Response of a continental fault basin to the global OAE1a during the Aptian: Hongmiaozi Basin, Northeast China

This paper presents new research on a lacustrine anoxic event (LAE). These data include stable carbon isotope (δ13Corg), pyrite sulfur isotope (δ34Spy), trace element and biomarker ratios from the Hongmiaozi Basin (North China) and unravel the response of continental lakes under the influence of early Aptian extreme climate conditions. According to the stratigraphic chronology (122–118 Ma) and carbon isotope correlations, terrestrial sediment was influenced by the early Aptian Oceanic Anoxic Event (OAE1a). The results show that the Xiahuapidianzi Group experienced a significant warming process under negative excursions in carbon isotopes due to the influence of increased carbon dioxide partial pressure (pCO2). The climate varied from warm and humid to hot and arid (high Sr/Cu, low Rb/Sr, calcareous mudstone), the evaporation and salinity increased (high Sr/Ba and B/Ga), and land input sources decreased (low Zr, Ti and Th). Moreover, high total organic carbon (TOC) content was source from bacteria, algae (n-alkanes), and euxinic depositional environments (Pr/Ph, Cu/Zn and U V Mo). In the stage of continuous carbon isotopes positive excursion, organic matter accumulated rapidly. A paleolake environment model has provided a better understanding of current global climate issues under global warming caused by increased carbon dioxide concentrations.

The distinctive features of the Cretaceous period are global warming, bio-events, shale deposition in marine environments due to intermittent oceanic anoxic events (OAEs) and associated isotopic anomalies 1,2 . The Aptian oceanic anoxic event 1a (OAE1a) at approximately 120 Ma is a typical OAE deposition event. OAE1a is characterized by sedimentary organic matter (Livello Selli) and dramatic fluctuations in the carbon isotope records of carbonate and organic carbon (δ 13 C carb and δ 13 C org ) 2 . These carbon isotope (δ 13 C) fluctuation records have been interpreted as evidence of perturbations of the global carbon cycle [2][3][4] . All δ 13 C excursions in δ 13 C carb and δ 13 C org records of OAE1a around the world are mainly preserved in marine basins [5][6][7] . Given the global significance of the perturbations and the ocean-atmosphere system, their record is found in marine sedimentary matter and terrestrial environments 8 . The terrestrial environment has remained almost unexplored except for the Xiagou Formation 9 in NW China and Yixian Formation 10 in NE China. Moreover, the associated research only employed field outcrop section samples and stable carbon isotope method. Drilling core samples with better integrity and continuity are more accurate for analysis of stable isotopes. Therefore, such work cannot fully reflect the detailed changes of the terrestrial sedimentary environment during the anoxic period, which further restricts the use of such data in determining the precise causes and changes mechanisms of the paleo-sedimentary environment under greenhouse.
The Xiahuapidianzi group (122-118 Ma) of the Hongmiaozi Basin in the northeastern part of China (Fig. 1) is a typical lacustrine sedimentary stratum. Lake sedimentation is the best carrier to reveal paleoenvironment and paleoclimatic changes because it is very sensitive to climate change 11  www.nature.com/scientificreports/ semi-deep lake facies dominated by dark mudstone and silty mudstone (Fig. 1). In this study, mudstone samples of the Cretaceous Xiahuapidianzi Formation were used to systematically explain the mechanism underlying the effect of anoxic events on the burial lacustrine organic matter in Hongmiaozi Basin by means of GC-MS, trace element analyses, stable carbon analyses and pyrite sulfur isotope geochemistry. This study is important for understanding the influence of OAE1a on the formation of lacustrine black mudstone.

