Paleoclimatic and Redox Condition Changes during Early-Middle Jurassic in the Yili Basin, Northwest China

Abstract

The Jurassic was mainly a “greenhouse” period characterized by global warming and by significant peat accumulations in some continental basins. However, studies of Jurassic climate and environments have mainly focused on marine records and only a few on terrestrial sediments. Yili Basin, a mid-latitude terrestrial basin in present Northwest China, included accumulation of the important recoverable coal seams. In this study, geological data, clay mineral analysis, and palynological assemblages were employed on fine-grained samples from the Su’asugou section in southern Yili Basin. The factors (paleoclimate, depositional conditions, and paleo-vegetation) impacting peat accumulation were investigated. The results suggest that the siliciclastics may have been derived from exposed Carboniferous rocks in a continental arc environment. A warm and humid paleoclimate in the Yili basin dominated during the early-Early Jurassic deposition of the Badaowan Formation and the Middle Jurassic deposition of the Xishanyao Formation. This climate contributed to high sedimentary rates and to a high productivity of peat-forming paleo-vegetation that was preserved under dysoxic conditions. In contrast, during the late-Early Jurassic between these two formations, the Sangonghe Formation was an interval of relatively aridity that included red beds preserved under more hypoxic sedimentary conditions, and with an interruption in peat formation and preservation.

1. Introduction

The paleoclimate of the Jurassic period is generally considered as a classical greenhouse with high atmospheric levels of CO₂ [1], a global perturbation of the carbon cycle [2], and major marine biological changes [3]. During the Lower and Middle Jurassic, data from clay mineralogy [4], geochemistry [5,6], and paleontology [7] generally point to a warm and equable climate, but with pronounced fluctuations. The Jurassic was also characterized by the worldwide preservation of organic-rich sediments [8] and by expanded tropical climate zones that feature considerable coal accumulation [9]. Possible explanations include elevated atmospheric CO₂ levels that were partly sequestered geological storage of carbon [10,11], and that triggered and/or amplified the climatic warming and anoxia conditions in global oceans and in terrestrial basin depositional environments [12,13]. Most Jurassic paleoclimate studies, however, have focused on marine sediments; only a few have examined middle-latitude terrestrial sediments [14]. Therefore, it is crucial to include studies of terrestrial ecosystems during the Jurassic in order to gain a better understanding of paleoclimatic fluctuations and redox changes in terrestrial environments.

Various proxies have been used to reconstruct paleoclimate and redox conditions [15,16]. In addition to geochemical proxies [17,18], clay mineralogy is a valuable paleoclimate proxy [19], especially to constrain variations in humidity and aridity in the hinterlands [20,21]. Palynological data enables reconstruction of terrestrial ecosystem responses to changes in environment [22]. However, any single proxy could be affected by factors in addition to local climate; therefore, lithologic features in the sediments, such as organic preservation and/or red-bed occurrences, should be included.

Coal is a significant terrestrial sink of organic carbon in the long-term global carbon cycle, and this sequestration is an important regulator of atmospheric CO₂ levels and paleoclimate [23,24]. Exploring the coal deposits is critical for understanding the global carbon cycle and climate change. Yet, the processes that control the peat accumulation in terrestrial basins remain poorly understood.

Jurassic coal-bearing strata are widely distributed throughout the world, especially in the mid-latitude terrestrial basins. The Jurassic stratigraphy in northern China is dominated by continental sediments [25], and a number of studies have noticed fluctuations of paleoclimate and redox conditions during the Early to Middle Jurassic transition in northern China [26]. The Yili Basin is a relatively large coal-bearing basin in the Xinjiang Province of Northwest China. Its Lower-Middle Jurassic terrestrial strata had a near-continuous deposition, thereby providing an ideal location and materials for studying the history of the terrestrial ecological system.

In this paper, we examined geological characteristics, clay minerals, and palynological data from Lower through Middle Jurassic strata in the Yili Basin in order to reconstruct the regional paleoclimatic and depositional redox history and the associated episodes of peat accumulation and preservation. Mineralogical and maceral composition of coal seams, coupled with sedimentological, palynological, and stable isotope data [27,28], provide other essential proxies for paleoclimate and depositional conditions and for the types of paleo-vegetation that contributed to peat accumulation.

2. Geological Setting

The triangular-shaped Mesozoic-Cenozoic continental Yili Basin is located in northwestern Xinjiang Province, and was separated from the adjacent Tarim and Kazakhstan plates by Paleozoic subduction zones (Figure 1A) [29,30]. The Yili Basin was the easternmost segment of the Kazakhstan–Yili microcontinent, which had a latitude of approximately 40° N during the Early and Middle Jurassic [31,32]. The basin formed as an intermountain graben basin derived from a former late Paleozoic post-collision extensional environment [33]. Due to the Cenozoic tectonic movement, the southwestern parts of the basin have a monoclonal structure that has not undergone tectonic deformation (Figure 1B) [34]. The pre-Mesozoic basement of the Yili Basin mainly consists of middle-upper Proterozoic through Paleozoic granite, felsic igneous rocks, and pyroclastic volcanics that are interbedded with carbonate rocks and clastic rocks [35,36,37]. The preserved Jurassic terrestrial strata in the Yili Basin contain abundant coal reserves [38].

