Investigation of Organic Matter Sources and Depositional Environment Changes for Terrestrial Shale Succession from the Yuka Depression: Implications from Organic Geochemistry and Petrological Analyses

0. INTRODUCTION

The Jurassic was a typical period of greenhouse climate (Bruce and Paul, 2008; Berner, 2003, 1994), with high sea levels and increased humidity in the atmosphere (Chivelet et al., 2019; Wierzbowski and Joachimski, 2007). Many lacustrine basins (e.g., Qaidam Basin, Turan Basin, Ordos Basin, and Junggar Basin in China) were formed in the Eurasia continent during the Jurassic, where organic matter (OM) rich sediments were deposited and may sever as excellent petroleum source rocks because of higher bioproductivity (Wang et al., 2022; Xiang et al., 2022; Gong et al., 2021; Li et al., 2021; Zhu et al., 2021; Zou et al., 2020; Guo et al., 2018; Sachsenhofer et al., 2003). The Qaidam Basin, which develops an integrated stratum of Jurassic terrestrial deposits with coal-bearing strata (Yang et al., 2006), as well as a typical lacustrine oil-producing basin northwest China (Shao et al., 2014; Liu et al., 2013), is a fine research area to investigate the evolution of the Middle Jurassic terrestrial biomes, paleoenvironment, and paleoclimate.

Paleoenvironment (e.g., temperature, salinity, and humidity) changes have a significant influence on paleofloral evolution and distribution (Chojnacka and Mikulewicz, 2018; Stefanova et al., 2013; Bauer et al., 1985), and diverse sedimentary facies are recorded and represented by different paleofloral communities. For example, green algae and Botryococcus are generally as indicators of the sediment of lacustrine-brackish-saline (Summons et al., 2002; Mckirdy et al., 1986). Higher plant-derived biomarkers (e.g., oleananes and retene) are commonly representative of terrigenous setting (Chattopadhyay and Dutta, 2014; Regnery et al., 2013; Moldowan et al., 1994; Woodhouse et al., 1992). Stable carbon isotopes and biomarkers have been widely utilized for palaeoflora and paleoclimate researches in terrestrial sediments (Liu et al., 2019; Wood and Hazra, 2017; Wang et al., 2015; Bechtel et al., 2008; Peters and Moldowan, 1993). Various vegetation would be contributed to different biomarkers recorded in sedimentary rocks (Hautevelle et al., 2006; Peters et al., 2005). Usually, gymnosperms produce large amounts of diterpenoids (e.g., abietane, pimarane, and retene), whereas angiosperms contribute largely to triterpenoids, such as oleananes, ursanes, and lupanes (Jiang et al., 2020; Otto and Wilde, 2001). Meanwhile, biomarker molecules are widely used for maturity assessment, paleo-sedimentary environment reconstruction, and for evaluation of source orga-nisms (Regnery et al., 2013; Peters et al., 2005). For instance, C₃₁ 22S/(22S+22R) hopanes and C₂₉ sterane isomerization ratios have been extensively applied to identify thermal maturity of sediments and crude oils (Peters et al., 2005; Seifert and Moldowan, 1986). The presence of mid-chain methylalkanes and 2α-methylhopanes in geological samples are generally associated with cyanobacteria (Summons et al., 1999). Jiang et al. (2020) proposed kinds of angiosperm/gymnosperm indices based on source-related aliphatic and aromatic compounds to reconstruct the Cretaceous–Eocene palaeoclimate of the Gipps- land Basin. Stable carbon isotopes of OM (δ¹³C_org) are commonly used to reconstruct sedimentary environment changes, make understand sediment organofacies (e.g., marine or lacustrine shales, coals) and characterize OM in source rocks (Lu et al., 2021, 2020a; Milkov et al., 2020; Bechtel et al., 2018). For instance, Lu et al. (2016) explained paleoclimate and depositional environment changes of the sediments from Middle Jurassic in the Qaidam Basin using the values of δ¹³C_org, and the changes of source of OM and sedimentary environment of the Lower Miocene oil shale in the Aleksinac Basin have been interpreted by changes in δ¹³C_org (Bechtel et al., 2018).

Previous studies have been primarily focused on the structural evolution, stratum division, shale gas potential, petroleum exploration potential, shale reservoirs characteristics, and kinetic mechanisms for petroleum generation in the Yuka Depression (Shao et al., 2014; Liu et al., 2013; Yang et al., 2006). How-ever, limited work about biomarkers and stable carbon isotopes in the sedimentary rocks from the Yuka Depression have been carried out (Hou et al., 2017; Liu et al., 2008). Lu et al. (2016) reported that the palaeoclimate of the 7th Member of Dameigou Formation (J₂d⁷) has obvious dry-wet fluctuations based on isotopic geochemistry evidence. Partial biomarkers (e.g., pristane, phytane, gammacerane index, and C₂₇–C₂₉ regular ste-ranes) were used to identify OM sources, types, and depositional conditions of the J₂d⁷ in the Yuka Depression (Qin et al., 2018). So far, detailed biomarkers and stable carbon isotope compositions of sediments from the J₂d⁷ have not been investigated in the Yuka Depression.

In this study, bulk organic geochemical parameters, biomarkers, organic petrology, and stable carbon isotopes were determined to reconstruct the sedimentary history of source rocks from the Middle Jurassic Yuka Depression, Qaidam Basin. The levels of paleowater column stratification, water salinity, redox conditions, and types of sources are also discussed to evaluate the relationships among palaeoenvironment conditions, OM input, and the sedimentation of source rocks from the Yuka Depression, providing more detailed evidence for the paleoenvironmental history of the Qaidam Basin.

1. GEOLOGICAL SETTING

The Qaidam Basin is the prolific oil and gas resource basin, which locates in the northwest China and covers an area of approximately 3.4 × 10⁴ km² (Fig. 1a). The basement of the Qai-dam Basin is the Proterozoic and Ordovician metamorphic series. The Jurassic continental sediments within the basin are extensively developed, which contain several high-quality source rocks sets, especially in the Middle Jurassic Dameigou Formation. The general stratigraphy of the Qaidam Basin within the Dameigou Formation in the Middle Jurassic (Fig. 2) includes the 4th Member (predominantly-fluvial deposition), the 5th Member (coal-bearing lacustrine and delta plain clastic deposition), the 6th Member (coal-bearing delta plain and fluvial deposition), and the 7th Member (lacustrine dominated shales and mudstone) (Yuan, 2007). In the northern Qaidam Basin, most exploration of petroleum, coal and related development activities occur in the Yuka Depression (Qin et al., 2018; Li et al., 2016; Ren et al., 2016; Liu et al., 2013).

Figure  1.  (a) Location and geological background of the Qaidam Basin; (b) sample location in the ZK6-1 well and thickness, Ro and TOC isogram of source rocks in the J₂d⁷ from the Yuka Depression, Qaidam Basin (modified from Liu et al., 2020).

DownLoad: Full-Size Img PowerPoint

Figure  2.  Stratigraphy, lithology and depositional facies of the Yuka Depression. Where the green stars in the profile represent sample locations (modified from Shao et al., 2016).

