The influence of organic sources and environments on source rock deposition during the periods of Cretaceous–Eocene and Oligocene–Miocene, northern Kalimantan

1. Introduction

The South China Sea (SCS) covers an area of approximately 350×10⁴ km², and it has experienced two tectonic cycles, in which 14 large basins were formed and are mainly distributed on the continental shelf and slope (Zhang et al., 2018, 2021a). Until 2020, hundreds of gas fields had been found, 37 of which are large gas fields, and they are mainly distributed in the Yinggehai Basin, the Qiongdongnan Basin and the Pearl Mouth River Basin in the northern SCS, the Zengmu Basin and the Baram-Sabah Basin in the southern SCS with reserves more than 300×10⁸ m³. These resources account for approximately 70% of the total gas reserves in the SCS.

As the most petroliferous basins in the southern SCS, the source rocks in the Zengmu Basin and the Baram–Sabah Basin have attracted enough attentions. According to earlier research, the source rocks in the Zengmu Basin and the Baram-Sabah Basin are the Oligocene and the Miocene strata which were deposited in the transitional facies (Lan, 2019; Li et al., 2021). These research have focused on the depositional environments and the hydrocarbon-generation potential of coal-measure source rocks (Wan Hasiah et al., 2013; Sia et al., 2014; Togunwa et al., 2015) and the controlling factors of organic matter accumulation (Li et al., 2021) but without thorough discussion on the biomarker signatures of the Oligocene and the Miocene shales and coal measures. Meanwhile, these Tertiary sediments were overlying the thick deep marine sediments of the Cretaceous–Eocene period. Previous studies on the Cretaceous–Eocene strata aimed to mark the sedimentary and stratigraphic architecture but the geochemical characteristics and hydrocarbon generation potential of the source rocks have received scant attention. Ahmed et al. (2021) has used Rock-Eval pyrolysis and total organic carbon analysis to evaluate the thermal maturity, organic matter richness and kerogen type of the Belaga Formation deposited during the late Cretaceous to the late Eocene. There is still no comprehensive studies have been carried out on the organic sources or hydrocarbon generation to discuss their possibility as source rocks.

The purpose of this study is to assess the geochemical variability of the Cretaceous–Eocene and the Oligocene–Miocene outcrop argillaceous rocks in the northwest of Kalimantan, to reveal the difference of the organic sources and hydrocarbon generation abilities of the source rocks in marine and transitional facies. Then, this paper tries to construct models for high quality marine source rock in the Cretaceous–Eocene period and high quality source rocks in transitional faicies in the Oligocene–Eocene period using palynological and geochemical data, and to provide references for petroleum resource evaluation and exploration in the southern SCS.

2. Geological setting

The SCS had experienced complicated tectonic activities (Zhang et al., 2021b). A unified basement of the SCS was initially formed in the early Cretaceous and then was separated into several sub-basement (Zhang et al., 2018). During the Cretaceous, with the closure of the Meso-Tethyan Ocean and the expansion of the neo-Tethyan Ocean, the western Kalimantan block and the eastern Kalimantan block drifted and collided (Hall, 2002). The Semitau Block in the Kuching zone in Kalimantan was separated from the southeast China continental margin, then drifted during the Paleocene, and was collided with the northern part of Kalimantan in the Eocene (Zahirovic et al., 2014). During the period of the late Eocene to the early Oligocene, the Zengmu Basin and the Baram-Sabah Basin collided with Borneo during the process of subduction and shrinking of the palaeo-SCS. This tectonic event resulted in the uplift and erosion of Borneo and the formation of the Zengmu Basin and the Baram-Sabah Basin as the compressional foreland basins (Wang et al., 2020). However, some research believed that the subduction of the palaeo-SCS started at Paleocene, or even at the late Cretaceous and the collisional event in the Zengmu Basin and Borneo started at the late Eocene (Zhang et al., 2018). During the upper Cretaceous to the middle Miocene, the terrestrial-marine siliciclastic sediments were deposited in the transitional part of the Zengmu Basin, where outcropping samples were obtained (Fig. 1a). After the middle Miocene, Kalimantan contra rotated and formed the current structural characteristics.

Figure  1.  Locations of the outcrop samples in Kalimantan (a) and the generalized stratigraphic column of the Sarawak area (b) (modified after Wang et al. (2016)).

DownLoad: Full-Size Img PowerPoint

Controlled by the tectonic evolution from the Cretaceous to the middle Eocene, transgression occurred and there was a large scope of deep-sea area in the Sarawak and Sabah area of Kalimantan (Briais et al., 1993). The Kuching zone was overlain by deep-sea sediments of the Lupar Formation which was dated to the upper Cretaceous based on foraminiferal evidence (Tan, 1982). The Sibu area, which is located in the onshore part of the Zengmu Basin, was underlain predominantly by the Lupar and Belaga Formation. Foraminiferal dating placed the Belaga Formation in the upper Cretaceous to the upper Eocene. It forms a broad outcrop band that is almost 200 km wide and trends eastward in a crescentic pattern into the Sabah Basin (Berhad, 1999). Since the late Eocene, regression occurred and only the northeast area remained to be the deposition center of the marine system in the northwest of Kalimantan (Hall and Breitfeld, 2017). Since the early Oligocene, the uplift of Kalimantan has provided abundant material sources, which were beneficial for the development of two large deltas from the southern Zengmu. Sediments in the Baram-Sabah Basin was continuously deposited under deep-water environment. Until the early−middle Miocene, large-scale deltas were constantly developed in the Zengmu Basin and multiple delta deposits in a delta-neritic-abysmal sedimentary sequence began to fill the Baram-Sabah Basin (Guo et al., 2006).

