A routine HR/ICP-MS method was used. For each analysis, a 50 mg sample of 200-mesh powder was accurately weighed and placed into a Teflon dissolving can. The sample was leached with 1 mL of HF at 150˚C and boiled to dryness to remove carbonate and calcium phosphate minerals. Then, the residues were fully dissolved in 1.0 mL of HF and 0.6 mL of HNO₃. The mixture was placed into a Teflon dissolving can and heated at 190˚C for at least 96 h. The solution was evaporated into an emulsion to remove excess HF. The residue was dissolved in 1 mL of concentrated HNO₃ and evaporated to an emulsion (this step was repeated twice). The residue was dissolved in 1.6 mL of HNO₃ and heated at 140˚C for 3 h to 5 h, and then transferred into a 50 mL centrifuge tube. The resultant heated residue was mixed with 1 mL of 500 ppb Rb internal standard, diluted to 50 mL, and analyzed by HR/ICP-MS at the State Key Laboratory for Mineral Deposits Research of Nanjing University. The analytical precision of elemental concentrations was generally better than 5%.

The same set of 200-mesh powder samples was analyzed for TOC composition. The powder samples were reacted with 1mol/L HCl in a water bath at 50˚C for at least 48 h (adding 1mol/L HCl twice) until no further reaction was observed. The solutions were washed with deionized water until pH 7, and the residue was dried and ground to 200-mesh. Samples were analyzed using a FLASH EA1112 elemental analytical instrument at the Nanjing Institute of Geography and Limnology of the Chinese Academy of Sciences.

A detailed description of the analytical techniques is given in Zhang et al. (2009) and Wen et al. (2010, 2011) [38] [39] [40] , and only a brief summary is presented here. Sample powders with an equivalent of >100 ng were oxidized at 600˚C for 8 h, and then transferred to a Teflon beaker. Samples were digested using a mixture of HF and HNO₃ (1:2) at 100˚C for at least 16 h until the samples were completely dissolved. An improved anion/cation exchange resin double- column procedure was used to separate Mo from natural samples [38] .

The Mo isotopic measurements were performed at the State Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, using an Isoprobe MC-ICP-MS. To correct isotopic drifts of the instrument, a sample-standard bracketing technique was employed [39] . The working Mo standard solution was prepared from a newly standard Mo solution (NIST SRM 3134, Merck, JMC, and Aldrich). For data presentation the δ^98/95Mo ratio is used. The external standard reproducibility is at or below 0.1‰ for the δ^98/95Mo ratio. The value of δ^98/95Mo is defined by the following Equation (1).

Mo isotopic data are presented in Table 1, and their stratigraphic trend is shown in Figure 2. The measured δ^98/95Mo values of the samples vary from −0.17‰ to 1.56‰. We obtained one negative shift at the base of the Yurtus formation (N1, sample SG0904-2, 3, 4, 5, 6), one positive shift (P1) immediately with one negative shift in the carbonaceous shale layer (N2) (Figure 2).

The trace element results and their selected ratios are shown in Table 1. Enrichment factors are calculated by normalizing each trace element to aluminium (Al) in concentration, which is assumed to represent the detrial influx, and

“/” means not measured.

then comparing these ratios to the normal shale [15] [41] . The data of standard shale are from McLennan et al. (1984) [42] . The EF values of several redox sensitive elements (such as U, Mo, and V) are given in Table 1. The Mo_EF values are from 6.22 to 438.46 (average 146.53), U_EF values are from 5.73 to 1431.41 (average 182.03), and V_EF values are from 3.15 to 159.89 (average 60.66).

The V/Cr ratios range from 0.34 to 5.71 in the chert-phosphorite beds, and from 2.20 to 8.04 in the carbonaceous shale layers; Ni/Co ratios are between 2.71 and 28.9 in the chert-phosphorite assemblages and from 13.4 to 24.0 in the carbonaceous shale beds (Table 1). Furthermore, δU (δU = 2U × (U + Th/3)) values reveal a little more consistency, and are from 1.76 to 2, indicate anoxic sedimentary conditions [43] [44] .

