Sr, S, and O Isotope Compositions of Evaporites in the Lanping–Simao Basin, China

: Evaporites are widely distributed within continental “red beds” in the Lanping–Simao Basin, west Yunnan, China. Sr (Strontium), S (Sulfur), and O (Oxygen) isotope compositions have been measured on 54 sulfate or/and sulfate-bearing samples collected from Lanping, Nuodeng, Jinggu, Mengyejing, Baozang throughout the Lanping–Simao Basin. The 87 Sr/ 86 Sr ratios of all samples (0.708081 to 0.710049) are higher than those of contemporaneous seawater, indicating a signiﬁcant continental contribution to the drainage basin. Sulfates in the Lanping Basin have higher 87 Sr/ 86 Sr ratios (0.709406 to 0.710049) than those (0.708081 to 0.709548) in the Simao Basin. Nevertheless, the δ 34 S values of gypsums (13.4‰ to 17.6‰) in Lanping and Baozang fall within the range of Cretaceous seawater. Gypsums from a single section in Baozang have trends of decreasing δ 34 S values and increasing 87 Sr/ 86 Sr ratios from base to top, indicating continental input played an increasingly signiﬁcant role with the evaporation of brines. High δ 34 S values (20.5‰ to 20.7‰) of celestites in Lanping are probably caused by bacterial sulfate reduction (BSR) process in which 34 S were enriched in residual sulfates and/or recycling of Triassic evaporites. The reduced δ 34 S values of gypsums (9.5‰ to 10.4‰) in Nuodeng could have been caused by oxidation of sulﬁdes weathered from Jinding Pb-Zn deposit. The complex O isotope compositions indicate that sulfates in the Lanping–Simao Basin had undergone sulfate reduction, re-oxidation, reservoir effects, etc. In conclusion, the formation of continental evaporites was likely derived from seawater due to marine transgression during the Cretaceous period. Meanwhile, non-marine inﬂows have contributed to the basin signiﬁcantly.


Introduction
The Lanping-Simao Basin hosts the only ancient potash deposit ever found in China [1], and a great quantity of metallic mineral resources which have a close relationship with evaporites, especially sulfates [2][3][4][5]. Consequently, the origin of those evaporites has attracted tremendous attention during the last decades [6][7][8][9][10][11][12][13]. The basin evolved from a remnant marine and marine-continental basin during the Triassic Period, through a continental depression basin during the Jurassic-Cretaceous Period, to a pull-apart continental basin during the Cenozoic Period [2]. Evaporites were primarily formed within the Triassic and Cretaceous periods during the evolution of the basin. The metallic-associated gypsums in the Sanhedong Formation, upper Triassic, was marine origin [2,14], whereas the sources of evaporites within Cretaceous continental "red beds" remain a subject of debate. The Br (bromide) geochemistry of rock salts in the Mengyejing potash deposit indicates a major seawater contribution [11,15]. On the contrary, Li et al. (2015) [16] suggested a major continental origin based on geochemical evidence of the Mengyejing potash deposit. Qu et al. (1998) [10] proposed that evaporites in the Lanping-Simao Basin formed during  [29]); (B) schematic geological map of Lanping-Simao Basin and sampling locations (marked by red dots) (after [27]).  [29]); (B) schematic geological map of Lanping-Simao Basin and sampling locations (marked by red dots) (after [27]).

Materials and Methods
A variety of samples, including layered and veined gypsums, rock salts were collected from Lanping, Nuodeng, Jinggu, Baozang, and Mengyejing from the north to the  [29]).

