The barium (Ba) isotopic composition of carbonate minerals (δ138Bacarb) is a promising new geochemical tracer for past Ba cycling. However, further study of the potential influence of early diagenetic processes on δ138Bacarb records remains an important avenue for this new tracer’s development. Marine methane seep environments are sites of intense Ba cycling during early diagenesis. Seeps may be responsible for much of the flux of Ba from the sedimentary subsurface into seawater, while the authigenic carbonate minerals precipitating at seeps may represent an archive of this potential Ba source. Here we present δ138Bacarb data from authigenic carbonates collected at eight methane seeps from the South China Sea, the Gulf of Mexico, the Congo Fan, and the Black Sea. The obtained δ138Bacarb values display a broad range from −0.15‰ to 0.37‰, indicating authigenic seep carbonates archive highly variable Ba isotope signals of local pore waters. Aragonite-dominated carbonates tend to show higher δ138Bacarb values than high-Mg calcite-dominated carbonates. This pattern suggests a greater influence of seawater during aragonite precipitation caused by its formation at shallower depths compared to high-Mg calcite. Accordingly, low δ138Bacarb values and relatively high Ba/Ca molar ratios are interpreted to reflect the release of Ba from sediments through barite dissolution. However, under conditions of increased upward flux of Ba reflected by high Ba/Ca ratios, seep carbonates display highly variable δ138Bacarb values independent of Ba/Ca molar ratios, suggesting that additional processes (i.e. local reprecipitation of 134Ba-enriched barite) control pore water barium isotope compositions. Taken together, these observations showcase the complexity of Ba isotope systematics at marine seeps, governed by the mixing of Ba from different sources, mineral dissolution in the subsurface, and local authigenesis (carbonate minerals and barite). These new findings highlight the dynamic nature of early diagenetic modifications on Ba isotopic fingerprints, while proposing a process-based conceptual framework for the combined use of Ba/Ca ratios and mineralogical analysis to assess diagenetic signals in carbonate-hosted Ba isotope archives.

Distinguishing methane and oil seeps using marine authigenic carbonates is crucial for reconstructing past seep activity and assessing their environmental impact. Current methodologies for such a discrimination lack direct analysis on the organic matter preserved within seep carbonates. Here we investigated the optical properties of iron-bound organic carbon (OC-Fe-R) in authigenic carbonates from methane and oil seeps from the Gulf of Mexico and the South China Sea. Methane-seep carbonates display a mean OC-Fe-R fraction (f(OC-FeR)) relative to total organic carbon (TOC) of 14.8 +/- 10.3%, which is significantly higher than the 3.1 +/- 2.2% observed for oil-seep carbonates. This difference is attributed to the predominance of low molecular weight carboxylic acids at methane seeps and the high aromaticity of the larger molecular weight compounds at oil seeps. Furthermore, a significant negative correlation was observed between the biological index (BIX) and TOC in oil-seep carbonates (R-2 = 0.40; p < 0.05), a relationship that was not evident for methane seeps. This result suggests that the TOC at oil seeps consisting of a recalcitrant component is resistant to biodegradation. Additionally, a strong negative correlation (R-2 = 0.60; p < 0.01) was identified between marine, microbial humic-like components and the delta C-13 compositions of oil-seep carbonates. We suggest that low f(OC-FeR) ratios, negative correlations between BIX and TOC and between marine, microbial humic-like components and delta C-13 values in authigenic carbonates serve as criteria for discriminating oil seeps from methane seeps. This optical fingerprinting enables rapid and straightforward discrimination of hydrocarbon sources in authigenic carbonates and advances our understanding of the contributions of seeps to the marine carbon cycle.