The studied Su’asugou section is at the southern margin of the Yili Basin (Figure 1B). The Lower Jurassic facies of the Badaowan Formation (Fm.) (SaJ1b) and the overlying Sangonghe Fm. (SaJ1s) were deltaic and braided-river floodplain deposits. The Badaowan Fm. is typified by mottled medium- to coarse-grained sandstone, siltstone, and mudstone with coal layers (Figure 2F–J). The Sangonghe Fm. consists of greyish white fine-grained sandstone, mudstone, and red siltstone, with a basal mottled medium- to coarse-grained sandstone unit, and lacks any coal seams (Figure 2C–E). The Middle Jurassic Xishanyao Fm. (SaJ2x) is composed of greyish white siltstone, coal, and carbonaceous siltstone, begins with a basal layer of greyish yellow coarse-grained sandstone, and is dominated by a facies of deltaic and meandering-stream floodplain deposits. The Xishanyao Fm. strata also include a total of approximately 20 m of red burnt rock (Figure 2A,B), which were interpreted by Shi (2020) as a later product from self-ignition of buried coals [40].

The Lower-Middle Jurassic boundary has been placed at different lithostratigraphic positions in the Yili Basin stratigraphy, but most researchers seem to assign that boundary to the top of the Sangonghe Fm., where red mudstone is overlain by coarse-grained sandstone at the base of the Xishanyao Fm. [41,42]. Plant fossils, including linear or ribbon-like leaves, are common in the outcrop (Figure 2G). The disappearance of the Concavisporites genus can be regarded as an indicator of the end of the Early Jurassic [43]. Therefore, according to the features of the palynofloral assemblage analyzed in our study section, this supports assigning the Badaowan and Sangonghe formations to the Lower Jurassic, and the Xishanyao Fm. to the Middle Jurassic.

3. Materials and Methods

A total of 21 fine-grained samples from the section in the southern Yili Basin (Figure 1) were examined using petrographic thin sections for investigating sedimentary fabric and using X-ray diffraction (XRD) for clay mineral assemblages. The samples are mainly mudstones and siltstones and were collected from fresh surfaces without roots, veins, or strongly weathered surfaces. Unfortunately, taking account of various circumstances of location and access transportation, there was only one sample for each layer, without highly accurate location data.

For trace element and clay minerals analyses, a portion of each sample from the Su’asugou section was ground into powder (200-mesh) using an agate mill. Clay minerals were studied using XRD of non-calcareous clay-sized particles on mounts oriented on glass sides. The XRD used a PANalytical X’Pert PRO diffractometer at ALS Chemic (Guangzhou) Ltd. with Cu Kα radiation, Ni filter with a divergence slit of 0.38 mm, under 40 kv voltage and 25 mA current. After powdered samples were deflocculated by successive washing with distilled water, clay fractions were concentrated by centrifugation. Oriented glass slides were made by smearing the clays unidirectionally. For each sample, three XRD runs were performed following air-drying, ethylene-glycol solvation (EG), and heating at 490 °C for 2 h. Identification of clay minerals was made by comprehensive comparison of these three XRD diffractograms using the software High score. Relative proportions of each clay mineral species were mainly calculated according to the area of the (001) basal reflections [45].

Trace elements were measured by inductively coupled plasma mass spectrometry (ICP-MS) with a Finnigan MAT Element II mass spectrometer at ALS Chemic PANnalytical (Guangzhou, China) Ltd. Then, 25-mg samples were digested using HF + HNO₃ (HF:HNO₃ = 1:2) in screw-top PTFE-lined stainless steel bombs at 190 °C for 48 h. Insoluble residues were dissolved at 130 °C using 5-mL 30% (v/v) HNO₃ for 3 h, and diluted to 25 mL. The detection limit for trace elements is 0.05 × 10⁻⁶ [46]. The analytical precision is better than 5%. A cluster analysis of 22 trace elements was performed using software of SPSS Statistics version 19 from International Business Machines Corporation (IBM).

To investigate things in more detail, we calculated the element enrichment factors (X_EF; Tribovillard, 2006) [20]. Trace element contents in sedimentary rocks may be affected by biogenic material, so it is necessary to normalize trace-element concentrations to aluminum content [46,47]. Enrichment factors (EFs) are an effective means to normalize elements, as calculated following the formula below:

X_EF = (Xsample/Alsample)/(X_AUCC/Al_AUCC)

(1)

where X is the observed elemental concentrations in the surface centimeter of a sample. For this study, the Average Upper Continental Crust (AUCC) values for Al (8.04 wt%), V (107 ppm), and U (2.8 ppm) are from McLennan (2001) [47].