The TOC concentrations in the laminated chert-phosphorite assemble are from 00.7% - 0.94%, and from 0.66% to 2.79% in the carbonaceous shale layer (Table 1).

Detrital, biogenic, and hydrogenous fractions show three independent concentrations of sediments [45] . Monitoring of the detrital fraction (i.e., crustal) is necessary to assess the enrichment of redox-sensitive elements relative to their detrital component. Th is stable in the water column, and occurs permanently in the insoluble Th⁴⁺ states, and largely independent of factors such as source area and grain size [16] . Therefore, it is used in the present study to monitor detrital input [46] . The lack of correlation between Th concentrations and U, and V ratios suggests no detrital influence on redox-sensitive elements (Figure 3(a), Figure 3(b)).

Trace elements (e.g., Mo, U, and, V et al.) and their ratios (including Ni/Co, V/Cr, and δU) have been widely used as indicators for paleoredox conditions

[15] [41] [47] [48] . In anoxic waters, the Mo, U, and V values of the sediments may become highly concentrated, compared to their concentrations in sediments formed under oxygenated water columns [49] [50] [51] . U enrichment is minor in oxic-suboxic environments, where U is present mainly as U (VI) in the form of chemically unreactive uranyl carbonate complex [] [52] [53] . Under anoxic or euxinic conditions, U (VI) can be reduced to U (IV) as the insoluble uranium dioxide(UO₂) or less soluble uramium fluoride complexes [16] [53] [54] . Just like U, Mo and V is also considered sensitive to conditions around the redox boundary during early diagenesis, and is preferentially enriched in sediments underlying anoxic or euxinic waters [49] [50] [51] . The ratios of Mo_EF (average 146.53), V_EF (average 60.66), and U_EF (average 182.03) (EF = Enrichment Factor) reveal the relative enrichment than that of average shale, which indicate the sedimentary environment may not be oxygenic environment (Table 2) [49] [50] [51] .

Wignall (1994a, b) proposed to use the δU (δU = 2U × (U + Th/3)) index to distinguish the sedimentary environment of shales, with δU > 1 representing anoxic environment and δU < 1 for normal marine sedimentary environment. Based on this criterion, all samples fall into the anoxic region (Table 2) [43] [44] . Therefore, this criterion is not suitable for environment analysis of our data. It needs more rigorous criteria to constrain the sedimentary environment of the black shale series we studied.

Previous studies have established standard values for trace element ratios to distinguish oxic, suboxic and anoxic conditions (Table 2) [9] [15] [41] [54] [55] . Although different redox indicators have diverse threshold to indicate the redox conditions, there is general trend that V/Cr and Ni/Co ratios increase with decreasing oxygenation levels in water columns [9] [15] [41] [54] [55] . Figure 2 reveals that the samples in the carbonaceous shale beds fall into the suboxic- anoxic sedimentary conditions area, and most of chert-phosphorite samples are also deposit in suboxic-anoxic sedimentary conditions. The same patterns could also be observed from Ni/Co vs. Mo and V/Cr, δU vs. Ni/Co and V/Cr (Figure 4). However, the lithology of the basal Yurtus Formation is kinds of phosphatic rocks, U enrichment, V enrichment, and V/Cr ratio could not be used as indicators

of depositional conditions, due to the disturbance of the system by the substitution of V and U into apatite [11] .