DownLoad: Full-Size Img PowerPoint

The Yuka Depression covers an area of approximately 430 km² with a NW direction, which located about 20 km northwest of Dachaidan Town (Fig. 1b). The Depression is filled by the complete Middle Jurassic sediments (Fig. 2), which has been known for more than 70 y due to coal mining and petroleum exploration activities. The Yuka Depression has a good prospect for coal, shale oil, and gas exploration and development (Hou et al., 2017; Shao et al., 2016; Liu et al., 2013), especially in the J₂d⁷, where there is a set of medium to excellent source rocks. Organic petrological research suggested that the J₂d⁷ is characterized by marginally mature to mature lacustrine sediments with high total organic carbon (TOC = 2%–10%), and vast shale oil and gas resource potential (Qin et al., 2018; Hou et al., 2017). Li et al. (2016) reported that sediments in the Yuka Depression with a depth lower than 1 980 m are favorable for shale oil exploration, whereas, in a depth higher than 1 980 m, sediments are favorable for shale gas exploration. Based on extant palaeoenvironment interpretation from sedimentological studies (e.g., Shao et al. (2014) and reference therein), the Yuka Depression experienced a change from semi-deep to deep lacustrine deposition during J₂d⁷. The sporopollen assemblage of J₂d⁷ is mainly characterized by Coniopteris and Phoenicopsis (Yang et al., 2006). The contents of gymnosperms, ferns and gingko are relatively high, indicating a warm and humid environment in the J₂d⁷ (Zhong et al., 2003).

2. SAMPLES AND METHODS

2.1 Samples

In this study, 33 core samples were collected from ZK6-1 well in the northwest part Yuka Depression, belonging to the interval of J₂d⁷. The depth of these samples varies from 1 040–1 203 m, and the stratigraphic dip is 35°. The samples represent a core interval 1–3 m long with composite samples, which corresponds roughly to 0.8–2.5 m of true thickness. The lithologies of studied samples include shale, mudstone, carbonaceous-mudstone, and siltstone. Note that mudstone and carbonaceous-mudstone are regarded as mudstone in this study. The TOC content, rock pyrolysis, organic petrology, and stable carbon isotope of all samples were determined and 19 samples were extracted for biomarker analysis. Organic geochemical analyses were conducted in the Organic Geochemistry Lab at Yangtze University.

2.2 Methods

2.2.1   Bulk organic analysis

All core samples were dealed with distilled water to remove surface impurities, and then dried at 60 ℃ till completely dry and crushed into fine powder (< 200 mesh). A representative amount (150 mg) of each sample was prepared for measurement of TOC content, by treating with 50% phosphoric acid. The TOC content was measured using an EA2000 carbon sulfur analyzer according to (PGEPSC, 2003), and the measurement error was lower than ±3%. Around 50 mg of each powdered sample was examined using a Rock-Eval VI instrument following the (PGEPSC, 2002) to obtain the parameters of S₁ (mg HC/g rock), S₂ (mg HC/g rock), and T_max (℃).

2.2.2   Organic petrologic analysis

The organic petrology was determined only on available samples, including vitrinite reflectance (Ro, %) and organic macerals. Polished samples were observed by reflected light and fluorescent light under a Leica DM2500P microscope with an oil immersion objective (50× magnification). The quantification of maceral components were conducted at least 300 valid points for each sample, and Ro value was detected on vitrinite with greater than 40 valid points for each sample.

2.2.3   Stable carbon isotope analysis

All 33 pulverized samples were conducted with hydrochloric acid (HCI) to remove inorganic carbon for carbon isotope analyses of TOC (δ¹³C_org). The value of δ¹³C_org was determined using a Finnigan Mat Delta Plus mass spectrometer interfaced with an HP 5890II chromatograph following the (PGEPSC, 2008), and the results are reported in ‰ relative to the Vienna Peedee Belemnite (V-PDB) standard with better than ±0.5‰ in analytical precision. International standard samples (IAEA-CH-7 and USGS24) were analyzed before and after sample analysis for calibration.

2.2.4   Biomarker analysis

Based on the lithology, TOC content, Rock Eval data (choosing from higher values in the same lithology) of the samples, 19 representative samples are selected for biomarker analysis. The extractable organic matters (EOM) of these samples were obtained by Soxhlet extraction for 72 h with a ratio 9 : 1 (v/v) of dichloromethane (DCM)/methanol (MeOH) solution. Then the EOM was separated into aliphatic, aromatic, and polar fractions via activated alumina/silica gel column chromatography by using n-hexane, n-hexane/DCM (2 : 1, v/v), and DCM/MeOH (1 : 1, v/v), respectively.

Aliphatic and aromatic fractions were analyzed by gas chromatography-mass spectrometry (GC-MS) using an Agilent gas chromatograph (7890B) coupled to an Agilent mass selective detector (5977B), equipped with an HP-5MS fused silica capillary column (30 m × 250 μm i.d.; 0.25 μm film thicknesses) and a flame ionization detector (FID). The GC temperature was initially held at 50 ℃ for 1 min, followed by a programmed increase to 100 ℃ at a rate of 20 ℃ /min, and then up to 315 ℃ (held for 16 min) at a rate of 3 ℃/min for detection. Helium was served as the carrier gas at a flow rate of 1.0 mL/min with a constant flow. For the analysis of biomarkers, the MS scan range was m/z 50 to 650. Isoprenoids and n-alkanes (m/z 85), terpanes (m/z 123 and m/z 191), hopanoids (m/z 191), and steroids (m/z 217) were recorded for biomarker analysis. The identification of biomarkers was performed by comparison with known retention times relative to the North Sea Oil standard (NSO-1; Weiss et al., 2000, library), and published data (Philp, 1985).

3. RESULTS

3.1 Bulk Geochemical Characteristics

The bulk geochemical results and stable carbon isotope composition are exhibited in Table 1. The TOC contents of all studied samples from the ZK6-1 well in the Yuka Depression range from 0.21% to 9.0% with an average of 3.0% (Table 1). Mudstones and shales generally have higher TOC values compared with siltstone samples in this study (Table 1). The measured vitrinite reflectance is detected only on available samples, ranging from 0.53% to 0.70% with a mean value of 0.66% (Table 1). T_max values of all samples do not show obvious variation in the whole well, ranging from 424 to 464 ℃ with a mean value of 437 ℃ (Table 1, Fig. 3). The highest T_max values in the samples YA-19 and YA-24 (464 and 461 ℃; Table 1) probably are the result of an unobvious S₂ peak. The hydrogen index (HI) varies from 20–408 mg HC/g TOC, with an average of 169 mg HC/g TOC (Table 1).

