The Formation of Authigenic Carbonates at a Methane Seep Site in the Northern Part of the Laptev Sea

: Authigenic carbonates from cold seeps are unique archives for studying environmental conditions, including biogeochemical processes associated with methane-rich ﬂuid migration through the sediment column. The aim of this research was to study major oxide, mineralogical, and stable isotopic compositions of cold-seep authigenic carbonates collected in the northern part of the Laptev Sea. These carbonates are represented by Mg-calcite with an Mg content of 2% to 8%. The δ 13 C values range from − 27.5% (cid:24) to − 28.2% (cid:24) Vienna Peedee belemnite (VPDB) and indicate that carbonates formed due to anaerobic oxidation of methane, most likely thermogenic in origin. The authigenic pyrite in Mg-calcite is evidence of sulfate reduction during carbonate precipitation. The δ 18 O values of carbonates vary from 3.5% (cid:24) to 3.8% (cid:24) VPDB. The calculated δ 18 O ﬂuid values show that pore water temperature for precipitated Mg-calcite was comparable to bottom seawater temperature. The presence of authigenic carbonate in the upper horizons of sediments suggests that the sulfate–methane transition zone is shallowly below the sediment–water interface.


Introduction
Methane cold seeps are a widespread phenomenon that occurs on the shelves and continental slopes of inland and marginal seas around the world [1][2][3][4][5][6][7], including the arctic seas [8][9][10]. They represent areas of the seafloor where methane-rich fluids in chemical non-equilibrium with seawater are transported through the sub-seafloor environment and are released at the bottom-sediment-seawater interface [11][12][13]. The variability of methane seepage activity over time leads to significant changes in sedimentary redox conditions [14,15] and vertical displacement of biogeochemical zones [16,17]. The anaerobic oxidation of methane (AOM) coupled with sulfate reduction (SR) is a key biogeochemical process at seep sites [12,17,18]. AOM and SR are mediated by a consortium of methane-oxidizing archaea (ANME) and sulfate-reducing bacteria (SRB). As a result of AOM, an abundance of dissolved inorganic carbon arises and the alkalinity of pore water increases. The presence of seawater cations (Ca 2+ , Mg 2+ ) and elevated carbonate alkalinity promote the precipitation of authigenic carbonates [7,19], which show a wide range of mineralogical compositions. The occurrence of Mg-calcite and aragonite in cold seep marine environments has been reported by various authors [1,4,20]. Other carbonate minerals, such as dolomite, ankerite, ikaite, and glendonite, are significantly less common [21,22].
Authigenic carbonates from cold seeps are unique archives for studying environmental conditions, including biogeochemical processes associated with the migration of methane-rich fluid through the sediment column [19,23,24]. The isotopic compositions of carbon and oxygen in authigenic carbonates  Baranov et al. [35] and Sekretov [37].

Analytical Methods
Grain size distribution was measured using a SALD-2201 laser diffraction particle size analyzer (Shimadzu, Japan). The samples were placed in a mixer bath with distilled water and dispersed using an ultrasonic device (40 W,40 kHz). Measurements were performed in a flow cell. Grain sizes were classified into three fractions: sand (>63 μm), silt (2-63 μm), and clay (<2 μm).
The bulk mineralogical composition was determined using a Bruker D2 Phase X-ray diffractometer (Billerica, MA, USA) with Cu-Kα radiation at a current of 10 mA and a voltage of 30 kV. Size fractions less than 10 μm of powdered samples were scanned from 5° to 70° 2θ, with a step size of 0.02° and a scanning rate of 2 s. The diverging, scattering, and receiving slits were 1 mm, 3 mm, and 0.3 mm, respectively. Microstructures of carbonates from sediments were studied under a TESCAN VEGA 3 SBU (TESCAN, Brno, Czech Republic) scanning electron microscope (SEM) and an OXFORD X-Max 50 (OXFORD Instruments, High Wycombe, UK) energy-dispersive adapter (EDS) with a 20 kV accelerating voltage, specimen current of 4-12 nA, and a spot diameter of approximately 0.1-2 μm. Carbonate fractions were collected using an optical microscope and fixed to a double-sided 9 mm carbon tab placed on an aluminum stub. It was carbon-coated (30 nm) prior to analysis.
The major element concentrations of the sediments were estimated using a HORIBA XGT 7200 X-ray fluorescence microscope (Kyoto, Japan) operated at a tube current of 0.5, with a 1.2 mm X-ray beam for 100 s and a voltage of 50 kV. The detection limit for major elements was 0.01 wt %. For carbon (δ 13 C) and oxygen (δ 18 O) stable isotope analysis, powdered samples of carbonates (approximately 100 μg) were treated with 100% phosphoric acid at 90 °C to release CO2 for measurements using a Delta V Advantage mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) at the Siberian Research Institute of Geology, Geophysics, and Mineral Resources. The values were expressed using the δ notation relative to the Vienna Peedee belemnite (VPDB) standard for δ 13 C and Vienna Standard Mean Ocean Water (VSMOW) for δ 18 O. The precision of the δ 13 C and δ 18 O measurements was on the order of 0.1‰. The δ 18 O values were converted from VSMOW to VPDB scales using the following conversion [41]:  Baranov et al. [35] and Sekretov [37].