Riverine sulfate, primarily derived from the weathering of continental sulfur-bearing minerals, plays a key role in shaping the isotopic composition of marine sulfate. In arid regions, gypsum dissolution is typically assumed to be the dominant sulfate source, but conventional sulfur and oxygen isotope systems (δ34S-SO42- and δ18O-SO42-) struggle to distinguish it from overlapping atmospheric and anthropogenic sources. Here, we apply the triple oxygen value (Δ'17O-SO42-) in the Yellow River Basin─characterized by widespread agriculture, abundant gypsum, and minimal atmospheric deposition─to resolve these ambiguities. We observe strongly negative Δ'17O-SO42- values (as low as -0.3‰) in upper basin waters during high-discharge periods, which we attribute to fertilizer-derived sulfate flushing from agricultural soils. These negative anomalies diminish downstream, where gypsum contributions become dominant. Isotopic modeling reveals that traditional approaches may overestimate gypsum inputs by up to 40%. Our results demonstrate that Δ'17O-SO42- is a powerful tracer of anthropogenic sulfate inputs in modern rivers. Moreover, its nonconservative behavior during transport challenges the assumption that negative Δ'17O values in paleo-sulfate archives strictly reflect contributions of atmospheric oxygen to sulfate.

Abstract Black carbon (BC), a refractory component of organic carbon, exhibits poorly constrained environmental behavior. Recent culture experiments suggested anaerobic methanotrophic archaea (ANME) produce BC via anaerobic oxidation of methane, a pathway supported by studies of seep carbonates — mineral deposits restricted to settings characterized by high methane flux. However, it remains unconfirmed in sediments with low methane flux, settings that are far more widespread in the global ocean. Here, we examine BC distribution and δ 13 C BC values in two seepage‐influenced sediment cores (PC‐01, PC‐02) and a reference core (C‐S05) from the Haima seeps in the South China Sea. Core PC‐01, influenced by higher methane flux, shows significantly 13 C‐depleted BC (−28.9‰ to −25.4‰) at the sulfate‐methane transition zone compared to global non‐seep sediments (−26.4‰ to −10.9‰). Furthermore, not all seepage‐affected sediments exhibit this signal, as demonstrated by core PC‐02. This discrepancy is likely governed by site‐specific variations in methane flux. Consequently, when methane flux is low, the limited production of ANME‐derived BC is insufficient to measurably shift the isotopic signature of the bulk BC pool. A Bayesian mixing model estimates that ANME‐derived BC accounts for ∼6.3% of the total BC at core PC‐01. Although quantitatively minor, ANME‐derived BC may represent a previously overlooked source of potentially aged BC in marine environments, consistent with microbial BC production in methane‐rich sediments.

In marine sulfidic surface sediments, the reduction of iron (Fe) oxides releases substantial Fe-bound phosphorus (P). However, the balance between its retention in solid phases versus its escape to the water column remains poorly constrained near the sediment-water interface (SWI). To investigate this, we analyzed authigenic carbonates formed close to the SWI at cold seeps in the South China Sea. Our results show that high-magnesium calcite forming under restricted sulfidic conditions with limited fluid mixing, as evidenced by its low S13C values and high S34S values of associated iron sulfides (S34SCRS), acts as a major sink of P, exhibiting higher contents of reactive P than nearby sediments unaffected by seepage. In contrast, aragonite, precipitating in more open, sulfate-reducing settings with prominent fluid mixing near the SWI, is depleted in reactive P relative to sediment unaffected by seepage. Critically, contents of reactive P, including Fe-bound P and authigenic P, correlate positively with Mg/Ca ratios and S34SCRS values but negatively with Sr/Ca ratios and S13C values, indicating that carbonate mineralogy and fluid-seawater exchange are the primary controls on P sequestration. Our findings suggest that the openness of the seabed interface, not just redox conditions, ultimately governs P efflux from sulfidic sediments. This insight from modern seeps offers a new framework for assessing P recycling and burial in ancient anoxic oceans.

The contents and distribution patterns of rare earth elements (REEs) are powerful tools for reconstructing fluid sources and redox conditions in marine sedimentary environments. However, their accurate determination in carbonate rocks is often complicated by low REE abundances and high barium (Ba) contents, which cause severe isobaric interference with europium (Eu) during conventional ICP-MS analysis due to the formation of BaO+ or BaOH+ polyatomic ions. In this study, we applied a validated DGA resin separation technique to analyze REEs from 30 Ba-rich authigenic carbonate samples from nine seep sites in the Gulf of Mexico. Before and after DGA separation, REE analyses were conducted on both bulk carbonate rock and carbonate fraction of carbonate rock (hereafter carbonate fraction). Ba contents ranged from 201 ppm to 91200 ppm in the bulk carbonate rock, and from 28 ppm to 1700 ppm in the carbonate fraction. Prior to separation, Eu/Eu* values ranged from 1.40 to 80.49 in bulk rock and from 1.44 to 17.65 in the carbonate fraction, with strong correlation to Ba/Sm ratios, indicating Ba-induced analytical artifacts. After DGA resin separation, Eu/Eu* values decreased to 1.10-1.16 in bulk rock and 1.11-1.22 in the carbonate fraction, confirming the elimination of false Eu anomaly. These results demonstrate that Ba-related interference can lead to substantial misinterpretation of REE data in Ba-rich samples. We advocate for routine chemical separation of REEs from matrix elements prior to ICP-MS analysis to ensure accurate paleoenvironmental reconstructions in seep and other Ba-enriched sedimentary settings.