Phosphorite genesis needs a favorable environment in the bottom water. Upwelling hydrothermal fluid and transgressive events could generate the proper conditions for the formation of phosphate nodules [16] . The REE results indicate that the lower Yurtus Formation shales are highly influenced by hydrothermal inputs [23] [27] . The phosphatic rocks and cherts of the basal Yurtus Formation exhibit oxic conditions with significantly negative Ce anomalies, and Ce/Ce* is in the range of 0.37 - 0.48 [27] . These significantly negative Ce anomalies are generally ascribed to organic matter derived from organisms in the euphotic zone [26] . However, the TOC concentrations of phosphatic rock and cherts are relatively low (Table 1, <0.94 wt %). Therefore, organic matter may not be the main trace element source for the phosphorites and cherts. The formation of phosphorite, especially phosphate nodules, in siliciclastic depositional systems might be related to high-energy hydrodynamic regimes, reworking and redeposition [56] [57] [58] . In addition, no sulfide was found at basal of the Yurtus Formation in the studied area. Thus, the bottom water during the early Cambrian could not be anoxic or euxin. Study on the Liuchapo Formation, southeastern Chongqing proposed by Li et al. (2015) further indicates the oxygen sedimentary conditions in early Cambrian [56] These consistent results from different areas demonstrate that oxic bottom seawater widespread in the whole basin around the Ediacaran-Cambrian boundary, and the following transgressive event induced an anoxic basin at the beginning of the early Cambrian.

Mo isotope fractionation patterns have been used to reconstruct the redox state of the Earth’s atmosphere and oceans [4] [7] . The δ^98/95Mo isotope composition of the black rock series from the Sugtebrak section, is from −0.17‰ to 1.56‰, and reveal two negative anomalies (N1 and N2) and one positive recovery (P1) over a few meters of stratigraphy thickness only (Figure 2, Table 1). Several mechanisms could contribute to the variabilities of Mo fraction, including: 1) the changes of the redox conditions (such as oxic, anoxic and strongly euxinic water-column). In oxygenated seawater, Mo is present as, and Mo isotope signature has light Mo isotopic composition (δ^98/95Mo = −0.7‰), which represent a large (~3‰) negative fractionation relative to the modern seawater composition (δ^98/95Mo_sw) of ~+2.3‰ [59] [60] [61] . A growing realization that Mo isotopes could also be fractionated during removal in intermediate redox environment where O₂ is scarce but H₂S is not abundant which is so-called suboxic or anoxic setting [14] ; 2) low-temperature hydrothermal systems. Mo may have been fractionated by precipitation of hydrothermal Fe-Mn oxides [5] [62] [63] [64] ; and 3) influx of riverine Mo in restricted ocean basins not fully connected to the global ocean circulation.

In the Sugtbrak section, the δ^98/95Mo ratios show lighter than that present-day euxinic sediments (2.3‰) [61] , and heavier than typical of Mo adsorbed onto Mn oxides (−0.7‰) [61] [65] , and fall into the range of suboxic sediments (Figure 5). The pattern of δ^98/95Mo curve has good relationships with trace element ratios (V/Cr and Ni/Co) trends (Figure 2). δ^98/95Mo negative anomaly (N1 and N2) could correspond to the suboxic-anxoic conditions which V/Cr and Ni/Co ratios indicated, and the δ^98/95Mo positive shift (P1, Figure 2) also could relative with the oxic-suboxic conditions. Wen et al. (2015) reported δ^98/95Mo values of early Cambrian black shales, which have a wide range of δ^98/95Mo values (0.27‰ - 1.79‰) in carbonaceous shales and cherts (interval 1), and are from 0.11‰ to1.70‰ in black shale layer (interval 2) in the Zunyi section, in South China [7] . Combining with Fe abundance in samples, redox elements characters, and δ^98/95Mo values, the redox conditions changed from anoxic and ferruginous in interval 1, and euxinic in interval 2 have been concluded [7] . Comparison with Zunyi section, no sulfide layer was occurred in the Sugetbrak section during the early Cambrian period. At the same time, the Early Cambrian sediments were affected by upwelling oceanic hydrothermal fluids [23] [27] . So the variation of Mo isotope in this study might not be merely connected with the suboxic or anoxic paleo-environment.