Materials and methods
The main lithology of the Xiahuapidianzi Formation in the Hongmiaozi Basin consists of mudstone, muddy siltstone, silty mudstone, and sandstone. The location of the Hongdi 1 well is shown in Fig. 1. A total of 35 mudstone core samples were collected to determine the stable carbon (δ 13 C org ), sulfur (δ 34 S py ) isotope geochemistry and total organic carbon (TOC) within the studied mudstone. In addition, 7 samples were selected for GC-MS analyses while 24 samples were selected for rare element analyses. Prior to this analysis, the selected samples were powdered to 200 mesh for Soxhlet extraction analysis to determine the bitumen and hydrocarbon content. The extracted bitumen was separated into asphaltene and maltene fractions. The maltene component separation was carried out using silica gel/alumina column chromatography. Elution with n-hexane, dichloromethane/n-hexane (3:1 by volume) and dichloromethane/methanol (2:1 by volume) was performed to separate saturated hydrocarbons, aromatics and nonhydrocarbon components. Then, the saturated and aromatic hydrocarbons were analyzed by GC-MS on an Agilent 6890GC-5975iMS at the China University of Petroleum (Beijing). Certain specific biomarker peaks, such as n-alkanes, tricyclic terpenoids, hopanes and steroids, were identified based on their retention time and the identification results of previous researchers 18 . The distributions of n-alkanes and isoprene were determined using chromatogram ion m/z 85.
Thirty-two samples were selected for stable carbon isotopic compositions and total organic carbon content (TOC). Firstly, thirty-two samples were crushed to 200 mesh, then weighed respectively. Secondly, the powder samples were acidified with 3 M HCl to completely remove the inorganic carbon (CaCO 3 ). Thirdly, washing the samples many times with deionized water to remove the residual HCl traces. Finally, the decarbonated samples were dried in an oven at 80 ℃ for 10 h and reweighed to calculate the percentage of carbonate (TIC) in the bulk samples. About 5-10 mg dried samples were wrapped with folded tin cups, then combustion to CO 2 with a Eurovector elemental analyzer (EA) in combination with an Element Isoprime isotope ratio mass spectrometer to determine the TOC abundance and carbon isotopic composition. The carbon isotopic values are expressed on a per mil (‰) basis relative to the Vienna Pee Dee Belemnite standard (V-PDB, δ 13 CV-PDB = 0) within 0.05 ‰ duplicate measurement precision.
The determination method for the sulfur abundance (TS pyr ) and pyrite sulfur isotope (δ 34 S py ) composition via Eurovector elemental analyzer in-line with Eurovector Isoprime isotope ratio mass spectrometer (IRMS). The sample of 10 mg was wrapped in a tin cup, and then blown into a quartz tube containing high purity reduced copper by a pure oxygen pulse, heated to 1130 °C. The quartz tube was connected to a Mg (ClO 4 ) 2 desiccant column (remove water), and connected a 3-m stainless steel GC column packed with Porapak-Q heated to 60 °C (separate SO 2 from other gases). Timed pulses of SO 2 reference gas (Airgas 99.999% purity, ~ 6 nA) were introduced at the beginning of the run using an injector connected to the IRMS with a fixed open ratio split. The isotope ratio of the reference gas and samples to the determined by ion beam intensities relative to background values. Each five test samples were corrected by two NBS127 and two NZ1 standard samples, and uncertainties for each analytical session based on these standard analyses were better than 1.0% and 0.3‰, respectively, for abundance and isotope compositions. Isotopic results are expressed in the delta notation as per mil (‰) deviations from the Canyon Diablo (V-CDT) standard. This method reference to Cao H et al., 2016 19 . The Stable carbon, pyrite sulfur isotopic and TOC were conducted at the Stable Isotope Laboratory of Oil Shale Test Center of Jilin University.
Trace element were measured at the Oil Shale Test Center of Jilin University by Thermo Scientific ELEMENT high-resolution inductively coupled plasma mass spectrometer (ICP-MS) according to GB/T 14506.30-2010. About 25 mg powder samples were reacted with 1 ml HF and 0.5 ml HNO 3 in a sealing beaker and dissolution at 185 °C for 24 h. Then, the residues were dissolved with 5 ml HNO 3 at 130 °C for 3 h. Finally, the remainders were diluted with distilled water to 25 ml for trace element measurement. The result show less than 5% measurement error base on duplicate analysis and standards.
