China,” Journal of Asian Earth Sciences, vol. 89, pp. 1–9, 2014.
- X. Li, X. Xiao, Y. Tang et al., “The generation and accumulation of natural gas from Yinan 2 gas pool in Kuqa Depression,” Chinese Science Bulletin, vol. 49, pp. 107–114, 2004.
- Y. Ju, X. W. Sun, L. W. Liu, Y. N. Xie, and H. X. Wei, “Characteristics of Jurassic tight gas reservoir in Dibei areaof Kuqa depression,Tarim Basin,” Xinjiang Petroleum Geology, vol. 35, pp. 264–267, 2014.
- G. Zhu, H. Wang, N. Weng et al., “Geochemistry, origin and accumulation of continental condensate in the ultra-deep-buried Cretaceous sandstone reservoir, Kuqa Depression, Tarim Basin, China,” Marine and Petroleum Geology, vol. 65, pp. 103–113, 2015.
- Z. Li, H. Guo, D. Wang, and W. Lin, “Mesozoic-cenozoic tectonic transition in kuqa depression-tianshan, Northwest China: evidence from sandstone detrital and geochemical records,” Science in China, Series D: Earth Sciences, vol. 48, no. 9, pp. 1387–1402, 2005.
- Y. G. Tang, J. H. Luo, Y. J. Ma, X. Z. Yang, M. L. Ye, and Q. L. He, “Effect of alkaline diagenetic environment of lower jurassic on reservoir property in Kuqa depression,” Xinjiang Petroleum Geology, vol. 32, no. 4, pp. 356–358, 2011.
- H. Guo and R. Zhu, “To research hydrocarbon migration and preservation conditions by fluid inclusion in the Eastern Kuqa depression, Tarim Basin,” Acta Petrologica Sinica, vol. 21, no. 5, pp. 1467–1472, 2005.
- W. Ju, G. T. Hou, S. Y. Huang, and K. X. Ren, “Structural fracture distribution and prediction of the Lower Jurassic Ahe Formation sandston in the Yinan-Tuzi area, Kuqa depression,” Geotectonica et Metallogenia, vol. 37, pp. 592–602, 2013.
- H. Zhao, X.-R. Luo, L.-K. Zhang et al., “A sign to the multi-phases hydrocarbon charge and adjustment: fluid inclusion study from the sangonghe formation in the no. 1 block, the middle junggar basin,” Natural Gas Geoscience, vol. 26, pp. 466–476, 2015.
- C. Z. Hu, L. K. Zhang, X. R. Luo et al., “Diagenesis and porosity evolution of the low-porosity and low-permeability sandstones: evidence from the lower jurassic sangonghe formation in Moxizhuang Area, Central Junggar Basin,” Natural Gas Geoscience, vol. 26, pp. 2254–2266, 2015.
- Z. Li, Z. X. Jiang, and X. Q. Pang, “Genetic types of the tight sandstone gas reservoirs in the Kuqa depression, Tarim Basin, NW China,” Earth Science-Journal of China University of Geosciences, vol. 38, pp. 156–164, 2013.
- J. L. Liu, Z. X. Jiang, and F. Li, “Identification of Paleo-oil-water contact of Jurassic Ahe Formation in Dibei area, Kuqa depression, Tarim Basin,” Xinjiang Petroleum Geology, vol. 36, pp. 153–158, 2015.
- K. Liu and P. Eadington, “Quantitative fluorescence techniques for detecting residual oils and reconstructing hydrocarbon charge history,” Organic Geochemistry, vol. 36, no. 7, pp. 1023–1036, 2005.
- K. Y. Liu, S. C. George, X. Lu, S. Gong, H. Tian, and L. Gui, “Innovative fluorescence spectroscopic techniques for rapidly characterising oil inclusions,” Organic Geochemistry, vol. 72, pp. 34–45, 2014.
- E. D. Pittman and R. E. Larese, “Compaction of lithic sands: experimental results and applications (1),” American Association of Petroleum Geologists Bulletin, vol. 75, no. 8, pp. 1279–1299, 1991.
- Y. K. Liu, G. H. Wu, J. F. Hu, J. N. Wang, D. M. Zheng, and J. G. Wu, “Analyzing the characteristics of Yinan-2 Gas reservoir in Kuche Depression of Tarim Basin,” Natural Gas Industry, vol. 24, pp. 12–18, 2004.
- N. H. Oxtoby, “Comments on: assessing the maturity of oil trapped in fluid inclusions using molecular geochemistry data and visually-determined fluorescence colours,” Applied Geochemistry, vol. 17, no. 10, pp. 1371–1374, 2002.