Sedimentary systems affected by high fluxes of reactive iron develop a so-called 'cryptic' sulfur cycle in which hydrogen sulfide is nearly fully reoxidized and its concentrations of hydrogen sulfide in the porewaters are in the submicromolar range. The sediments of the Gulf of Aqaba represent a classic example of such a system. The goal of this work was to provide quantitative constraints on hydrogen sulfide concentrations in the sediments of the Gulf of Aqaba. Sulfate reduction rates in the sediments of the Gulf of Aqaba were found to be lower than in marine sediments that had not been affected by high fluxes of the reactive iron, while the rate constants of hydrogen sulfide oxidation were found to be higher than in the highly reactive iron-rich sediments of the Svalbard fjord. A combination of slow rates of sulfate reduction and fast rates of hydrogen sulfide oxidation results in concentrations of hydrogen sulfide in the porewater, both measured and calculated, that are below 100 nmol l(-1). We suggest that a similar cycling of sulfur species may occur in organic-matter-poor marine systems situated in dry environments with highly reactive iron mineral delivery, such as the Red Sea and the Atlantic Ocean in the vicinity of the Sahara.

Cadmium (Cd) isotopes are emerging tracers of marine biogeochemical cycling, but their incorporation into seep carbonates-abundant yet underutilized archives in the sedimentary record-remains poorly constrained. Here we report Cd contents and S114/110Cd values of seep carbonates from the South China Sea and the Gulf of Mexico. Sequential extraction experiments demonstrate that Cd is predominantly hosted in carbonate minerals rather than in sulfides, organic matter, or Fe-Mn oxides. Distinct S114/110Cd signatures are observed across carbonate mineral types, indicating that formation depth rather than mineralogy primarily controls Cd isotope partitioning in seep carbonates. Specifically, aragonite is isotopically lighter than or similar to bottom seawater, high-Mg calcite (HMC) exhibits S114/110Cd values consistent with bottom seawater, whereas dolomite exhibits lower values than co-occurring HMC. This distribution pattern is primarily controlled by the depth of the sulfatemethane transition zone rather than carbonate mineral facies. Because aragonite and HMC form at shallow seafloor depths while dolomite precipitates at greater depths, the S114/110Cd values of seep carbonates provide valuable records of pore-water conditions during carbonate formation. Compared with inorganic precipitation experiments, seep carbonates show a reduced magnitude of Cd isotopic fractionation, implying stronger controls from pore-water processes in natural settings. These results indicate that aragonite- and HMC-dominated seep carbonates can archive bottom seawater Cd isotopes, whereas dolomite preferentially records deeper pore-water signals. Seep carbonates thus provide a complementary archive for reconstructing Cd cycling and bottom seawater-pore-water gradients in early diagenetic environments.