Table  1.  Bulk geochemical characteristics and stable carbon isotope composition of source rocks from the Yuka Depression, Qaidam Basin

Sample No.	Depth (m)	Lithology	EOM	TOC (%)	Ro (%)	Tmax (℃)	S1	S2	HI	δ13Corg (‰)	Liptinite (%)	Vitrinite (%)	Inertinite (%)	Saprope- linite (%)
YA-1	1 040.0–1 041.5	Shale	2.3	2.9	0.69	435	0.99	11.78	402	-24.38	58	7	1	34
YA-2	1 041.5–1 043.0	Shale	3.8	4.4	0.70	431	1.21	18.03	408	-24.49	66	12	0	22
YA-3	1 043.0–1 044.8	Siltstone	2.1	0.78	0.68	432	0.13	0.65	83	-24.32	54	29	6	11
YA-4	1 044.8–1 047.5	Siltstone	3.3	0.95	0.68	431	0.09	0.42	44	-23.74	58	26	8	8
YA-5	1 048.0–1 049.6	Mudstone	0.75	4.8	0.63	431	0.15	6.18	129	-23.61	66	28	6	0
YA-6	1 049.6–1 050.4	Mudstone	0.80	8.3	0.66	425	0.31	16.40	198	-28.63	78	15	2	5
YA-7	1 053.2–1 055.2	Mudstone	0.46	5.7	0.67	435	0.24	7.66	134	-24.47	58	19	0	23
YA-8	1 055.2–1 058.2	Siltstone	0.10	0.76	0.66	438	0.03	0.39	51	-24.12	49	35	2	14
YA-9	1 058.2–1 060.0	Siltstone	0.06	0.58	n.d.	445	0.02	0.13	23	-23.89	n.d.	n.d.	n.d.	n.d.
YA-11	1 060.0–1 065.0	Mudstone	0.54	3.0	0.61	435	0.15	3.24	108	-27.86	61	12	0	27
YA-12	1 065.0–1 068.0	Mudstone	0.09	0.63	0.61	436	0.03	0.26	42	-31.27	69	10	1	20
YA-13	1 072.3–1 074.3	Mudstone	0.50	1.9	0.64	437	0.09	4.93	267	-31.32	77	11	0	12
YA-14	1 074.3–1 076.4	Mudstone	0.18	0.90	n.d.	438	0.04	2.32	257	-28.65	n.d.	n.d.	n.d.	n.d.
YA-15	1 082.3–1 085.8	Mudstone	0.58	3.4	0.67	433	0.13	9.91	290	-29.12	46	18	0	36
YA-16	1 085.8–1 092.9	Mudstone	0.41	3.0	0.65	437	0.08	2.74	92	-25.12	54	26	0	20
YA-17	1 092.9–1 093.5	Siltstone	0.07	0.48	n.d.	438	0.02	0.12	25	-23.45	n.d.	n.d.	n.d.	n.d.
YA-18	1 093.5–1 099.5	Mudstone	0.33	2.0	0.67	439	0.06	1.73	89	-24.56	52	21	0	27
YA-19	1 115.9–1 126.1	Siltstone	0.04	0.21	n.d.	464	0.01	0.05	26	-23.72	n.d.	n.d.	n.d.	n.d.
YA-20	1 133.3–1 134.5	Siltstone	0.08	0.68	n.d.	438	0.03	0.19	27	-24.78	n.d.	n.d.	n.d.	n.d.
YA-21	1 137.0–1 138.5	Mudstone	0.09	0.58	0.66	440	0.03	0.35	61	-31.38	77	7	1	15
YA-22	1 138.5–1 140.0	Mudstone	0.33	1.9	0.67	436	0.07	2.08	109	-28.94	72	13	4	11
YA-23	1 140.0–1 148.0	Mudstone	0.17	1.3	0.68	437	0.04	0.88	68	-23.68	19	62	19	0
YA-24	1 148.0–1 153.7	Siltstone	0.05	0.23	n.d.	461	0.01	0.05	20	-24.16	n.d.	n.d.	n.d.	n.d.
YA-25	1 155.5–1 156.6	Siltstone	0.09	0.55	n.d.	435	0.02	0.26	47	-24.68	n.d.	n.d.	n.d.	n.d.
YA-26	1 156.6–1 159.6	Siltstone	0.08	0.37	n.d.	448	0.02	0.11	30	-23.75	n.d.	n.d.	n.d.	n.d.
YA-27	1 159.6–1 164.0	Mudstone	0.51	1.6	0.66	437	0.12	1.62	100	-26.54	65	18	1	16
YA-28	1 166.0–1 168.0	Mudstone	1.4	4.4	0.65	428	0.29	11.53	260	-24.12	67	21	0	12
YA-29	1 168.0–1 172.8	Mudstone	0.11	0.58	n.d.	444	0.03	0.19	33	-26.76	n.d.	n.d.	n.d.	n.d.
YA-30	1 186.9–1 189.7	Mudstone	1.2	9.0	0.68	427	0.40	14.73	164	-23.39	43	45	12	0
YA-31	1 189.7–1 191.7	Mudstone	0.31	2.8	0.66	436	0.09	3.24	114	-23.07	50	44	6	0
YA-32	1 191.7–1 192.8	Mudstone	0.89	2.9	0.64	428	0.18	6.25	215	-24.56	64	32	4	0
YA-33	1 192.8–1 194.0	Mudstone	0.90	2.9	0.53	424	0.20	4.62	159	-24.12	76	19	5	0
YA-34	1 198.5–1 203.0	Mudstone	0.50	2.8	0.67	430	0.24	6.33	224	-23.54	52	38	2	8
EOM: Extractable organic matter yield; TOC. total organic carbon content; Ro. vitrinite reflectance; HI. hydrogen index (S2/TOC × 100); n.d. not detected. EOM, S1, S2, HI in mg HC/g TOC)

 | Show Table

DownLoad: CSV

Figure  3.  Depth profiles of TOC, T_max, HI, and δ¹³C_org of source rocks from the Yuka Depression, Qaidam Basin. For abbreviations refer to Table 1.

DownLoad: Full-Size Img PowerPoint

The extractable organic matters (EOM) of the studied samples have a range of 0.04–3.8 mg HC/g rock (Table 1). The group components were varied in samples of the well. Overall, saturated hydrocarbons are generally higher than aromatic hydrocarbons for samples from the upper part of ZK6-1 well with saturated hydrocarbons/aromatic hydrocarbons mostly > 1.0 (Table 2). The NSO compounds and asphaltenes of samples show obvious variation with the range of 3.1%–50.7% and 0.3%–44.5%, respectively (Table 2).

Table  2.  Organic characteristics and parameters calculated from n-alkanes and isoprenoids of source rocks from the Yuka Depression, Qaidam Basin