- F. Li, Z. X. Jiang, and Z. Li, “Enriched mechanism of natural gas of Lower Jurassic in Dibei area, Kuqa depression,” Earth Science-Journal of China Uiversity of Geosciences, vol. 40, no. 9, pp. 1538–1548, 2015.
- D. Tiab and E. C. Donaldson, Petrophysics Theory and Practice of Measuring Reservoir Rock and Fluid Transport Properties, Gulf Professional Publishing, 2011.
- S. N. Ehrenberg, “Relationship between diagenesis and reservoir quality in sandstones of the garn formation, haltenbanken, mid-Norwegian continental shelf (1),” AAPG Bulletin, vol. 74, pp. 1538–1558, 1990.
- S. P. Dutton and T. N. Diggs, “Evolution of porosity and permeability in the Lower retaceous Travis Peak Formation, East Texas,” AAPG Bulletin, vol. 76, no. 2, pp. 252–269, 1992.
- S. T. Paxton, J. O. Szabo, J. M. Ajdukiewicz, and R. E. Klimentidis, “Construction of an intergranular volume compaction curve for evaluating and predicting compaction and porosity loss in rigid-grain sandstone reservoirs,” AAPG Bulletin, vol. 86, no. 12, pp. 2047–2067, 2002.
- S. Morad, K. Al-Ramadan, J. M. Ketzer, and L. F. De Ros, “The impact of diagenesis on the heterogeneity of sandstone reservoirs A review of the role of depositional facies and sequence stratigraphy,” AAPG Bulletin, vol. 94, no. 8, pp. 1267–1309, 2010.
- L. F. De Ros, S. Morad, and I. S. Al-Aasm, “Diagenesis of siliciclastic and volcaniclastic sediments in the Cretaceous and Miocene sequences of the NW African margin (DSDP Leg 47A, Site 397),” Sedimentary Geology, vol. 112, no. 1-2, pp. 137–156, 1997.
- G. Berger, J.-C. Lacharpagne, B. Velde, D. Beaufort, and B. Lanson, “Kinetic constraints on illitization reactions and the effects of organic diagenesis in sandstone/shale sequences,” Applied Geochemistry, vol. 12, no. 1, pp. 23–35, 1997.
- R. H. Worden and S. Morad, Clay Minerals in Sandstones Controls on Formation, Distribution and Evolution, Wiley Online Library, 2003.
- J. D. Kantorowicz, “The influence of variations in illite morphology on the permeability of Middle Jurassic Brent Group sandstones, Cormorant Field, UK North Sea,” Marine and Petroleum Geology, vol. 7, no. 1, pp. 66–74, 1990.
- R. H. Lander and L. M. Bonnell, “A model for fibrous illite nucleation and growth in sandstones,” AAPG Bulletin, vol. 94, no. 8, pp. 1161–1187, 2010.
- S. P. Dutton, “Calcite cement in permian deep-water sandstones, delaware basin, west Texas: origin, distribution, and effect on reservoir properties,” AAPG Bulletin, vol. 92, no. 6, pp. 765–787, 2008.
- S. Morad, J. M. Ketzer, and L. F. De Ros, “Linking diagenesis to sequence stratigraphy an integrated tool for understanding and predicting reservoir quality distribution,” in Linking Diagenesis to Sequence Stratigraphy, S. Morad, J. M. Ketzer, and L. F. De Ros, Eds., John Wiley & Sons, West Sussex, UK, 2013.
- E. F. McBride, M. D. Picard, and R. L. Folk, “Oriented concretions, Ionian Coast, Italy; evidence of groundwater flow direction,” Journal of Sedimentary Research, vol. 64, pp. 535–540, 1994.
- S. M. C. Dos Anjos, L. F. De Ros, R. S. De Souza, C. M. De Assis Silva, and C. L. Sombra, “Depositional and diagenetic controls on the reservoir quality of Lower Cretaceous Pendencia sandstones, Potiguar rift basin, Brazil,” AAPG Bulletin, vol. 84, no. 11, pp. 1719–1742, 2000.
- J. P. Hendry, M. Wilkinson, A. E. Fallick, and N. H. Trewin, “Disseminated ‘jigsaw piece’ dolomite in Upper Jurassic shelf sandstones, Central North Sea: an example of cement growth during bioturbation?” Sedimentology, vol. 47, no. 3, pp. 631–644, 2000.
- X. R. Luo, L. K. Zhang, Y. H. Lei, C. Z. Hu, H. Shi, and B. F. Cao, “Structural heterogeneity of reservoirs and its implication on hydrocarbon accumulation in deep zones,” China Petroleum Exploration, vol. 21, pp. 28–36, 2016.
Copyright © 2017 H. Shi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.