The palynology samples were subjected to KOH (10%) for 10 min at 80 °C and acetolysis treatment for 3 min at 90 °C, following standard pollen preparation techniques [48]. Microscope slides were prepared from the residue and mounted in glycerol. A Zeiss light microscope at 400× and 1000× magnification was used for the identification of palynoflora.

4. Results

4.1. Trace Elements and Rare Earth Elements

The concentrations and ratios of trace elements and rare earth elements (REEs) of the samples are shown in

Table 1. Trace element and rare earth element contents (ppm) of samples from the Su’asugou section in Yili Basin, NW China.

Sample	Li	Be	Sc	V	Cr	Co	Ni	Cu	Zn	Ga	Rb	Sr	Ta	Th
SaJ1b-2	53	2.3	16.	133	77	12	32	41	138	19	106	75	0.94	10
SaJ1b-3	129	2.7	22	117	59	3.9	27	24	118	31	112	79	1.5	14
SaJ1b-6	36	3.3	14	105	60	21	24	27	121	22	141	66	1.2	11
SaJ1b-7	44	3.2	16	92	50	9.5	20	38	186	25	148	90	1.5	13
SaJ1b-11	20	1.9	11	66	27	10	16	21	177	16	135	69	1.5	12
SaJ1b-12	48	1.9	18	130	82	15	44	43	105	20	95	74	0.71	9.9
SaJ1b-14	55	2.1	19	151	87	18	54	46	111	23	107	75	0.74	11
SaJ1b-18	65	2.3	21	163	92	19	55	54	124	25	113	83	0.79	11
SaJ1b-21	11	0.83	5.	41	27	5.8	8.7	10	29	9.4	70	40	0.49	5.8
SaJ1b-24	10	0.77	3.5	31	20	4.5	7.9	12	22	8.2	59	43	0.39	5.0
SaJ1b-25	17	0.97	5.0	45	19	9.6	9.8	15	40	8.9	57	37	0.41	5.7
SaJ1s-2	34	2.3	14	102	83	13	33	34	60	19	115	49	0.95	12
SaJ1s-3	42	1.6	12	82	63	6.3	18	30	80	18	90	94	1.0	10
SaJ1s-4	31	1.9	20	144	92	20	33	56	105	19	70	128	0.78	7.4
SaJ1s-5	12	0.85	6.8	47	37	9.6	20	11	59	12	85	35	0.66	6.9
SaJ1s-6	20	0.49	8.8	73	57	1.6	6.2	19	27	15	18	19	1.8	14
SaJ2x-3	32	0.65	11	94	67	1.8	7.8	25	30	21	28	32	2.2	19
SaJ2x-8	64	1.03	13	108	70	2.4	12	21	31	27	45	38	2.2	17
SaJ2x-12	56	0.76	11	90	70	2.4	12	16	28	22	33	29	2.1	16
SaJ2x-13	11	1.5	8.7	59	57	4.6	13	14	72	13	33	56	1.1	9.2
SaJ2x-15	9	1.8	10	63	77	5.6	14	19	84	15	41	138	1.2	12
	Lu	U	W	Zr	Mo	Nb	Cs	Ba	Hf	La	Ce	Pr	Nd	Yb
SaJ1b-2	0.53	4.5	19	253	0.76	14	12	426	6.4	28	61	7.1	27	3.2
SaJ1b-3	0.70	4.0	23	407	0.79	20	18	318	10	21	35	3.9	13	4.2
SaJ1b-6	0.54	3.1	11	339	0.78	17	17	462	8.1	34	89	7.6	29	3.4
SaJ1b-7	0.87	4.2	18	461	0.66	23	15	786	10	51	133	15.3	64	5.6
SaJ1b-11	0.77	5.0	13	593	4.8	24	7.1	799	13	42	90	10	41	4.8
SaJ1b-12	0.42	3.5	13	151	1.7	9.8	8.3	337	4.3	27	55	6.6	26	2.6
SaJ1b-14	0.47	3.4	20	162	1.6	10	9.8	350	4.4	30	64	7.4	29	2.9
SaJ1b-18	0.5	3.5	12	165	1.1	10	10	350	4.7	29	63	7.5	28	3.0