Sample No.	Sat.	Aro.	NSO	Asphalt.	Sat./Aro.	Cmax	%Short-chain n-alkanes	%Mid-chain n-alkanes	%Long-chain n-alkanes	Pr/Ph	Pr/ n-C17	Ph/ n-C18	GI	CPI20-30	OEP	Wax index
	(wt.%)
YA-1	48.3	20.7	13.4	8.8	2.3	C21	42	39	19	1.89	0.24	0.12	0.16	1.17	1.17	3.55
YA-2	46.7	17.0	18.1	9.4	2.7	C16	71	21	8	2.76	0.38	0.19	0.04	1.32	1.29	4.80
YA-3	69.8	16.0	9.0	0.8	4.4	C23	34	43	23	1.92	0.31	0.14	0.24	1.17	1.15	2.88
YA-4	68.2	14.9	4.5	0.3	4.6	C23	32	45	23	1.85	0.34	0.15	0.20	1.18	1.18	2.99
YA-5	42.5	18.1	17.0	14.0	2.3	C16	80	15	5	3.79	0.37	0.13	0.03	1.41	1.32	5.62
YA-6	18.5	23.3	22.2	27.5	0.8	C23	25	46	29	3.96	2.01	0.43	0.01	1.45	1.34	1.65
YA-7	20.8	21.4	17.2	33.8	1.0	C23	32	41	27	3.02	0.87	0.26	0.02	1.39	1.28	1.62
YA-12	24.8	18.6	26.5	26.5	1.3	C23	21	53	26	0.69	0.44	0.42	0.09	1.27	1.23	2.53
YA-13	54.5	20.1	11.0	13.0	2.7	C23	29	45	26	1.43	0.80	0.48	0.04	1.50	1.34	2.30
YA-14	62.1	23.3	6.8	3.4	2.7	C25	13	55	32	1.11	0.24	0.13	0.02	2.20	1.92	1.57
YA-15	23.8	20.9	21.4	28.9	1.1	C25	17	55	28	2.83	0.70	0.17	0.02	1.80	1.55	1.95
YA-21	22.0	51.6	3.3	13.2	0.4	C25	16	45	39	0.72	0.51	0.48	0.03	1.32	1.16	1.17
YA-22	16.2	21.3	28.7	25.9	0.8	C25	30	40	30	2.86	0.93	0.34	0.01	1.54	1.29	1.29
YA-23	12.7	14.8	50.7	20.4	0.9	C18	57	32	11	0.61	0.33	0.41	0.02	1.22	1.08	4.90
YA-27	17.0	28.5	7.6	44.5	0.6	C23	14	55	31	0.90	0.59	0.33	0.03	1.34	1.27	1.82
YA-28	18.9	27.3	14.9	33.6	0.7	C25	17	42	41	3.30	2.07	0.45	0.02	1.59	1.37	0.84
YA-29	17.6	10.4	22.4	43.2	1.7	C23	18	54	28	0.65	0.68	0.58	0.06	1.24	1.18	2.21
YA-30	13.5	21.9	19.7	37.9	0.6	C25	19	55	26	3.06	1.09	0.27	0.01	1.62	1.38	2.70
YA-33	24.3	26.1	3.1	39.4	0.9	C23	30	39	30	2.71	1.60	0.56	0.02	1.49	1.33	1.47
Sat./Aro. Saturated hydrocarbons/aromatic hydrocarbons; NSO: polar compounds; Asphalt. asphaltenes; Cmax. maximum carbon number of n-alkane; GI (gammacerane index) = gammacerane/C30 αβ hopane; Pr/Ph. pristane/phytane ratio; CPI20-30 (carbon preference index) = ((n-C21 + n-C23 + n-C25 + n-C27 + n-C29)/(n-C20 + n-C22 + n-C24 + n-C26 + n-C28) + (n-C21 + n-C23 + n-C25 + n-C27 + n-C29)/(n-C22 + n-C24 + n-C26 + n-C28 + n-C30))/2; OEP (odd-to-even predominance) = 1/4 × (n-C21 + n-C23 + n-C25)/(n-C22 + n-C24); Wax index = (n-C21 + n-C22)/(n-C28 + n-C29); n.d. not detectable.

 | Show Table

DownLoad: CSV

The stable carbon isotope compositions of OM (δ¹³C_org) of source rocks from the Yuka Depression range from -31.38‰ to -23.07‰ (average = -25.58‰) (Table 1). Most samples have a δ¹³C_org value within the scope of -26‰ to -24‰, whereas obvious negative offset of δ¹³C_org have been observed in some samples (Table 1, Fig. 3), probably reflecting a change of depositional environment or OM input of during the deposition of the source rocks.

3.2 Petrography and Maceral Composition

The lithologies of analyzed samples in the J₂d⁷ include shale, mudstone, carbonaceous-mudstone, and siltstone (Fig. 3). The core from ZK6-1 well contains abundant stem and leaf fossils of plants distributed in mudstone and siltstone (Figs. 4a, 4b), as well as relatively integrated gingko leaf fossils (Fig. 4c) and clearly identifiable ferns fossils (Fig. 4d).

Figure  4.  Images from ZK6-1 well in the Yuka Depression. (a) Phytoclasts in mudstone; (b) plant fragments in siltstone; (c) gingko leaf preserved in mudstone; (d) ferns found in siltstone.

DownLoad: Full-Size Img PowerPoint

Liptinite, vitrinite, sapropelinite, and inertinite macerals have been observed from the samples under microscope observations (Fig. 5). These macerals composition are varied in the studied samples, and not detected in a small part of samples owing to inadequate valid points (Table 1). Liptinite (19% to 78%, ave. 60%) is the predominant maceral in most of the samples, following by vitrinite (7% to 62%, ave. 24%), sapropelinite (0 to 36%, av. 13%), and inertinite macerals (0 to 19%, av. 3%) (Table 1). Liptinite and vitrinite macerals are dominant components in all samples. Sapropelinite maceral is varied in the studied samples with an absence in the lower part of the ZK 6-1 well (YA-30 to YA-33). Inertinite is the lowest component of maceral composition for most samples with an absence in 8 samples, and only two samples (YA-23 and YA-30 samples) contain inertinite macerals with more than 10% (Table 1).

Figure  5.  Macerals in samples from ZK6-1 well of J₂d⁷. (a) Vitrinite, sample No. YA-23; (b) algae and amorphous sapropel, sample No. YA-12; (c) algae, sample No. YA-1; (d) vitrinite and inertinite, sample No. YA-30, using reflected-light oil immersion, and (e) and (f) liptinite, sample No. YA-5, using reflected-light and fluorescent light.

DownLoad: Full-Size Img PowerPoint

3.3 Biomarker Compositions

3.3.1   n-Alkanes and isoprenoids

The representative distributions of n-alkanes in sediments from the Yuka Depression are shown in Fig. 6. The carbon number of most samples varies from C₁₃ to C₃₇ for n-alkanes and the maximum carbon number of most samples are n-C₂₃ or n-C₂₅ (Table 2, Fig. 6), with the exception of samples YA-1, YA-2, YA-5, and YA-23. Sample YA-1 has an n-C₂₁ dominated profile, while samples YA-2, YA-5 and YA-23 having maximum carbon number at n-C₁₆ or n-C₁₈ (Table 2, Fig. 6). Most samples have slight odd over even predominance characterized by odd-to-even predominance (OEP) ranging from 1.08–1.92 (Table 2). The carbon preference index (CPI_20-30) has a range of 1.17–2.20 (Table 2). The mid-chain n-alkanes (n-C₂₁–n-C₂₅) are dominant in most samples, ranging from 15%–55% of the total n-alkane concentrations with an average of 43% (Table 2). Four samples (i.e., YA-1, -2, -5, and -23) show a dominance of short-chain n-alkanes (< n-C₂₁) with a pattern of %short-chain n-alkanes > %mid-chain n-alkanes > %long-chain n-alkanes (Table 2; Fig. 7). The wax index [calculated from (n-C₂₁ + n-C₂₂)/(n-C₂₈ + n-C₂₉)] for all samples are high (> 1.17) expect for sample YA-28, which has a wax index of 0.84 (Table 2). The Pr/Ph values of the studied samples have a variation from 0.61 to 3.96 (Table 2).

Figure  6.  The m/z 85 mass chromatograms exhibiting the distribution of n-alkane and isoprenoids in the aliphatic fractions of representative source rocks from the Yuka Depression, Qaidam Basin. Pr. Pristane; Ph. phytane, the n-alkane chain is marked by the carbon number.