Marine carbonates are widely regarded as excellent archives for recording the geochemical evolution of seawater. However, the long-standing “dolomite problem” has hindered the reconstruction of formation mechanisms and depositional settings of the widespread occurrences of dolomite throughout Earth’s history. Cold seeps, where sulfate-driven anaerobic oxidation of methane (SD-AOM) promotes the formation of dolomite, are key to deciphering low-temperature dolomitization processes. Here, we investigate late Miocene to early Pleistocene seep dolomite from Chiahsien, Taiwan, together with modern seep dolomite from the Gulf of Mexico and South China Sea, using a coupled molybdenum (Mo) and magnesium (Mg) isotope approach to constrain the dolomitization environment and processes within seep systems. The wide range of authigenic Mo isotope compositions (δ98Moauth: 0.29 to 2.94‰), correlating with Fe/Al ratios, tracks changes of methane flux spanning from diffusion-dominated (enrichment of heavy Mo) to oxide-shuttle-dominated (enrichment of light Mo) regimes. This contrasts with the homogeneous δ26Mg values of seep dolomite (−2.72 ± 0.21‰), reflecting equilibration with seawater during dolomitization in a shallow, porewater environment with high replenishment of seawater Mg. The Chiahsien seep dolomite further displays higher crystallographic ordering expressed as I(015)/I(110), generally lower Sr/(Mg + Ca) ratios, and more euhedral crystals than modern seep dolomite. Taken together, our new observations suggest that precursor carbonate minerals (aragonite, calcite) were dolomitized during early diagenesis at shallow depth. The lack of Rayleigh fractionation effects of Mg isotope distinguishes seep dolomite from dolomite formed in more restricted, deeper settings, such as environments dominated by organic sulfate reduction and methanogenesis. This study demonstrates that combined Mo–Mg isotope systematics can effectively decipher the redox history and the process of dolomitization. Furthermore, it suggests that seep dolomite, which tends to form in open, seawater-buffered porewater environments, is a promising archive for reconstructing the Mg isotope composition of ancient seawater.

The primary controls on sedimentary dolomite formation remain unresolved. The transformation from disordered to ordered dolomite at low temperatures is one of the key issues in the dolomite problem, as sedimentary dolomites in the geological record show varying degrees of cation ordering, whereas dolomites synthesized at ambient temperatures and those found in modern natural settings lack cation ordering. A primary obstacle in deciphering the mechanism of low-temperature dolomite ordering is the inability of existing proxies to decouple the effects of temperature from recrystallization. Here we show that a coupled oxygen-magnesium (O-Mg) isotope approach can separate the temperature effect from recrystallization signals. By compiling theoretical calculation data and experimental datasets, we demonstrate that temperature variations generate a negative correlation between delta 18O and delta 26Mg, whereas the increasing degree of recrystallization produces a positive correlation. This framework was then applied to Neogene dolomites from Well NK-1, Nansha Islands, South China Sea, that are inferred to have formed between 16 and 32 degrees C. Integrating petrographic, mineralogical, and geochemical data with principal component analysis, we found that dolomite ordering is primarily driven by recrystallization, with only a few samples showing a significant influence of temperature. Consequently, our results indicate that dolomite ordering can proceed at ambient temperatures, likely driven by thermodynamic stability and/or reaction kinetics, following Ostwald's step rule. This suggests that the fundamental pathways of dolomite formation and ordering in laboratory settings are intrinsically analogous to those in natural environments. Such mechanistic consistency validates the use of experimental data to decipher the recrystallization history of geological samples. Applying our approach to published sedimentary dolomite data indicates that the recrystallization signal is detectable mainly at low temperatures and can be obscured at higher temperatures or under large temperature variations. We propose that delta 18O serves as a proxy to constrain the dominant controls on dolomite ordering, whereas delta 26Mg may be used to indicate the conditions and stages of dolomite recrystallization.

Active nitrogen cycling, including microbially mediated nitrogen fixation and nitrogen loss, occurs at marine cold seeps and may influence nitrogen budgets in the deep ocean. However, the record of these processes in seep environments remains poorly constrained. Here, we investigate the nitrogen isotopic composition of organic nitrogen (delta 15NON) along with carbon isotopes and elemental composition of authigenic carbonates from four methane-seep sites and three brine-seep sites in the Gulf of Mexico and the South China Sea. Methane-seep carbonates exhibit significantly lower delta 13C values of total organic carbon (delta 13CTOC; mean: -34.7 +/- 8.0 parts per thousand, n = 32) than authigenic carbonates from brine seeps (mean: -27.9 +/- 8.4 parts per thousand, n = 24). In contrast, delta 15NON values are generally low for both types of seeps, with mean values of 0.1 +/- 2.2 parts per thousand (n = 32) for methane seeps and -1.2 +/- 0.8 parts per thousand (n = 24) for brine seeps. delta 15NON values below zero are common and delta 15NON values are largely decoupled from delta 13CTOC values at most sites, which is interpreted to reflect dynamic nitrogen cycling near the sediment-water interface at seeps, involving processes such as dissimilatory nitrate reduction to ammonium, denitrification, and anaerobic ammonium oxidation. The nitrogen involved in these processes is subsequently taken up by methanotrophic consortia and associated microorganisms, preserving its isotopic signature in the organic matter incorporated into seep carbonates. Our results demonstrate that delta 15NON values of seep carbonates represent a novel archive of a hotspot of benthic nitrogen cycling, providing new insight into nitrogen cycling in deep-sea seep environments.