3. Materials and methods

3.1 Samples

Outcrop samples obtained in Kalimantan is representative of various strata deposited from the Cretaceous to the Neogene in the Zengmu and the Baram-Sabah Basins. A total of thirty-six outcrop samples in different sedimentary facies were collected along the northwest margin of Kalimantan (Fig. 1). Age constraint of the outcrops is based on the geological map of Sarawak, calcareous nannofossils and sporopollen. Calcareous nannofossils were only found in K12, Q7, Q12, Q14 and Q16. Common nannofossils, widely distributed in the Cretaceous period, such as Chiastozygus plicatus, Cretarhabdus crenulatus, Cretarhabdus conicus, Cribrosphaerella ehrenbergii, Eiffellithus eximius, Eiffellithus turriseiffelii, etc. were detected in K12 samples (Fig. 2), while in other 4 samples, the appearance of the diagnostic nannofossils, such as Cyclicargolithus floridanus, Discoaster deflandrei, Reticulofenestra pseudoumbilica, Reticulofenestra minuta, Sphenolithus moriformis, etc. manifested the Oligocene–Miocene of stratigraphic age (Fig. 2). Although the majority of the observed spores and pollen were widely distributed in the Paleogene and the Neogene, there were still some pollen that had stratigraphic significance. Cicatricosisporites, Schizaeoisporites and Tricolpopollenites have been observed in Sample K12. Cicatricosisporites and Schizaeoisporites that were widely found in the early Cretaceous strata in Asia, suggested the deposition age of the Cretaceous–Eocene period. The appearance of Tricolpopollenites has also proved that the depositional time was later than the early Cretaceous period (OláníyìOdébòdé, 1987). Magnastriatites hawardtii was found in Samples Q5 and Q12. The first appearance of Magnastriatites hawardtii in Kalimantan was in the very late period of the Eocene and the continuous distributed from the Oligocene. Florschuetzia, an important marker genus in dating the Miocene sediments, especially within Kalimantan and adjacent areas (Morley, 1977; Lei, 1998), was found in Q7, Q9, Q15, Q16, Q20 and Q21 sample in high abundance. Therefore, samples from K1 to K12 and Q1 to Q4 were regarded as the sediments of the Cretaceous–Eocene period, K13 and K14 and Q5 to Q22 were deposited during the Oligocene to the Eocene period. Fteen of the collected samples are the Cretaceous to the Eocene shales in the Kuching zone from the Cretaceous to the Eocene strata and one is in the Sibu area. Lithologies of the other samples are coal, carbonaceous shales and shales that are mostly distributed in the north of Sibu in the Oligocene and the Miocene strata. Detailed information of these samples is provided in Table 1.

Figure  2.  Main calcareous nannofossils detected in K12 (a–d) and Q14 (e–h) samples. a. Cretarhabdus crenulatus; b. Cretarhabdus conicus; c. Cribrosphaerella ehrenbergii; d. Chiastozygus plicatus; e. Cyclicargolithus floridanus; f. Discoaster deflandrei; g. Reticulofenestra minuta; h. Sphenolithus moriformis.

DownLoad: Full-Size Img PowerPoint

Table  1.  Geochemical parameters of the outcrops in the north margin of Kalimantan