Analytical Methods
Grain size distribution was measured using a SALD-2201 laser diffraction particle size analyzer (Shimadzu, Japan). The samples were placed in a mixer bath with distilled water and dispersed using an ultrasonic device (40 W, 40 kHz). Measurements were performed in a flow cell. Grain sizes were classified into three fractions: sand (>63 µm), silt (2-63 µm), and clay (<2 µm).
The bulk mineralogical composition was determined using a Bruker D2 Phase X-ray diffractometer (Billerica, MA, USA) with Cu-Kα radiation at a current of 10 mA and a voltage of 30 kV. Size fractions less than 10 µm of powdered samples were scanned from 5 • to 70 • 2θ, with a step size of 0.02 • and a scanning rate of 2 s. The diverging, scattering, and receiving slits were 1 mm, 3 mm, and 0.3 mm, respectively. Microstructures of carbonates from sediments were studied under a TESCAN VEGA 3 SBU (TESCAN, Brno, Czech Republic) scanning electron microscope (SEM) and an OXFORD X-Max 50 (OXFORD Instruments, High Wycombe, UK) energy-dispersive adapter (EDS) with a 20 kV accelerating voltage, specimen current of 4-12 nA, and a spot diameter of approximately 0.1-2 µm. Carbonate fractions were collected using an optical microscope and fixed to a double-sided 9 mm carbon tab placed on an aluminum stub. It was carbon-coated (30 nm) prior to analysis.
The major element concentrations of the sediments were estimated using a HORIBA XGT 7200 X-ray fluorescence microscope (Kyoto, Japan) operated at a tube current of 0.5, with a 1.2 mm X-ray beam for 100 s and a voltage of 50 kV. The detection limit for major elements was 0.01 wt %. For carbon (δ 13 C) and oxygen (δ 18 O) stable isotope analysis, powdered samples of carbonates (approximately 100 µg) were treated with 100% phosphoric acid at 90 • C to release CO 2 for measurements using a Delta V Advantage mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) at the Siberian Research Institute of Geology, Geophysics, and Mineral Resources. The values were expressed using the δ notation relative to the Vienna Peedee belemnite (VPDB) standard for δ 13 C and Vienna Standard Mean Ocean Water (VSMOW) for δ 18 O. The precision of the δ 13 C and δ 18 O measurements was on the order of 0.1% . The δ 18 O values were converted from VSMOW to VPDB scales using the following conversion [41]:

Grain Composition of Sediments
The results of the grain-size analysis for the studied sediment samples show similar particle size distribution. Three size fractions were identified: sand (>63 µm), silt (2-63 µm), and clay (<2 µm). The content of sand, silt, and clay size fractions varied from 0.0% to 5.3%, 65.2% to 89.1%, and 9.6% to 34.9%, respectively, and indicated the predominance of silt among the studied samples. The only exception was station AMK-6056, where sediments were represented by clay silt (the content of the clay size fraction is 34.9%). Sand and clay were not found among the studied samples. The maximum content of the clay size fraction (34.9%) was noted in the sample taken at the AMK-6056 station. The sediments sampled at the cold seep site (stations AMK-6027, AMK-6045, LV78-17, and LV78-12) were represented by dense black silt with a smell of hydrogen sulfide.