Chemosymbiosis may have been a key driver of early animal evolution and diversification. Yet evidence for animal-microbe nutritional partnerships in the fossil record of early metazoans remains scarce, largely owing to the limited availability of robust geochemical proxies and suitable fossilization pathways. Here, we report extremely low molybdenum isotope (δ98Mo) values in pyritized fossils of the terminal Ediacaran Conotubus, a nonbiomineralized cloudinomorph representing some of the earliest metazoans. Such low δ98Mo values are otherwise only known in modern cold-seep chemosymbiotic tubeworms hosting sulfur-oxidizing bacterial symbionts. Homogeneously low δ34S signatures of fossils indicate that rapid pyritization in early diagenesis under euxinic taphonomic settings preserved primary Conotubus chitinous δ98Mo. Substantial δ34S fractionation relative to seawater sulfate and high Δ33S values indicate that active microbial H2S oxidation occurred within their habitat. Our findings provide evidence that Conotubus may have engaged in chemosymbiosis with sulfur-oxidizing bacteria-the earliest geochemically supported case in the fossil record. This trophic strategy likely enabled Conotubus to thrive in the redox-stratified Ediacaran ocean where sulfide and oxidants coexisted, facilitating its ecological success.

Barite is an important archive for tracing fluid-rock interactions in marine systems. However, accurate determination of rare earth elements (REEs) in barite by ICP-MS is strongly hindered by polyatomic ion interferences derived from high barium (Ba) matrix, particularly affecting europium (Eu) measurements. This analytical limitation has restricted the application of REE geochemistry in studies of barites from seeps (hereafter seep barites). Here we investigate barite-carbonate pairs from seeps in the Gulf of Mexico and develop a robust analytical protocol for reliable REE determination. The procedure integrates (1) selective removal of carbonate phases by mild acetic acid leaching, (2) complete digestion of the purified barite fraction using a mixed HNO3-HF-HClO4 acid system, and (3) efficient separation of Ba from REEs using DGA resin chromatography prior to ICP-MS analysis. Comparison of REE data before and after Ba removal demonstrates that extremely positive Eu anomalies observed in untreated samples (Eu/Eu* up to 122) may result from BaO+/BaOH+ interferences, highlighting the importance of careful matrix separation for reliable Eu determination. After matrix separation, the barite samples display moderate positive Eu anomalies (Eu/Eu* = 1.25-3.85), consistent negative Ce anomalies, and heavy REE-enriched patterns. These corrected REE signatures indicate precipitation from pore fluids interacting with surrounding sediments under suboxic to oxic conditions, rather than from high temperature hydrothermal fluids. Our results suggest that some previously reported extreme Eu anomalies in barite, particularly where Ba removal or interference correction was not documented, may warrant careful re-evaluation as potential analytical artifacts, and emphasize the necessity of matrix separation for reliable REE analysis in Barich minerals.

Natural cold seeps in the marine environment are significant contributors to atmospheric greenhouse gases. To accurately quantify the methane flux from cold seeps, a novel in situ acoustic measuring device was designed to determine the methane bubble flux using the attenuation in acoustic signal intensity caused by bubbles passing through a set pipe. The acoustic measuring device consists of an inverted conical tent and an acoustic measurement pipe with a rectangular cross-section. The acoustic measurement pipe comprises a bubble-fragmented homogenization unit and an acoustic cross-correlation sensor subsystem. The bubble-fragmented homogenization unit adjusts the rising gas bubbles to a diameter of less than 3 mm with a uniform spatial distribution. The acoustic cross-correlation sensor subsystem measures the bubble-rise rate and section void fraction to calculate bubble flux. The accuracy of the bubble rising rate is +/- 3%, and the relative error of bubble flux is 4.6%. The bubble flux of a single vent in the Lingtou promontory seep area is 2358.1 L, and the bubble rising rate ranges from 0.29 to 0.52 m/s, controlled by changes in ocean tides. Therefore, we believe this device can be a powerful tool for quantifying bubble flux in deep-sea environments.