DownLoad: Full-Size Img PowerPoint

Figure  7.  The bar diagram showed the distribution of short-chain (< C₂₁), mid-chain (C₂₁–C₂₅) and long-chain (> C₂₅) n-alkanes of source rocks from the Yuka Depression, Qaidam Basin.

DownLoad: Full-Size Img PowerPoint

3.3.2   Sesquiterpenoids, diterpenoids, and triterpenoids

Sesquiterpenoids were identified in the samples based on their relative retention times and mass spectra characteristics, and they are comprised of 4β(H)-eudesmane, 8β(H)-drimane, 8α(H)-drimane, 8β(H)-homodrimane, C₁₅-sesquiterpane (one isomer), and C₁₆-sesquiterpanes (two isomers). They are in relatively low contents in most samples (Fig. 8) with ∑sesquiterpenoids/∑hopanoids < < 0.1 expect for sample YA-5, in which ∑sesquiterpenoids/∑hopanoids is 0.24 (Table 3). Overall, 8β(H)-homodrimane predominates in all sesquiterpenoids for all analyzed samples. The 8β(H)-homodrimane/(8α(H)-drimane + 8β(H)-drimane) is high, ranging from 2.89–15.79 (Table 3).

Figure  8.  The m/z 123 mass chromatograms showing distribution of sesquiterpenoids and diterpenoids in representative source rocks from the Yuka Depression, Qaidam Basin.

DownLoad: Full-Size Img PowerPoint

Table  3.  Partial biomarker parameters of source rocks from the Yuka Depression, Qaidam Basin

Sample No.	∑sesquiterpenoids/∑hopanoids	8β(H)-homodrimane/(8α(H)- + 8β(H)-drimane)	C24Te/C26TT	∑TT/∑H	Ts/(Ts+Tm)	C29Ts/(C29Ts+C29 αβ hopane)	30M/30H	C30 αβ/(αβ+βα) hopanes	C32 αβ 22S/(22S+22R) hopanes	C27/C29 ααα 20R steranes	C27 βα/(αββ+ααα) steranes	C29 ααα 20S/(20S+20R) steranes	C29 βα/(αββ+ααα) steranes	Steranes/hopanes	C27 steranes (%)	C28 steranes (%)	C29 steranes (%)
YA-1	0.08	5.02	3.53	0.07	0.33	0.28	0.19	0.84	0.59	1.45	0.38	0.30	0.24	0.21	51	7	42
YA-2	0.10	4.11	2.26	0.03	0.22	0.23	0.35	0.74	0.55	0.97	0.41	0.33	0.18	0.12	40	7	53
YA-3	0.06	4.12	2.95	0.05	0.35	0.26	0.14	0.88	0.60	1.35	0.58	0.25	0.40	0.16	52	6	42
YA-4	0.04	5.09	3.02	0.05	0.32	0.25	0.14	0.88	0.61	1.30	0.57	0.29	0.28	0.18	52	10	38
YA-5	0.24	2.89	2.77	0.16	0.05	0.05	0.41	0.71	0.43	0.12	0.57	0.17	0.13	0.19	15	7	78
YA-6	0.07	5.74	2.28	0.08	0.04	0.02	0.47	0.68	0.47	0.11	0.32	0.17	0.07	0.18	15	4	81
YA-7	0.04	7.80	2.23	0.08	0.04	0.02	0.48	0.67	0.43	0.11	0.28	0.22	0.08	0.14	19	3	78
YA-12	0.01	9.88	0.78	0.21	0.19	0.12	0.28	0.78	0.47	0.53	0.43	0.38	0.25	0.25	32	12	56
YA-13	0.03	7.53	1.37	0.03	0.09	0.18	0.29	0.77	0.46	0.42	0.41	0.34	0.14	0.14	28	7	65
YA-14	0.01	15.79	2.86	0.03	0.10	0.09	0.29	0.78	0.41	0.11	0.34	0.31	0.11	0.09	13	9	78
YA-15	0.02	10.53	2.61	0.04	0.05	0.03	0.48	0.67	0.36	0.07	0.23	0.20	0.14	0.17	15	6	79
YA-21	< 0.01	10.24	0.78	0.23	0.14	0.09	0.34	0.75	0.42	0.32	0.35	0.32	0.19	0.32	29	4	67
YA-22	0.07	8.69	1.22	0.10	0.04	0.01	0.51	0.66	0.39	0.06	0.46	0.17	0.07	0.18	10	3	87
YA-23	0.01	9.39	0.98	0.26	0.06	0.04	0.30	0.77	0.45	0.09	0.34	0.17	0.15	0.27	18	7	75
YA-27	< 0.01	9.70	0.91	0.15	0.10	0.04	0.34	0.75	0.46	0.16	0.27	0.21	0.13	0.31	22	8	70
YA-28	0.02	6.01	1.48	0.10	0.02	0.06	0.30	0.77	0.50	0.02	0.18	0.13	0.21	0.33	18	9	74
YA-29	< 0.01	n.d.	0.60	0.37	0.22	0.14	0.22	0.82	0.52	0.38	0.42	0.30	0.25	0.32	33	6	61
YA-30	0.05	4.50	2.24	0.14	0.03	0.01	0.45	0.69	0.42	0.06	0.28	0.20	0.12	0.27	13	8	80
YA-33	0.06	3.84	1.55	0.06	0.06	0.04	0.30	0.77	0.45	0.20	0.37	0.23	0.21	0.17	28	7	66
For abbreviation see in Figs. 8, 9 and 10.

 | Show Table

DownLoad: CSV

The C₁₉-diterpane is the only detected diterpenoid (Fig. 8), and it is trace/absent in the studied samples.

A series of tricyclic terpanes (C₁₉–C₂₆) have been detected in the samples, but with very low content compared to hopanoids (Fig. 8). The total tricyclic terpanes/total hopanes (∑TT/∑H) is low (< 0.26), and C₂₄ tetracyclic terpane/C₂₆ tricyclic terpane (C₂₄ Te/C₂₆ TT) ranges from 0.60–3.53 with an average of 1.92 (Table 3).

3.3.3   Hopanoids

The aliphatic fractions of all analyzed samples from the Yuka Depression exhibit the presence of hopanoids, such as C27–C35 hopanes, without exception. Notably, the absence of the C28 member is evident from the data presented in Fig. 9. C30 17α-hopane is the dominant compound of hopanoids for all analyzed samples. C₂₇ 17α (H)-22, 29, 30-trisnorhopane (Tm) generally have a higher abundance compared with C₂₇ 18α(H)-22, 29, 30-trisnorneohopane (Ts) with Ts/(Ts + Tm) ranging from 0.02–0.35 (Table 3). C₂₇ 17β (H)-22, 29, 30-trisnorhopane (βTm) is recorded in most of samples (Fig. 9). The C₂₉Ts/(C₂₉Ts + C₂₉ αβ hopane) ratios in all samples are low and vary from 0.01–0.28 (Table 3). The ratios of C₃₀ 17β, 21α(H)-moretane to C₃₀ 17α, 21β(H)-hopane (30M/30H) in all samples fall into the range of 0.14–0.85 (Table 3). The C₃₀ αβ/(αβ + βα) hopane and C₃₂ 22S/(22S + 22R) hopane ratios of the Yuka Depression hydrocarbon source rocks are 0.66–0.88 and 0.36–0.61, respectively (Table 3). The C₃₂ 22S/(22S + 22R) hopane ratios for the analyzed samples is generally lower than the equilibrium value of 0.57–0.62 (Seifert and Moldowan, 1980). Although it is interesting to note that the samples from the upper part of the well (YA-1 to YA-4) have relatively higher Ts/(Ts + Tm), C₂₉Ts/(C₂₉Ts + C₂₉ αβ hopane), C₃₀ αβ/(αβ + βα) hopane, and C₃₂ 22S/(22S + 22R) hopane ratios, relatively lower 30M/30H values than the lower part samples (YA-5 to YA-33, Table 3, Fig. 10).