The Bulk Mineral Composition
The bulk mineral composition of the studied samples was represented mainly by quartz, feldspars, and clay minerals (illite + chlorite + montmorillonite; Figure 2a). Small amounts of amphiboles were present, the content of which did not exceed 3% on average (with the exception of sample LV78-23, where their concentration increases to 4.5%). The total content of quartz and feldspars varied between 62.7% to 89.3%, the average being 79.0%. Illite-the content of which ranged from 2.6% to 30.9%-prevailed among clay minerals in all studied samples. In general, the content of clay minerals was within the range of 9.5-37.8%. The maximum content of clay minerals was noted in sample AMK-6056, and the minimum was in the sample from station LV78-23 (depth 22 m).

Grain Composition of Sediments
The results of the grain-size analysis for the studied sediment samples show similar particle size distribution. Three size fractions were identified: sand (>63 μm), silt (2-63 μm), and clay (<2 μm). The content of sand, silt, and clay size fractions varied from 0.0% to 5.3%, 65.2% to 89.1%, and 9.6% to 34.9%, respectively, and indicated the predominance of silt among the studied samples. The only exception was station AMK-6056, where sediments were represented by clay silt (the content of the clay size fraction is 34.9%). Sand and clay were not found among the studied samples. The maximum content of the clay size fraction (34.9%) was noted in the sample taken at the AMK-6056 station. The sediments sampled at the cold seep site (stations AMK-6027, AMK-6045, LV78-17, and LV78-12) were represented by dense black silt with a smell of hydrogen sulfide.

The Bulk Mineral Composition
The bulk mineral composition of the studied samples was represented mainly by quartz, feldspars, and clay minerals (illite + chlorite + montmorillonite; Figure 2a). Small amounts of amphiboles were present, the content of which did not exceed 3% on average (with the exception of sample LV78-23, where their concentration increases to 4.5%). The total content of quartz and feldspars varied between 62.7% to 89.3%, the average being 79.0%. Illite-the content of which ranged from 2.6% to 30.9%-prevailed among clay minerals in all studied samples. In general, the content of clay minerals was within the range of 9.5-37.8%. The maximum content of clay minerals was noted in sample AMK-6056, and the minimum was in the sample from station LV78-23 (depth 22 m).

Major Element Composition in Sediments
Major element concentrations in the studied samples are listed in Table 1. Sediments were characterized by the high concentrations of SiO 2 (56.9-65.5%), Al 2 O 3 (12.6-14.6%), and Fe 2 O 3 (4.2-9.2%), as well as the relatively low contents of CaO (0.9-2.9%). The concentrations of MgO, MnO, P 2 O 5 , TiO 2 , Na 2 O, and K 2 O were no more than 3%. In general, all studied samples had a relatively homogeneous chemical composition. However, in sediments containing authigenic carbonates, CaO concentrations were significantly higher than non-carbonate sediments.

Authigenic Carbonates
Authigenic carbonates were found in three samples collected at stations AMK-6045 and LV78-12. Typical macroscopic features of authigenic carbonate samples are shown in Figure 3. The host sediments were represented by dense black silt with an admixture of clay particles (13.28%). Carbonates were represented by isometric concretions with sizes up to 2 cm in diameter and nodules with a diameter of 2-3 mm (Figures 3 and 4). They have yellow-brown color and a consertal structure and are dense, bioturbated, and strongly cemented in texture. In concretions, we noted cavities up to 4 mm in diameter and hollow channels with a diameter of 1-2 mm.