SaJ1b-21	0.18	1.7	10	105	0.49	6.1	2.3	430	2.9	16	33	3.9	14	1.2
SaJ1b-24	0.15	1.3	9.7	85	0.61	4.5	1.6	445	2.3	19	43	4.7	17	1.0
SaJ1b-25	0.33	1.5	8.8	109	0.44	4.9	1.8	351	3.2	18	41	4.5	17	2.0
SaJ1s-2	0.5	6.2	5.0	288	1.9	13	10	386	7.7	35	71	8.1	31	3.2
SaJ1s-3	0.35	3.9	9.9	217	2.4	14	6.9	410	6.0	24	52	5.8	22	2.3
SaJ1s-4	0.51	2.5	21	180	1.7	11	4.4	475	5.0	25	55	6.6	26	2.9
SaJ1s-5	0.29	1.6	1.1	127	0.7	8.5	2.7	435	3.2	22	41	4.7	18	1.7
SaJ1s-6	0.51	3.9	3.6	736	1.3	23	1.5	69	18	20	41	4.7	16	3.2
SaJ2x-3	0.57	4.7	4.2	734	1.6	27	2.4	257	18	27	57	6.3	22	3.6
SaJ2x-8	0.51	4.9	6.2	459	1.4	28	4.7	156	12	28	55	6.1	21	3.2
SaJ2x-12	0.55	4.6	7.1	701	1.2	27	3.5	116	17	26	50	5.6	19	3.4
SaJ2x-13	0.38	2.7	3.0	388	0.78	19	6.6	253	9.8	24	53	5.9	21	2.3
SaJ2x-15	0.47	3.4	3.1	667	1.2	21	8.3	241	16	33	61	6.6	23	2.9
	Sm	Eu	Gd	Tb	Dy	Ho	Er	Tm	∑REE	Sr/Cu	V/(V+Ni)	U/Th	MoEF	VEF
SaJ1b-2	5.6	1.4	4.9	0.82	4.9	1.1	3.2	0.48	150	1.8	0.80	0.43	0.46	1.14
SaJ1b-3	3.2	0.91	3.9	0.87	6.2	1.4	4.3	0.65	101	3.3	0.81	0.28	0.34	0.71
SaJ1b-6	5.8	1.2	5.3	0.87	5.7	1.2	3.5	0.52	188	2.4	0.81	0.26	0.49	0.92
SaJ1b-7	14.	2.8	12	1.8	10	2.0	5.8	0.89	322	2.3	0.82	0.31	0.37	0.72
SaJ1b-11	8.1	1.7	6.8	1.2	7.3	1.6	4.6	0.75	223	3.2	0.80	0.39	3.57	0.69
SaJ1b-12	5.3	1.3	4.6	0.76	4.5	0.93	2.7	0.41	139	1.7	0.75	0.35	0.97	1.07
SaJ1b-14	5.7	1.2	5.1	0.79	4.9	1.0	3.0	0.46	156	1.6	0.73	0.31	0.84	1.14
SaJ1b-18	5.8	1.2	4.8	0.80	4.9	1.0	3.0	0.46	155	1.5	0.75	0.30	0.54	1.17
SaJ1b-21	2.6	0.6	1.9	0.33	1.9	0.40	1.2	0.18	79	3.8	0.82	0.30	0.57	0.67
SaJ1b-24	3.3	0.73	2.4	0.37	2.4	0.40	1.0	0.16	97	3.5	0.80	0.25	0.69	0.28
SaJ1b-25	3.6	0.84	3.3	0.52	3.1	0.68	2.0	0.30	99	2.5	0.82	0.25	0.50	0.71
SaJ1s-2	6.1	1.4	5.3	0.83	5.1	1.0	3.0	0.48	173	1.4	0.76	0.48	1.22	0.93
SaJ1s-3	4.2	0.8	3.7	0.57	3.7	0.79	2.3	0.35	124	3.1	0.82	0.40	1.59	0.76
SaJ1s-4	5.8	1.5	5.2	0.81	4.8	1.0	2.9	0.45	139	2.3	0.81	0.34	0.99	1.2
SaJ1s-5	3.4	0.78	3.4	0.55	3.2	0.66	1.9	0.26	103	3.2	0.69	0.22	0.65	0.58
SaJ1s-6	3.2	0.57	3.2	0.56	3.9	0.89	2.8	0.48	103	1.0	0.92	0.28	1.45	1.17
SaJ2x-3	4.3	0.75	3.9	0.72	4.8	1.02	3.3	0.51	136	1.3	0.92	0.24	1.27	1.08
SaJ2x-8	4.0	0.83	3.6	0.63	4.4	0.99	2.9	0.46	133	1.8	0.90	0.28	2.91	3.12
SaJ2x-12	3.8	0.69	3.6	0.62	4.3	0.97	3.1	0.48	123	1.8	0.88	0.28	1.94	1.99
SaJ2x-13	4.4	1.1	3.6	0.64	3.9	0.83	2.4	0.37	126	2.3	0.79	0.47	0.75	1.53
SaJ2x-15	4.8	1.1	4.1	0.68	4.4	0.94	2.9	0.45	148	4.1	0.84	0.73	1.07	0.56