Stable carbon isotope and pyrite sulfur isotope. The isotopic analysis results of samples from the Xiahuapidianzi Formation are shown in Fig. 3 and Table 1. The organic carbon isotopic (δ 13 C org) values in the study samples are ranging from -27.2% to − 22.7%, with an average of − 24.9%, n = 35. The pyrite sulfur isotope (δ 34 S py ) values are between 3.7 ‰ and 17.0 ‰, with an average of 10 ‰, n = 33. The TOC values are range from 0.41% to 2.35%, with an average of 1.12%, n = 34. The percentages of TS pyr values are between 0.16% and 2.30%, with an average of 0.69%, n = 33. In Fig. 3, the sampling depth of the δ 13 C org , δ 34 (Fig. 3).
The interrelationship of stable carbon isotopes (δ 13 C org ), pyrite sulfur isotopes (δ 34 S py ), and TOC can be well correlated by the microbial sulfate reduction (MSR) reaction process as reported by Gill et al. and Berner 13,20 . The MSR can affect most of the organic matter burial and mineralization 21 : The reaction equations show that sulfate, organic matter, dissolved iron and depositional environment are the decisive factors in the process.   Fig. 4. The biomarker distribution could well reflect the source input and depositional environment condition of sediments 18 . The n-alkanes distribution of the Xiahuapidianzi Formation is illustrated in Fig. 5. The higher molecular weight compounds range from C17 to C23. Tissot and Welte reported that low to medium molecular weight compounds usually received a substantial input of long chain alkanes, either from algae of plant curricular waxes 22 .
The concentration of pristine (Pr) and phytane (Ph) in acyclic isoprenoids are the most important terms 23 . The ratio of pristane to phytane (Pr/Ph) is an indicator of the redox conditions for paleoenvironment conditions 18 . It is generally believed that higher Pr/Ph ratios (> 1) reveal oxidation conditions and low Pr/Ph ratios (< 1) indicate a reduction environment 24  Gammacerane was originally reported as an environmental salinity index for both marine and lacustrine environemts 18 , and the Gammacerane Index (gammacerane/C30 hopane) is believed to reflect the extent of the salinity stratified water column during deposition 26 . The gammacerane/C 30 hopane ratios range from 0.16 to 0.27, with an average value of 0.22, in the Xiahuapidianzi Formation samples ( Table 2).
The distributions of diasteranes and steranes (C 27 , C 28 and C 29 ) are shown in the m/z 217 ion chromatograms (Fig. 5). The relative abundances of C 27 , C 28 and C 29 regular steranes are calculated and the results are given in   (Table 2). Figure 4A shows that the Xiahuapidianzi Group was sedimentary in a suboxic-anoxic environment and algae were the main source of organic matter. Figure 4B shows the triangular plot of the three end members of the regular steranes and suggests the predominance of planktonic, algal and bacterial organic matter source input to the sediment of the Xiahuapidianzi Formation.
