A natural gas hydrate-oil-gas system in the Qilian Mountain permafrost area, northeast of Qinghai-Tibet Plateau

1.   Introduction

Natural gas hydrate or hydrate, is commonly known as “flammable ice”. It is composed of water and light-weighted gas molecules (such as methane, ethane, propane, isobutane, hydrogen sulfide, carbon dioxide, etc.), a kind of crystalline solid substance formed with sufficient gases at low temperature (generally around 273.15 K) and high pressure (generally greater than 3 MPa to 5 MPa) (Sloan ED and Koh CA, 2008). In nature, hydrates are usually distributed in subsurface sediments with a water depth of greater than 300 m (Kvenvolden KA, 1993) or in permafrost with a stratum depth of greater than 130 m below the surface (Shi D and Zheng JW, 1999). In addition to drilled hydrates (Wu NY et al., 2009) even with test-production on the northern slope of the South China Sea (Li JF et al., 2018; Ye JL et al., 2018, 2020), China also collected hydrate samples by drilling for the first time in the Qilian Mountain permafrost, achieving a breakthrough in the investigation of hydrates in permafrost regions (Zhu YH et al., 2009; Lu ZQ et al., 2011). This is the first time in the world that hydrate has been discovered in the mid-latitude permafrost.

Besides hydrates, various oil and gas indications (oil spots, oil immersions, etc.) associated phenomena are common in the Qilian Mountain permafrost, e.g. in drilling holes of DK-1, DK-2, DK-3, DK-7, DK-8, DK-9, DK-12, etc. (Lu ZQ et al., 2018, 2013a, 2013b, 2015a; Cheng B, 2018; Tang SQ, 2015a, 2020) and DK8-19, DK10-16, DK10-17, DK10-18, DK11-14, DK12-13, DK13-11, etc. (Wen HJ et al., 2015; Lu ZQ et al., 2015b; Wang WC et al., 2015; Li YH et al., 2015; Tang SQ et al., 2015b). In particular, in DK-9 borehole, apart from the thick layered hydrates, thick oil-bearing layers were encountered, and in DK-10 borehole an abnormally shallow natural gas layer was also encountered. This is the first time in the South Qilian Basin to obtain the oil and gas bearing layer by drilling (Lu ZQ et al., 2018). On the one hand, this new discovery indicates that the Qilian Mountain permafrost contains a good potential of hydrate resources and has a good prospect for oil and gas exploration, which can provide a new national backup option for hydrate and oil and gas resource security (Wu CG et al., 2011); on the other hand, it also poses a new important scientific question whether there is an inherent genetic link between hydrate and oil-bearing layer, abnormal gas layer or various oil and gas indications in the Qilian Mountain permafrost.

In the past, a detailed analysis of key elements was conducted in the world’s typical marine hydrate accumulations, such as in the Blake Ridge offshore the United States, the Hydrate Ridge offshore Canada, the northern slope of the Gulf of Mexico, and the Storegga landslide offshore Norway and it revealed that they all developed a good combination of hydrocarbon gas-generation series, gas-fluid migration series, and hydrocarbon gas-accumulation series, which together reflected the geological process and geological elements from the formation to preservation of hydrates. The combination constitutes a natural gas hydrate geological system (Lu ZQ et al., 2008). Collett TS et al. (2009, 2011), Max MD and Johnson AH (2014) published articles that clearly put forward the concept of “natural gas hydrate and petroleum system”, and made a keynote report at the 8^th International Gas Hydrate Conference to further clarify that the geological control over hydrate occurrences was referred to as “natural gas hydrate and petroleum system” (Collett TS, 2014). Since then, the natural gas hydrate geological system or natural gas hydrate and petroleum system has gradually gained attention and application in practice, such as the “natural gas hydrate system” in the Krishna-Godavari basin offshore India (Riedel M et al., 2010; Shankar U and Riedel M, 2010; Badesab FP et al., 2017) or “natural gas hydrate petroleum system” (Vedachalam N et al., 2016), “natural gas hydrate system” in the southwest of the Barents Sea (Rajan A et al., 2013; Vadakkepuliyambatta S et al., 2015), “natural gas hydrate system” in the Niger Delta Basin off Nigeria (Akinsanpe OT et al., 2017) or “natural gas hydrate petroleum system” offshore Angola (Nyamapfumba M and McMechan GA, 2012), “natural gas hydrate system” in Black Sea (Hillman JIT et al., 2018), “natural gas hydrate system” offshore Northwest Taiwan in the northern South China Sea (Han WC et al., 2019) or “natural gas hydrate system” (Sha ZB et al., 2015; Liang YX et al., 2013) and its related “seepage system” in Dongsha waters (Wu SG et al., 2010; Zhang W et al., 2018).