Overprint by fractioned Mo from low-temperature hydrothermal systems onto sediments could also result in the variabilities of Mo isotope composition [62] [63] . REE and Os isotope of the black shales give the clues of the hydrothermal overprint [23] [27] , indicating that the hydrothermal fluid might affect the variability of Mo isotope and the Mo enrichment in our study. At the end of Neoproterozoic, the southern Tianshan Ocean began to pull-apart because of the breakup of the Rodinia suppercontinent [23] [66] . The northern Tarim passive continental margin and southern Tianshan Ocean were developed in the Cambrian [66] . There are no deep faults developed on the northern Tarim passive continental shelf and no possibility to form local hydrothermal activity. The hydrothermal fluid in the northern Tarim continental shelf might come from extensional setting during the pattern of transgression and a rapid regional sea- level rise in the early Cambrian [23] .

Kirschvink et al. (1997) argued that marine circulation was reorganized repeatedly during the early Cambrian [67] . Nevertheless, the regional restricted basin with occasional seawater replenishment and intermittent dominance by continental input into restricted basins via riverine transport was preferred in our study. Mo contents of black rock series are generally within dozens of ppm range (Table 1), similar to black shales of the Niutitang Formation (a stratigraphic equivalent of the Yuetus Formation) from the Dingtai profile investigated by Xu et al. (2012), who argued that the Dingtai profile represents a restricted basinal environment [5] . At the same time, geophysics, geochemistry, and stratigraphic characters also support that the Tarim Basin is a kind of regional restricted basin with passive continental margin during the early Cambrian period [23] [27] [68] . The remarkable variability of δ^98/95Mo composition in the black rock series could be the expression the interplay between open seawater and continental input into restricted basins via riverine transport [59] . In the restricted basin, Mo influx is dominated by continental input, and is replenished by new seawater with heavy Mo restores the normal seawater situation [5] . The comparatively large number of samples with δ^98/95Mo below average present-day dissolved river load (0.7‰) [69] , and belong to the characters of continental margin suboxic sediments (from −0.7‰ to 1.6‰) [59] .

There is a long standing debate about the reason for significant changes in oceanic geochemistry during the Ediacran-Cambrian transition [26] [29] [30] [70] [71] . Our study researches the redox sensitive elements, TOC, and Mo isotope characters of the black rock series (cherts-phosphorite assemblages and carbonaceous shale) to evaluate the changes of sedimentary condition during the early Cambrian in the Tarim Basin. A schematic model of the depositional environment of the black rock series is shown in Figure 6. The sedimentary condition where black rock series deposited was the restricted basin and near the continental margin. And the δ^98/95Mo values also reveal the characters of continental

margin suboxic sediments (from −0.7‰ to 1.6‰) [59] .

At the beginning of the early Cambrian (Nemakit-Daldynian age), the black rock series was drowned by a transgressive event [33] . And the sea level stayed a relative low level with the first episode hydrothermal fluid. Oxic seawater conditions in combination with nutrition provided by upwelling hydrothermal fluids induced the formation of phosphorite. Trace element and Mo isotope data suggest that interbedded chert-phosphorite assembles were scavenged from seawater under restricted conditions. The inferred scavenging process could occurred under replenishment of nutrition to the restricted basins by upwelling oxidize seawater [25] [72] . High current or wave energy in combination with oxic seawater did not favor the preservation of organic matter, thus , the TOC contents of the basal Yurtus Formation are quiet low (0.05% - 0.94%) in cherts-phospho- rite assemblages layer. When the relatively sea level increase, the early Cambrian shallow sea became lack oxygen and become suboxic/anoxic. In the suboxic/anoxic conditions of the bottom water, the organic matter that fell from the photic zone partly oxidized because the residual oxygen in the seafloor. Thus, the TOC concentrations of upper Yurtus Formation carbonaceous shale are low (1.15% - 2.79%). In combination with the suboxic/anoxic conditions, another episode of hydrothermal activities promoted the enrichment of trace elements.