Strontium isotopes of authigenic carbonate potentially record sediment provenance, fluid sources, and fluid–rock interactions, little was studied on this topic in clastic strata. This study investigated clastic rocks containing authigenic calcite in the Lower Triassic Baikouquan Formation in the Junggar Basin, northwestern China. Mineral compositional and fluid inclusion analyses were conducted to constrain the precipitation processes of authigenic calcite, and the Sr contents and isotope ratios of the calcite were also measured. The authigenic calcite was precipitated at 80–140°C as the final product of thermochemical oxidation of hydrocarbons and thus has high Mn contents and highly negative δ13CVPDB values (as low as −70‰). The calcite also exhibits anomalously low 87Sr/86Sr values (0.704827, 0.706612), which are lower than contemporaneous seawater and published 87Sr/86Sr values of carbonate cements in clastic sediments, and also much lower than 87Sr/86Sr values (0.722027, 0.736750) of alkali feldspar in the strata. These low 87Sr/86Sr values record the low 87Sr/86Sr of the dominant rocks in the provenance area, such as volcanic rocks. During diagenesis, especially mesodiagenesis, the charging of hydrocarbon-bearing fluids promoted abundant dissolution of orthoclase in the alkali feldspar detritus, releasing radiogenic 87Sr into the pore waters, and eventually increasing the 87Sr/86Sr values in the late-stage calcite that precipitated after this reaction. This inference is consistent with the positive correlation between the calcite 87Sr/86Sr ratios and the dissolution intensity of orthoclase. In regions that do not undergo hydrocarbon-charging and where orthoclase remains stable, the lower 87Sr/86Sr ratios of the calcite generally record the provenance. For authigenic calcite associated with intense fluid–rock interactions, the higher 87Sr/86Sr ratios reflect the enhanced dissolution intensity of 87Sr-rich minerals such as orthoclase. Therefore, combined with a petrological study, Sr isotopes of authigenic carbonate in clastic sediments can trace sediment provenance and intensity of fluid–rock interactions.

Strontium isotope composition is a robust tracer of sediment provenance [1], fluid sources [2, 3], and fluid–rock reactions [4, 5]. The 87Sr/86Sr signature of ancient seawater can serve as a proxy for understanding the tectonic evolution of the Earth system [6, 7] as well as a tool for stratigraphic correlation [8-10]. The 87Sr/86Sr of seawater records the relative importance of two major strontium fluxes: (a) the riverine input of radiogenic Sr due to continental weathering and (b) the “mantle Sr” from hydrothermal circulation at mid-ocean ridges [1, 10]. Strontium isotopic composition of pore water mainly reflects fluctuations of fluid sources, pathways, mixing [2, 3], and fluid–rock reactions [4, 5]. Therefore, Sr isotope was also used in studies investigating the evolution of hydrology and paleohydrology [2, 11], diagenetic reactions, and the burial history of the sediments [12]. The 87Sr/86Sr ratios of primary sedimentary carbonate, particularly those not modified by fluid–rock interactions, potentially record continental weathering and the properties of the initial sedimentary pore-waters. For example, the 87Sr/86Sr ratios of carbonates and stream waters in a Paleocene–Eocene Lake basin allow the evolution of the provenance area and paleohydrology to be characterized [13]. The 87Sr/86Sr values of marine carbonate that has experienced weak or no diagenesis reflect the properties of ancient seawater [14].

Authigenic carbonates, including calcite, dolomite, and siderite, precipitated inorganically at the sediment–water interface or within sediment pore waters, when diagenetic reactions result in supersaturation of carbonate minerals [15, 16]. The origins of authigenic carbonate are complex in clastic strata, and the timing of precipitation varies from early to late diagenesis due to variations in pore fluids, diagenetic processes, input of exogenous fluids, and formation temperatures and pressures [15-18]. During burial, in situ precipitated authigenic carbonate inherits pore-water Sr and its isotopic compositions [6, 19]. The dissolution and reprecipitation of carbonate minerals generally do not significantly change the 87Sr/86Sr ratios of pore waters in a shallow burial system lacking exogenous fluids [13, 20]. However, the input of exogenous fluids, such as meteoric waters or mantle-derived fluids, can affect the properties of pore water and modify the Sr isotopic composition of authigenic carbonate. The fluid–rock interactions involving noncarbonate sediments, predominantly silicate minerals, can also obviously change Sr isotopes of pore water [8, 12]. The spatial heterogeneity of rock porosity and permeability results in spatial variations in the intensity of fluid–rock interactions in clastic strata [21]. The Sr isotope ratios of authigenic carbonate in sedimentary rocks possibly retain information on sediment provenance, but also likely record fluid charging and fluid–rock interactions.