Figure  9.  The m/z 191 mass chromatograms showing the distribution of hopanes in representative samples from the Yuka Depression. TT. Tricyclic terpane; Te. tetracyclic terpane; Ts. C₂₇ 18α(H)-22, 29, 30-trisnorneohopane; T. C₂₇ 17α(H)-22, 29, 30-trisnorhopane; βT_m. C₂₇ 17β(H)-22, 29, 30-trisnorhopane; H. hopanes, M. moretane; G. gammacerane; a. C₂₉ 18α(H)-30-neohopane (C₂₉Ts); b. C₃₀ 17α(H)-diahopane; c. C₂₉ moretane.

DownLoad: Full-Size Img PowerPoint

Figure  10.  Depth profile of selected geochemical parameters in the source rocks from Yuka Depression.

DownLoad: Full-Size Img PowerPoint

The gammacerane index (GI = gammacerane/C₃₀ αβ hopane) in all source rocks from Yuka Depression is generally low with a variation of 0.01–0.24 (Table 2). Relatively higher GI values are observed in the source rocks from the upper part of the studied well (e.g., YA-1, -3, and -4) (Fig. 10).

3.3.4   Steroids

The distribution of diasteranes (C₂₇–C₂₉) and regular ste-ranes (C₂₁–C₂₂, C₂₇–C₂₉) in the source rocks from the Yuka Depression is shown in Fig. 11. Pregnane (C₂₁) and homopregnane (C₂₂) have been detected in low concentrations. Overall, diaste-ranes are relatively low that regular steranes. For example, C₂₇ βα/(αββ+ααα) sterane ratios are lower than 0.58 for all studied samples (Table 3). C₂₉ ααα 20R sterane predominates in C₂₇ to C₂₉ regular steranes for most samples (Figs. 11b–11c, Table 3), and few samples are dominated by C₂₇ ααα 20R sterane (e.g., YA-1, YA-3, and YA-4; Fig. 11a) with C₂₇/C₂₉ ααα 20R sterane > 1 (Table 3). The ratios of C₂₉ ααα 20S/(20S+20R) and C₂₉ βα/(αββ+ααα) in all analyzed samples are low, with a range of 0.13–0.38 and 0.24–0.43 respectively (Table 3). The steranes/hopanes are generally low in all samples, ranging from 0.09–0.33 (Table 3).

Figure  11.  The m/z 217 mass chromatograms showing the distributions of diasteranes and steranes in representative source rocks from the Yuka Depression, Qaidam Basin. a. C₂₇ βαS; b. C₂₇ βαR; c. C₂₇ αβS; d. C₂₇ αβR; e. C₂₈ βα20S 24β; f. C₂₈ βα20S 24α; g. C₂₈ βα20R 24α; h. C₂₈ βα20R 24β; i. C₂₈ αβS; j. C₂₇ aaaS; k. C₂₇ aββR; l. C₂₉ αβS; m. C₂₇ aββS; n. C₂₈ αβ20R; o. C₂₇ aaaR; p. C₂₉ βαR; q. C₂₈ aaa20S; r. C₂₉ αβR; s. C₂₈ aββR; t. C₃₀ βαR; u. C₂₈ aββS; v. C₂₈ aaaR; w. C₂₉ aaaS; x. C₂₉ aββR; y. C₂₉ aββS; z. C₂₉ aaaR; αβ, βα, ααα and αββ represent 13α(H), 17β(H)-diasteranes, 13β(H), 17α(H)-diasteranes, 5α(H), 14α(H), 17α(H)-steranes, and 5α(H), 14β(H), 17β(H)-steranes, respectively.

DownLoad: Full-Size Img PowerPoint

4. DISCUSSION

4.1 Thermal Maturity Evaluation

Various geochemical parameters, such as vitrinite reflectance, T_max, and biomarker ratios could be used to evaluate the thermal maturation of OM in source rocks and petroleum (Peters et al., 2005). The measured vitrinite reflectance (Ro = 0.53–0.70%; Table 1) indicates immature to early oil window for the source rocks from the Yuka Depression, Qaidam Basin, although it is not determined in some samples. This is consistent with the variation of T_max values (424–444 ℃, Table 1). The cross-plot of T_max and hydrogen index (Fig. 12) further consolidates this interpretation. Slightly higher T_max values of YA-19 and YA-24 samples may imply a higher-maturity, which is not reliable owing to the unobvious recorded S₂ peak, as suggested in Section 3.1.

Figure  12.  Cross-plot of T_max vs. HI indicating the kerogen type of source rocks from the Yuka Depression. For abbreviations refer to Table 1.

DownLoad: Full-Size Img PowerPoint

Biomarker data, such as relatively high CPI and OEP values (> 1, Table 2), low ratios of Ts/(Ts+Tm), C₂₉Ts/(C₂₉Ts+C₂₉ αβ hopane), and C₃₀ αβ/(αβ+βα) hopane, together with non-equilibrium C₃₂ αβ 22S/(22S + 22R) hopane and C₂₉ sterane isomerisation parameters, to suggest a relatively low maturity for the samples (Table 3). This result is also confirmed by the presence of βTm in the studied samples (Fig. 9) and the plot of C₂₉ ααα 20S/(20S + 20R) vs. C₂₉ αββ/(αββ + ααα) sterane (Fig. 13a), in which all samples fall into the immature-low mature zone. However, the samples of YA-1 to YA-4 have higher values of these biomarker parameters than other analyzed samples in the sedimentary sequence of ZK 6-1 well, suggesting that the enhanced thermal maturity of OM in the upper part samples (Table 3). Previous studies of shale gas or coalbed methane from the CY-1, Y34, and YQ-1 wells, which located in the southeast part of Yuka Depression (Fig. 1b) (Qin et al., 2018; Hou et al., 2017; Li H et al., 2016; Li M et al., 2014), found that the thermal maturity of OM in shales increases with depth, this is different with the present results from the ZK6-1 in the northwest part of the Yuka Depression, mostly because of the changes of depositional environment and source input, especially for the upper part of the northwest area in this Depression (see discussion in sections 4.2 and 4.3). Although the OM has a low maturity at immature to oil window in the studied samples, the abundant liptinite especially suberinite observed in these samples could be the dominant maceral for hydrocarbons generation, implying a good shale gas and oil potential in the J₂d⁷ of the Yuka Depression (Liu et al., 2020; Qin et al., 2018).

Figure  13.  (a) Cross-plot of C₂₉ αββ/(αββ + ααα) sterane vs. C₂₉ ααα 20S/(20S + 20R) sterane; (b) cross-plot of Ph/n-C₁₈ vs. Pr/n-C₁₇; (c) cross-plot of C₂₇/C₂₉ ααα 20R steranes vs. Pr/Ph, and (d) ternary diagram of C₂₇-C₂₈-C₂₉ steranes of source rocks from the Yuka Depression, Qaidam Basin.