Major Element Composition in Sediments
Major element concentrations in the studied samples are listed in Table 1. Sediments were characterized by the high concentrations of SiO2 (56.9-65.5%), Al2O3 (12.6-14.6%), and Fe2O3 (4.2-9.2%), as well as the relatively low contents of CaO (0.9-2.9%). The concentrations of MgO, MnO, P2O5, TiO2, Na2O, and K2O were no more than 3%. In general, all studied samples had a relatively homogeneous chemical composition. However, in sediments containing authigenic carbonates, CaO concentrations were significantly higher than non-carbonate sediments.

Authigenic Carbonates
Authigenic carbonates were found in three samples collected at stations AMK-6045 and LV78-12. Typical macroscopic features of authigenic carbonate samples are shown in Figure 3. The host sediments were represented by dense black silt with an admixture of clay particles (13.28%). Carbonates were represented by isometric concretions with sizes up to 2 cm in diameter and nodules with a diameter of 2-3 mm (Figures 3 and 4). They have yellow-brown color and a consertal structure and are dense, bioturbated, and strongly cemented in texture. In concretions, we noted cavities up to 4 mm in diameter and hollow channels with a diameter of 1-2 mm.   Petrographic observations show that the carbonate concretions are made up of a homogeneous microcrystalline carbonate matrix (Figure 5a), in which detrital grains of quartz, feldspars, and opaque minerals, are interpsersed (Figures 4 and 5c,d). The XRD analysis of the carbonate nodules ( Figure 2b) revealed that the carbonate phase is represented by Mg-calcite (25-36%). The accessory non-carbonate minerals were mainly quartz (22-28%), feldspar (23-29%), clay minerals (illite + chlorite + montmorillonite, 18-23%), amphiboles (~1%), and halite (~1%). Quartz and feldspars were represented by semi-rounded silt-sized grains randomly scattered within the carbonate matrix (Figures 4 and 5d). The results of the EDS analysis show that the Mg content in calcite varied from 2% to 8%.
Authigenic pyrite was found in samples AMK-6045, AMK-6045/2, and LV78-12, and was in paragenesis with Mg-calcite (Figure 5b,e). Moreover, it was the main form of iron disulfide in carbonate concretions. Pyrite was represented by individual microcrystals in cubic and occasionally octahedral forms embedded into Mg-calcite nodules. The size of pyrite crystals varied from 2 to 10 μm. Sample LV78-12 contained individual pyrite framboids no more than 5 μm in size that were adsorbed on the surface of clay minerals. Petrographic observations show that the carbonate concretions are made up of a homogeneous microcrystalline carbonate matrix (Figure 5a), in which detrital grains of quartz, feldspars, and opaque minerals, are interpsersed (Figures 4 and 5c,d). The XRD analysis of the carbonate nodules ( Figure 2b) revealed that the carbonate phase is represented by Mg-calcite (25-36%). The accessory non-carbonate minerals were mainly quartz (22-28%), feldspar (23-29%), clay minerals (illite + chlorite + montmorillonite, 18-23%), amphiboles (~1%), and halite (~1%). Quartz and feldspars were represented by semi-rounded silt-sized grains randomly scattered within the carbonate matrix (Figures 4 and 5d). The results of the EDS analysis show that the Mg content in calcite varied from 2% to 8%.
Authigenic pyrite was found in samples AMK-6045, AMK-6045/2, and LV78-12, and was in paragenesis with Mg-calcite (Figure 5b,e). Moreover, it was the main form of iron disulfide in carbonate concretions. Pyrite was represented by individual microcrystals in cubic and occasionally octahedral forms embedded into Mg-calcite nodules. The size of pyrite crystals varied from 2 to 10 µm. Sample LV78-12 contained individual pyrite framboids no more than 5 µm in size that were adsorbed on the surface of clay minerals.