In this study, the internal connections between hydrate and oil-bearing layer, shallow gas layer are revealed by the analysis of the sources of hydrate, oil-bearing layer, and shallow gas layer in the Qilian Mountain permafrost under the natural gas hydrate geological system theory. For the first time, different natural gas hydrate-oil and gas subsystems have been established in this area, and the understanding of natural gas hydrate accumulation rules are gradually improved in the Qilian Mountain permafrost, which provides a new theoretical guidance for hydrate and oil and gas investigations in permafrost regions in China.

2.   Geological background

It is dominated by mountainous permafrost in the Qilian Mountain permafrost, covering an area of about 100×10³ km². The permafrost thickness is generally 60–95 m. It is mainly distributed in the middle and western sections of the Qilian Mountain. The lower boundary of the permafrost is roughly the same as the isotherm equivalent of annual average temperature –2 − –2.5 °C (Zhou YW et al., 2000).

The study area is located in Muli Town, Tianjun County, Qinghai Province, in the third mining pit of Juhugeng Coal Mine area in Muli Coal Field, with an altitude of 4026–4128 m. In tectonics, it is in the western section of the Central Qilian Block formed during the Caledonian tectonic movement period (513–386 Ma) (Wen HJ et al., 2006), adjacent to the South Qilian structural belt, belonging to the Muli Depression of the South Qilian Basin. During the Yanshan Movement, the stratum in the study area was strongly uplifted, and the older fault blocks in the flanking strata gradually rose. Its main fault properties changed from tension to compression, resulting in a series of secondary fault structures. Experiencing a series of the tectonic evolution it eventually formed the present-day thrust-nappe structure framework with NWW and NW (Guo JN et al., 2011). As a result of the tectonic movement and evolution, the central part of the study area is composed of an anticline consisting of Triassic, and the north and south sides are individually syncline consisting of Jurassic coal-bearing strata. Along the north and south sides of the anticline-syncline complex, large scaled thrust-nappe faults develop, controlling the boundaries of the existing depression. In the north and south synclines a set of large-scale northeast-trending shear faults develop, which cuts the depression into intermittent blocks with different size. These faults become the natural boundary that divides the mining pits in the Juhugeng coal mine area, making the study area present a structural feature of various north-south zones and east-west blocks (Fig. 1).

Figure  1.  Geological characteristics and borehole locations in the study area.

DownLoad: Full-Size Img PowerPoint

In addition to the Quaternary, drilling results reveal that the exposed strata in the study area also include the Middle Jurassic and part of the Upper Triassic. The Upper Triassic is widely exposed in the north and south of the study area. The lithology is dominated by black mudstone and siltstone with thin coal seams, and is in angular unconformity contact with the overlying Jurassic. The Middle Jurassic can be divided into Muli Formation (J₂m) and Jiangcang Formation (J₂j) from bottom to top. The Muli Formation can be subdivided into upper and lower lithological sections: The lower section is mainly braided-river sediments, dominated by gray-white medium-coarse-grained sandstone, developed with bottom-conglomerate; the upper section is the main coal-bearing section with two sets of exploitable coal layers and locally with thin coal seams, and is mainly gray fine-medium-grained sandstone and dark gray fine-grained to siltstone in lake-marsh environment. The Jiangcang Formation (Fm.) can also be subdivided into two lithological sections: The lower section is composed of dark gray mudstone, siltstone and gray fine sandstone in a delta-lake environment, and contains multiple thin coal seams; the upper section is composed of deposits in shallow lakes and semi-deep lakes, and dark mudstone and gray-black oil shale are developed.