Materials and Methods
A variety of samples, including layered and veined gypsums, rock salts were collected from Lanping, Nuodeng, Jinggu, Baozang, and Mengyejing from the north to the south of the Lanping-Simao Basin. Sulfate samples (mostly gypsums) stemmed from outcrops located in Lanping, Nuodeng, Jinggu, and Baozang. Rock salt samples were sampled from the underground mine lane of the Mengyejing potash deposit (Figure 3). All samples collected from the Lanping-Simao Basin were cut, polished, and thinned using an oil system. The thin sections were examined using a polarizing microscope.
The SEM analysis was carried out at the Key Laboratory of Deep-Earth Dynamics, Institute of Geology using the FEI Nova NanoSEM 450. The back scattered electron (BSE) images were taken under operating voltage of 15-20 KV and working distance of 13.5 mm.
Sr isotope analyses of all samples were performed at the State Key Laboratory of Isotope Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences. Powdered gypsum samples (~5 to 10 mg) were dissolved with 4M HNO3 after washing with milli-Q water. Five to 10 g of rock salt samples were washed with milli-Q water to eliminate chloride salts and accumulated sulfates and subsequently dissolved with 4M HNO3. Sr was extracted from the samples using a Sr-specresin. The detailed procedure about Sr separation is given by [30]. Samples were analyzed in a Neptune Plus multicollector (MC) ICP-MS. The 87 Sr/ 86 Sr ratios were corrected for mass discrimination using 87 Sr/ 86 Sr ratio = 0.1194. NBS 987 standard yields 87 Sr/ 86 Sr values of 0.71022 (30) and 0.71030 (7) for MC-ICP-MS spectrometers. Uncertainties in the 87 Sr/ 86 Sr are quoted in 1σ.
Oxygen isotopic composition of sulfates were performed at the State Key Laboratory Four samples were collected from gypsum laminae in Lanping; three samples were collected from vein-shaped gypsums in Nuodeng; four samples were collected from gypsum laminae in Jinggu; 37 samples were collected in Baozang, including 32 laminated gypsum samples and five veined gypsum samples. Six rock salt samples were collected from the underground mine lane of the Mengyejing potash deposit.
In order to select samples with sufficient sulfate for analyzing S isotope composition, all samples were tested by X-ray diffraction (XRD) analysis. To eliminate the effect of sulfides, all examples were examined under binoculars.
All samples collected from the Lanping-Simao Basin were cut, polished, and thinned using an oil system. The thin sections were examined using a polarizing microscope. The SEM analysis was carried out at the Key Laboratory of Deep-Earth Dynamics, Institute of Geology using the FEI Nova NanoSEM 450. The back scattered electron (BSE) images were taken under operating voltage of 15-20 KV and working distance of 13.5 mm.
Sr isotope analyses of all samples were performed at the State Key Laboratory of Isotope Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences. Powdered gypsum samples (~5 to 10 mg) were dissolved with 4M HNO 3 after washing with milli-Q water. Five to 10 g of rock salt samples were washed with milli-Q water to eliminate chloride salts and accumulated sulfates and subsequently dissolved with 4M HNO 3 . Sr was extracted from the samples using a Sr-specresin. The detailed procedure about Sr separation is given by [30]. Samples were analyzed in a Neptune Plus multicollector (MC) ICP-MS. The 87 Sr/ 86 Sr ratios were corrected for mass discrimination using 87 Sr/ 86 Sr ratio = 0.1194. NBS 987 standard yields 87 Sr/ 86 Sr values of 0.71022(30) and 0.71030 (7) for MC-ICP-MS spectrometers. Uncertainties in the 87 Sr/ 86 Sr are quoted in 1σ.
Oxygen isotopic composition of sulfates were performed at the State Key Laboratory of Lithosphere Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences. Sulfates were dissolved and reacted with BaCl 2 solutions. The precipitated BaSO 4 was washed and dried. Solid barite samples were weighed into a silver capsule and introduced into a graphite furnace where BaSO 4 is converted to CO gas at 1400 • C in helium gas using a thermal combustion elemental analyzer (TCEA). Oxygen isotope ratios were measured by a continuous flow isotope ratio monitoring mass spectrometry system using a Flash HT 2000 high temperature pyrolysis furnace coupled with a Finnigan Conflo IV open split interface to a Thermo Scientific DELTA V Advantage mass spectrometer. Measurements were calibrated using the two-point linear normalization method based on international sulfate standards NBS 127 (+8.6‰, VSMOW), IAEA SO-5 (+12.13‰, VSMOW). Repeated measurements of international standards (four measurements per standard per run) yield reproducibility of better than 0.2‰ (1σ) for oxygen isotope measurements.
For the S isotope measurement, the sulfate was combusted at 980°C in a Flash Element Analyzer and the resulting sulfur dioxide (SO 2 ) was measured with continuous flow GS-IRMS (Thermo, Delta V Plus) at the Beijing Research Institute of Uranium Geology. δ 34 S values are reported vs. the Canyon Diablo Troilite (CDT), and the error was determined using the standard deviation of the standard (GBW-04414 and GBW-04415) at the beginning and the end of each run (<0.5‰).
The detailed procedure for ICP-MS trace element analysis is given in [31], and the two-sigma error for the 87 Rb/ 86 Sr ratio was estimated at ±2.6%.