Stable Carbon and Oxygen Isotope Data
The collected carbonate samples had very similar carbon-and oxygen-stable isotopic compositions. All samples had negative δ 13

Discussion
Petrological, mineralogical, and isotope geochemical characteristics of the authigenic carbonates provide important information for understanding their formation mechanisms as well as the source and temperature of fluids.
Recent investigations of sediment grain size in the study area [44,45] have shown that the main type of surface sediments was silt and clay. However, in some sites of the study area, poorly sorted sediments composed of multiple grain sizes (including grain >1 mm) were found. Methane seepage, which can contribute to the "washout" of the clay and silt size fractions has been proposed as one of the hypotheses that explain such a distribution of grain size in the sediment. The results of the grain size analysis obtained in this study show a rather uniform grain size composition of sediments and practically the absence of the sand fraction. The grain size, as well as the bulk mineral composition, of the host sediments from the studied cold seep sites and background sediments are very similar.
A distinctive feature of modern bottom sediments of the Arctic seas, as compared to the sediments of the seas of humid and arid zones, is the almost complete absence of carbonate minerals in their composition. The reason for this is the increased solubility of CO2 caused by low seawater temperatures [46].
Among the studied samples, authigenic carbonate, represented by Mg-calcite, was found in samples AMK-6045, AMK-6045/2, and LV78-12. The dissolved methane concentrations in the surface seawater and hydroacoustic data confirm active methane seepage at the site where these samples were collected [9,34,47,48]. Mg-calcite-associated pyrite indicates the activity of the sulfate reduction process during carbonate precipitation in an anaerobic environment [24,49]. Pyrite formation probably occurred during the interaction of hydrogen sulfide produced due to sulfate reduction with iron ions (the source of which is detrital minerals). As a result of this reaction, metastable iron monosulfides (mackinawite, greigite) were formed, which subsequently recrystallized to pyrite [6,50].