3.   Natural gas hydrate and oil and gas display characteristics

3.1   Natural gas hydrate characteristics

Since 2008, the China Geological Survey has deployed 12 hydrate drilling holes in the study area, including DK-1, DK-2, DK-3, DK-7, DK-8, DK-9, DK-12, etc. Hydrate was encountered in these boreholes. In addition, Shenhua Group also implemented 14 hydrate boreholes in the study area and hydrate was also obtained by drilling in DK8-19, DK11-14, DK12-13, DK13-11 and other boreholes. Among them, in DK-9 borehole, hydrate has the most obvious characteristics and the largest thickness.

The main characteristics of hydrate observed in the field include (Fig. 2): (1) White and milky white aggregates are found on the fracture surface of the hydrate-bearing core; (2) the hydrate-bearing core can be directly burned upon ignition; (3) the temperature of the hydrate-bearing core is measured by infrared thermal imaging camera and then obvious low temperature anomalies are shown; (4) bubbles and water droplets can continuously emerge on the surface of the hydrate-bearing core; (5) the hydrate-bearing core continuously bubbles in the water; (6) abnormal gas is often encountered in the hydrate-bearing section during drilling and when the sample is put into the gas tank, a large amount of gas can be resolved; (7) the honeycomb structure remains on the surface of the hydrate-bearing core after being placed for a certain time; (8) the authigenic carbonate crystals and pyrite particles are associated on the fracture surface of the hydrate-bearing core (Lu ZQ et al., 2010).

Figure  2.  Characteristics of natural gas hydrate collected from drilling holes in the study area.

DownLoad: Full-Size Img PowerPoint

In many boreholes in the study area the drilling results revealed that hydrate mainly occurs at a shallow depth of 400 m or less. It is visible with the naked eyes as a thin white ice-like layer (smoky gray when mixed with mud) in fissures of strata or occurs as a fine disseminated state in pores of rock formations. The reservoir lithology is mainly mudstone, oil shale, siltstone, fine sandstone, etc. On the contrary, hydrate is rarely seen in medium sandstone and coarse sandstone (Lu ZQ et al., 2010; Zhu YH et al., 2010; Wang WC et al., 2015).

3.2   Oil and gas display characteristics

Preliminary statistics showed that oil and gas displays were encountered in 15 hydrate drilling holes such as in DK-1, DK-2, DK-3, DK-4, DK-5, DK-8, DK-9, DK-10, DK8-19, DK10-16, DK10-17, DK10-18, DK11-14, DK12-13, DK13-11 etc. in the study area. In particular, a nearly 8 m thick oil-bearing layer was encountered in DK-9 borehole. The oil-bearing layer is within 362.79–370.58 m with the middle-sized sandstone of the Middle Jurassic. On the surface of the oil-bearing core it can be seen to be grayish brown to light brown, showing a large area of oil immersion or oil staining (Fig. 3). On-site gas logging shows that the content of total hydrocarbons and that of methane have increased significantly, with 2.19%–8.35% and 1.01%–5.62% respectively, while the content of total hydrocarbons and that of methane in non-oil-bearing layers are only 0.10%–0.40% and 0.035%–0.225%.

Figure  3.  Characteristics of oil-bearing cores in DK-9 in the study area.

DownLoad: Full-Size Img PowerPoint

In the meantime, in the hydrate scientific drilling test hole DK-10 in the study area, when drilled to a siltstone interval at a 52.9 m depth, a strong gas eruption occurred. The height of the burning flame exceeded 10 m. When the erupted gas was introduced to the area about 200 m away from the well through a simple pipeline and was ignited, and gases burned violently. The flame height arrived at 3–4 m (Fig. 4). According to on-site estimates, the flow rate of high-pressure abnormal gas encountered during drilling is greater than 4800 m³/d. Due to the abnormally high-pressure gas layer, DK-10 borehole was finished and sealed. The final hole depth is 52.9 m. The main formation lithology is: Quaternary gravel sand at 0–6.1 m, gray siltstone partially mixed with coal at 6.1–10.2 m, coal inter-bedded with silt-bearing mudstone, carbonaceous mudstone at 10.2–37.21 m, and inter-beds of argillaceous siltstone, fine sandstone, silty mudstone at 37.21–52.9 m. According to the characteristics of lithology combination and the stratigraphic comparison among boreholes, apart from Quaternary gravel-bearing sand layer, the strata encountered by the borehole drilling are coal-bearing strata of the upper part of Muli Formation (to the depth of 37.21 m) in the shallower part, and it is the Triassic siltstone and argillaceous siltstone of Galedesi Formation in the lower part.