.

The concentrations of trace elements in sediments are mainly controlled by terrigenous and/or by sedimentary and diagenesis processes and by some unknown factors [49]. A multivariate statistical method was applied to distinguish these complex effects, factors and processes. The Pearson correlation for cluster analysis of the 22 trace elements yielded 3 main categories (Figure 3), which could be applied to demonstrate the similarities and differences of elements [50,51].

Category (a) is mainly high-field-strength elements, including Zr, Th, Hf, Ta, and Nb, indicating the influence of terrigenous matter [52,53]. Category (b) includes Li, Ga, Sc, V, Cu, Co, and Cr, and Category (c) includes Ba, Sr, Rb, Zn, W, Cs, and Be.

The total rare earth elements (∑REE) in the Badaowan Fm. show significant variability from 78 to 322 ppm (average = 155 ppm). The range of ∑REE from the Sangonghe Fm. is 102–173 ppm, with an average value of 128 ppm. In the Xishanyao Fm., the ∑REE values vary between 123 and 147 ppm, with an average value of 132 ppm. The average REE composition of the three formations are not much different in normalized concentrations. From the similarity in the pattern, it appears that the sediment sources remained constant during the period of deposition, although relative contributions may have varied [54].

The upper continental crust (UCC)-normalized REEs and trace elements’ distribution patterns of the samples for each Early-Middle Jurassic formation in the Yili Basin are presented in Figure 4. The UCC-normalized REES pattern in the Badaowan and the Xishanyao formations exhibit positive Eu anomalies. These significant Eu anomalies were previously interpreted as being related to hydrothermal circulation [40].

In the Badaowan Fm., Sr/Cu varies from 1.5 to 3.8, with an average value of 2.5. V/(V+Ni) ranges from 0.73–0.82, and the average value is 0.79. The ranges of U/Th are 0.25–0.43, and the average value is 0.31. V_EF and Mo_EF range from 0.28–1.2 and 0.34–3.6, respectively, with average values of 0.84 for V_EF and 0.85 for Mo_EF.

In the Sangonghe Fm., Sr/Cu varies from 1.0 to 3.2, with an average value of 2.2. V/(V+Ni) ranges from 0.69 to 0.92, and the average value is 0.80. The ranges of U/Th are 0.22–0.48, and the average value is 0.34. V_EF and Mo_EF range from 0.58–1.2 and 0.65–1.59, respectively, with average values of 0.93 for V_EF and 1.2 for Mo_EF.

In the Xishanyao Fm., Sr/Cu undergoes a positive excursion from 1.3 to 4.1, with an average value of 2.3. V/(V+Ni) ranges from 0.79–0.92, and the average value is 0.87. The ranges of U/Th are 0.24–0.73, and the average value is 0.40. V_EF and Mo_EF range from 0.56–3.1 and 0.75–2.9, respectively, with average values of 1.7 for V_EF and 1.6 for Mo_EF.

4.2. Clay Minerals

The XRD analysis of the clay mineralogy of the samples yielded a high variability in the proportion of chlorite (3%–26%), kaolinite (9%–79%), illite (1%–93%), and I/S (illite–smectite) (2%–60%), with rare smectite in several samples (

Table 2. Clay mineral data of samples from the Su’asugou section in Yili Basin, NW China.

Sample Number	Lithology	Chlorite (wt%)	Kaolinite (wtw%)	Illite (wt%)	Illite-Smectite (wt%)
SaJ1b-2	Siltstone	10.9	15.8	11.7	56.7
SaJ1b-3	Mudstone	24.4	10.6	12.3	52.3
SaJ1b-6	Siltstone	12.2	10.1	21.1	57.4
SaJ1b-7	Mudstone	11.5	37.2	18.2	32.6
SaJ1b-11	Mudstone	5.1	21.8	16.4	56.7
SaJ1b-12	Siltstone	4.1	24.0	20.1	51.9
SaJ1b-14	Mudstone	4.4	17.5	25.3	52.7
SaJ1b-18	Mudstone	8.1	67.9	6.8	17.2
SaJ1b-21	Mudstone	3.1	40.3	30.6	26.1
SaJ1b-24	Siltstone	-	-	93.4	6.6
SaJ1b-25	Mudstone	5.6	34.2	16.2	34.7
SaJ1s-2	Siltstone	-	79.1	18.8	2.1
SaJ1s-3	Mudstone	10.1	65.9	1.1	21.0
SaJ1s-4	Mudstone	16.6	53.8	2.5	26.0
SaJ1s-5	Mudstone	20.8	62.9	8.5	7.9
SaJ1s-6	Mudstone	14.1	76.8	3.7	5.4
SaJ2x-8	Siltstone	22.3	57.0	15.0	5.7
SaJ2x-12	Mudstone	17.6	54.6	12.9	8.1
SaJ2x-13	Mudstone	9.7	68.4	14.3	7.6
SaJ2x-15	Siltstone	26.2	27.3	40.6	5.9

).

The Lower Jurassic Badaowan Fm. (SaJ1b) is characterized by a dominance of I/S, almost equal proportions of illite and kaolinite, and minor chlorite. From the Badaowan Fm. towards the Sangonghe Fm. (SaJ1s), a gradual increase in the kaolinite content and a corresponding decrease in illite and I/S content is observed. The change from the Lower Jurassic Sangonghe Fm. to the Middle Jurassic Xishanyao Fm. (SaJ2x) is characterized by a disappearance of I/S and a slightly increased content of illite, although kaolinite remains the dominant clay mineral.