To explore the potential of Sr isotopic ratios of authigenic calcite as a proxy for sediment provenance and fluid–rock interactions in clastic rocks, we investigated the Lower Triassic Baikouquan (T1b) Formation in the Junggar Basin, northwestern China. This formation has a clearly defined provenance area and distribution of source rock types. In addition, the 87Sr/86Sr ratios of the various source rock types vary widely from 0.703074 to 0.745750 in the provenance area [22, 23]. Notably, the charging intensity of hydrocarbon-bearing fluids and fluid–rock interactions in the Baikouquan Formation were spatially variable. The extent to which the 87Sr/86Sr ratios of authigenic calcite were affected by these external geological fluids and diagenesis is unclear. Therefore, based on the mineral compositions, fluid–rock interactions, and Sr elemental and isotopic data, this paper focuses on (a) whether the 87Sr/86Sr ratios of authigenic calcite still record the sediment provenance of the clastic strata and under what conditions this is the case and (b) whether the 87Sr/86Sr ratios of authigenic calcite can record the intensity of typical fluid–rock interactions.

The Junggar Basin is located in northwestern China and covers an area of 1.3 × 105 km2 (Figure 1(a)). The basin is located in the center of the Central Asian Orogenic Belt (CAOB). The CAOB is one of the largest and longest-lived accretionary orogens in the world and was responsible for Neoproterozoic and Phanerozoic crustal growth [24, 25]. In this orogenic belt, three collage systems were generated in the Neoproterozoic and amalgamated in the Permian–Triassic [24]. The study area is located in the western Mahu Sag adjacent to the West Junggar terrane. The West Junggar terrane was the sediment source for the study area (Figure 1) and was folded into the present-day orocline during the late Permian to Early Triassic [26, 27]. West Junggar and the adjacent western Junggar Basin are interpreted to have been a Mariana-type arc system driven by northwestward subduction in the Junggar Ocean during the Paleozoic [28]. In the eastern part of West Junggar, this tectonism formed thrust nappe structures [29], voluminous Ordovician to Carboniferous volcanogenic sediments, early Paleozoic ophiolites (561, 332 Ma), late Paleozoic granitoids (A- and I-type; 436–276 Ma), and Carboniferous basalts (345, 297 Ma) [23, 29-33]. Most of these rocks are characterized by low initial 87Sr/86Sr ratios (0.7023, 0.7054) [22, 31]. During the early Permian (ca. 280 Ma), the rotation of West Junggar modified the geometry of the convergent zone into a convex shape toward the Junggar Basin, which may have favored the formation of strike-slip faults and lateral displacement of units in West Junggar [26]. During the end-Permian to Early Triassic, activity on Permian strike-slip faults (e.g., the Dalabute Fault) produced the present-day geometry of the basin [26, 34]. The Zhayier and Hala’alate mountains adjacent to the study area were also uplifted at this time (Figure 1(a)) [24].

The Mahu Sag is a NE–SW-trending tectonic depression in the northwestern Junggar Basin, adjacent to West Junggar (Figure 1(a)). The basement of the sag is a late Carboniferous arc–basin system, which is overlain by Permian–Cenozoic terrestrial strata [35]. The Carboniferous to Permian strata are distributed across the entire study area, whereas the Triassic to Cretaceous strata thin gradually from the depocenter to marginal fault belts. The studied Baikouquan Formation unconformably overlies the upper Permian Upper Wuerhe Formation (P3w) and underlies the Karamay Formation (T2k), with no depositional hiatuses.

The uplift of the eastern part of West Junggar led to the deposition of coarse-grained clastic sedimentary rocks (mainly pebbly conglomerates) in the Baikouquan Formation. The deposits are fine eastward and comprise gravelly conglomerates, sandstones, and mudstones deposited in lacustrine and retreating fan delta settings in the sag area (Figure 1(a)) [21]. Three fan delta complexes are located in the study area, which are the Karamay Fan (K. Fan), Huangyangquan Fan (H. Fan), and Xiazijie Fan (X. Fan), from south to north. The Baikouquan Formation can be divided into three fining-upward members, which are T1b1, T1b2, and T1b3 from base to top. T1b1 comprises conglomerates interbedded with thick massive mudstones and some thin sandstones; T1b2 contains more mudstones and sandstones, although conglomerates are also present; T1b3 consists mainly of massive mudstones with only a few conglomerates and sandstones. Authigenic calcite occurs widely in variable contents in the T1b conglomerates and sandstones, and is characterized by highly negative δ13C values (−70‰ to −22.5‰) and high Mn contents (MnO = 0.79–14.67 wt%) [36]. It is thought to be the final product of the thermochemical oxidation of hydrocarbons, including methane, caused by high-valence metal oxides. During diagenesis, hydrocarbons from deeper source strata were abiotically oxidized by high-valence Mn–(Fe) oxides at 90–135°C, releasing 13C-depleted CO2 and soluble Mn2+ (Fe2+). Both CO2 and metal ions were then incorporated into authigenic calcite with elevating pH values of pore water [36].