Figure  4.  Blowout of gases from the abnormal gas layer in DK-10 in the study area.

DownLoad: Full-Size Img PowerPoint

4.   Sources of natural gas hydrate and oil and gas

4.1   Source of natural gas hydrate

Studies have shown that the gas from hydrate itself is mainly light hydrocarbons in the study area, with the characteristics of moisture gas, and its isotopes are characterized by a series of positive carbon isotopes, indicating that the gas of hydrate is of organic origin, and is mainly sourced by pyrolysis, with a small amount of microorganisms. Among them, the pyrolysis sourced gas is mainly related to crude-oil cracked gas and crude-oil associated gas, and a small part is related to condensate-oil associated gas, coal-derived gas, and kerogen cracked gas. The organic geochemical analysis of mudstone, oil shale, and coal in the hydrate layer in the study area shows that the mudstone, oil shale, and coal in the hydrate layer cannot be the main gas source rock for hydrate by the abundance, type, thermal evolution degree and other parameters of organic matter, indicating the source for gas of hydrate may be mainly from deep oil or crude oil-associated gas or mature/over-mature gas from deep gas source rock formations.

With the help of thermal simulation experiment methods, cores such as mudstone, oil shale and coal are select to conduct thermal simulation experiments. Under thermal simulation conditions, the composition and carbon isotope composition of newly produced hydrocarbon gases will be further analyzed. Their gas composition and isotopic characteristics are compared with those of hydrate to explore the source for gas of hydrate. The results show that under the condition of low temperature below 300°C, the gas products are mainly non-hydrocarbon CO₂, and the content of hydrocarbon gases is small. The amount of hydrocarbon gas-produced in mudstone is less than the amount of that produced in oil shale. And the latter is less than that produced in coal. As the thermal simulation temperature increases, the amount of produced hydrocarbon gases increases significantly, reaching the highest at 500°C; on the contrary, the amount of CO₂ gas production does not change much. As the temperature of the thermal simulation increases, the carbon isotopes of the hydrocarbon gas-produced in mudstone, oil shale, and coal show the characteristics of firstly become lighter and then become heavier, indicating a positive carbon isotope sequence like δ¹³C₁<δ¹³C₂<δ¹³C₃.

The hydrocarbon gas-composition, carbon isotope composition produced by thermal simulation are compared with gas composition and isotopic characteristics of gas from hydrate. The results show that the gas composition and carbon isotope composition of the hydrocarbon gas-produced by mudstone at 350–400°C or oil shale at 380–400°C are similar to those of gas from hydrate (Figs. 5, 6). It is speculated that the gas source for hydrate corresponds to the deep mudstone or oil shale according to the equivalent thermal simulation temperature. It has a geochemical relationship. On the contrary, although the hydrocarbon gas-produced in coal is similar in composition to that of gas in hydrate, the isotopic composition of them is quite different. It is deemed that the source for gas of hydrate is not much related to coal (Xue XH et al., 2013; Lu ZQ et al., 2013c). This conclusion is basically consistent with the view of other scholars (Zhai GY et al., 2014).

Figure  5.  Comparison of gas composition between natural gas hydrate and thermal simulation products.

DownLoad: Full-Size Img PowerPoint

Figure  6.  Comparison of carbon isotopes between natural gas hydrate and thermal simulation products.

DownLoad: Full-Size Img PowerPoint

4.2   Source of oil-bearing layer

DK-9 hole is taken as an example to compare conventional source rock with oil source and to compare thermal simulation product with oil source. The Middle Jurassic and Upper Triassic low-maturity source rock samples are selected to conduct thermal simulation experiments at five temperature points of 300°C, 350°C, 390°C, 410°C and 460°C to simulate the process of hydrocarbon gas-generation and expulsion from deep source rocks. The newly-produced hydrocarbon gases are then compared with the oil and gas display by their components to further explore the oil and gas display sources.

Conventional oil source analysis shows that the oil and gas in this area can be divided into two categories. Type I oil and gas may suffer from bio-degradation and have a slightly higher maturity, while Type II oil and gas have a slightly lower maturity (Fig. 7); source rocks are mainly divided into three classes, corresponding to depths of 163.30–207.42 m, 207.42–348.50 m, 357.90–586.50 m. Comparison of conventional oil sources shows that Type I oil and gas are homologous to Type I source rocks, and Type II oil and gas may be mainly homologous to Type II source rocks, and may also be similar to or related to Type III source rocks or deeper sources.