Sample ID	Period/series	Facies	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
K1	C.–E.	B	–24.0	1.1	468	2.6	43	0.83	0.08	0.11	0.98	0.10	0.12	0.11	0.40	6.88	1.95
K2	C.–E.	B	–23.9	0.6	436	2.5	22	N.A.	0.07	0.72	1.04	0.11	0.12	0.13	0.58	8.79	2.06
K3	C.–E.	B	–23.8	0.5	449	1.7	31	0.83	0.09	0.75	0.85	0.14	0.24	0.10	0.29	13.00	1.95
K4	C.–E.	B	N.A.	0.7	484	1.3	56	N.A.	0.07	0.14	0.39	0.27	0.46	0.16	0.88	6.09	1.96
K5	C.–E.	B	–22.0	0.1	593	2.1	3	1.47	0.06	0.06	0.25	0.41	0.80	0.14	0.23	6.87	1.74
K6	C.–E.	B	–23.5	0.1	588	2.7	5	1.12	0.08	0.00	0.39	0.40	0.71	0.15	0.31	5.99	1.86
K7	C.–E.	B	–24.6	0.1	369	0.9	8	2.27	0.07	0.04	0.34	0.47	0.74	0.15	0.25	5.27	1.64
K8	C.–E.	B	–24.8	0.0	477	0.8	4	3.61	0.05	0.21	0.36	0.43	0.82	0.15	0.34	5.05	1.74
K9	C.–E.	B	–26.2	0.1	597	1.5	5	2.42	0.07	0.04	0.25	0.51	0.77	0.15	0.19	4.74	1.52
K10	C.–E.	B	–26.7	0.1	596	2.6	4	2.26	0.07	0.03	0.36	0.43	0.73	0.14	0.22	4.71	1.77
K11	C.–E.	B	–27.3	0.1	598	3.0	4	2.57	0.07	0.05	0.38	0.33	0.57	0.17	0.38	5.15	1.40
K12	C.–E.	B	–24.6	0.2	448	0.6	30	N.A.	0.02	0.07	0.56	0.52	0.73	0.06	0.35	1.57	0.99
Q1	C.–E.	B	–27.2	0.0	576	0.8	4	1.50	0.07	0.01	0.36	0.60	1.11	0.12	0.34	4.52	1.54
Q2	C.–E.	B	–27.8	0.0	477	0.6	5	3.64	0.07	0.02	0.47	0.56	0.91	0.14	0.25	3.81	1.64
Q3	C.–E.	B	–27.9	0.3	461	0.4	77	N.A.	0.11	0.07	0.78	0.36	0.43	0.15	0.48	4.95	1.44
Q4	C.–E.	B	–26.8	0.1	597	0.5	10	2.66	0.06	0.02	0.54	N.D.	N.D.	0.12	0.44	4.65	1.54
Q5	O.–M.	B	–26.8	0.6	457	0.9	67	N.A.	0.46	8.72	2.53	1.48	0.57	0.11	0.48	4.40	0.28
Q6	O.–M.	B	–26.5	0.3	439	0.4	73	N.A.	0.16	2.12	2.43	2.53	0.98	0.10	0.68	0.71	0.76
Q7	O.–M.	B	–27.0	0.3	444	0.3	84	0.81	0.21	1.19	1.97	1.69	0.95	0.11	0.49	0.52	1.33
Q8	O.–M.	B	–28.4	0.3	415	1.0	29	0.79	0.66	1.21	1.27	1.29	1.12	0.05	0.35	2.68	0.77
K13	O.–M.	D-N	–26.0	1.0	527	3.2	33	0.95	0.13	0.17	0.12	0.37	0.76	0.11	0.26	2.40	1.51
K14	O.–M.	D-N	N.A.	2.6	443	2.7	98	N.A.	0.21	3.35	3.10	2.11	0.62	0.05	0.39	1.44	0.47
Q9	O.–M.	D-N	–27.8	4.5	448	2.9	154	0.47	1.63	3.69	3.11	2.06	0.61	0.10	1.39	0.63	0.58
Q10	O.–M.	D-N	–27.0	0.4	433	0.4	120	N.A.	0.26	1.12	3.13	3.86	1.10	0.11	0.73	0.49	0.27
Q11	O.–M.	D-N	–26.5	0.2	435	0.3	70	1.66	0.08	1.36	1.39	1.34	0.89	0.07	0.68	0.94	0.78
Q12	O.–M.	D-N	–27.2	0.7	440	0.6	107	0.39	0.19	1.20	2.49	2.21	0.89	0.08	1.01	0.32	0.62
Q13	O.–M.	D-N	–27.5	0.8	442	0.7	109	0.49	0.19	1.99	2.65	2.12	0.87	0.07	0.75	0.36	0.62
Q14	O.–M.	D-N	–27.1	0.5	451	0.5	99	1.06	0.41	1.43	2.56	2.07	0.79	0.08	1.40	0.30	0.57
Q15	O.-M.	D-N	–27.1	1.2	411	1.7	75	0.65	4.39	1.21	11.98	N.D.	0.36	0.06	0.69	0.82	0.34
Q16	O.–M.	D-N	–28.3	1.0	441	0.7	133	N.A.	0.31	0.98	2.97	3.10	1.28	0.11	0.48	0.67	0.48
Q17	O.–M.	C	–27.4	29.8	445	18.7	159	0.61	0.14	0.11	4.55	1.09	0.20	0.07	3.47	0.39	0.77
Q18	O.–M.	C	–28.0	12.3	455	9.2	134	0.78	0.76	36.64	3.45	1.60	0.53	0.03	1.12	1.78	1.42
Q19	O.–M.	C	–27.2	145.7	441	53.3	273	0.71	0.13	0.80	3.66	3.14	1.35	0.05	3.10	0.09	0.89
Q20	O.–M.	C	–26.7	29.4	433	14.9	197	0.63	0.20	2.26	3.37	4.09	1.86	0.08	1.42	0.09	1.02
Q21	O.–M.	C	–27.9	31.9	426	21.1	152	N.A.	1.27	0.57	3.60	0.44	1.24	0.11	0.58	0.59	0.00
Q22	O.–M.	C	–26.8	1.8	412	7.3	25	0.40	2.03	0.80	2.54	2.82	1.68	0.04	0.16	3.59	0.27
Note: C.–E.=Cretaceous–Eocene; O.–M.=Oligocene–Miocene; B=bathyal; N-D=delta-neritic; C=coaly; N.A.=not analyzed; N.D.=no data; P1=$ \text{δ}^{13} $Ckerogen; P2= S2, mg/g; P3=Tmax, ℃; P4=total organic carbon (TOC), %; P5=HI=hydrocarbon index, mg/g; P6=Ro, %; P7=Oleanane/C30 $\alpha \beta $ hopane; P8=bicadinane T/C30 $\alpha \beta $ hopane (m/z 412); P9=Pr/Ph; P10=Pr/nC17; P11=Ph/nC18; P12=Gam/C30 $\alpha \beta $ hopane; P13=(C19+C20)/(C23+C24) tricyclic terpanes; P14=(C19−C29) tricyclic terpanes/C30 $\alpha \beta $ hopane; P15=C27/C29 $\alpha \alpha \alpha $ 20R steranes.

 | Show Table

DownLoad: CSV

3.2 Analytical methods

After removing external surfaces using a rock saw, the freshly-cut samples were washed with methanol and deionized water to remove the possible surficial contamination, the outcrop samples were then used for the following analyzing processes.