The current depth of the Baikouquan Formation increases from 2850 m in the west to 4500 m in the east of the Mahu Sag (Figure 2(a)). Over 100 million tons of proven oil reserves have been discovered in the T1b Formation in recent years [21]. The oil and gas are thought to be mainly derived from the organic-rich Lower Permian Fengcheng Formation (P1f) [37]. The Baikouquan Formation reservoir rocks were charged during the Early Jurassic and Early Cretaceous along steeply dipping faults, although hydrocarbon charging in the Early Jurassic was limited in scale compared with the Early Cretaceous [37].

It is noted that these strata host different oil and gas accumulations. Within the study area, the Mahu Fault was the main migration conduit from the source rock into the T1b reservoir, while the Eric Lake Fault to the west was not large enough to serve as a substantial hydrocarbon conduit (Figure 2(a) and (b)). Since the well M18 region is located on the hanging wall of the Mahu Fault, it is enriched in oil and gas. However, to the west of the Eric Lake Fault, petroleum accumulations in the well AH2 region decreased substantially. The T1b reservoir rocks are unconventional reservoirs with strong spatial heterogeneity in both porosity and permeability [21]. There is no obvious oil–water interface in the reservoirs. No water is produced from the wells, except for individual wells in the sag center like well Mzh 1 (Figure 2(a)). Limited formation water basically exists as irreducible water on the debris surface [37, 38]. These reservoirs are tight unconvetional hydrocarbon accumulations [39]. Light oil-containing gas is concentrated in isolated sandy conglomerates in the fan-delta front subfacies with higher porosities and permeabilities in each member (Figure 2(b)). The precipitation of calcite took place randomly in the strata overall. Strontium elements could not be transported during gas diffusion. Thus, the 87Sr/86Sr ratios of the T1b calcite were not mainly controlled by the hydrocarbon-bearing fluids from the P1f source rock.

The sedimentary units in the Mahu Sag experienced a relatively simple burial history. After its deposition, the Baikouquan Formation did not experience significant uplift or tectonic activity [40]. However, the geothermal gradient decreased gradually throughout its geological evolution from 36.3°C km–1 in the late Permian to 22.8°C km–1 by the Neogene [41]. The Baikouquan Formation temperature was ~80°C during the first interval of hydrocarbon charging and >100°C in the second interval during the Early Cretaceous.

3.1. Samples

Detailed lithological observations and logging were conducted on 154.2 m of cores from nine wells (Figure 1(a)). Of these wells, the well M18 allowed the most detailed sampling because it has the longest cores. Eighty-six samples were collected, including thirty-nine samples from well M18 and forty-seven from the other wells.

3.2. Petrological and Mineral Analysis

To identify the rock textures and mineral assemblages, thirty-five samples were injected with blue epoxy and then made into polished thin sections. The thin sections were examined with a Nikon Eclipse E600WPL optical microscope. Fifteen samples were further subjected to microstructural and mineralogical study using a field emission scanning electron microscope (FE-SEM) and back-scattered electron (BSE) imaging with an electron probe microanalyzer (EPMA). The major element contents of authigenic minerals were also determined with the EPMA to determine the geochemical characteristics of the pore waters. The FE-SEM was a Carl Zeiss Supra 55 instrument operated at 5 kV, 30 μm standard grating, and 40 s counting time. The EPMA was a JEOL JXA-8230 instrument operated at 15 kV, 20 nA beam current, 30 s scanning time for BSE imaging, 20 s counting time, and 2 μm beam size.

3.3. Fluid Inclusions in Calcite

The homogenization temperatures of two-phase aqueous inclusions in authigenic calcite record the temperature at which calcite cement precipitates [42]. Aqueous inclusions without hydrocarbons were identified by fluorescence imaging. Microthermometric analyses were carried out using a Linkam THMS600 heating–freezing stage, with a temperature range of −195 to +600°C, which is mounted on a DMLB Leica microscope. The stage was calibrated by measuring the melting points of pure water inclusions (0°C), pure CO2 inclusions (−56.6°C), and potassium dichromate inclusions (398°C). The accuracy of the measured homogenization temperatures was approximately ±2°C between room temperature and 600°C.