Figure  7.  Characteristics of normal alkanes (a), terpanes (b), and steranes (c) from oil samples in DK-9 in the study area.

Pr–Pristine; Ph–Phytane; Ts–22, 29, 30-Trisnorneohopane-II; Tm–22, 29, 30-Trisnorneohopane.

DownLoad: Full-Size Img PowerPoint

Thermal simulation experiments show that the bio-marker parameters from the Middle Jurassic thermal simulation samples begin to coincide with those of Type II oil and gas from 390°C. It is inferred that the liquid hydrocarbon produced by the Middle Jurassic source rocks above 390°C are equivalent to Type II oil and gas (Fig. 8). At the same time, starting from 410°C, the thermal simulation biomarker parameters of the Upper Triassic samples began to coincide with the parameters of those of Type II oil and gas. It is inferred that the liquid hydrocarbon produced by the Upper Triassic samples above 410°C are equivalent to Type II oil and gas (Fig. 9).

Figure  8.  Contrast of oil and gas displays with thermal simulation products from the Middle Jurassic samples in DK-9 in the study area.

1–Ts/(Ts+Tm); 2–C₃₀RH/C₂₉(H+Ts); 3–C₂₉(H+Ts)/C₃₀H; 4–C₂₇D/C_27-29-St; 5–C₂₇αααR/C₂₇-St; 6–C₂₈αααR/C₂₈-St; 7–C₂₉αααR/C₂₉-St; 8–C₂₉-S/(S+R); 9–C₂₉-ββ/(αα+ββ).

DownLoad: Full-Size Img PowerPoint

Figure  9.  Contrast of oil and gas displays with thermal simulation products from the Upper Triassic samples in DK-9 in the study.

1–Ts/(Ts+Tm); 2–C₃₀RH/C₂₉(H+Ts); 3–C₂₉(H+Ts)/C₃₀H; 4–C₂₇D/C_27-29-St; 5–C₂₇αααR/C₂₇-St; 6–C₂₈αααR/C₂₈-St; 7–C₂₉αααR/C₂₉-St; 8–C₂₉-S/(S+R); 9–C₂₉-ββ/(αα+ββ).

DownLoad: Full-Size Img PowerPoint

Combining conventional oil source comparison, thermal simulation experiments and geological conditions analysis, it is speculated that Type I oil and gas are mainly homologous to Type I source rocks; Type II oil and gas are mainly homologous to Type II source rocks, and there may also be some contribution from Type III source rocks or deeper source rocks. Namely the parent material sources of Type II oil and gas are related to both the Middle Jurassic source rocks and the Upper Triassic source rocks (Lu ZQ et al., 2015a; Tang SQ et al., 2015a).

4.3   Source of shallow gas layer

It was encountered with abnormally high pressure shallow gas at 52.9 m in the DK-10 borehole in the study area. Although in the upper coal-bearing formation, gas logging has a total hydrocarbon and methane value of about 3%, the high-pressure abnormal shallow gas encountered in drilling can be excluded from the influence of coal-bed methane because the surface casing with a diameter of 146 mm is put down till about 41 m. And the possibility of direct decomposition of hydrate can also be excluded within this depth range.

According to on-site mud gas logging, the total hydrocarbon content of the drilling mud gas is about 60% in DK-10 in the study area. The hydrocarbon gas-component is mainly methane, and its content is about 60%. The ethane content is about 0.25%, and the other components are less than 0.01% (Table 1), showing the composition characteristics of pyrolysis gas.

Table  1.  Gas composition recorded by gas logging in DK-10 in the study area (%).