3.2.1   Rock-Eval pyrolysis, total organic content and vitrinite reflectance measurements

The samples were powdered to 200 mesh for Rock-Eval pyrolysis and total organic carbon (TOC) measurement. Rock-Eval pyrolysis was performed on a Rock-Eval VI to determine the amounts of free hydrocarbons (S₁), potential hydrocarbons (S₂), and the temperature of maximum generation (T_max). The TOC was determined on a Leco CS-344 analyser, after pre-treatment of the samples with 5% dilute hydrochloric acid for 12 h and rinsing with deionised water 50 times to completely remove residual hydrochloric acid.

For the vitrinite reflectance (R_o) measurements, samples were crushed and embedded in liquid epoxy resin, polished to a smooth surface, and desiccated for 12 h. R_o was measured using a Leica MPV-SP microphotometer under an oil-immersion objective lens that was calibrated using a double calibration method using two R_o standards. The samples were re-measured if the measurement deviation of the standard samples was greater than ±0.02%. Between 35 to 50 vitrinite reflectance measurements were recorded for each sample, with the average R_o reported in Table 1.

3.2.2   Gas chromatography-mass spectrometry and gas chromatography-triple quadrupole mass spectrometry

The aliphatic hydrocarbons of the outcrop samples were obtained from solvent extraction and fractionation and then were analysed by gas chromatography-mass spectrometry (GC-MS) on an Agilent 6890 GC interfaced to an Agilent 5973 mass selective detector, with the ionisation source operated at an electron impact energy of 70 eV. An HP-5ms fused silica capillary column was used (30 m×0.25 mm i.d., film thicknesses 0.25 μm). The oven temperature was programmed to start at 20℃ (hold for 1 min), increased to 100℃ at 20℃/min, and from 100℃ to 310℃ at 3℃/min, with a final hold of 18 min. Helium was used as the carrier gas at a flow rate of 1.0 mL/min. Biomarker ratios were calculated based on their relative peak areas of individual compounds.

GC-MS-MS analyses were performed on a TSQ 8000 system (Thermo Fisher Scientific, USA). Samples were injected into the GC system using the AS 3000 auto-sampler. The GC was equipped with a HP-5ms used silica capillary column (60 m×0.25 mm i.d.×0.52 μm film thickness). Splitless mode was used with the inlet temperature set at 300℃, and the split flow at 15 mL/min following splitless time for 1.00 min. Helium (99.999%) was used as the carrier gas under constant flow mode at a rate of 1.0 mL/min. The GC oven temperature was initially set at 50℃ for 1 min, then programmed at 20℃/min to 120℃, then at 4℃/min to 250℃, then at 3℃/min to 300℃ and held for 30 min. The triple quadrupole mass spectrometer was operated in EI mode at 70 eV. The temperatures of the MS transfer line and ion source were set at 280℃ and 230℃, respectively. The filament emission current was 75 μA, and the filament-multiplier delay was set at 7 min. The tandem mass spectrometric detection was operated in the MRM mode for the highest possible selectivity and sensitivity. Collision-induced dissociation was performed using ultrapure nitrogen as the collision gas in the second quadrupole (Q2) collision cell. The first quadrupole (Q1) and third quadrupole (Q3) were tuned at unit mass resolution.

3.2.3   Palynological analysis

The sporopollen analyses were processed under the well-established Chinese industrial standard procedure for pollen extraction (SY/T 5522-2018), using 10% hydrochloric acid to eliminate carbonates and 40% hydrofluoric acid to eliminate silicates. A heavy liquid (HBr, KI, Zn) with a specific gravity of 2.0 (d=2) was used to separate the pollen from the remaining inorganic fraction. The palynological residues were sieved using 10 μm mesh nylon sieves in an ultrasonic bath. Sporopollen were identified and counted on a binocular microscope (Leica DM4000B) using 400×magnification. An average of 38 and 84 pollen grains were counted respectively in K samples and Q samples. The proportion of individual sporopollen taxa were determined relative to the total amount of spore and pollen (%).

4. Results

4.1 Bulk geochemical characteristics of the source rocks

The Cretaceous–Eocene Formation samples have total organic carbon (TOC) contents of 0.5%–3.0% (average 1.54%), Rock-Eval S₂ values (the content of HC in rock) of 0.03–1.10 mg /g (average 0.3 mg /g), hydrogen indices of 6–77 mg /g (average 20.5 mg /g) and vitrinite reflectance of 0.65%–3.64% (Table 1, Fig. 3). The positive correlation between R_o and S₂ in the Cretaceous–Eocene samples suggests that the high thermal evolution degree is associated with the decrease in S₂. The cross-plot of R_o and TOC indicates that the high maturity had minor influence on TOC value, proving that TOC can be used to reflect the hydrocarbon generation potential of the source rocks. The TOC contents of the Oligocene–Miocene Formation samples range from 0.3% to 53.3% (average (the content of HC in rock) 7.0%), Rock-Eval S₂ values range from 0.2 mg /g to 145.7 mg /g (average 13.3 mg /g), hydrogen indices range from 25 mg /g to 273 mg /g (average 109.5 mg /g), and vitrinite reflectance ranges from 0.39% to 1.65%. Most of the Oligocene–Miocene samples have R_o lower than 1.2%, suggesting that the samples are in the immature to peak stages of hydrocarbon generation. The relative low values of hydrogen indices suggest that the organic matters are dominated by terrestrial higher plants.