3.4. Strontium Contents and Isotopes

Trace element contents, including Sr, in authigenic calcite were determined with a Finnigan Element II HR-ICP-MS at the State Key Laboratory for Mineral Deposits Research, Nanjing University. About 100 mg of powdered sample was leached by 1 M acetic acid to dissolve the carbonate fraction. The detailed procedure followed reference [43]. The analytical precision for Sr content is estimated to be ±10% based on duplicate analyses of samples and GSR-12 (GBW-07114; a Chinese national rock standard).

The strontium isotopes of authigenic carbonate were determined by thermal ionization mass spectrometry (Finnigan Triton TI) in the same laboratory. To obtain the carbonate fractions, the sample powders (~200 mg) were reacted with 2 mL of 1 M acetic acid for 15 minutes in an ultrasonic bath at room temperature and then centrifuged. The dissolution was repeated three times in Teflon beakers, and the supernatants of the last two steps were dried down for further analysis. The first pre-leaching is necessary as it generally reduces, or completely avoids the contaminant of radiogenic Sr on clay ion-exchange sites [44, 45]. The dried carbonate fractions were re-dissolved in 1 mL of double-distilled 4 M HCl, centrifuged, and then the supernatant was loaded directly on cation exchange columns (AG 50WX8; 2 mL columns; 2.5 M HCl eluent). The Sr fraction was collected in ~7 mL of 4 M HCl and dried down. Given that orthoclase likely contributes the most radiogenic Sr due to its high Rb content [46], alkali feldspar (50 mg) picked from the Baikouquan Formation rocks was dissolved in Teflon beakers using HF + HNO3 and Sr was also separated by the cation exchange method. The detailed procedures are the same as those described by reference [47]. After Sr separation, the mass spectrometry measurements followed the procedures of reference [48]. The precision on the 87Sr/86Sr ratios was <1 × 10−5. Measurement of the 87Sr/86Sr ratios for the NIST9 87Sr standard during this study yielded an average of 0.710246 ± 0.000006 (2σ; n = 11).

4.1. Petrological Features

Because West Junggar contains voluminous igneous rocks, the Baikouquan Formation conglomerates and sandstones are rich in igneous detritus, including detritus derived from tuff, andesite, basalt, and granite (Figure 3). The sand component (grain size <2 mm) in the conglomeratic lithofacies comprises 45%–90% rock fragments (average = 59%), 8%–38% quartz (average = 19%), and 7%–32% feldspar (average = 21%; n = 17). The rock fragments consist of 60%–75% tuff, 6%–28% granite, 6%–15% andesite, and 2%–5% rhyolite. The content of detrital granite is higher in the H. and K. fans (15%–28%), as compared with the X. Fan (<10%). The detrital components of the sandstone lithofacies are similar to those of the conglomerates. The contents of rock detritus in the sandstones from the X. Fan are 53%–91% (average = 78%; n = 3) and exhibit a low maturity. In contrast, the rock detritus contents in the sandstones from the H. Fan are 38%–80% (average = 56%; n = 6), and the feldspar content is higher. Within a given fan system, the detrital components are similar for the same lithofacies. In addition, the feldspar detritus in the Baikouquan Formation is almost all alkali feldspar, including perthite, antiperthite, microcline with cross-hatched twinning, and rare anorthoclase.

4.2. Diagenesis

With a gradual increase in burial depth, the Baikouquan Formation experienced obvious mechanical compaction (Figure 3). With the increase in formation temperature, the unstable minerals in the volcanic debris were altered into authigenic minerals, such as chlorite, and the smectite in the rock matrix was gradually altered to mixed-layer illite–smectite (Figure 4). A typical diagenetic phenomenon was the selective dissolution of orthoclase within alkali feldspar, whereas albite remained stable and was even authigenically precipitated. Orthoclase exhibits a honeycomb-like texture as the result of dissolution, with a final resultant dissolved layer parallel to the (001) lattice plane (Figure 4(a)–(c)), which eventually formed abundant secondary pores along the cleavage planes (Figure 3). In contrast, the lattice planes of albite are smooth and without any traces of dissolution. In fact, authigenic albite is present (Figure 4(a)–(d)). Most of the feldspar dissolution pores are distributed along the orthoclase lamellae in perthite or antiperthite, and microcline (Figure 4(a)–(c)). The orthoclase dissolution exhibits differences throughout the formation, with the dissolution intensity decreasing gradually upward from the T1b1 to T1b3 members. As such, the orthoclase content also increases up-section, whereas the albite content remains largely constant (Table 1).