Discussion
Petrological, mineralogical, and isotope geochemical characteristics of the authigenic carbonates provide important information for understanding their formation mechanisms as well as the source and temperature of fluids.
Recent investigations of sediment grain size in the study area [44,45] have shown that the main type of surface sediments was silt and clay. However, in some sites of the study area, poorly sorted sediments composed of multiple grain sizes (including grain >1 mm) were found. Methane seepage, which can contribute to the "washout" of the clay and silt size fractions has been proposed as one of the hypotheses that explain such a distribution of grain size in the sediment. The results of the grain size analysis obtained in this study show a rather uniform grain size composition of sediments and practically the absence of the sand fraction. The grain size, as well as the bulk mineral composition, of the host sediments from the studied cold seep sites and background sediments are very similar.
A distinctive feature of modern bottom sediments of the Arctic seas, as compared to the sediments of the seas of humid and arid zones, is the almost complete absence of carbonate minerals in their composition. The reason for this is the increased solubility of CO2 caused by low seawater temperatures [46].
Among the studied samples, authigenic carbonate, represented by Mg-calcite, was found in samples AMK-6045, AMK-6045/2, and LV78-12. The dissolved methane concentrations in the surface seawater and hydroacoustic data confirm active methane seepage at the site where these samples were collected [9,34,47,48]. Mg-calcite-associated pyrite indicates the activity of the sulfate reduction process during carbonate precipitation in an anaerobic environment [24,49]. Pyrite formation probably occurred during the interaction of hydrogen sulfide produced due to sulfate reduction with iron ions (the source of which is detrital minerals). As a result of this reaction, metastable iron monosulfides (mackinawite, greigite) were formed, which subsequently recrystallized to pyrite [6,50].
The presence of carbonate and sulfide authigenic minerals in the upper horizon of sediments (~5 cm) suggests the shallow position of the sulfate-methane transit zone [51,52]. AOM often occurs in the upper horizon of sediments or even directly at the sediment-seawater interface at high fluid flows [53][54][55]. In this case, the pore waters should be characterized by high concentrations of sulfate ions, which would cause aragonite precipitation [4,49]. However, Mg-calcite was the only form of calcium carbonate in the studied samples. In addition to SEM and XRD results, the Mg/Ca ratio in the host sediments confirms this [56,57]. The Mg-calcite precipitation probably occurred in a reducing environment with a sulfate depletion in pore water and the participation of dissolved sulfide, as was shown by the example of carbonates from the Arabian Sea [58]. These geochemical conditions differ from that of hydrocarbon seeps in the Gulf of Cadiz [42,59,60], where pyrite forms in the association with carbonates and Fe-Mn nodules [61].
The isotopic composition of carbonates can often show effects of mixing of carbon from different sources (hydrocarbon gases including methane, crude oil, dissolved carbon in seawater, marine biogenic carbonate, and organic sediment matter), which makes it more difficult to identify the composition of the parent carbon [32,42,43]. An example of mixing methane sources has been reported for hydrocarbon-derived carbonate chimneys in the Gulf of Cadiz [42,43]. The carbon isotope composition of authigenic carbonates from stations AMK-6045 and LV78-12 was characterized by a very small variation of the δ 13 C values ( Figure 6). The narrow range variation of the δ 13 C (from −27.5% to −28.2% ) values indicate the predominance of a single carbon source during the precipitation of these authigenic carbonates. Moreover, the δ 13 C values suggest that the carbonates formed due to anaerobic oxidation of 13 C-depleted methane [4,13,32,62].
The negative carbon isotope compositions of studied carbonates samples (average −27.7% ) indicate that the most likely carbon source is thermogenic methane (δ 13 C values of thermogenic methane range from −30% to −50% [42,67]). Carbonates derived from this methane type have δ 13 C values varying from −20% to −50% [42,68,69]. The source of thermogenic methane is likely to be gas condensate or oil reservoir in underlying sedimentary sequences. The molecular composition of sediment organic matter in the study area suggests hydrocarbon migration from underlying deposits [70]. Faults of the Laptev Sea Rift System can serve as pathways for upward migration of methane. Gas hydrate deposits, which hypothetically accumulate under submarine permafrost [34,39,48] can also be a potential source of thermogenic methane. It should be noted that carbonates resulting from oxidation of crude oil have δ 13 C values ranging from −25% to −30% [71], but in this case, they showed spots of biodegraded crude oil within the carbonate samples [26,[72][73][74]. In our case, oil spots were not observed. The contribution of sedimentary organic matter was not considered because the total organic carbon content in the sediments collected at or near the sites where the carbonates were recovered did not exceed 1% [45].
The oxygen isotope composition of the authigenic carbonates provides additional information on the seawater/pore water temperature during precipitation and the δ 18 O values of the pore-fluid from which the carbonates precipitated [26,64,65,75]. In this research paper, the theoretical values of δ 18 O fluid were calculated using the bottom seawater temperatures, using the equation proposed by Anderson and Arthur [75]. The average temperature of the bottom seawater at the stations where carbonates were sampled was about −1.5 • C (unpublished data). The calculated δ 18 O fluid values range from −0.5% to −0.2% VSMOW and were comparable to the δ 18 O values of the bottom seawater in the northern part of the Laptev Sea (from −0.7% to 0.4% VSMOW; [76]). Assuming that δ 18 O fluid was in equilibrium with δ 18 O seawater , the authigenic carbonates of the present study precipitated under the same pore water temperature as the bottom water temperature.

Conclusions
In this study, we characterized the major oxide, mineralogical, and stable isotopic compositions of authigenic carbonates from methane seep site of the northern part of the Laptev Sea. These carbonates were represented by Mg-calcite with a magnesium content between 2% to 8%. The negative carbon isotope compositions of studied carbonate samples suggest that carbonate formation was a result of the anaerobic oxidation of thermogenic methane (average carbonate δ 13 C −27.7% ). Authigenic pyrite within the carbonate matrix is evidence of sulfate reduction during anaerobic oxidation of methane. The hydrogen sulfide produced in this process interacted with the iron ions contained in pore water, resulting in precipitation of iron monosulfides, which recrystallized into pyrite. The narrow range of the δ 13 C values indicates the predominance of a single carbon source during the precipitation of these authigenic carbonates, which was probably thermogenic methane. The calculated δ 18 Ofluid values show that pore water temperature during Mg-calcite precipitation was comparable to bottom seawater temperature. The presence of authigenic carbonate in the upper horizons of sediments suggests that the sulfate-methane transition zone is located shallowly below the sediment-water interface.