DownLoad: Full-Size Img PowerPoint

4.2 Sources of OM

The different distribution patterns of n-alkane could suggest different origins of OM in deposits and crude oils (Tissot and Welte, 1989; Moldowan et al., 1985). For example, land plants-derived OM commonly contains more n-alkanes than aquatic matter (Peters et al., 2005). Immature source rock with high contribution of land plants-derived OM is controlled by the odd-carbon-numbered n-alkanes distribution, particularly with high contents of n-C₂₇, n-C₂₉, and n-C₃₁ (Eglinton and Hamilton, 1967). The short chain n-alkanes (mainly represented by n-C₁₅, n-C₁₇, and n-C₁₉) present in hydrocarbons may suggest a direct heritage from algae and the related acids-derived OM (Tissot and Welte, 1989). As discussed in Section 3.3.1, most samples show a mid-chain n-alkanes-dominated alkane profile, but samples YA-1, -2, -5, and -23 contain relatively high amount of short-chain n-alkanes and low content of long-chain n-alkanes (Fig. 7). This result indicates a higher contribution from aquatic organisms/algae (Bechtel et al., 2012; Ficken et al., 2000; Nott et al., 2000) but relatively lower terrigenous plants (Tissot and Welte, 1989; Eglinton and Hamilton, 1967) for OM in samples YA-1, -2, -5 and -23 compared to the other analyzed samples. Typical lacustrine-derived oil contains high abundance of n-C₂₃ to n-C₃₅ alkanes with a slight predominance of odd carbon-chain lengths (Farrimond et al., 2015). The n-C₂₃ to n-C₃₅ alkanes in these oils are thus regarded as potential indicators of land plant contribution to the OM. Philippi (1974) proposed that the wax index for marine-derived oils is in the range of 1.5–5.0, while a range of 0.6–1.2 is typically for oil reservoirs rich in terrigenous OM. However, some lacustrine-derived oils could also be waxy as they maybe originate from oil-prone algal biomass with type I-dominated kerogens (e.g., North Falkland Basin oils; Farrimond et al., 2015). Due to samples from the ZK6-1 well in this study are Type-II/Type-III kerogens and dominated by tType-III kerogen based on Rock-Eval data (Fig. 12) and isoprenoid ratios (Fig. 13b), the source of high molecular weight n-alkanes in the studied area may be mainly from higher plants instead of lacustrine algae. Overall, the variable distributions of short-/mid-/long-chain n-alkanes, OEP, CPI, and wax index show that both aquatic organisms (e.g., macro red alga, Cao et al., 2009; lower aquatic alginites; Liu et al., 2008; Jiao et al., 2005; Wang et al., 2004) and land plants (e.g., conifers, Wang et al., 2005) are sources of the OM for these source rocks. In addition, abundant stem and leaf fossils of plants distribute in the core samples from this study which can also demonstrate the contribution of higher plants to OM (Fig. 4).

Sesquiterpenoids, diterpenoids, and C₂₄ Te mostly originate from terrigenous higher plants (Otto et al., 1997; Simoneit, 1985; Alexander et al., 1983; Grantham and Douglas, 1980), but drimanes and rearranged drimanes are suggested to have a microbial origin (Alexander et al., 1983). The terrigenous OM has been supported by the presence of 4β(H)-eudesmane, C₁₉-diterpane, and C₂₄ Te, while relatively high 8β(H)-homodrimane in sesquiterpenoids may suggest a considerable contribution of microorganisms to OM (Figs. 8 and 9). This result is coincident with the previous n-alkane analysis.

The distribution of regular steranes in sediments and source rocks could be another parameter for the sources of OM (Kodner et al., 2008; Volkman, 1986). Generally, the predominant biological source for C₂₇ sterols has algae, and for C₂₉ sterols is photosynthetic organisms (e.g., land plants) (Volkman, 1986). However, Kodner et al. (2008) suggested that the early divergent prasinophyte green and red algae can also produce relatively high amount of C₂₉ sterols. Algae-derived C₂₉ sterols have been found in the Cretaceous oil shales and source rocks in the Songliao Basin (Tong et al., 2018; Bechtel et al., 2012). The proportions of C₂₉ steranes to Ts/C₂₇-C₂₉ steranes for most samples are relatively high, showing a distribution of C₂₉ (> 50%) > > C₂₇ > C₂₈ steranes. However, three samples (i.e., samples YA-1, -3, and -4) from the upper part of the well are characterized by C₂₇ (> 50%) > C₂₉ > C₂₈ steranes (Table 3). The C₂₇/C₂₉ ααα 20R sterane ratios for most samples are less than 1 expect for samples YA-1, -3, and -4 (1.30–1.45; Table 3). The discrimination diagram of C₂₇/C₂₉ ααα steranes vs. Pr/Ph has shown the source of OM for majority samples are land plants, while samples YA-1, -2, -3, and -4 of the upper part of the well, have a mixed/algae source (Fig. 13c). This result is consistent with the C₂₇-C₂₈-C₂₉ steranes diagram (Fig. 13d) and organic maceral types and contributions (Table 1), which shows less land plants contribute to OM in samples YA-1, -2, -3, and -4. However, this finding is slightly different from results based on the n-alkane analysis. For example, samples YA-5 and -23 that have high abundance of short-chain n-alkanes (more aquatic organism/algae contribution) do not exhibit a C₂₇-dominated sterane distribution as expected. Instead, they have high content of C₂₉ steranes. Considering that the biological precursor of C₂₉ steranes in these source rocks could come from both algae and vascular plants, the relatively low C₂₇/C₂₉ ααα 20R steranes may because algae contribute to the high content of C₂₉ steranes for these two samples.

Hopanoids are commonly thought to be derived from bacterial activities (Rohmer et al., 1992; Ourisson et al., 1979), as well as some mosses and ferns (Bechtel et al., 2018, 2012). According to Peters et al. (2005), a low steranes/hopanes ratio is indicative of a lacustrine environment or facies that is impacted by bacteria, while a high steranes/hopanes ratio implies the dominance of marine and algal-derived organic facies (Moldowan et al., 1985). The steranes/hopanes ratios are relatively low (< 0.33, Table 3), likely indicating land-derived and/or bacteria-influenced OM in these source rocks. In addition, organic maceral compositions in the studied samples are mainly liptinite and vitrinite (Table 1), result indicating that land plants are dominant. However, sapropelinite maceral is absence in the lower part of the ZK 6-1 well (YA-30 to YA-33), and showing higher percentage in the upper part may suggest that algae source increased during deposition (Table 1), which is consistent with the liptinite variations and biomarker results. In summary, the OM in source rocks from the Yuka Depression have multiple origins possibly including aquatic organisms/algae, microorganisms/bacteria, and land plants, which is also supported by petrologic characteristics of the OM (Table 1, Figs. 4, 5).