Trace elements. Trace element data of the Xiahuapidianzi Formation samples and relevant ratios (e.g., V/ (V + Ni), Cu/Zn, and Sr/Ba) are list in Table 3. These elements (e.g., V, Sr, Ba, B, Rb, and Ni) and their ratios are always useful in various studies to decipher the paleoclimate, paleosalinity, and redox conditions during the deposition of sedimentary rocks in basins 14,[29][30][31][32][33] . Terrigenous debris are common components in sediments or sedimentary rocks. The impact of terrigenous detrital must be evaluated before the sedimentary environment is judged. The best way to check whether the content of a given element is dominantly controlled by the detrital flux is to crossplot the trace element versus aluminum or titanium, which are commonly overwhelmingly of detrital origin. Here we choose Ti (titanium) for evaluation whether the influence by the detrital flux. Because Paleoclimate. Climate is a major factor controlling sedimentary organic matter (OM) input 35 . The climatic conditions can influence the geochemical characteristics of sediments in lacustrine environments by controlling the terrigenous material flux and diagenetic processes 32 . Trace elements Sr, Cu, and Rb are considered to be sensitive to paleoclimate, and Sr/Cu and Rb/Sr ratios may provide useful clues to infer paleoclimatic changes 31 . In dry and hot climates, the lake water area shrinks. In response, Sr becomes enriched in the lake water, while Cu precipitates out. Thus, the Sr/Cu ratio is used to reflect the climate conditions during deposition 36 . Previous studies evaluating palaeoclimate observed consistent results on the ratio of Sr/Cu. When the ratio of Sr/Cu in the range of 1to10, it indicates the climate is humid to semi-humid. When the ratio of Sr/Cu range from 10 to 20, it reveals the climate is semi-humid to semi-arid. When the ratio of Sr/Cu larger than 20, it manifests the climate is arid and hot 37 . The Sr/Cu ratios of this study vary greatly from 5.39 to 29.96, with an average of 11.14, n = 24. The Rb/Sr ratios vary from 0.09 to 1.36, with an average of 0.68, n = 24. The ratios of Sr/Cu and Rb/Sr are generally < 10 and > 0.6, respectively (Table 3), although an obvious change occurs from 710.20 to 741.93 m and 595.65-464.50 m, with Sr/Cu (> 10) and Rb/Sr (< 0.60) (Fig. 7).
Paleosalinity. Trace elements Sr, Ba, Ga, and B are also sensitive to salinity. Sr/Ba and B/Ga ratio can be useful indicators for assessing paleosalinity 29,30,33 . Sr/Ba values < 0.6 indicates a freshwater setting, between 1.0 and 0.6 represent brackish water, and > 1 indicates saline water in an arid climate 38 . The Sr/Ba ratios in all samples vary from 0.20 to 1.72, with a mean of 0.62 (Table 3). The ratios of Sr/Ba are lower than 1.0 except from 710-736 m, which exceeds 1.2. B/Ga values greater than 5.0 indicate salt water, from 3.0 to 5.0 indicate brackish water, and less than 3.0 indicate fresh water 39 . The B/Ga ratios in all samples are 1.99-8.66, with a mean of 4.24 (Table 3). A significant increase of B/Ga values occurs from 710 to 736 m, with values greater than 5.0 (Fig. 7).
Paleoredox conditions. Trace elements such as V, U, Ni, Co, Cr, and Ce are generally considered "redoxsensitive metals" and could be used as paleoredox proxies with minimal detrital influence in lake environments 40 . It has been widely accepted that the trace element ratios of U/Th, Cu/Zn, and V/ (V + Ni) can be used as an index for paleoredox conditions 40 . Hatch and Leventhal documented that an environment characterized by strongly reducing and euxinic conditions led to V/ (V + Ni) ratios greater than 0.84, while that under suboxic-anoxic environments led to V/ (V + Ni) ratios of 0.60-0.84 40,41 . The V/ (V + Ni) ratios for all the Xiahuapidianzi Formation samples vary between 0.64 and 0.79, with a mean of 0.74 (Table 3). A ratio of Cu/Zn greater than 0.63 indicates an oxidizing environment, a value of 0.50-0.63 indicates a weakly oxidizing environment, and a value less than 0.21 indicates a reducing environment. The Cu/Zn ratios for all the Xiahuapidianzi Formation samples vary between 0.20 and 0.52, with a mean of 0.32 (Table 3). Furthermore, U, V and Mo have a variety of chemical valence states, which are significantly affected by the redox state during deposition; most of them are authigenic components in sediments or sedimentary rocks, and almost no migration occurs during diagenesis, keeping the original record of deposition. U and V are reduced and can accumulate under denitrifying conditions, whereas Mo are enriched mainly under sulfate-reducing conditions alone. Based on the difference in geochemical properties of these two elements, U and V enrichment without Mo enrichment, we could infer suboxic/anoxic depositional without free H 2 S. Conversely, sediments exhibiting concurrent enrichments in U, V and Mo reflect euxinic conditions 34 .