Depth/m	Total HC	CH4	C2H6	C3H8	iC4H10	nC4H10	iC5H12	nC5H12	CO2
50	57.27	55.99	0.25	0.004	0.0004	0	0	0	0.44
51	63.15	62.02	0.25	0.005	0.0006	0.0002	0.0002	0.0001	0.20
52	52.23	51.61	0.22	0.004	0.0007	0.0005	0.0004	0.0004	0.21

 | Show Table

DownLoad: CSV

The collected shallow gas samples were further sent to the laboratory for testing, and the results showed that the gas samples contained lots of nitrogen and oxygen (Table 2), which was caused by the inevitable mixing of a certain amount of air during the collection process. According to the gas composition and isotopic characteristics, the gas from DK-10 borehole seemingly shows the properties of biogenic gas (Fig. 10). Considering that the sample mixed with a certain amount of air may affect the results of hydrocarbon gas-analysis and testing, it is speculated that the abnormally high pressure shallow gas may be due to the microbial transformation from the lower or deep part during the pyrolytic gas migration to the shallow part.

Table  2.  Gas constituents (V-%) and isotopes (V-‰) of shallow gas in DK-10 in the study area.

Sample No.	N2	O2	CO2	Ar	He	CH4	C2H6	C3H8	δ13C-C1 (V-PDB)	δD-C1 (V-SMOW)
DK10-K-01	26.66	2.54	1.43	0.32	0.12	68.81	0.12	0.00	–61.7	–248.6
DK10-K-02	30.89	0.76	2.40	0.36	0.11	65.37	0.11	0.00	–60.1	–244.6
DK10-K-03	18.35	0.51	1.03	0.21	0.14	79.62	0.14	0.00	–61.7	–247.3
DK10-K-04	35.79	6.56	0.94	0.43	0.09	56.09	0.10	0.00	–61.6	–247.4
DK10-K-05	33.17	5.41	0.71	0.39	0.10	60.12	0.10	0.00	–60.6	–248.1
DK10-K-06	17.07	0.77	0.92	0.20	0.14	80.76	0.14	0.00	–61.6	–247.9
DK10-K-07	36.12	1.18	2.74	0.46	0.03	59.47	0.00	0.00	–59.8	–240.0
DK10-K-08	27.08	0.58	2.46	0.36	0.04	69.35	0.14	0.00	–60.5	–245.4
DK10-K-09	30.36	1.08	2.48	0.39	0.04	65.53	0.12	0.00	–60.5	–242.5
DK10-K-10	16.55	0.98	1.37	0.22	0.05	80.68	0.15	0.00	–60.9	–247.7
Notes: Analyzed and tested by the laboratory of Lanzhou Oil and Gas Resources Research Center, Chinese Academy of Sciences.

 | Show Table

DownLoad: CSV

Figure  10. 

This page contains the following errors:

error on line 1 at column 1: Start tag expected, '<' not found

Below is a rendering of the page up to the first error.

I₁‒microbial; I₂‒mixture of microbial and sub-microbial; I₃‒sub-microbial; II₁‒crude oil associated; II₂‒oil-typed cracking; III₁‒mixture of oil-typed cracking and coal-generagted; III₂‒mixture of condensate-oil associated and coal-generated; IV‒coal-generated; V₁‒inorganic; V₂‒mixture of inorganic and coal-generated. A‒biogenic; B‒transtion from microbial to thermal catalyzed; C‒oil associated; D‒condensate-oil associated; E‒coal-typed; F‒marine-phase transitional.

DownLoad: Full-Size Img PowerPoint

5.   Natural gas hydrate-oil and gas system

5.1   Natural gas hydrate and hydrocarbon gas-generation serials

According to the analyses of hydrate and oil and gas sources, although they have very few microbial sources of organic matter, they are mainly sources of pyrolysis of source rocks. Previous studies on the hydrocarbon gas-generation potential of the South Qilian Basin have shown that there are four main sets of potential source rocks in the study area, namely the Middle Jurassic coal-bearing strata, the Upper Triassic mudstone of Galedsi Formation, the Lower Permian limestone of Caodigou Formation and the Carboniferous mudstone or limestone (Fu JH and Zhou LF, 1998, 2000). The four sets of source rocks have relatively high abundance of organic matter. The organic matter types are mainly Type II₂ and Type III, and the degree of organic matter evolution is generally mature or over-mature with some in immature stage. They have good hydrocarbon generation potentials (Hao AS et al., 2016; Tang SQ et al., 2015c; Ren YJ and Ji YL, 2000; Xie QF et al., 2011, 2015; Cheng QS et al., 2016; Zhang JZ et al., 2017; Gong WQ et al., 2013).