Characteristics of Evaporite Minerals
In Lanping, the gypsum section is interbedded with the underlying and overlying mudstones ( Figure 3) with relatively sharp contact boundary. The mudstones show massive structure. Gypsum aggregates are usually present as two forms, i.e., gypsum laminae with fine-grained crystals (alabaster) and selenite macrocrystallines. The common millimetric gypsum laminae show slightly wavy features ( Figure 4A) and are seen as alabaster on planar direction ( Figure 4B). Selenite crystals are sporadically interbedded or embedded with gypsum laminae. The boundary between selenite and gypsum laminae are sharp ( Figure 4A). The gypsum laminae are composed of microcrystalline gypsum. The microcrystalline gypsums display a variety of textures ranging from xenotopic to idiotopic. The crystals are present primarily as xenotopic ameboid gypsums with minor embedded euhedral crystals ( Figure 5A). The gypsum crystals showed no orientation and variation in sorting. The sizes of the microcrystalline gypsums are equant and not varied along with the changes of the laminae. The scanning electron microscopy (SEM) analysis showing that euhedral pyrite ( Figure 5B) and celestite are present within gypsum laminae ( Figure 5C). of gypsum laths show herringbone pattern ( Figure 5J).
In Mengyejing, the underground potash deposit consists of rhythmically alternating clastic rocks and evaporites. The clastic rocks are composed of unconsolidated mudstones and siltstones. The chloride salts, namely, halite sylvite or carnallite are crosscutting or cementing most mudstones and siltstones. The layers of the orebody are dipping nearly vertically due to tectonic deformation ( Figure 4K). The rock salt samples are primarily composed of halite, with trace amounts of euhedral anhydrites ( Figure 5K) and sylvites ( Figure 5L).  In Nuodeng, fractures are developed within mottled (reddish-brown and greenishgrey) clastic rocks, and filled by nearly pure gypsum veins ( Figure 4C). These veins consist of fibers from 1 to 10 cm long. The gypsum fibers are curved and show perpendicular or oblique orientation to the wall rocks ( Figure 4D). The elongated gypsum crystals within veins are curved and aligned with minor distorted relative short gypsum crystals ( Figure 5D). The clastic rocks are unconsolidated and showing stratified structure with no cross-bedding detected. Some parts of the clastic rocks contain clayey breccias ( Figure 4C,D).
In Jinghong, gypsum laminae are normally organized in millimetric to centimetric beds ( Figure 4E), alternating with calcareous clastic layers. The stratified gypsum laminae and intercalated calcareous clastic layers are plane-parallel. The lamination is clearly differentiated by white and gray thin layers ( Figure 4F). In thin section the white laminae were seen as small gypsum nodules displaying ameboid texture ( Figure 5E). Anhydrite relics with jagged edges surrounded by granular gypsums are detected within the thin layer gypsums, showing dehydration and rehydration processes ( Figure 5F).
In Baozang, gypsum laminae are mingled with mudstones and/or siltstones ( Figure 4G). The bedding structure is distinct based on color banding ( Figure 5G). Some parts of the gypsum laminae are replaced by nodular selenite ( Figure 4H). In nodular selenite, swallowtail- shaped gypsum crystals ( Figure 5H) and corroded anhydrite crystals ( Figure 5I) are detected, suggesting a similar dehydration-rehydration process to that in Jinghong. Interstratified and vertical fractures ( Figure 4H) are filled with satin spar gypsums (veined gypsums). In addition to those veined gypsums within laminae fractures, some wired gypsum veins are present in clayey or silty breccias ( Figure 4I). The gypsum fibers are perpendicular or oblique to the surface of the wall rocks. Relatively long gypsum laths occur within clastic rocks, showing slightly curved features ( Figure 4J). Microphotographs of gypsum laths show herringbone pattern ( Figure 5J).

XRD Results
The mineral components are measured using XRD analyses. The result shows that gypsum samples in Lanping consist of approximately 90% gypsum and 10% calcite, with a trace amount of quartz. Two celestite-bearing samples in Lanping contain approximately 40% and 10% celestite, 55% and 90% calcite, respectively. Samples in Nuodeng consist exclusively of gypsum, except for sample LP-SM-G3 with a small amount of quartz and albite. The major mineral of samples in Jinggu are gypsum (90% to 100%), and subordinate amounts of calcite, dolomite, and magnesite (Table 1). Halite is the most abundant mineral in samples from Mengyejing, with NaCl content ranging from 65% to 100%. Anhydrites and gypsums are present as a minor constituent. Some samples contain a certain In Mengyejing, the underground potash deposit consists of rhythmically alternating clastic rocks and evaporites. The clastic rocks are composed of unconsolidated mudstones and siltstones. The chloride salts, namely, halite sylvite or carnallite are crosscutting or cementing most mudstones and siltstones. The layers of the orebody are dipping nearly vertically due to tectonic deformation ( Figure 4K). The rock salt samples are primarily composed of halite, with trace amounts of euhedral anhydrites ( Figure 5K) and sylvites ( Figure 5L).

XRD Results
The mineral components are measured using XRD analyses. The result shows that gypsum samples in Lanping consist of approximately 90% gypsum and 10% calcite, with a trace amount of quartz. Two celestite-bearing samples in Lanping contain approximately 40% and 10% celestite, 55% and 90% calcite, respectively. Samples in Nuodeng consist exclusively of gypsum, except for sample LP-SM-G3 with a small amount of quartz and albite. The major mineral of samples in Jinggu are gypsum (90% to 100%), and subordinate amounts of calcite, dolomite, and magnesite (Table 1). Halite is the most abundant mineral in samples from Mengyejing, with NaCl content ranging from 65% to 100%. Anhydrites and gypsums are present as a minor constituent. Some samples contain a certain amount of carbonates and potash minerals ( Table 1). Samples from Baozang are composed of nearly pure gypsums, with a trace amount of bassanite, carbonates, and quartz (Table 1).