Terrigenous detrital. Rivers carry large amounts of terrigenous detrital matter into lakes. Thus, an increase in detrital input probably results from increased river input. The concentrations of high field strength elements, such as Ti (titanium) and Th (thorium), can reflect changes in the input of terrigenous detrital matter. Because Ti (titanium) and Th (thorium) are commonly overwhelmingly of detrital origin and usually immobile during diagenesis 42 . The content of Ti (titanium) and Th (thorium) are 2729-6041 μg/g mean = 4706 μg/g (n = 24) and 5.5-16.7 μg/g (mean = 12.6 μg/g, n = 24) ( Table 3). www.nature.com/scientificreports/ Palaeoproductivity. The nickel (Ni) and copper (Cu) are dominantly delivered to the sediments in association with OM (organometallic complexes). Moreover, they are often referred to as 'sulfide forming' , because there are released through OM decay and can be trapped by pyrite if sulfate-reducing conditions prevail. Consequently, the high contents of Ni and Cu in the sediments indicate not only the high organic matter fluid, but also the reducing sedimentary environment. Moreover, even if OM be remineralized by bacterial activity, Ni and Cu could be retained within the sediments, being hosted by pyrite (most frequently). In that way, Ni and Cu may speak to the original presence of OM and consider a representative proxy to OM 34 . The content of Ni and Cu was 26.2-73.0 μg/g, mean = 42.5 μg/g and 29.9-48.5 μg/g, mean = 37.9 μg/g respectively. The barium (Ba) accumulation rate generally shows a positive correlation with the OM content and primary productivity 43 . The content of Ba was 477-1555 μg/g, mean = 704 μg/g ( Table 3). The TOC content in the sediment can more sensitively reflect the changes of lake organic carbon production and can be used to estimate the ancient productivity of ancient lakes 44 . The TOC of the Xiahuapidianzi Formation varied from 0.48 to 2.35%, with an average of 1.16%.

Discussion
Volcanic activity during OAE1a. The OAE1a is derived by two mechanisms: (1) the release of massive volcanic CO 2 emission ( 13 C ~ − 5 %) 45,46 and (2) 5,45,50 . In this study, three volcanic ash layers were found (Fig. 2) within the Xiahuapidianzi Formation (121-118 Ma) as indicated by the intense and frequent volcanic activities during the Xiahuapidianzi Formation deposition. The excursion of pyrite sulfur isotopes (δ 34 S py ) and the sulfur abundance (TS) in the Xiahuapidianzi Formation (Fig. 3 and Table 1) reflect the exogenous SO 4 2− input, which has been associated with volcanic activities 20 .    (Fig. 8), which have been widely acknowledged 2,47,48,51 . Figure 8 shows that the δ 13 C org curve of the lacustrine Xiahuapidianzi Group can be correlated well with the marine curve (C1-C8) (Fig. 8). We divide the lacustrine anoxic event (LAE) period into the following three phases according to the changes in isotopes (Fig. 3) and trace elements (Fig. 7) to facilitate interpretations of changes in the terrestrial phase under the influence of global anoxia events (LAE) and comparisons with marine changes (OAE) (Fig. 8 Pre-LAE stage. From depths of 860.00 to 767.23 m (Fig. 8 Pre-LAE), the lithology is argillaceous siltstone, pebbled sandstone, silty mudstone and thin-layer mudstone, thus reflecting a strong hydrodynamic sedimentary environment. The values of δ 13 C org change from − 26.3 to − 24.8 ‰, showing an increase of 1.5‰, which may be related to increased photosynthesis. Studies on modern photosynthetic plants indicated that moderate increases in pCO 2 are beneficial to bio-photosynthesis 52 . However, if the pCO 2 gradually increases and exceeds a certain CO 2 concentration value (e.g., − 20‰ to − 35‰), then photosynthesis will be inhibited or even decease 53 . Most aquatic organisms use dissolved carbon dioxide as an enzyme substrate. When CO 2 (aq) is higher, the selectivity of organisms to 12 C is enhanced, and the δ 13 Corg value is negative excursion. When the high productivity caused by algae in the water body, the consumption of CO 2 is large and CO 2(aq) is low. When CO 2(aq) is in short supply, organisms will use 13 C-rich HCO 3 -(normal lake water temperature HCO 3 − is 7-10‰ heavier than the δ 13 Corg value of dissolved CO 2 ) as the enzyme substrate. The δ 13 Corg position excursion may prove this 54 .