Extensive dissolution can result in significant mineral precipitation later in time or at other locations. The orthoclase dissolution was generally accompanied by precipitation of kaolinite, quartz, and chlorite (Figure 4(d)–(f)). Kaolinite is randomly distributed on the surfaces of partly dissolved grains of orthoclase or in cleavage cracks as hexagonal tablets and also occurs as book- or worm-like aggregates (Figure 4(e)). Small flakes of chlorite occur mainly in feldspar cleavage cracks or as fish-scale-like grains that fill inter-particle pores (Figure 4(f)). Authigenic quartz is also common in the inter-particle pores, occurring as short, columnar, hexagonal crystals associated with kaolinite or chlorite (Figure 4(d) and (f)).

Authigenic calcite occurs widely in the Baikouquan Formation conglomerates and sandstones (up to 6 vol%, but mostly <2 vol%). Calcite occurs mainly as poikilotopic cement in primary intergranular pores, and locally as subhedral crystals filling secondary pores that resulted from feldspar dissolution in the conglomerates and sandstones (Figure 4(g)–(h)).

4.3. Fluid Inclusions

To constrain the precipitation stages of calcite, the homogenization and freezing point temperatures of fluid inclusions in calcite were determined. Before measurement, the calcite stages were identified via BSE imaging. The homogenization temperatures of primary, two-phase, aqueous inclusions in the two calcite stages cluster into two ranges (80–105°C and 105–140°C; online Supplementary Material Table 2, Figure 5), corresponding to their precipitation temperatures.

4.4. Strontium Contents and Isotopes

In the Baikouquan Formation, the Sr contents of calcite range from 3152 to 66,056 ppm, while their 87Sr/86Sr ratios are generally low (0.704827, 0.706612; Table 1), compared with contemporaneous seawater (0.707270, 0.707876) [49]. The 87Sr/86Sr ratios of the two stages of calcite do not exhibit obvious differences. Amongst the three fan systems, the 87Sr/86Sr ratios of the calcite from the H. and K. fans are slightly higher than those of the X. Fan. For the H. Fan, the 87Sr/86Sr values of the calcite from the oil-producing sections are generally higher than those from the non-oil-producing sections. In contrast to the authigenic calcite, the alkali feldspar detritus in the studied strata typically have high 87Sr/86Sr values (0.722027, 0.745311; Table 2).

5.1. Calcite Formation During Diagenesis

Based on petrological features and fluid inclusion data, the calcite in the Baikouquan Formation can be divided into two major stages: late-stages I and II. Late-stage I calcite is dark gray in BSE images, and occurs as subhedral crystals filling poorly connected inter-particle or feldspar dissolution pores (Figure 4(g)–(h)). Its content is low in whole-rock samples (mostly <2 vol%). This calcite stage is characterized by high Mn contents (MnO = 1.26–4.66 wt%). The homogenization temperatures of aqueous inclusions in this calcite stage vary from 80 to 105°C (Figure 5). In contrast, late-stage II calcite is bright gray in BSE images, and occurs as coarse-grained crystals with well-developed crystal planes in large inter-particle pores (Figure 4(g)). This calcite stage fills the center of inter-particle pores, whereas the late-stage I calcite fills the rim areas of inter-particle pores (Figure 4(g)), clearly demonstrating that late-stage II calcite precipitated after late-stage I calcite. This calcite stage is characterized by higher Mn contents (MnO = 4.92–13.84 wt%; online Supplementary Material Table 1). The precipitation processes of the calcite stages can be determined based on the burial history and associated diagenetic phenomena.

The Baikouquan Formation underwent early diagenesis in the Early Triassic to Early Jurassic at formation temperatures of <80°C, prior to hydrocarbon charging [21, 41]. Due to the increasing thickness of the overlying strata, the rocks underwent mechanical compaction, and pore fluids from the mudstones were released into the reservoirs. The feldspar clasts include perthite, antiperthite, and microcline, and were not, or only slightly, dissolved due to the lack of fluid acidity. As such, little authigenic kaolinite formed.

After the first hydrocarbon charging event during the Early Jurassic, CH4 and C2+ hydrocarbons were thermochemically oxidized by high-valence Mn–Fe oxides, which led to the formation of 13C-depleted CO2 at 80–105°C (Figure 6) [36] and probably other alteration products like short-chain alkanes or carboxylic acids [49]. Adjacent to permeable faults, minor organic acids generated in source rocks during initial thermal maturation probably also migrated into the reservoirs together with hydrocarbons [50, 51]. The CO2 and organic acids reduced the pH of the formation waters and accelerated the dissolution of alkali minerals, including orthoclase, while albite remained stable. This released large amounts of K+ into the formation waters, which caused continuous illitization of smectite in the reservoirs. However, due to the limited hydrocarbon charging and relatively low formation temperature as compared with the second charging interval, the intensity of orthoclase dissolution remained limited as well as the kaolinite precipitation [52]. Alkali mineral dissolution buffered the H+ content of carbonic and carboxylic acids in the formation waters, and the increasing pH of the formation waters triggered the precipitation of late-stage I, Mn-rich, and 13C-depleted calcite.