Sr, S and O Isotopes
The 87 Sr/ 86 Sr ratios of all samples in the Lanping Basin range from 0.709406 to 0.710049, which are much higher than those of the Late Cretaceous seawater [22]. The celestite-bearing samples have slightly higher 87 Sr/ 86 Sr ratios compared with the gypsum samples ( Table 2). The 87 Sr/ 86 Sr ratios of the gypsum samples in the Simao basin range from 0.708081 to 0.709548 (Jinggu: 0.708081 to 0.708792, Baozang: 0.708114 to 0.7.9548). There is considerable overlap between the 87 Sr/ 86 Sr ratios of Baozhang and Jinggu. The 87 Sr/ 86 Sr ratios of the rock salt samples range from 0.709717 to 0.710071, which are higher than those of gypsum samples in Jinggu and Baozang ( Figure 6A).

Discussion
The ages of the samples are constrained to be the Middle to Late Cretaceous (ca. 110 to 65 Ma). Albeit the uncertainty of the sedimentary ages, comparison between the S, O, and Sr isotope compositions of these samples with values of seawater [19,21,22] reveals the origin of those parent brines in which evaporite minerals were precipitated.  Figure 6C).

Discussion
The ages of the samples are constrained to be the Middle to Late Cretaceous (ca. 110 to 65 Ma). Albeit the uncertainty of the sedimentary ages, comparison between the S, O, and Sr isotope compositions of these samples with values of seawater [19,21,22] reveals the origin of those parent brines in which evaporite minerals were precipitated.

Sr Isotopes
The Sr isotope compositions of the evaporite minerals reflect the sources of Sr to the basin, together with possible interactions between the brines and rocks within the drainage basin [18]. The elevated 87 Sr/ 86 Sr ratios of rock salts in Mengyejing ( Table 2) was very likely caused by the accumulation of radiogenic 87 Sr due to high Rb/Sr ratios ( Table 2) and/or continental waters with high 87 Sr/ 86 Sr ratios. The 87 Sr/ 86 Sr ratios of gypsums in halite crystal fall within those of chloride salts (halite, sylvite, and carnallite, 0.708697-0.710956, except two anomalous low values [1]) in the Mengyejing potash deposit. The Rb/Sr ratios of those chloride salts are much higher than those of gypsums [1], indicating that the salt minerals in the Mengyejing potash deposit were formed by recent recrystallization process, and high Rb contents of chloride salts have not accumulated sufficient radiogenic 87 Sr to generate higher 87 Sr/ 86 Sr ratios compared with gypsums.
The influence of radiogenic 87 Sr on gypsum samples was negligible because of the extremely low Rb/Sr ratios (Table 2). Thus, the Sr isotope composition of gypsum could represent the Sr isotopic signature of parent brine in which gypsum precipitated. The 87 Sr/ 86 Sr ratios of all gypsum samples shown in Table 2 and Figure 7 are higher than the range of coeval seawaters [22]. This is consistent with the conclusion that the 87 Sr/ 86 Sr ratios of evaporites in continental setting are generally higher than those of seawater [32]. Apparently, the parent brines were derived, at least partly, from continental water. The weathering of the Lincang granite ( Figure 1) could have supplied dissolved components with high 87 Sr/ 86 Sr ratios. The 87 Sr/ 86 Sr ratios of biotite granite in Lincang range from 0.730006 to 0.743494, corresponding to an initial 87 Sr/ 86 Sr ratio range of 0.713566 to 0.728476 when granite formed during Triassic based on Rb/Sr ratios [33]. In addition to the Lincang granite, other regions may have provided weathering products with varying Sr isotope compositions. The Sr isotope compositions and Sr concentrations of those continental waters were unknown. We postulate that the continental freshwater had similar Sr concentration and Sr isotopic ratios to those of present river water. Noh et al. (2009) [34] presented Sr concentrations and isotope compositions of two major rivers enclosing the Lanping-Simao Basin: The Jinshajiang River, 0.33-11.46 µM, 87 Sr/ 86 Sr = 0.70891-0.71494, and the Lancang River, 0.28-6.73 µM, 87 Sr/ 86 Sr = 0.70888-0.72678. The low Sr concentrations of river waters necessitate a large amount of continental fluvial input to produce the elevated 87 Sr/ 87 Sr ratios of gypsums in the Lanping-Simao Basin.
The 87 Sr/ 86 Sr ratios of ankerites within the Mesozoic strata in the Lanping-Simao Basin range from 0.70874 to 0.71332 [35]. The brines in which those ankerites precipitated were thought to have been formed by circulation of basinal fresh waters. Sr was leached out from Mesozoic strata [35], thus the 87 Sr/ 86 Sr ratios of ankerites could represent those of the Mesozoic sedimentary rocks. Consequently, it is likely that the parent brine for forming evaporites could also have derived from adjacent clastic rocks to some extent.
As a whole, the Sr isotope compositions of gypsums in the Lanping Basin are higher than those of gypsums in the Simao Basin. Recent provenance studies show that the Late Cretaceous sediments from these basins have an overall S-directed paleocurrent that flowed from the Lanping Basin to the Simao Basin [36]. The Lanping Basin could have trapped more continental waters compared with the Simao Basin. As a whole, the Sr isotope compositions of gypsums in the Lanping Basin are higher than those of gypsums in the Simao Basin. Recent provenance studies show that the Late Cretaceous sediments from these basins have an overall S-directed paleocurrent that flowed from the Lanping Basin to the Simao Basin [36]. The Lanping Basin could have trapped more continental waters compared with the Simao Basin.