The Sr/Cu and Rb/Sr ratios show a warm and humid paleoclimate condition, which is suitable for biophotosynthesis 55 . Therefore, the increase in carbon isotopes corresponds to increased photosynthesis. The TOC and rare element Ba concentrations are good proxies of paleoproductivity 44 . The TOC varied from 0.43 to 1.43 % (approximately 1.00% increase), and Ba varied from 477.4 to 794.3 µg/g (Table 3), which indicated that the increased photosynthesis led to an increase in paleoproductivity. The Mo abundance (Table 3) further suggested the increasing organic carbon burial in this abnormal interval 56 .
The δ 34 S py value increases from ~11.0 to ~13.4‰ (about 2.4‰) and then decreased to ~6.0. The TS pyr values did not show obvious change (<0.1‰) (Fig. 3 and Table 1). The δ 34 Spy increasing values were mainly caused by OM rising according to the microbial sulfate reduction (MSR) reaction. With OM growth, 32 SO 4 2− will be www.nature.com/scientificreports/ consumed, which in turn leads to increases in δ 34 S py , show a reducing environment. However, the decrease of δ 34 S py (after the short increase) may suggest an external SO 4 2− input, which Berner ascribed to volcanic activities 20 . The average Pr/Ph value is 0.59 at 860.00 m and 0.25 at 798.00 m, thus indicating a reducing environment, which was probably caused by the promotion of pCO 2 . The values of V/(V + Ni) < 0.84 and Cu/Zn < 0.52 also indicated a suboxic-dysoxic environment ( Table 3). The distribution patterns of n-alkanes are considered good indicators of the depositional environment and OM source. Peters et al. demonstrated that n-alkane ≤ nC19 is primarily associated with planktonic organisms and/or bacterial communities 18 . Cranwell thought that nC20-nC25 are mainly derived from aquatic macrophytes both marine and non-marine 57 and > nC25 are mainly derived from terrestrial plants 18,58,59 . The n-alkane distribution (860.00 m and 798.00 m in Fig. 5) of the mudstone are < nC25, indicating mixed OM source inputs from planktonic organisms, bacterial communities and aquatic macrophytes at depths from 860.00 m to 767.23 m.
The biological pump 65 66 . Large amounts of CO 2 were converted to particulate organic carbon (POC) via the biological pump. As CO 2 was consumed, the OM increased in this process. With more organic matter produced in the surface water, it consumes lots of oxygen to degradation before the settlement which will aggravate the formation of suboxic-dysoxic environment.
The initial volcanic activity moderately enhanced pCO 2 and important nutrient fluxes (such as biolimiting metals) to the continental lake, which promoted plant photosynthesis. which sustained high biological productivity 3,45 . The productivity increased in the short-term, and the TOC gradually increased. At this stage, the environment was mainly warm and humid sedimentary, and the water bodies were mainly low-oxygen freshwater. Based on these results and analyses, a pattern diagram of the pre-LAE stage (Fig. 8. pre-LAE) was established.