4.3. Palynological Data

Palynological data is an important source of quantitative terrestrial paleoclimate data. For example, Deng (2017) interpreted the climate change within the Lower Jurassic coal-bearing Hongqi Formation in the Xilinhot Basin based on plant fossils [44]. The studied flora is characterized by abundant Ginkgopsida and Cycadopsida. These changes in floral compositions indicate that a dramatic temperature rise and increase in aridity occurred during the late Early Jurassic in North China. Ptilophyllum is usually considered as an important indicator of hot and relative arid environments [56].

The palynofloral assemblages from the study area (

Table 3. List of the palynological data identified from the Su’asugou section in Yili Basin, NW China.

Formation	Sample Number	Palynologica Data
Sangonghe Formation	SaJ1s-1	Chasmatosporites elegans, Vittatina sp., Striatoabieites multistriatus, Piceites arxanensis, Quadraeculina limbata, Pinuspollenites spp., Podocarpidites arxanensis, non-striate bisaccate
Badaowan Formation	SaJ1b-25	Dictophyllidites mortoni, Densoisporites sp., Cycadopites sp., Chasmatosporites elegans, Concavisporites toralis, Pseudopicea sp., Piceites arxanensis, Pinuspollenites spp., Piceaepollenites sp., Podocarpidites multesimus, non-striate bisaccate
	SaJ1b-14	Cyathidites minor, Concavisporites bohemiensis, Concavisporites toralis, Osmundacidites wellmani, Aratrisporites granulatus, Chadmatosporites elegans, Taeniaesporites pellucidus, Pseudopicea sp., Piceites arxanensis, Pinuspollenites spp., Piceaepollenites sp., Podocarpidites multesimus, Podocarpidites arxanensis, non-striate bisaccate
	SaJ1b-12	Concavisporites toralis, Aratrisporites granulatus, Pseudopicea sp., Vitreisporites sp., Klausipollenites sp., Pinuspollenites spp., Podocarpidites multesimus, non-striate bisaccate
	SaJ1b-7	Stereisporites sp., Vittatina sp., Protohaploxypinus limpidus, Striatoabieites multistriatus, Gardenasporites sp., Pseudopicea sp., Vitreisporites sp., Piceites arxanensis, Parvisaccites sp., Pinuspollenites spp., Podocarpidites multesimus, non-striate bisaccate
	SaJ1b-6	Podocarpidites sp. non-striate bisaccate
	SaJ1b-3	Osmundacidites wellmani, Perinopollenites sp., Pseudopicea sp.

) show that the Badaowan Fm. was dominated by Gymnospermae with very few fern spores. Most of the identified taxa have known distributions that were limited to the Jurassic, with only a small number having reported first appearances during the Late Triassic [56,57]. Cycadopites sp. and Osmundacidites from our studied section were similar to Deng (2017), which are usually adapted to a humid and warm climate. The main palynomorphs found in Sangonghe Fm. are similar to those in the Badaowan Fm.

5. Discussion

5.1. Provenance

Aspects of the provenance of the sediments were interpreted based on the discrimination diagram of Zr/Sc versus Th/Sc (Figure 5A). It shows that most of these sources themselves were apparently a previous product of recycling and transport sorting that resulted in a concentration of zircons [58,59]. The discrimination diagram of Hf versus La/Th [60] indicates that the Lower through Middle Jurassic sediments in the Yili Basin were mainly derived from acidic igneous rocks from a volcanic arc (Figure 5B). Values from most samples fall within the region of sedimentary rock to the intersectional region of alkaline basalt and sedimentary rock, but with a few samples falling within the region of continental tholeiitic basalts (Figure 5C). The diagram of La-Th-Sc shows that the provenance for most of the sediments was located within a continental acidic arc (Figure 5D).

Huang (2017) confirmed that the South Tianshan oceanic crust was subducting beneath the Yili–Central Tianshan block from Late Devonian to late Carboniferous (380–310 Ma), followed by a final amalgamation that resulted in the uplift and denudation of the southern portion of the Yili–Central Tianshan block and of a (U)HP metamorphic belt along the northern margin of the South Tianshan orogenic belt [31]. Combining this tectonic history with the geochemical features of the fine-grained samples from the section, it can be concluded that the provenance for the Lower-Middle Jurassic sedimentary rocks in this region of the Yili Basin were from eroding Carboniferous rocks that were initially derived from a continental volcanic arc at an active continental margin.

5.2. Redox Condition

Redox-sensitive trace metals have been used extensively as geochemical proxies to infer the redox conditions for the sediments during and immediately following deposition [18,19,20,21]. The authigenic enrichment of these trace metals in sediments is driven by the different solubility and/or affinity for particulates of the various redox states, which in turn can be related to the redox conditions at the time of sediment deposition [19,20].

Trace element ratios V/(V+Ni), U/Th, V_EF, and Mo_EF were applied to decipher the redox conditions (Figure 6).