Figure  3.  Correlation between major bulk geochemistry indices, showing the organic matter types (a) and the high maturity influence of R_o on S₂ (b) and TOC (c).

DownLoad: Full-Size Img PowerPoint

4.2 GC-MS and GC-MS-MS analysis

Aliphatic hydrocarbons of the outcropping samples were analyzed by GC-MS. Representative m/z 191, m/z 412 and m/z 217 mass chromatograms for the analyzed outcrops are shown in Fig. 4. All the samples have various distributions of tricyclic terpanes, hopanes and steranes (Table 1), suggesting obvious different origins and/or depositional environments for the organic matter in the samples. Twenty-eight samples were selected for GC-MS-MS analysis to further identify the diterpanes in aliphatic fraction (Fig. 5). Due to their low abundances, peak areas and the related diterpane parameters are not calculated. Nevertheless, the distinct patterns between the Cretaceous–Eocene and the Oligocene–Miocene samples provide evidence of the original discrepancy of the organic matter in source rocks.

Figure  4.  The m/z 191, m/z 412 and m/z 217 mass chromatograms of the representative Cretaceous–Eocene abysmal sample (Sample K7, TOC=0.91%) (a) and the Oligocene–Miocene delta sample (Sample Q19, TOC=53.3%) (b).

DownLoad: Full-Size Img PowerPoint

Figure  5.  The m/z 123 mass chromatograms and the parent-daughter m/z 274−123 transition of representative samples from the Cretaceous–Eocene (a) and the Oligocene–Miocene (b), showing the distributions of diterpanes. 1: β-labdane; 2: 19-norisopimarane; 3: 18-norisopimarane; 4: rimuane; 5: 17-nortetracyclane; 6: beyerane; 7: isopimarane; 8: β-phyllocladane; 9: β-kaurane; 10: Atisane; 11: ent-kaurane.

DownLoad: Full-Size Img PowerPoint

4.3 Palynological records

Gymnosperm pollen are the dominant in flora, with an average proportion of 63% in the Cretaceous–Eocene samples. Fern spores have the average proportion of 32% in the coeval palynological records. In contrast, there are less angiosperms and algae, with the proportion of 3% and 2% respectively. Gymnosperm pollen are mainly dominated by Cheirolepidiaceae (42%) and Cupressaceae (21%). Cheirolepidiaceae are mainly composed of Classopollis annulatus. Exesipollenites tumulus and E. pseudotriletes are the two primary species of the Cupressaceae family. In addition, Abietineaepollenites, Pinuspollenites, Taxodiaceaepollenites hiatus, Ephedripites, and Cycadopites nitidus appeared sporadically. Fern spore types are characterized by Deltoidospora and Cyathidites minor, accounting for nearly 19% of all the spores and pollen. 8% of the sporepollen are Lygodiaceae spores, mainly Cicatricosisporites, Toroisporis and Klukisporites. Schizaeoisporites, Lygodiumsporites pseudomaximus, and other species make up 5% of the total (Fig. 6a).

Figure  6.  Difference in sporopollen composition between the Cretaceous–Eocene (a) and the Oligocene–Miocene (b) sediments.

DownLoad: Full-Size Img PowerPoint

The sporopollen in the palynological records from the Oligocene and the Miocene outcrops include 51% of angiosperms, 31% of ferns, and 16% of gymnosperms (Fig. 6b). In the pollen of angiosperms, Florschuetzia, the Moraceae pollen, accounts for 35.9% of the total sporopollen fossil, followed by Fagaceae at 5.4%. Other angiosperms speices are mainly Tiliaepollenites, Betulaceoipollenites, Alnipollenites verus, Caryapollenites triangulus, Liquidambarpollenites, etc., accounting for 9.7%. The pollen species of gymnosperms are monotonous. Most of them are Pinaceae (Abietineaepollenites, Pinuspollenites, Tsugaepollenites igniculus, etc.), with a proportion of 14%. Polypodiaceae is the dominant species in Ferns, with the abundance of 22%. Other ferns speices are mainly Magnastriatites hawardtii, Cyathidites minor, and Lygodiumsporites pseudomaximus.