S Isotopes
The δ 34 S values of gypsum laminae samples in Baozang show a narrow range, from 13.4‰ to 15.2‰ (Table 2, Figure 8) which is consistent with those of Cretaceous seawater [19,21]. Wang et al. (2014a) [12] implied that there was a marine transgression during the Late Cretaceous based on geochemical, palaeogeographical, and paleomagnetic studies ( [12] and references therein). The crystal of vein-shaped gypsum was corroded and dissolved by external fluids or internal waters from dehydration of gypsum ( Figure 5I), indicating that vein-shaped gypsums had undergone dissolution and recrystallization process. The secondary veined gypsum samples in Baozang have similar S isotope compositions to those of bedded gypsum samples ( Table 2), denoting that the sulfate-bearing fluids for forming the secondary veined gypsums mainly stemmed from the dissolution of bedded gypsums.
The δ 34 S values of rock salt samples in Mengyejing range from 8.0‰ to 15.5‰, which are slightly lower than those of gypsum samples near Mengyejing (Baozang). During evaporation, 34

S Isotopes
The δ 34 S values of gypsum laminae samples in Baozang show a narrow range, from 13.4‰ to 15.2‰ (Table 2, Figure 8) which is consistent with those of Cretaceous seawater [19,21]. Wang et al. (2014a) [12] implied that there was a marine transgression during the Late Cretaceous based on geochemical, palaeogeographical, and paleomagnetic studies ( [12] and references therein). The crystal of vein-shaped gypsum was corroded and dissolved by external fluids or internal waters from dehydration of gypsum ( Figure 5I), indicating that vein-shaped gypsums had undergone dissolution and recrystallization process. The secondary veined gypsum samples in Baozang have similar S isotope compositions to those of bedded gypsum samples ( transgression. Gypsum laminae ( Figure 4F) and amenoid microcrystalline gypsums (Figure 5E) suggest a likely primary origin. However, the anhydrite relics with surrounding gypsum crystals ( Figure 5F) indicate that dehydration of gypsum and hydration of anhydrite cycle have occurred. The S isotopes of sulfates suggests that the dehydration-hydration process did not affect the isotopic signatures significantly.   Figure 4F) and amenoid microcrystalline gypsums ( Figure 5E) suggest a likely primary origin. However, the anhydrite relics with surrounding gypsum crystals ( Figure 5F) indicate that dehydration of gypsum and hydration of anhydrite cycle have occurred. The S isotopes of sulfates suggests that the dehydration-hydration process did not affect the isotopic signatures significantly.
In Lanping, δ 34 S values of two gypsum samples are 14.5‰ and 17.6‰, respectively; slightly higher than those of Cretaceous seawater and gypsum samples in the Simao Basin. Two celestite samples have δ 34 S values ranging from 20.5‰ to 20.7‰, which are much higher than those of gypsum samples. Reservoir effect and continental contribution could lower the δ 34 S values, which is not the case here. Therefore, the elevated δ 34 S values of gypsum and celestite samples in Lanping could have contributed to other factor(s), such as bacterial sulfate reduction (BSR). During BSR, the lighter isotopes 32 S and 16 O are preferentially metabolized by microorganisms, causing an enrichment of heavy isotopes 34 S and 18 O in the remaining sulfate [38]. Organic matters are widely distributed in the Jinding Pb-Zn deposit in Lanping area. In reducing environment, sulfates were reduced to sulfides. Pyrite is commonly developed in gypsums ( Figure 5B). It was suggested that S 2in sulfides (mainly consist of sphalerite and galena) were generated by sulfate reduction [3] and resulted in 34 S-enriched fluids. The euhedral celestite crystal ( Figure 5C) could have been formed by 34 S-enriched fluids in combination with Sr-bearing metal fluids. Gypsums and celestites are distributed within the Triassic marine sequence in Lanping area with δ 34 S values ranging from 15.3‰ to 17.5‰ [39]. It was possible that the recycling of Triassic evaporites could have contributed and affected the composition of S isotopes of evaporites formed in the non-marine setting during Cretaceous.
The reduced δ 34 S values of gypsum samples in Nuodeng were not controlled by BSR. It was not likely caused by reservoir effect either because it engenders negligible depletion of 34 S in sulfates during gypsum precipitation stage [37]. Therefore, only continental input with isotopically light 32 S could account for this result. The Jinding Pb-Zn deposit comprises a great amount of sulfide minerals, including sphalerite and galena. Approximately 600 million tons of Pb + Zn were eroded [39]. The sulfide minerals show a wide range of δ 34 S values, from −54.9‰ to +3.5‰ [39]. The δ 34 S values of sulfates formed via sulfide oxidation are generally equivalent to those of the parent sulfide minerals [40]. There is no or insignificant fractionation during the oxidation process of sulfide. Nuodeng is only 60 km to the south of Lanping. Weathering products of the Jinding Pb-Zn deposit could be easily transported from Lanping to Nuodeng. The re-oxidation of reduced sulfides with low δ 34 S values resulted in relatively low sulfate δ 34 S values in Nuodeng.
A giant marine evaporite deposit occurred within the Maha Sarakham Formation, the Khorat Basin, Thailand. Qu (1998) [10] suggested that the evaporites within the Lanping-Simao Basin have a close relationship with evaporites within the Khorat Basin based on sedimentary sequences comparison and salt mineral assemblages. The δ 34 S values of anhydrites intercalated with rock salt layers within the Maha Sarakham evaporite deposit range from 14.8‰ to 17.7‰ [41]. The δ 34 S values of layered anhydrite in Baozang are consistent with those of anhydrites in the Khorat Basin. Qin et al. (2020) [42] proposed a Cretaceous seawater recharge model that the paleoseawater flowed from Bangong-Nujiang Ocean (West Meso-Tethys Ocean) through the Qiangtang and Lhasa blocks to the Lanping-Simao Basin and the Khorat Basin. Alternatively, the paleoseawater could have derived from East Meso-Tethys Ocean and recharged the Lanping-Simao and Khorat Basins through Tengchong-Baoshan Blocks [42].