Lewan (1984) observed that the relative proportions of V and Ni are usually enriched in comparison with Ni in anoxic environments [62]. A high V/(V+Ni) value is an indication of anoxic depositional conditions [63,64]. The averages of the V/(V+Ni) values from three formations are 0.79, 0.80, and 0.81, respectively (Figure 3). The V/(V+Ni) ratios for all the samples are slightly high (0.69–0.92), suggesting that the sediments were deposited under dysoxic conditions.

U/Th ratios are another proxy for redox conditions, in which higher values indicate anoxic depositional conditions, and lower values result from dysoxic conditions [65,66]. The relatively low values of the U/Th ratios from the three formations (averages of 0.31, 0.34, and 0.40) indicate deposition in an environment where a moderate amount of oxygen was available (dysoxic conditions) [21].

The trace metal enrichment factors (V_EF, Mo_EF) in sediments have been intensively studied [18,67,68]. The averages of V_EF from three formations are 0.84, 0.93, and 1.66, respectively. Most of these values are not consistent with strongly euxinic conditions in the Black Sea [18]. The averages of V_EF gradually increase through the three formations, indicating a dysoxic depositional environment [69]. Mo_EF values in the three formations range from 0.34 to 3.6, 0.65 to 1.59, and 0.75 to 2.9, respectively (Figure 6). Mo_EF reaches high values in the Badaowan Fm. (maximum of 3.6) and the Xishanyao Fm. (2.9). However, these levels do not indicate anoxic conditions [21,70,71], such as those in the strongly anoxic Cariaco Basin, where Mo_EF exceeds 100 and can reach 1000 [70].

Overall, these geochemical proxy results indicate a dysoxic environment for depositional setting in the Yili Basin during the Early-Middle Jurassic. A caveat is that any single geochemical proxy for depositional redox conditions can be affected by additional factors; therefore, additional geologic features should be included. The presence of well-preserved floral fossils (Figure 2G) is consistent with leaf litter being preserved in acidic and anoxic conditions, especially below a surficial peat containing living plants (acrotelm) [72].

In contrast, the reddish color of the silty sediments in the Sangonghe Fm. (Figure 6) is notably different from the other formations. Oxidizing depositional conditions are widely used to explain the coloration of red sandstones in the Yili Basin [73,74] and elsewhere [75]. These sedimentary features are supported by the Mo_EF values to imply that the sediments in the Sangonghe Fm. were deposited under a more oxic condition. Such changes in redox conditions can result from paleoclimate fluctuations [76,77,78].

5.3. Paleoclimate Conditions Inferred from Geochemistry and Clay Mineralogy

Clay mineralogy and geochemical ratio (Sr/Cu) in sediments are proxies for paleoclimate and weathering conditions in the provenance region. For example, when the climate is more humid and warm to promote stronger weathering, Sr in the sediment being produced leaches out, and the Sr/Cu ratio in the deposited sediment is lowered [79]. It has been suggested that a low Sr/Cu ratio in the range of 1.3–5.0 in the sedimentary record indicates a warm-humid climate, whereas ratios greater than 5.0 reflect a more hot-arid condition [80,81]. The samples from the Yili Basin have low Sr/Cu ratios between 0.6 and 4.1, with an average of 2.1 (Figure 7, Table 1).

Clay mineralogy has been applied to interpret Jurassic paleoclimate [4,22,23]. Sedimentary kaolinite is a clay mineral that usually forms in soils under a hot and wet tropical climate conducive for strong hydrolyzing processes [22,81]. By contrast, illite and I/S are associated with hot semi-arid to arid conditions [82]. Chlorite is considered as detrital clay mineral indicative of active mechanical erosion and limited soil formation under dry climatic conditions [83,84], although it cannot be excluded that some chlorite is a later diagenetic product [85,86]. Therefore, the K/I ratio is traditionally used as an indicator for variations in humidity/aridity conditions [87,88]. There is a complication in a progressive illitization of kaolinite with burial depth [89,90], but rapid changes in K/I in the succession, such as those in our outcrop (Figure 7), generally are indications of primary original paleoclimatic control [91].

Within the Lower Jurassic Badaowan Fm., the K/I ratio gradually increases upward, while the I/S ratio is decreasing upward (Figure 7). This would indicate a gradual decrease in aridity [84]. A sharp increase in the kaolinite content and K/I ratio at the horizon of sample SaJ1b-18 may be the result of a more moderate temperate-humid climate [81]. The overlying Sangonghe Fm. is markedly kaolinite-rich, which suggests intense leaching under hot-wet climatic conditions [86]. Kaolinite reaches a peak in abundance in the uppermost part of the Sangonghe Fm., thereby suggesting an increase in hot-wet weathering conditions in the provenance region. However, the red color of the siltstone in the Sangonghe Fm. was interpreted as an indicator of semi-arid climate conditions at the depositional site [87]. Passing upward into the Middle Jurassic, there is a strong decrease in I/S, while kaolinite remains the dominant clay mineral, thereby suggesting humid-subtropical conditions [88].