5. Discussion

5.1 Palaeo-environment changes

5.1.1   Palaeoclimate evolution

Cheirolepidiaceae, an extinct taxa of conifers, was found in records of spores and pollen during the Cretaceous and the Eocene, and its thriving is suggestive of the arid and hot climate (Axsmith and Jacobs, 2005; Watson, 1988). Schizaeoisporites, one taxa of herbaceous fern, have been mainly distributed in a tropic and dry area in the southern hemisphere (Umetsu and Matsuoka, 2003; Umetsu and Sato, 2007). Ephedripites, a kind of dwarf shrub, are common in the dessert, arid steppe and saline lakeside zone (Hoorn et al., 2012). Cicatricosisporites and Klukisporites are herbaceous ferns and have a preferential adaptation to the tropical and subtropical environments (Ding and Zhang, 2004; Li and Liu, 1994; Zhong et al., 2012). These floral characteristics, taken as a whole, suggest an arid and tropical or subtropical environment throughout the Cretaceous–Eocene epoch (Fig. 7).

Figure  7.  Typical taxa of spores and pollen detected in outcrop samples. a–g. Classopollis annulatus, flourished in the Cretaceous; h. Schizaeoisporites sp, flourished in the Cretaceous, especially from late period of the early Cretaceous to early period of the late Cretaceous; i. Ephedripites sp., common in the Cretaceous–Neogene period; j–k. Cicatricosisporites spp., flourished in the Cretaceous; l–n. Cicatricosisporites minor, flourished in the Cretaceous; o–p. Klukisporites sp., widely distributed from the Barremian to the Cenomanian in the Cretaceous; q–s. Florschuetzia semilobata, flourished in the early Miocene; u–v. Florschuetzia levipoli, flourished in the early Miocene to early stage of the middle Miocene in Southeast Asia; t. Magnastriatites hawardti, first found in the very late Eocene and became extinct since the Oligocene.

DownLoad: Full-Size Img PowerPoint

The Oligocene–Miocene palynological records during the Oligocene and the Miocene are quite different from the Cretaceous–Eocene spore and pollen. The Florschuetzia group pollen, which were identified in the samples, including F.semilobata, F. trilobat and F.levipoli, are derived from the mangrove genus Sonneratia which is philotherm (Mao and Foong, 2013; Mustapha et al., 2017; Zhou and Lin, 2001). Magnastriatites hawardi are diagnostic spores of ferns, have the ecologically preferential adaptation to the tropical and subtropical climate (Mkpong et al., 2019). According to the sporepollen compositions, the Oligocene and the Miocene epochs were marked by a steady warming and humidification of the climate, which culminated in a tropical setting. This conclusion is in line with previous work from Morley (2018).

5.1.2   Sedimentary environments

Gammacerane is believed to be formed by the reduction of tetrahymanol, hence the gammacerane/C₃₀ αβ hopane (Gam/C₃₀H) is used to indicate the stability of the stratified water column and perhaps water salinity in marine and non-marine source rock depositional environments (Damsté et al., 1995). Relative abundance of pristine/phytane (Pr/Ph), Pr/nC₁₇ and Ph/nC₁₈ have been widely used to evaluate the depositional environments of the organic matter (Didyk et al., 1978; Shanmugam, 1985). There was an obvious variation of sedimentary conditions from the Cretaceous–Eocene to the Oligocene–Miocene periods. In comparison to samples in other ages (Fig. 8a), the Cretaceous-Eocene samples show the lowest Pr/Ph ratios and the greatest Gam/C₃₀H ratios (>0.1), which suggests that anoxic and saline water conditions were more prevalent in the more still water during this time. It is consistent with previous study in which the restricted water was caused by the weakening influence of the palaeo-Pacific on the palaeo-SCS and weak interaction between the palaeo-Pacific and the Tethys Ocean in the Cretaceous (Zhang et al., 2018). In contrast, the Oligocene–Miocene samples have the highest Pr/Ph ratios and the lowest Gam/C₃₀H ratios, reflecting less reducing and larger influx of freshwater during deposition of the Oligocene–Miocene sediments. Hence, the water body in the shallow sea is constantly disturbed due to the adequate influx of terrestrial materials from delta to sea, causing the more oxidizing condition of the water environment. Additionally, the cross-plot of Ph/nC₁₈ versus Pr/nC₁₇ demonstrates that the majority of the Cretaceous–Eocene samples fall in the reducing district with a greater contribution from marine algae, whilst the Oligocene and the Miocene samples are deposited in a reducing to oxidizing environment, with most of them falling in the oxic zone (Fig. 8b).

Figure  8.  Correlations between various geochemical parameters reflecting types of organic matter and/or depositional environments in the southern SCS, showing the differences in biomarker compositions and TOC between the Cretaceous–Eocene and the Oligocene–Miocene samples. For compound abbreviations see Table 1.

DownLoad: Full-Size Img PowerPoint

5.2 Changes in organic matter inputs

5.2.1   Algal organic matter

Sterane parameters are effective to reflect primary producers in marine and lacustrine systems (Knoll et al., 2007). C₂₇ steranes are believed to be originated from phytoplankton and metazoan, C₂₉ steranes are mainly derived from terrigenous higher plants (Volkman, 1986). C₂₇/C₂₉ sterane ratios have high values (>1.5) in the Cretaceous–Eocene samples and exhibit low values (<1.5) in the Oligocene–Miocene samples (Table 1, Fig. 4). This suggests that the Cretaceous–Eocene to Oligocene–Miocene organic matter received reduced contribution from algae. Additionally, there is a positive correlation between (C₁₉–C₂₉) tricylic terpanes/C₃₀ αβ hopane and C₂₇/C₂₉ ααα 20R steranes, which is indicative of a greater contribution from algae in the period of Cretaceous to Eocene compared to the period of the Oligocene and the Miocene (Fig. 8c).