O Isotopes
The history of seawater δ 18 O sulfate is less well-defined compared with S and Sr isotope compositions [18]. Thus, the δ 34 S-δ 18 O relationships are presented for the comparison with S and O isotope values of Cretaceous seawater [19]. The global isotopic evolution through time of marine sulfates has been well documented worldwide. The oxygen isotopic compositions of sulfates from Mesozoic to present-day are within a range of approximately +10‰ to +15‰ [19]. to the south of Lanping. Weathering products of the Jinding Pb-Zn deposit could be easily transported from Lanping to Nuodeng. The re-oxidation of reduced sulfides with low δ 34 S values resulted in relatively low sulfate δ 34 S values in Nuodeng.
A giant marine evaporite deposit occurred within the Maha Sarakham Formation, the Khorat Basin, Thailand. Qu (1998) [10] suggested that the evaporites within the Lanping-Simao Basin have a close relationship with evaporites within the Khorat Basin based on sedimentary sequences comparison and salt mineral assemblages. The δ 34 S values of anhydrites intercalated with rock salt layers within the Maha Sarakham evaporite deposit range from 14.8‰ to 17.7‰ [41]. The δ 34 S values of layered anhydrite in Baozang are consistent with those of anhydrites in the Khorat Basin. Qin et al. (2020) [42] proposed a Cretaceous seawater recharge model that the paleoseawater flowed from Bangong-Nujiang Ocean (West Meso-Tethys Ocean) through the Qiangtang and Lhasa blocks to the Lanping-Simao Basin and the Khorat Basin. Alternatively, the paleoseawater could have derived from East Meso-Tethys Ocean and recharged the Lanping-Simao and Khorat Basins through Tengchong-Baoshan Blocks [42].