The association of organic carbon (OC) to reactive iron oxides (Fe R ), forming OC‐Fe R complexes, represents a significant OC sink in marine sediments. However, the impact of diagenetic processes, such as sulfate reduction and iron sulfide formation, on the stability of OC‐Fe R in marine sediments remains poorly understood. Here, we compare sulfidic sediments from three cores taken at methane seeps with a non‐sulfidic sediment record from a nearby site. Our results show that an overall 6.3% decrease in OC‐Fe R is associated with a 42% reduction in Fe R during the transformation from iron oxides to iron sulfides, suggesting that OC‐Fe R is resistant to sulfidization. We observed highly 13 C‐depleted OC‐Fe R in the sulfidic sediments, likely due to the interaction between OC and Fe R during anaerobic oxidation of methane. Our findings highlight the stability of OC‐Fe R in natural sulfidic sediments, offering new insights into the role of OC‐Fe R in continental margin sediments.

The Mg isotope composition (delta Mg-26) of shallow-water carbonates provides valuable insights into the marine Mg cycle, past climate, and diagenesis. However, the effects of the mixing zone and meteoric diagenesis on the delta Mg-26 of carbonates remain poorly understood, and the variations of delta Mg-26 during diagenesis differ globally under diverse environmental settings. In this study, we investigate the controlling factors affecting delta Mg-26 of shallow-water carbonates during diagenesis across different mineralogies (aragonite, calcite, and dolomite) and diagenetic realms (meteoric, marine, and mixing zone diagenesis, as well as dolomitization) using bulk carbonate samples from Well NK-1, Meiji Atoll, South China Sea, by integrating mineralogical, geochemical analyses with numerical modeling. Our results reveal that delta Mg-26 of carbonate is primarily controlled by both mineralogy and diagenetic realms, with varying degrees of diagenetic alteration. Dolomites with a fluid-buffered origin and well-preserved high-Mg calcite exhibit delta Mg-26 offsets of similar to -2 parts per thousand and similar to -2.4 parts per thousand from coeval seawater, respectively, making them reliable archives for seawater delta Mg-26 reconstruction. In contrast, the delta Mg-26 of aragonite-dominated samples is easily influenced by mixed calcite due to the low Mg content in aragonite. Moreover, calcite displays a large delta Mg-26 variation from -5.1 parts per thousand to -3.1 parts per thousand, with the constrained Mg isotope fractionation during meteoric, mixing zone, and marine diagenesis overlapping between -4.5 parts per thousand and - 4 parts per thousand. This suggests that aragonite and low-Mg calcite are not ideal archives of seawater delta Mg-26, whereas the least altered limestone sample may represent a lower limit for coeval seawater delta Mg-26 with an offset of similar to -2.4 parts per thousand. By comparing our geochemical data with other modern shallow-water carbonates, we propose that variations of limestone delta Mg-26 from global sites are primarily controlled by both diagenetic realms and the degree of alteration. This finding underscores the importance of using more severely altered samples to assess the fidelity of carbonates in recording seawater chemistry. Our study enhances the understanding of the behavior of Mg isotopes in carbonates during diagenesis, evaluates the reliability of carbonate minerals as seawater delta Mg-26 archives, and clarifies key controls on carbonate delta Mg-26 across global settings.

Marine authigenic carbonates are considered a major long-term carbon sink, yet the role of silicate weathering in their formation remains unclear. Here we use lithium isotope compositions of authigenic carbonates from the Gulf of Mexico to trace the coupling between silicate weathering and carbonate authigenesis in marine sediment. In contrast to the positive delta 7Licarb values in seep carbonates formed near the seafloor, carbonates originating from organoclastic sulphate reduction in deeper burial settings exhibit negative delta 7Licarb values (-6.6 parts per thousand to -1.2 parts per thousand). Calculated delta 7Lifluid values (-4 parts per thousand to +1.4 parts per thousand) in pore fluids closely match the delta 7Lisilicate values of the silicate component in the rocks, suggesting congruent silicate weathering is approached. A positive correlation between delta 7Lifluid and delta 13Ccarb values indicates organoclastic sulphate reduction enhances silicate weathering, which together promote carbonate authigenesis in anoxic sediments. Our findings demonstrate that lithium isotopes are a valuable tool for tracing carbon-silicate interactions and reconstructing carbon cycling in anoxic sediments.