LAE stage. The LAE stage represents the zone at depths between 767.38 and 703.25 m (Fig. 8 LAE). The lithology is mainly silty mudstone, calcareous mudstone and argillaceous siltstone interbedded, and the hydrodynamic environment is lower than that of the previous stage. The TIC content of calcareous mudstone increased significantly and was within the range of 3.38-48.0%. Bischoff and Cummins documented that calcareous mudstone mainly formed in low-lying areas, which experienced seasonal rains and intense evaporation caused by an arid or semiarid climate 60 . In arid climate conditions, the atmospheric precipitation decreases, the water quality becomes salty, the pH value is higher, the chemical precipitation of calcite is enhanced, and "calcium" is formed. The solubility of calcium carbonate increases with the depth of water. At a certain depth of water, the rate of dissolution of carbonate will be equal to the rate of precipitation. This depth is called carbonate compensation depth (CCD). When the lake bottom is above the compensation depth of carbonate, calcium carbonate is easy to precipitate to form a calcareous layer. Otherwise, the calcium carbonate will dissolve and the calcareous layer cannot be formed. Therefore, the content of calcareous components can reflect changes in lake level up and down or changes in water depth, evolving from a pure mudstone layer to a calcareous layer, and the water depth may have a shallower trend. These may all reflect the change from warm to hot climate.
The δ 13 C org values range from − 24.8 to − 27.2‰, indicating an approximately 2.4‰ maximum decrease ( Fig. 3 and Table 1). The negative excursion of carbon isotope is a typical case of OAE1a 2,46,61 . The previous studies on marine isotopes (Fig. 8) indicated an isotopic decline of ~ 3 ‰ in the C2-C3 stage 2 . There is little doubt that the δ 13 C org variations in this stage may have resulted from the early Aptian global carbon disturbance. The volcanic activity during this period is probably the main reason for the ~ 2.4‰ carbon isotopic decline in the study area. This assumption is supported by the high atmospheric pCO 2 content during the Cretaceous due to the Ontong Java and SB-V volcanic activities 6,49,50,56 , which may contribute to the global greenhouse 45,46,50,61 .
As the proxy of paleoclimate, the ratios of Sr/Cu and Rb/Sr (Table 3 and Fig. 7) rise up, which reflect a semihumid to hot arid climate. But the ratios of Sr/Cu and Rb/Sr could be affected by the ion substitution due to similar radius of Sr and Ca 62 . Therefore, we combine biomarker and the enrichment factor (EF) for comprehensive analysis and verification the paleoclimate variation.
The n-alkane is C19 at 772.7 m and C17 at 720 m (Fig. 5), which are characterized predominantly by planktonic organisms and/or bacterial communities and little with terrestrial high plants. The C29 (m/z = 217, depth = 720.00) in Fig. 5 show an obvious low, reflect the litter source of terrestrial plants input. The biomarker results reflect that the decrease of terrestrial input, which may cause by the change of climate from warm and humid to hot and arid.
The enrichment factor (EF) could well eliminate the dilute effect of the biogenic carbonate rocks. The enrichment factor of a certain trace element can be expressed as: element factor (EF) = (C/X) sample /(C/X) PAAS , C = element, X = usually are Al or Th (PAAS = post Archean average Australian shale 63 ). If the enrichment factor is greater than 1, it indicates that the element is enriched, otherwise it is depleted. Th only presents four valences, generally not easily soluble in water, and inactive under oxidative conditions, so Th is inactive or very low in the epigenetic zone. The Sr EF, Cu EF and Ba EF increase in LAE stage (Fig. 9), show a hot and arid climate change. The ratio of B/Ga was originally reported to be related to water salinity 30,33,41 , and the average ratios were 6.69, which reflect high evaporation and increased salinity in the shrinking water body. The terrigenous detrital proxy, Zr, Ti, Th (Table 3 and Fig. 7) show a significant decrease at this stage. The increase in salinity and reduce of terrestrial debris also verify the hot and arid climate changes in the depositional environment together with calcareous mudstone petrographic analysis. Zhang, X., and Li, S got the same conclusions (hot and arid climate condition) on mudstone and shale samples by the methods of trace elements, quantitative palaeosalinity, carbon and oxygen isotopes in Jiaolai Basin, Shandong Province, Easten China 37 .
The values of δ 34  2− values for pyrite will gradually increase and even exceed the sulfate level of the lake water. This MSR reaction further reflected a reducing environment 64 . The high TS content may imply euxinic environment, support by the enrichment of U, V and Mo (Fig. 7), which could infer euxinic deposition environment with H 2 S.