O Isotopes
The history of seawater δ 18 Osulfate is less well-defined compared with S and Sr isotope compositions [18]. Thus, the δ 34 S-δ 18 O relationships are presented for the comparison with S and O isotope values of Cretaceous seawater [19]. The global isotopic evolution through time of marine sulfates has been well documented worldwide. The oxygen isotopic compositions of sulfates from Mesozoic to present-day are within a range of approximately +10‰ to +15‰ [19].   [19]. The dark black symbols indicate layered gypsum samples, whereas grey symbols indicate veined gypsum samples.
In Baozang, the narrow range of δ 34 S values suggests that the inflow of dissolved sulfates in the continental waters was insignificant. Because in general cases, sulfates from continental waters will add light sulfur and δ 34 S values decreases accordingly [37]. As discussed above, S isotope compositions of gypsum samples in Baozang denote a marine origin. S and O isotope compositions of marine sulfates are insensitive to minor non-marine contributions because seawater hosts much higher SO4 concentration than most freshwaters [43], and dissolution of sulfates results in little or negligible isotopic fractionation  [19]. The dark black symbols indicate layered gypsum samples, whereas grey symbols indicate veined gypsum samples.
In Baozang, the narrow range of δ 34 S values suggests that the inflow of dissolved sulfates in the continental waters was insignificant. Because in general cases, sulfates from continental waters will add light sulfur and δ 34 S values decreases accordingly [37]. As discussed above, S isotope compositions of gypsum samples in Baozang denote a marine origin. S and O isotope compositions of marine sulfates are insensitive to minor non-marine contributions because seawater hosts much higher SO 4 concentration than most freshwaters [43], and dissolution of sulfates results in little or negligible isotopic fractionation of S and O [19,44]. Thus, the variation of O isotope compositions of those gypsum samples was not controlled by inflow of continental dissolved sulfates and dissolution process. The scattered O isotope compositions could not either be accounted for by reservoir effect which fails to induce such a wide variation.
The changes in oxygen isotopic composition of sulfate are related to more complex processes than those affecting sulfur isotopes [45]. In a restricted basin, BSR process produced sulfides with relatively negative δ 34 S values and residual dissolved sulfates with positive δ 34 S values. The resulted sulfides from BSR diffused into shallow water and reoxidized to sulfates with the incorporation of O from water and/or molecular oxygen [45].
Sulfates formed by re-oxidation of sulfide would cause no fractionation on S but varied fractionation on 18 O through the incorporation of dissolved oxygen and/or oxygen of hypersaline water [46]. In the presence of molecular oxygen, the δ 18 O value of the resulting sulfate oxidized by sulfide is considered to shift towards heavy values. Whereas, under anaerobic conditions, the oxidation of sulfide yields a sulfate with isotopically light δ 18 O values equal or very close to that of environmental water [45]. Moreover, the exchange of oxygen atoms of intermediate anions like SO 3 2and/or HSO 3 complicates the oxidation processes that may affect the final O isotope composition of sulfate. The proportions of water-derived oxygen and molecular oxygen incorporated into sulfates during the reoxidation process of sulfide were subject to environments ( [47] and references therein). Mangalo et al. (2007) [48] performed an experiment which proved that the δ 18 O value of sulfate during BSR could be affected by isotope exchange with water. They supported a fractionation mechanism of re-oxidation of sulfite to sulfate rather than that of reaction from sulfate-enzyme complex back to sulfate ( [48] and references therein).
At the water-sediment interface, the constant cycling between sulfate reduction and sulfide reoxidation has no net effect on the burial of reduced sulfur, but greatly affects the oxygen isotope composition of marine sulfate (δ 18 O SO4 ; [49] and references therein). Therefore, this redox cycle of sulfur only affects the sulfate O-isotope ratio significantly, but not the sulfate S-isotope ratio [50,51]. The S vs. O pattern of sulfates in Baozang is consistent with that produced by sulfate reduction and sulfide re-oxidation process.
Organic matters were widely distributed in the Lanping-Simao Basin, usually presented as debris (Lanping, Nuodeng, Baozang, and Mengyejing, Figure 4L) and/or banded layer (Lanping and Jinggu) [52]. It was very likely that BSR occurred under anoxic conditions and sulfates were reduced to sulfides. The sulfides were then re-oxidized in the "red bed" environment. This process causes little variation on S isotope compositions if there was no extraneous S. Whereas, reduction-re-oxidation process changes the O isotope compositions of sulfates drastically under different environments.
The S-O isotope compositions of sulfates in rock salt samples from Mengyejing are located to the lower left of marine isotope compositions (Figure 9). This pattern could have been the result of reservoir effect and/or reduction-re-oxidation process. It is practically impossible to determine what process(es) was predominant.
The S-O isotope compositions of sulfates from Lanping are located to the upper right of marine isotope compositions (Figure 9), which are similar to those of Messinian gypsums in the Nijr Basin [53]. The elevated δ 34 S and δ 18 O values resulted by redox cycling involving BSR and re-oxidation in stratified brines [53]. BSR produced a great amount of S 2which incorporated into metal cations and formed metal sulfides in Lanping. The residual sulfates would be enriched in 34 S and 18 O.