Based on analyses of current geothermal field characteristics, and basin thermal history, sedimentary burial history, tectonic subsidence history, source rock maturity evolution history, hydrocarbon gas-generation and expulsion history, etc., the source rocks of the Middle Jurassic and the Upper Triassic Galedesi Formation were simulated in the study area. The results showed that the thermal evolution of the source rocks was controlled by the paleo-temperature field and reached the maximum in the late Middle Jurassic. The Muli and Jiangcang Formations of the Middle Jurassic experienced one stage of hydrocarbon gas-generation and expulsion in the late Middle Jurassic, and the degree of maturity was from immature to mid-mature stage, mainly with oil generation and very little gas production. The Upper Triassic Galedesi Formation experienced two stages of hydrocarbon gas-generation and expulsion, respectively in the Late Triassic and the Late Middle Jurassic. However, the source rock did not experience the peak of hydrocarbon gas-generation in the first stage, and the most of the source rock experienced a peak of hydrocarbon gas-generation in the second stage. The middle and lower parts of source rock reached the gas generation stage, and the gas generation was dominant on the whole (Zuo YH et al., 2016).

5.2   Natural gas hydrate and oil and gas migration serials

According to the regional geological structure evolution data, the NW-SE thrust fault was formed from the late Middle Jurassic to the early Early Cretaceous and is the most important fault in the study area. The spatial mutual cutting relationship of these different faults indicates that the F₁ and F₂ faults were formed at the relatively late stage of faulting in this period. Judging from the current structural framework of the study area, the Himalayan tectonic activities have an inherited influence on the F₁ and F₂ faults in this area. Therefore, the thrust nappe faults of the F₁ and F₂ faults formed in the Yanshanian period in the study area have the characteristics of continuous compression, which can be served as a good blocking and sealing effect on the liquid and gaseous hydrocarbons migrated from the deep part, together with the Middle Jurassic mudstone or shale. These faults are beneficial to the final formation of hydrate and the accumulation of oil and gas. They are the main control faults for hydrate and oil and gas display in the study area.

The drilling data in the study area show that the hydrate distribution is closely related to the F₁ and F₂ faults not only on the plane but also on the borehole profile, and is particularly obviously controlled by the F₂ fault. For example, the hydrate intervals are mainly distributed in the foot wall of the F₂ fault. Hydrate is more obviously controlled by the foot wall of the F₁ and F₂ thrust faults (Lu ZQ et al., 2015c, d).

The analysis and test results of the headspace of borehole cores in the study area also show that methane in the headspace components is often accompanied by heavy hydrocarbon components such as ethane, propane, isobutane, n-butane, isopentane, and n-pentane. The appearance of butane often indicates the leakage and diffusion of deep gas, indicating that hydrocarbon gases are characteristic of deep migration (Han WC et al., 2019; Sha ZB et al., 2015; Liang YX et al., 2013; Wu SG et al., 2010).

The drilling revealed that faults and fracture zones are generally developed in the boreholes in the study area. Many boreholes show that the deep hydrocarbon gases have the characteristics of migrating upward along the faults or fracture zones. In the shallow fracture zones, hydrate and its anomalies (Han WC et al., 2019; Sha ZB et al., 2015; Liang YX et al., 2013; Wu SG et al., 2010) indicate that different fault systems in the study area can provide upward migration channels for hydrocarbon gases in the deep, and shallow faults or fracture zones can also provide space for hydrate to accumulate.

5.3   Natural gas hydrate and hydrocarbon gas-accumulation serials

According to the evolution history of hydrocarbon gases, hydrate accumulation in the study area is mainly limited to the permafrost formation stage. The formation of permafrost in the Qilian Mountains is mainly related to the Mesozoic-Cenozoic tectonic activity and uplift process. Through the study of moraine sediments on the periphery of the study area, it was found that there existed the penultimate or maximum glacial period formed between 0.5 Ma and 0.7 Ma, the bottom third glacial period of 0.13 Ma to 0.30 Ma, and the last glacial period of 10 Ka to 70 Ka. The average annual temperature of the bottom third glacial period was not higher than –7.5− –10°C, which was at least 9.2−11.7°C lower than the present. The main part of the study area entered the cryosphere. The annual average temperature of the penultimate glacial period was not higher than –9.5−–10°C, which was at least 11°C lower than the present (Qi BS et al., 2014). It is inferred that the hydrate formation in the study area was not later than the early Middle Pleistocene.