Shallow-water carbonates record seawater sulfate and barium isotope signatures, offering valuable insight into ocean biogeochemistry. However, the robustness of these proxies can vary with carbonate origin, mineralogy, and the extent of diagenesis. Here, we assess delta S-34 and delta O-18 compositions of carbonate-associated sulfate (CAS; hereafter delta S-34(CAS) and delta O-18(CAS)) and barium isotope compositions of carbonates (delta Ba-138(carb)) of three lithologies, aragonite-rich mud, high-Mg calcite-rich clusters with coralline algae remains, and coral-microbialite facies from a reef environment in the South China Sea. Aragonitic mud shows lower CAS/Ca and higher Ba/Ca molar ratios than calcitic components, reflecting mineralogical control on trace element incorporation. While delta S-34(CAS) values of carbonate clusters and coral-microbialite facies closely resemble seawater values, delta O-18(CAS) values are about 1 parts per thousand higher, consistent with moderate oxygen isotope fractionation. In contrast, aragonite-rich mud displays lower delta S-34(CAS) values (similar to 0.9 parts per thousand lower) and delta O-18(CAS) values (similar to 2.0 parts per thousand lower) than seawater, likely due to incorporation of sulfate deriving from sulfide oxidation. Notably, delta Ba-138(carb) values remain constant across all carbonate facies - regardless of mineralogy, biological origin, or diagenesis - highlighting the robustness of the barium isotope proxy. This study recognizes different degrees of primary fractionation during formation and preservational biases among the analyzed proxies: delta Ba-138(carb) values exhibit constant isotope offset with seawater composition irrespective of the archive or the degree of alteration, while delta S-34(CAS) values and even more so delta O-18(CAS) values are compromised by different degrees of primary fractionation upon incorporation of sulfate into the carbonate mineral lattice and secondary alteration. Calibrated offsets from seawater composition for reef carbonates are ca. +/- 1 parts per thousand (2 sigma) for delta S-34, 1.0 +/- 2.2 parts per thousand (2 sigma) for delta O-18, and - 0.23 +/- 0.08 parts per thousand (2 sigma) for delta Ba-138. Our findings help to assess the reliability of isotope-based reconstructions for mixed carbonate lithologies and underscore the robustness of delta S-34(CAS) and delta Ba-138(carb) values as proxies for past seawater chemistry.

Oil seeps are important hydrocarbon sources to the ocean and sustain diverse chemosynthesis-based ecosystems. Distinguishing oil- from methane-dominated seeps in the geological record is crucial for reconstructing the evolution of chemosynthetic communities, yet remains challenging. Authigenic carbonates precipitating at seeps offer a valuable archive of the composition of seep fluid. While previous work has focused on fibrous aragonite cement, the geochemical potential of volumetrically dominant microcrystalline analog remains underexplored. Here, we present a comprehensive study of the petrography and stable isotope as well as element composition of oil-derived seep carbonates from the northern Gulf of Mexico (sites GC232, GC185), compared to methanederived seep carbonates from the northern Gulf of Mexico (site AT340) and the South China Sea (site GMGS208). All studied samples are dominated by microcrystalline aragonite. Carbon isotope data (average delta 13C = -20.9 +/- 4.5 parts per thousand; VPDB) and the presence of oil residues agree with crude oil degradation as the main carbon source of the oil-seep carbonate. Compared to methane-seep carbonates, oil-seep carbonates exhibit higher rare earth elements (REE) contents, slight light-REE enrichment, and elevated contents of uranium (U), molybdenum (Mo), zinc (Zn), and cobalt (Co), signatures linked to microbial oil oxidation. These element patterns preserved in microcrystalline aragonite offer a novel proxy for identifying oil seepage in the rock record, particularly in cases where fibrous cement is absent.