Tricyclic terpanes are widespread in oils and source rock extracts. The extremely high tricyclic terpanes/C₁₇α-hopanes is a source parameter that assesses the relative contribution from a group of bacterial or algal lipids (tricyclics) and various prokaryotic species (hopanes) (Peters et al., 2005). The patterns of tricyclic terpanes have a distinguishing difference between the Cretaceous–Eocene and Oligocene–Miocene samples. The Cretaceous–Eocene samples have a pronounced high content of tricyclic terpanes, with almost similar peak heights with pentacyclic triterpanes (Fig. 4a) compared to low relative abundance of tricylic terpanes in the Oligocene–Miocene samples (Fig. 4b). Some C₁₉ and C₂₀ tricyclic terpanes were thought to be derived from higher plants including gymnosperms and C₂₃ tricyclic terpanes were in association with a possible marine algae source (Peters et al., 2005). The Cretaceous–Eocene samples have low (C₁₉+C₂₀) TT/(C₂₃+C₂₄) TT and high (C₁₉–C₂₉) TT/C₃₀H compared to the Oligocene–Miocene samples (Fig. 8d), respectively, which supports the theory that higher plant-derived C₁₉ and C₂₀ tricyclic terpanes are not the causes of the high (C₁₉+C₂₀) TT/(C₂₃+C₂₄) TT ratio. Hence, such high ratios can reflect the major contribution of algae (probably Tasmanites algae) during the Cretaceous–Eocene period. In contrast, the Oligocene-Miocene samples have relatively lower ratios of (C₁₉–C₂₉) TT/C₃₀H (mostly<1.6) but higher (C₁₉+C₂₀) TT/(C₂₃+C₂₄) TT than the younger samples, indicating large algal contributions of C₁₉ and C₂₀ tricyclic terpanes to the C₁₉–C₂₉ tricyclic terpanes. It further suggests the limited contribution from algae to the Oligocene–Miocene samples.

Abundance of tricyclic terpanes in Tasmanites rock extracts indicates a possible tricylic terpane origin from Tasmanites algae (Azevedo et al., 1992). Tasmanites alage is an important organic contributor in good-excellent quality source rocks, these algae are unexpectedly enriched in ¹³C (Peters et al., 2005). In addition, Tasmanites algae are known to have adaptation in a variety of environments with various water salinity and temperature in large range of geological periods (Guy-Ohlson, 1988). Blooms of Tasmanites alage happen two or three times annually in the western British Channel (Vigran et al., 2008). In the cross-plot of δ¹³C_kerogen versus (C₁₉–C₂₉) TT/C₃₀ αβ hopane, samples with (C₁₉–C₂₉) TT/C₃₀H>6 have heavier kerogen carbon isotopes (δ¹³C_kerogen>–24‰), while samples with (C₁₉–C₂₉) TT/C₃₀ αβ hopane between 4 and 6 have lighter carbon isotopes of kerogen (most δ¹³C_kerogen<–26‰) (Fig. 8e). The samples with high (C₁₉–C₂₉) TT/C₃₀H and heavy ¹³C_kerogen may shed light on the phenomenon that Tasmanites or other types of prasinophyte green algae that are similar to Tasmanites intermittently dominated the algal population as the primary producers during the Cretaceous–Eocene period. When the particular planktonic algae in the marine ecosystem declined, other planktonic algae with less ¹³C, such as blue-green algae and diatoms, were probably the main producers.

5.2.2   Terrigenous organic matter

Angiosperms did not appear until the early-Cretaceous (Burger, 1980). Gymnosperms and ferns were still the dominant vegetation types during the Cretaceous (Dettmann et al., 1992). In study area, the gymnosperm-derived diterpanes were presented in very low abundance in all samples. Nevertheless, distribution of the detectable deterpanes include β-labdane, 19-norisopimarane, 18-norisopimarane, rimuane, 17-nortetracyclane, beyerane, isopimarane, β-phyllocladane, β-kaurane and ent-kaurane were classified into two patterns (Fig. 5). The Cretaceous–Eocene samples have similar patterns with high abundance of isopimarane, β-phyllocladane and β-kaurane, while the Oligocene–Miocene samples are characterized by relatively higher contents of rimuane, beyerane, isopimarane, β-kaurane and Atisane. Rimuane was originated from rimuene which was formed from the epicuticular wax and was rich in Cupressaceae, Taxodiaceae, Podocarpaceae as well as Phyllocladaceae (Salasoo, 1984). Beyeranes has been found in several species of conifers, including Cupressaceae, Araucariaceae and Podocarpaceae (Jiang and George, 2018). Isopimaranes were reported in all conifer communities (Otto and Wilde, 2001). β-phyllocladane was identified in resins from Podocarpus and Dacrydium (family of Podocarpaceae). Kaurane was detected in Agathis, a family of Araucariaceae (Killops et al., 1995). Unlike other kinds of diterpanes, atisane was isolated from the bark of the mangrove plant Excoecaria agallocha which was angiosperm-derived (Wang et al., 2009). Thus, these diterpanes, such as rimuane, beyerane, isopimarane and kaurane, which were chiefly rich in conifer resins, indicate the contribution of gymnosperms to the organic matter during the Cretaceous–Eocene and the Oligocene–Miocene periods. The difference in the relative abundance of rimuane and phyllocladane is likely due to the thermal evolution of the organic matter (Jiang and George, 2018). The low abundance of gymnosperm-derived diterpanes in the Cretaceous–Eocene samples was in accordance with the bathyal-abysmal sea environment, which was interpreted to have received limited supplies of terrigenous organic matter. In addition, the low abundance of gymnosperm-derived diterpanes in the Oligocene–Miocene samples is suggestive of the subordinate contribution of gymnosperm to the organic matter. The relatively abundant atisane detected in the Oligocene and the Miocene samples illustrates the increase in contribution of angiosperms, especially mangroves, compared to the Cretaceous–Eocene samples.