The TOC values increase by 1.04% (from 0.71 to 1.75%). The content of, Ni, Cu increasing, indicate a high paleoproductivity which mainly contribute from planktonic organisms and/or bacterial communities differ from Pre-LAE stage. The Pr/Ph ratio value was 1.01 at 720.0 m, the Cu/Zn ratio was 0.50-0.63, which further indicated an anoxic condition.
The warm humid to hot arid climate weather due to the high release of pCO 2 by intense volcanic activity. The euxinic deposition environment provide a favorable environment for the storage of organic matter. Based on the above evidence and analyses, a pattern diagram of the LAE stage (Fig. 10 LAE) was established.
LAE-recovery stage. The third stage at 703.25 m and 665.10 m (Fig. 8 LAE-recovery) mainly included black mud deposits in a lacustrine sedimentary environment. In this stage, the δ 13 C org of the Xiahuapidianzi Group has undergone a successive two-stage increase with the marine C4-C6 stages (Fig. 8) after the intense decrease of δ 13 C org . The marine C4-C6 stages usually accompany black mudstone deposits (Livello Selli). A 40 m thick black mudstone deposit was observed in the OAE-recovery stage (Fig. 8), which shows a deeper and more stable sedimentary environment than the OAE stage. The value of δ 13 C org is within the range of -26.5 ‰ to -22.9 ‰, which indicates an increase of 2.6 ‰. The increase in the δ 13 C org isotope reflected the continuous consumption of CO 2 to produce OM and the termination of OAE 1a under the less active (or ceased) SB-V and OJP volcanism 49 .
The value of δ 34 S py is in the range of 3.7 to 9.1 ‰, and the value of TS pyr is between 0.42 ‰ and 0.93 ‰ ( Table 1). δ 34 S py and TS pyr still show increases as observed in the LAE stage, although the change is less than that in the LAE stage (Fig. 3) because less OM is available to support the MSR reaction. The decreased TOC value and Ba, Ni, Cu content indicate a decrease in paleoproductivity as the pCO 2 value declined. A large amount of CO 2 consumption also led to paleoclimate changes from an arid hot to warm humid climate according to the decreased ratios of Sr/Cu and Rb/Sr (Fig. 7). In the warm humid climate, the detrital input increased and paleosalinity of the lake decreased as indicated by changes in the Zr, Ti, and Th contents and Sr/Ba and B/Ga ratios ( Table 3). The ratios of Cu/Zn, U/Th, and V/V(V + Ni) indicate suboxic-anoxic paleoredox conditions. The n-alkanes distribution (C17 to C23) (Fig. 5) reflected the change in OM source from mixed resources 18,58,59 . Based on these evidences and analyses, a pattern diagram of the LAE-recovery stage (Fig. 10 LAE recovery) was established.

Conclusions
We report a new lacustrine anoxic event (LAE) under the influence of the global OAE1a event. The carbon isotope stratigraphic comparison and chronological data show that the Xiahuapidianzi Formation in the Hongmiaozi Basin is a valuable terrestrial research region for studying extreme climate changes (e.g., greenhouse and anoxic) in the Cretaceous.
The comprehensive analysis (e.g., stable carbon sulfur isotope, GC-MS and rare element analyses) of the Xiahuapidianzi Formation indicated that during the deposition period, volcanic activity led to the input of a large amount of greenhouse gases (CO 2 ), sulfate, nutrients, etc., which resulted in a significant decrease of carbon and sulfur isotopes. Due to the large amount of greenhouse gases (high pCO 2 ), the climate was significantly warm, even dry and hot climate conditions. Under these conditions, the surface runoff decreased, the lake evaporation and salinity increased, calcareous mudstone was deposited. Moreover, the dry and hot climate and high salinity www.nature.com/scientificreports/ may lead to organisms to die and quickly accumulate to form an anoxic to euxinic environment, which are beneficial to the preservation of OM (2.0% TOC). In the LAE-recovery stage, lake expansion, high productivity and rapid OM accumulation led to the favorable suboxic-anoxic environment for thick black mudstone storage.