Combined with geological analysis and previous research, the following hydrate-oil and gas accumulation relationship model can be preliminarily established (Fig. 11): The gas source of hydrate in the study area is dominated by oil-typed pyrolysis gas, and only a small part is mixed with some microbial origin in the shallow part and a small amount of coal-type gas. The gas source of this oil-typed pyrolysis origin is mainly provided by the lower Upper Triassic or deeper source rocks, secondarily followed by the Middle Jurassic mudstone or shale source rocks. When the source rock generates liquid and gaseous hydrocarbons and then migrates to the shallow part, they are directly or indirectly blocked to accumulate by the F₁ and F₂ compressive faults formed in the late Middle Jurassic to the early Early Cretaceous together with mudstone or oil shale. Oil and gas accumulation in the shallow part was partially supplied with microbial hydrocarbon gases or coal-typed gas. Coupled with the island-like permafrost formed no later than the early Middle Pleistocene, it was combined with water in hydrate stability zone to form hydrate. When gases were outside the hydrate stability zone, they would exist in shallower abnormally high pressure gas layer or as free gas or adsorbed gas in formations at different depths. The abnormally high pressure gas layer in the shallow part would be modified by microorganisms and had the characteristics of microbial gas (Lu ZQ et al., 2019).

Figure  11.  Illustration for natural gas hydrate-oil and gas system in the study area.

DownLoad: Full-Size Img PowerPoint

In particular, from the view point of natural gas hydrate-oil and gas system, it can be subdivided into several main natural gas hydrate-oil and gas subsystems, according to the hydrocarbon gas-generation serials in the Middle Jurassic and Upper Triassic source rocks, and hydrocarbon gas-accumulation serials in the study area. The first one is the conventional Upper Triassic gas-bearing sub-system with peak hydrocarbon gas-generation in the late Middle Jurassic. The second one is the conventional Middle Jurassic oil-bearing sub-system with low to mature hydrocarbon gas-generation in the late Middle Jurassic. The third one is the natural gas hydrate sub-system with main gas source from the Upper Triassic gas-bearing sub-system and minor gas source from the conventional Middle Jurassic oil-bearing sub-system as well as little gas source from the Middle Jurassic coal-formed gas and the shallower microbial gas. The fourth one is the shallower gas sub-system with microbial alteration of the main gas source from the Upper Triassic gas-bearing sub-system. Once the gas hydrate-oil and gas system is further applied in the Qilian Mountain permafrost, it will play a more important role in the exploration of natural gas hydrate, oil and gas, even other energy resources in the study area. Accordingly, it is of important theoretical and practical significance.

6.   Conclusions

(i) Natural gas hydrate, oil, and gas have been discovered in the Qilian Mountain permafrost area. They are closely related with each other in temporal, spatial, and genetic terms, and they mutually form a natural gas hydrate-oil-gas system.

(ii) This natural gas hydrate-oil-gas system in the Qilian Mountain permafrost area can be at least subdivided into four major natural gas hydrate-oil and gas sub-systems: A conventional Upper Triassic gas-bearing sub-system with peak hydrocarbon gas-generation in the late Middle Jurassic; a conventional Middle Jurassic oil-bearing sub-system with low to mature hydrocarbon gas-generation in the late Middle Jurassic; a natural gas hydrate sub-system with main gas source from the Upper Triassic gas-bearing sub-system and minor gas source from the Middle Jurassic oil-bearing sub-system as well as little gas source from the Middle Jurassic coal-bed gas and the microbial gas; a shallower gas sub-system with microbial alteration of the main gas source from the Upper Triassic gas-bearing sub-system.

CRediT authorship contribution statement

Zhen-quan Lu and Gang-yi Zhai conceived of the presented idea. Shi-qi Tang and Dong-wen Fan carried out the experiments. Zhen-quan Lu and Shi-qi Tang wrote the manuscript with support from Xiao-ling Luo. Hua Liu and You-hai Zhu helped supervise the project. Ting Wang contributed to the interpretation of the results. Shi-qi Tang and Rui Xiao designed the figures. All authors provided critical feedback and helped shape the research, analysis and manuscript.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgment

This work was supported by the projects of China Geological Survey (DD20160223, DD20190102).