The stable carbon isotopes of OM with a variation of more than ∼2‰–3‰ are generally thought to be the results from different sources input to oils/sediments (Peters et al., 1993), although thermal maturation can also affect the stable organic carbon isotopes (Lewan, 1983). Considering that the maturity of the samples in present study varies slightly (early-oil generation window), thus the influence of maturity on the δ¹³C_org values for the samples in this study can be negligible. The variation of the δ¹³C_org values of all determined samples in the present study is located in the range of stable carbon isotopes for freshwater phytoplankton, land plants (C₃), and lacustrine algae (Lu et al., 2020b; Cloern et al., 2002; Mook, 2001; Meyers, 1994), however, not any obvious correlation have been found in this study, probably suggesting a multiple influence on the stable carbon composition of OM in these source rocks.

4.3 Depositional Environment Changes

Geological survey and previous research reports suggested that the depositional environment of the Middle Jurassic deposits from northern Qaidam Basin are formed mainly in fluvial and lacustrine conditions, and no marine deposits have been found in this area during the Middle Jurassic (Guo et al., 2018; Qin et al., 2018; Li et al., 2014; Wang et al., 2005; Ritts et al., 1999).

The Pr/Ph ratio has been widely used to evaluate redox conditions of sediments, source rocks, and oils (Jiang and George, 2018; Escobar et al., 2011; Peters et al., 2005; Didyk et al., 1978). Generally, Pr/Ph < 1 indicates an anoxic condition, while Pr/Ph > 3 is indicative for an oxic condition (Didyk et al., 1978). However, this ratio could also be affected by thermal maturity (Ten Haven et al., 1987; Tissot and Welte, 1984; Connan and Cassou, 1980), higher-plant input (Didyk et al., 1978), and salinity (Peters et al., 2005). In addition, similar sources for pristane and phytane have been suggested recently for sediments from the Upper Cretaceous Songliao Basin, which was supported by the consistent range of δ¹³C values (Wang et al., 2015; Bechtel et al., 2012). Variable Pr/Ph ratios (Table 2) and the discrimination diagram of Pr/Ph vs. C₂₇/C₂₉ ααα 20R steranes (Fig. 13c) in this study suggest a variation of depositional environment from reducing to oxidizing environment, which is also supported by inertinite maceral variations in the studied samples (Table 1). In addition, C_org/P ratio can be a robust proxy for assessing the redox conditions of sediments: C_org/P = (TOC/12) / (P/30.97), where the C_org is organic carbon in samples, and the numbers are molar mass of TOC and P, respectively (Algeo and Ingall, 2007). Generally C_org/P ratios are greater than 100 for anoxic, between 50 and 100 for suboxic, and less than 50 for oxic conditions (Algeo and Li, 2020). Majority of the studied samples fall within oxic and suboxic domains, and only three samples are exception, indicative of oxic to suboxic conditions are prevailing during deposition (Fig. 14a), as well as this result is supported by cross plot of Pr/Ph vs. GI (Fig. 14b). The variation of depositional environment conditions is more likely resulted from the change of water depth in sedimentary facies, as investigated by Yue et al. (2011), Shao et al. (2014), Qin et al. (2018), and Liu et al. (2020), who suggested that the depositional conditions of Middle Jurassic Dameigou Formation was a periodical flooded river-influenced aquatic condition. The Pr/Ph, Pr/n-C₁₇ and Ph/n-C₁₈ ternary diagram shows that most of samples fall into A and B zones, indicating fresh water lacustrine to lacustrine-swamp facies (Fig. 15a). Another cross-plot of Pr/Ph vs. dibenzothiophene/phenanthrene (DBT/P) suggests also lacustrine and fluvial-deltaic sedimentary conditions for the studied samples (Fig. 15b).

Figure  14.  Discrimination diagrams for palaeoredox and palaeosalinity (Wei and Algeo, 2020). (a) C_org vs. P; (b) Pr/Ph vs. gammacerane index (GI) for the 7th member of the Dameigou Formation.

DownLoad: Full-Size Img PowerPoint

Figure  15.  (a) Ternary diagram of Pr/Ph, Pr/n-C₁₇ and Ph/n-C₁₈; (b) cross-plot of Pr/Ph versus dibenzothiophene/phenanthrene (DBT/P) of source rocks from the Yuka Depression, Qaidam Basin.

DownLoad: Full-Size Img PowerPoint

High abundance gammacerane usually indicates highly reducing and hypersaline conditions (Sinninghe Damsté et al., 1995; Moldowan et al., 1985). Peters et al. (2005) found that the increasing salinity of water body contributes to high GI and low Pr/Ph ratios based on the analyses of oils derived from lacustrine source rocks in Angola. Unlike previous reports (e.g., Peters et al., 2005), there has not obviously negative relationship between Pr/Ph and GI for the studied samples (Fig. 14b). Overall, the GI values in most samples are low, indicating a fresh water condition for the study area. Three samples (i.e., YA-1, -3 and -4) have slight higher GI values (Table 2; Fig. 14b). The relatively higher GI values in the source rocks from the upper part may indicate an increased salinity stratification during the deposition of these samples. Increased salinity and anoxic condition from the upper part of ZK6-1 could possibly favor the formation and enrichment of algae, contributing to higher proportion of short-chain n-alkanes and C₂₇ steranes that are associated with an enhanced input of algae to OM observed in these samples. This assumption is coincident with the reports from previous studies (e.g., Liu et al., 2020, and reference therein) that the change of depositional environment from a semi-lacustrine to a lacustrine-swamp deposition in the upper part of the ZK6-1 well. In addition, seldom inertinite macerals in the samples (except for YA-23 and YA-30 samples) suggest that a relatively reductive environment during mudstones deposition (Table 1).

5. CONCLUSIONS

The black mudstones from the Middle Jurassic coal-bearing Dameigou Formation are the most significant hydrocarbon source rocks in Qaidam Basin. In present study, the thermal maturity, OM source, and sedimentary environments of source rocks in the J₂d⁷ from the northwest part of Yuka Depression have been well understood based on organic petrological features, Rock-Eval pyrolysis, TOC concentrations, biomarkers, and stable carbon isotopes.

Our study results provide detailed evidence for the assessment of hydrocarbon generation potential of mudstones, especially in the northwest part of Yuka Depression, which is different with the southeast part of Yuka Depression and have not been well studied previously. All analyzed samples stay between marginally low mature to mature stage indicated by thermal maturity parameters and biomarker indexes. The distribution of n-alkanes and biomarkers in these source rocks suggested multiple origins of the OM, possibly including algae, microorganisms/bacteria, and land plants, with a dominance of land plants. The depositional environment of the J₂d⁷ mainly was characterized by fluvial-deltaic and lacustrine facies with fresh water.

A lacustrine-swamp condition with elevated salinity, thermal maturity, and high abundance of algae contribution to OM is observed in the source rocks from the upper part of ZK6-1 well, suggesting a higher potential of hydrocarbon generation in the upper part of the sedimentary sequence from the northwestern part of Yuka Depression, which is different with the southeastern area of this Depression. The results are with great significance for further shale oil and gas exploration and exploitation of the Yuka Depression in the Qaidam Basin, especially for the northwest region that has not been studied in detail.

The observed depositional environment change of the source rock suggest that a lacustrine-swamp condition in the upper part of the sequence is more favor for the generation of hydrocarbon compared with the lower part of the sequence in the northwestern area of the Yuka Depression, meanwhile, more detailed studies need to carry out in the future to consolidate the observation, and this is also with great importance for the understanding of palaeoenvironment and palaeovegetation change during the Middle Jurassic in Qaidam Basin.