During the large-scale emplacement of hydrocarbons in the Early Cretaceous, the formation temperature reached 105–140°C [41]. The higher temperature facilitated intense oxidation of hydrocarbons (mainly methane), and caused the formation of CO2, which reacted with dissolved Mn2+ and Fe2+ in the formation waters [36]. At this higher temperature, the long reaction period (>120 Myr) after hydrocarbon emplacement caused Fe2O3 reduction [53] and precipitation of Fe-rich chlorite. The decrease in formation water pH values due to dissolution of CO2 caused the nearly complete dissolution of orthoclase in the high-permeability sandy conglomerates, whereas albite remained stable due to the continuing conversion of smectite into illite [52]. Associated with orthoclase dissolution, abundant authigenic kaolinite was precipitated along with authigenic chlorite, quartz, and albite. Late-stage I calcite was also partly dissolved during this process. The alteration of mineral assemblages consumed the H+ in the formation waters, which increased the pH and led to the precipitation of late-stage II 13C-depleted calcite. Although the two calcite stages precipitated in different diagenetic sequences, there are no obvious differences in 87Sr/86Sr ratios.

5.2. Low Calcite 87Sr/86Sr Ratios Due to Sediment Provenance

In both continental and marine clastic strata, the 87Sr/86Sr values of authigenic calcite are generally higher than that of contemporaneous seawater [1]. However, the 87Sr/86Sr ratios of the Baikouquan Formation calcite are significantly lower than those of contemporaneous Early Triassic seawater (0.707270, 0.707876) [49]. The values are lower than all currently reported 87Sr/86Sr values of authigenic carbonate in clastic rocks within sedimentary basins worldwide and in modern marine clastic sediments (Figure 7). The low 87Sr/86Sr values indicate that the Baikouquan Formation may have experienced the injection of geological fluids with low 87Sr/86Sr ratios, such as fluids from the deep mantle or hydrocarbon-bearing fluids, or that the rocks in the provenance area are generally depleted in radiogenic Sr [6, 54]. However, there is no evidence for the former presence of mantle-derived fluids in the Baikouquan Formation. Deep faults do not occur in the study area [37]. As a result, the formation was not susceptible to the effects of mantle-derived fluids. Typical hydrothermal minerals related to deep mantle-derived fluids, such as fluorite and saddle dolomite, have not been found. In addition, the emplacement of hydrocarbon-bearing fluids would not cause the systematically low 87Sr/86Sr ratios of the Baikouquan Formation calcite. The authigenic carbonate in the main source rocks (Fengcheng Formation) for the Baikouquan Formation reservoirs has low 87Sr/86Sr ratios, ranging from 0.7041 to 0.7059 [54, 55]. After the Fengcheng Formation source rocks entered the oil generation window, the hydrocarbon-bearing fluids migrated into the Baikouquan Formation in two main stages during the Early Jurassic and Early Cretaceous [37]. This may have resulted in the low 87Sr/86Sr ratios of calcite in the oil-charged areas, because the hydrocarbon-bearing fluids were depleted in radiogenic Sr. However, in the whole study area, the scale of oil charging was limited, and in the areas not affected by hydrocarbon charging, this process could not cause the low 87Sr/86Sr values of the calcite. The 87Sr/86Sr values of calcite from the non-oil-producing sections, such as well regions of AH2 and AH4, also have lower values of 0.704849–0.706055. In contrast, the 87Sr/86Sr values of calcite from sections with higher oil saturation, such as the T1b1 Member, are slightly higher than the two overlying members with low oil saturation (Table 1). This indicates that the emplacement of hydrocarbon-bearing fluids and the charging intensity were not the main reasons for the anomalously low calcite 87Sr/86Sr values.

In addition, formation water could carry abundant Sr with lower 87Sr/86Sr ratios [3, 11], likely causing low 87Sr/86Sr ratios of calcite. However, no water is produced from the wells in the study area, and limited formation water exists in the form of irreducible water on the debris surface [37, 38]. Thus, the action of formation water alone could not form the calcite low 87Sr/86Sr ratios.