Globally, angiosperms became the dominant type in flora from the late Cretaceous. Oleanane, originated from betulins, taraxerene and other pentacyclic triterpenoids in angiosperms, is highly specific for angiosperm plant inputs during the Cretaceous or younger age (Peters et al., 2005). Bicadinanes are mostly derived from a resinous polymer produced by angiosperms such as the tropical/subtropical plants Dipterocarpaceae (van Aarssen et al., 1992). In recent years, many authors have reported abundant oleanane and bicadinanes in crude oils, source rocks and outcrops in the SCS and South East Asia (Wan Hasiah, 1999; Sia et al., 2014). High abundance of bicadinanes is considered to be associated with mangroves (Lan, 2019). As shown in Fig. 8f, lowest oleanane/C₃₀ hopane and lowest T-bicadinane/C₃₀hopane (T/C₃₀H) ratios in the Cretaceous–Eocene samples indicate minor angiosperm contribution during this period. In contrast, the Oligocene–Miocene samples were plotted in the area with high Ol/C₃₀H and T/C₃₀H ratios. High abundance of oleanane, bicadinane and the negative correlation between Ol/C₃₀H and C₂₇/C₂₉ ααα 20R (Fig. 8g) suggest that angiosperms were dominant types in flora. Abundant bicadinanes in the samples are also consistent with the presence of atisane in the Oligocene–Miocene samples.

In conclusion, despite gymnosperms being the main flora type during the Cretaceous–Eocene period, their contribution to organic matter was relatively low due to the lengthy onshore distance where scant terrigenous organic matter was deposited. Angiosperms contributed more organic matter than gymnosperms throughout the Oligocene–Miocene epoch.

5.3 Models for source rocks deposition in response to co-evolution of environments and organisms

Two models for source rock developments were constructed from the Cretaceous–Eocene period to the Oligocene–Miocene period.

During the Cretaceous–Eocene period, gymnosperms flourished. Due to the distance offshore, the organic matter that is received in the bathyal-abysmal marine is limited, and the majority of the organic matter is sea-derived algae. Tasmanites or other varieties of prasinophyte green algae may occasionally bloom and offer a rather high productivity. This can be substantiated by the strong positive correlations between (C₁₉–C₂₉) TT/C₃₀H ratios and TOC, and between kerogen carbon isotopes and TOC (Figs 8h–i). Additionally, the good reduction of seawater allows the preservation of algae, which can be dedaced from the high C₂₇/C₂₉ content and low Pr/Ph of the Cretaceous–Eocene sample in Fig. 8j, which in turn allows the development of high-quality source rocks of the oil-prone.

Angiosperms dominated higher plants during the Oligocene–Miocene epoch, and their growth was facilitated by the humid climate. Mangroves that were hydrogen-rich served as a significant source of higher plant parent material in the deltaic-shallow marine sedimentary system. The coal developed in the delta with such organic parent material has a significantly stronger potential to produce oil than normal coal. In the shallow sea, the stronger hydrodynamic force will accelerate the dilution effect on land-derived organic matter and the enhanced oxidation of water body, which promote up the oxidation and degradation of the hydrogen-rich components in sea-derived algae and land-derived organic matter. As a result, the shallow marine source rocks from the Oligocene and the Miocene are gas-prone.

6. Conclusions

(1) The distinguishing biomarker assemblages of the 36 outcrop samples suggest the differences of organic matter inputs and preservation conditions of the Cretaceous–Eocene and the Oligocene–Miocene source rocks. Due to the weakened impact of the palaeo-Pacific on the palaeo SCS and the weakening connection between the palaeo-Pacific and Tethys Ocean, the sea was semi-closed and the water was salty and reducing over the Cretaceous-Eocene period. In addition, phytoplankton algae were the important contributors to the organic matter communities. During the Oligocene–Miocene, large-scale deltas developed forwards into the sea caused by the Sarawak orogeny. Strong hydrodynamic force in the shallow sea was not beneficial for the preservation of organic matter. Abundant angiosperms and terrestrial organic matter were carried into the delta, shallow sea or even deep sea environments. Higher plants became the dominant organic matter source compared to algae.

(2) The moderate to high primary productivity and reducing water conditions were beneficial for the formation of high-quality marine source rocks during the Cretaceous–Eocene period, although the low Rock-Eval S₂ values and low hydrogen indices of the source rocks which were influenced by high thermal maturity. Abundant angiosperm-derived terrestrial organic matter, especially hydrogen-rich mangrove-origin plant materials, contributed to the formation of high-quality source rocks in marine-continental transitional facies during the Oligocene–Miocene period.