The Origin of Evaporites and Paleoenvironmental Significance
Although Sr isotope compositions of all samples and S isotope compositions of samples from Nuodeng corroborate a major continental contribution to the formation of the evaporites, S isotope compositions of sulfates from Lanping, Jinggu, and Baozang indicate a marine origin. Trace elements of chloride salts in the Mengyejing potash deposit also suggested a marine contribution [15]. In conclusion, the parent brine in which evaporite minerals precipitated were derived from a mixture of seawater and continental waters.
In Baocang, the narrow range of δ 34 S values vs. wide range of 87 Sr/ 86 Sr ( Figure 10) indicate that continental input imposes a greater effect on Sr than on S. The seawater dominated in terms of S isotopes, whereas continental input controlled the Sr isotopes. In Jinggu, the 87 Sr/ 86 Sr ratios vs. δ 34 S values show a similar pattern to that of evaporites in Baozang, which suggest a similar origin and formation process. In Mengyejing, the 87 Sr/ 86 Sr ratios are relatively steady but δ 34 S values show a large variation. This pattern indicates that reservoir effect controlled the S isotopes which is consistent with preceding discussion. In Nuodeng, the low δ 34 S values, high 87 Sr/ 86 Sr ratios, and limited distribution of S vs. Sr suggest that the parent brines were only controlled by continental waters derived from weathering. In Lanping, the positive relationship between the 87 Sr/ 86 Sr ratios and the δ 34 S values indicates that the formation of evaporites could have been controlled by continental influx, recycle of older evaporites, and BSR synergistically.
indicate that continental input imposes a greater effect on Sr than on S. The seawater dominated in terms of S isotopes, whereas continental input controlled the Sr isotopes. In Jinggu, the 87 Sr/ 86 Sr ratios vs. δ 34 S values show a similar pattern to that of evaporites in Baozang, which suggest a similar origin and formation process. In Mengyejing, the 87 Sr/ 86 Sr ratios are relatively steady but δ 34 S values show a large variation. This pattern indicates that reservoir effect controlled the S isotopes which is consistent with preceding discussion. In Nuodeng, the low δ 34 S values, high 87 Sr/ 86 Sr ratios, and limited distribution of S vs. Sr suggest that the parent brines were only controlled by continental waters derived from weathering. In Lanping, the positive relationship between the 87 Sr/ 86 Sr ratios and the δ 34 S values indicates that the formation of evaporites could have been controlled by continental influx, recycle of older evaporites, and BSR synergistically.  [54] suggested that continental input dominates Sr and S isotopic signature when 87 Sr/ 86 Sr ratios and δ 34 S values shift in opposition. In this section, the decreasing trend of δ 34 S values in combination with the increasing trend of 87 Sr/ 86 Sr ratios suggest that continental input played an increasingly significant role with the progressive evaporation of brines. We suggest that the major waterbody for forming the evaporites were remnant seawater due to marine transgression. In the late stage of evaporation, continental waters played a more important role and predominated with respect to isotope signatures.
As discussed above, redox conditions, O isotope compositions of parent brine, and molecular O could determine the final O isotope compositions of dissolved sulfate, thus complicating the O isotope compositions of precipitated sulfates. The greatly varied S and O isotope compositions of gypsums in Messinian evaporites [55] may result from repeated processes of evaporite dissolution and re-precipitation as well as from bacterial activities during redox variations [45,56]. The wide variation of O isotopes compositions of gypsums in such a short section (15 m) in Baozang also suggests a drastic sedimentary environmental change during evaporite deposition. S and O isotope compositions of gypsums in Jinggu have a similar pattern to that of in Baozang, denoting a similar process.  [19], δ 34 S values [21], and 87 Sr/ 86 Sr ratios [22] of contemporaneous marine sulfates.

Conclusions
(1) The 87 Sr/ 86 Sr ratios of sulfate samples (including gypsum and celestite) in the Lanping-Simao basin are higher than those of contemporaneous seawater, indicating continental contribution; elevated 87 Sr/ 86 Sr ratios of rock salt samples were caused by continental contribution and radiogenic 87 Sr accumulation. (2) The δ 34 S values of gypsum samples in the Simao basin are consistent with those of Cretaceous seawater, suggesting a marine origin; the reduced δ 34 S values of rock salts samples might be due to reservoir effect and continental contribution; the relatively higher δ 34 S values of sulfates in Lanping were likely caused by BSR or/and recycling  [19], δ 34 S values [21], and 87 Sr/ 86 Sr ratios [22] of contemporaneous marine sulfates. Kristall et al. (2018) [54] suggested that continental input dominates Sr and S isotopic signature when 87 Sr/ 86 Sr ratios and δ 34 S values shift in opposition. In this section, the decreasing trend of δ 34 S values in combination with the increasing trend of 87 Sr/ 86 Sr ratios suggest that continental input played an increasingly significant role with the progressive evaporation of brines. We suggest that the major waterbody for forming the evaporites were remnant seawater due to marine transgression. In the late stage of evaporation, continental waters played a more important role and predominated with respect to isotope signatures.
As discussed above, redox conditions, O isotope compositions of parent brine, and molecular O could determine the final O isotope compositions of dissolved sulfate, thus complicating the O isotope compositions of precipitated sulfates. The greatly varied S and O isotope compositions of gypsums in Messinian evaporites [55] may result from repeated processes of evaporite dissolution and re-precipitation as well as from bacterial activities during redox variations [45,56]. The wide variation of O isotopes compositions of gypsums in such a short section (15 m) in Baozang also suggests a drastic sedimentary environmental change during evaporite deposition. S and O isotope compositions of gypsums in Jinggu have a similar pattern to that of in Baozang, denoting a similar process. In summary, the parent brines in which evaporites precipitated within the Mesozoic "red bed" of the Lanping-Simao Basin mainly stemmed from remnant seawater due to marine transgression and continental water. During the evaporation, the paleoenvironment changed dramatically based on O isotopic compositions of sulfates, and continental water played an increasingly important role compared with remnant seawater.