The sediment provenance, weathering, and sedimentation processes determine the lithofacies and mineral composition of clastic strata, and also affect the chemical compositions of initial pore waters inherited from ancient sedimentary environments and the fluid–rock interactions during diagenesis [12, 18]. Given the similar detrital components in the same lithofacies, the provenance supply likely dominated the initial mineralogy of the Baikouquan Formation rocks. Its provenance area was the ancient mountains and adjacent areas in the eastern part of West Junggar [21]. West Junggar hosts voluminous Ordovician to Carboniferous volcanic rocks, early Paleozoic ophiolites, late Paleozoic A- and I-type granitoids, and rare late Carboniferous basalts [23, 29, 31]. Their 87Sr/86Sr values exhibit significant differences. For the Silurian–Devonian and early Carboniferous volcanic rocks, the values are 0.7079–0.7161 [56] and 0.7044–0.7083 [23]; for the Ordovician–Devonian ophiolites, the values are 0.7047–0.7067 [57]. The ratios vary more widely from 0.7042 to 0.7565 for the late Carboniferous granites [22, 58] but are 0.7047–0.7067 for the late Carboniferous basalts [31]. The 87Sr/86Sr values of the late Carboniferous granites also vary in different regions, and are 0.7042–0.7086 in the Karamay region, 0.7046–0.7310 in the Miaoergou region, and 0.7208–0.7565 in the Akbastao region (Figure 1(a)) [22, 58]. The initial 87Sr/86Sr values for these rocks in the Early Triassic (250 Ma) were calculated (Figure 8). Based on the relative weathering areas of different rock types in Western Junggar (Figure 1(a)), the low 87Sr/86Sr values of the calcite (0.704827, 0.706612) indicate the provenance of the Baikouquan Formation was mainly from the early Carboniferous volcanic rocks and late Carboniferous granites in the Karamay region. Small amounts of Sr might have been derived from basalt and ophiolite with a limited exposure area, while the inputs from Silurian–Devonian volcanic rocks, and the Miaoergou and Akbastao granites were limited. This is consistent with the detrital components and abundances in the strata. The abundant occurrence of tuff derived mainly from early Carboniferous volcanic rocks likely dominates the overall low 87Sr/86Sr values of the calcite in the three fan systems. Moreover, within the study area, the 87Sr/86Sr values of calcite in the H. and K. fans are similar and slightly higher than that in the X. fan. This is likely due to the higher input flux of late Paleozoic granites in the H. and K. fans (Figure 1(a)). In general, compared with carbonate sediments and authigenic carbonate in marine environments, the authigenic calcite in continental clastic strata is more affected by the sediment provenance.

5.3. Increasing Calcite 87Sr/86Sr Values Due to Orthoclase Dissolution

In the Baikouquan Formation, the typical fluid–rock interaction was the selective dissolution of alkali feldspars (i.e., orthoclase underwent dissolution, whereas albite remained stable) (Figure 4(a)–(c)). This was accompanied by the precipitation of abundant secondary minerals, such as kaolinite, Fe‐rich chlorite, quartz, and albite (Figure 4(d)–(f)). Orthoclase is a typical Rb-rich mineral, in which the radioactive decay of 87Rb to 87Sr increases its 87Sr/86Sr ratio with time [59]. As described above, the 13C-depleted Baikouquan Formation calcites precipitated after oil and gas charging during late diagenesis. The abundant orthoclase dissolution likely modified the Sr isotopic composition of the late-stage calcite in the studied strata.

The intensity of orthoclase dissolution is variable throughout the Baikouquan Formation (i.e., the dissolution is most extensive in the T1b1 Member), and the dissolution intensity decreases gradually upward from member T1b1 to T1b3. As such, the orthoclase contents also increase up-section, whereas the albite contents remain constant. The detrital and mineral compositions of the rocks in the different members are similar, showing that the provenance did not change much up-section. From the base of the formation upward, orthoclase was completely dissolved in most samples from member T1b1, with its content increasing from below the detection limit to 4 wt% (average = 2.2 wt%). Up-section into member T1b2, the orthoclase contents of the sandy conglomerates increase from 3 to 11 wt% (average= 7.5 wt%). In member T1b3, the orthoclase contents vary from 6 to 13 wt% (average = 9.9 wt%). This can be explained by the gradual weakening of the charging intensity of acidic hydrocarbon-bearing fluids [52]. In the H. Fan, the 87Sr/86Sr values of the T1b1 calcites vary from 0.705370 to 0.706377 (average = 0.706105), which decrease from 0.705220 to 0.706612 (average = 0.705887) and 0.704849 to 0.705962 (average = 0.705242) for the T1b2 and T1b3 samples, respectively. From member T1b1 to T1b3, with decreasing dissolution intensity, the orthoclase contents increase gradually upward, whereas the 87Sr/86Sr values of calcite decrease. This feature is typical in oil-producing wells like well M18 (Figure 9). In the whole study area, the orthoclase content is also negatively correlated with the 87Sr/86Sr values of authigenic calcite (Figure 10). For the H. Fan, the correlation coefficient reaches −0.83 (P value < 0.01). This demonstrates that the pore fluids that precipitated calcite were also modified by strong fluid–rock interactions in sections where orthoclase dissolution was intense, which increased the 87Sr/86Sr values of the calcite.