5.4. Paleoclimate Conditions Inferred from Palynology

The palynological data collected in the present study reveal a generally warm and humid climate, which is consistent with the warm temperate conditions interpreted for the Early-Middle Jurassic of North China [90]. When combined with the other multi-proxy paleoclimate proxies, it suggests a prevalence of a warm-wet climate at this site and in the provenance region of its sediments, although the climate during deposition of late-Early Jurassic Sangonghe Fm. at this site may have been more semi-arid (Figure 7). One would expect regional climate fluctuations during Early Jurassic [1,91], and such changes are reported in other basins in Northwest China, such as the Qaidam, Ordos, and Junggar basins (Figure 4) [89], based on plant fossils [92] and isotopes [93]. Vakhrameev (1991) found more Classopollis (a warm/dry environmental indicator) and thinner coal strata during intervals of the Early Jurassic in central Asia [94]. There is evidence for the Early-Middle Jurassic “Climatic Optimum” interpreted in Northwest China also recorded to the north in Siberia [95,96].

5.5. Implication for the Peat Accumulation

The Yili Basin is one of the largest producers of Jurassic coal in China, with the Badaowan and Xishanyao formations containing the main peat accumulations and regionally widespread coal seams [97]. The occurrence and the termination of peat accumulations are governed by many factors, including regional climate, local topography, and depositional environment conditions [98,99]. In particular, paleoclimate governs the growth and reproduction of plants, hence the flourishing of peat-forming vegetation [72]. Our multi-proxy analyses indicate warm and humid climatic conditions during most of the Early-Middle Jurassic within the Yili Basin, but these were interrupted by a relatively more arid interval during a portion of the late-Early Jurassic during deposition of the intervening Sangonghe Fm. Other factors for the preservation of peat through burial and surface drainage conditions can also reflect climate changes [24,97]. For example, a lowering of the local water table from changes in precipitation-evaporation and runoff conditions can result in an increase in degradation of the organic material in the peat regardless of the depositional setting [98,99]. Dai (2020) summarized the relationship between the precipitation-evaporation balance and organic matter deposition and preservation as a “dry-light” and “wet-dark” model [100].

The geochemical results show that the sediments during most of the Early-Middle Jurassic at this site were deposited in a dysoxic environment. The relatively rapid accumulation episodes for the clay-dominated sediments (less than 50% of the deposits are coarse-grained) helped to reduce the exposure times of the organic matter to the low levels of oxygen in the dysoxic environment. These episodes were both more conductive to the preservation of organic matter, and to help in the periodic sealing of the peat layers from short-lived fluctuations to more oxic conditions of surficial water. Peat accumulation is common at the terminal infilling of lacustrine basins. Sedimentary environments play an important role in enrichment of elements in coal. The paleoclimate controlled the growth and reproduction of plants, and the peat-forming vegetation that flourished could contribute to the peat accumulation. Overall, the combination of palaeoclimate and redox conditions resulted in peat accumulation (Figure 8).

In contrast to the abundance of coal within the Badaowan Fm. and coal-associated burnt rock within the Xishanyao Fm., the red coloration of the siltstone in the intervening Sangonghe Fm. indicates a more oxic average depositional condition. A possible analogous situation occurred during the Early Jurassic in the present Qinghai-Tibet Plateau region of China, where Lu (2020) documented that the onset of a more arid climate resulted in the occurrence of red beds and the termination of coal deposition [93].

6. Conclusions

Fine-grained samples from the Early-Middle Jurassic sediments in Yili Basin were analyzed to infer the paleoclimate changes and redox conditions during their deposition. Based on cluster analysis of trace elements; UCC-normalized patterns; diagrams of Th/Sc-Zr/Sc, La/Yb-∑REE, La-Th-Sc, and La/Th-Hf; and on the regional tectonic history, it is concluded that the provenance of the siliciclastic components of the Yili Basin sediments were derived from the rosion of exhumed Carboniferous rocks, which in turn were products of the erosion of an active continental margin and a continental acidic arc.

The conditions of weathering of these source rocks and the redox conditions within the depositional environment were studied by a combination of field observations and laboratory methods that included XRD analysis of the proportions of clay minerals and the values and ratios of redox-sensitive elements of U, Th, and V. The field observations and geochemical ratios suggest dysoxic conditions during deposition of the Lower Jurassic Badaowan Formation (Fm.) and the Middle Jurassic Xishanyao Fm, in contrast to a more oxidizing situation during deposition of the intervening Lower Jurassic Sangonghe Fm. Multi-proxies of clay ratios and palynofloral assemblages indicate a relatively warm and humid paleoclimate with an abundance of vegetation. However, under the more arid conditions during late-Early Jurassic when the Sangonghe Fm. was deposited, there was a reduction in the intensity of chemical weathering in the sediment provenance region, a reduction in the vegetation, and a lowering of the groundwater table at the depositional region within the Yili Basin. These factors, plus the increased oxidation levels within the deposits, resulted in siltstones with a reddish coloration and a lack of significant preserved peat deposits. Therefore, this temporary cessation in peat accumulation in the Yili Basin was primarily caused by a regional climate fluctuation in Northwest China to more arid conditions.