The 87Sr/86Sr ratios of formation waters could be significantly modified by different fluid–rock interactions and reaction intensities, especially during the dissolution of minerals enriched in radiogenic Sr, such as orthoclase [59]. The 87Sr/86Sr values of detrital feldspar in the studied strata are 0.7250–0.7360. After dissolution, the radiogenic Sr in the orthoclase crystals was released into the pore waters, causing an increase in its 87Sr/86Sr ratio, which in turn led to an increase in the 87Sr/86Sr values of the late-stage calcite. During early diagenesis, prior to calcite precipitation, the Baikouquan Formation temperature increased gradually due to the increasing burial depth. The alteration of unstable minerals in the volcanic rocks and granites with low 87Sr/86Sr values from West Junggar resulted in the low 87Sr/86Sr values of the pore fluids in the Baikouquan Formation. This is evident in the sections where obvious orthoclase dissolution did not occur, such as the T1b3 Member, which contains calcite with low 87Sr/86Sr values (0.704849, 0.705962; average = 0.705242). During late diagenesis, the orthoclase was dissolved heterogeneously due to the differential charging of oil–gas-bearing fluids in the Early Jurassic and Early Cretaceous [52]. In the sections with strong orthoclase dissolution, such as T1b1, large amounts of radiogenic Sr were released into the pore fluids by orthoclase dissolution and eventually caused the higher 87Sr/86Sr values of the calcite (0.705370, 0.706377; average = 0.706105). These effects of fluid–rock interactions on the Sr isotopic compositions of authigenic calcite are typical in the studied strata. Therefore, the 87Sr/86Sr ratios of authigenic calcite can record the dissolution intensity of 87Sr-rich minerals, such as orthoclase, in clastic sediments.

Based on a detailed petrological study, and Sr contents and isotope ratios of calcite in the Baikouquan Formation, this study investigated the validity of using 87Sr/86Sr values of authigenic calcite as a tracer for sediment provenance and fluid–rock interactions in clastic strata. This formation is enriched in volcanic and granite detritus, which was derived from the adjacent West Junggar terrane. In the Baikouquan Formation, authigenic calcite formed in two late stages and has low 87Sr/86Sr values of 0.704827–0.706612. These values are not only lower than that of contemporaneous seawater but also lower than currently known 87Sr/86Sr values for authigenic carbonates in clastic strata worldwide and 87Sr/86Sr values of detrital feldspar in the strata. During diagenesis, before hydrocarbon-bearing fluid charging, the Sr isotopes of the pore waters were constrained by the initial mineralogy of the provenance rocks. These were mainly early Carboniferous volcanic rocks and late Carboniferous granites around the Karamay area. All these rocks are depleted in radiogenic 87Sr. As a result, the 87Sr/86Sr values of the T1b authigenic calcite are generally low. After hydrocarbon charging, orthoclase was preferentially dissolved in the study area, and the Sr isotopes of late-stage calcite precipitated after this reaction were controlled by both the initial provenance and Sr isotopes released by orthoclase dissolution. The 87Sr/86Sr values of the calcite are systematically higher in the sections that experienced more intense orthoclase dissolution, indicating that orthoclase dissolution supplied radiogenic 87Sr to the pore waters. This study demonstrates that the 87Sr/86Sr values of authigenic carbonate can effectively trace the sediment provenance and intensity of fluid–rock interactions in clastic strata.

All data mentioned in this work are available in the Supplementary materials.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This study was supported by the National Natural Science Foundation of China (Grant nos. 42272179, U23B20155, and 41830425). We extend our gratitude to Huan-Ling Lei and Yi-Bo Lin for their assistance with strontium isotope analysis, and Wen-Lan Zhang for her assistance with EPMA element analysis. We also thank associate editor Jiyuan Yin, and two anonymous reviewers for their constructive comments on this paper.

The data used in this study are available in the supplementary files. The supplementary dataset contains two Excel tables. Appendix Table 1 provides major element contents of the calcite in the Baikouquan Formation. Appendix Table 2 exhibits the homogenization temperatures (Th) and calculated salinity of aqueous inclusions in calcite.

Exclusive Licensee GeoScienceWorld. Distributed under a Creative Commons Attribution License (CC BY 4.0).

Supplementary data