Coupling and Decoupling Between Sedimentary Mercury and Organic Carbon Preservation in the Oxygenated Marine Environment

Mercury (Hg) enrichment relative to total organic carbon (TOC) in sedimentary records has been widely used as a volcanism proxy. However, the depositional and diagenetic effects on Hg burial are not well understood, limiting the reliability of the proxy. Here, we report a systematic investigation of Hg sedimentation under well‐oxygenated bottom water. Most of the studied cores show total mercury content (THg) to TOC ratio between 50 and 300 ppb/%, including northwest Pacific ODP Site 1208, East Equatorial Pacific ODP Sites 677 and 1241, the Antarctic Zone ODP Site 1094, and East China Sea sediment core EC2005. The consistent THg/TOC ratio confirms the strong coupling between Hg and particulate organic matter (POM) despite large differences in geographic locations, sedimentation rates, TOC content, POM sources, and early diagenetic environments. Nevertheless, the THg/TOC ratio is higher in sediments under well‐oxygenated bottom water than in those under oxygen‐depleted waters, probably as a result of a higher degree of organic matter degradation in oxygenated sediments during early diagenesis. The THg in the high TOC variability section at Site 1208 is abnormally high, resulting in decoupling between Hg and POM. Mercury in this section is still leachable by oxidizing solution and has a similar trend of variability with easily reducible iron content, implying that the metal oxide‐organic matter association might act as Hg bounding phase. We suggest that the enriched Hg is probably supplied by the neighboring high TOC sediment. Therefore, while THg/TOC > 300 ppb/% could be considered as distinct Hg enrichment in sedimentary records, the diagenetic mobilization effect should be first excluded as the possible cause as demonstrated by the records of site 1208. Our study therefore provides new insights into the Hg cycle in the modern ocean and the utilization of Hg enrichment as a volcanism proxy.


Introduction
The major primary source of mercury (Hg) in surficial ecosystems is volcanic degassing before intense anthropogenic emission (Mason & Sheu, 2002;Selin, 2009).As a result, Hg enrichment is widely viewed as an evidence of enhanced volcanic degassing (Bergquist, 2017).Mercury enrichment in sedimentary records has been reported on all of the "big five" mass extinctions (Grasby et al., 2019), ocean anoxic events (OAE) (Percival et al., 2015), and Paleocene-Eocene Thermal Maximum (PETM) (Kender et al., 2021).Most of these Hg enrichments are explained to be related to magmatic CO 2 release due to volcanic activity or sedimentary organic carbon burning due to magmatic intrusion, building up a link between volcanic degassing and climate perturbation in sedimentary records (Grasby et al., 2019;Shen et al., 2022).Hg in the surficial ecosystem is actively transported among the atmosphere, ocean, pedosphere, and biosphere, with the atmosphere being an important hub (Selin et al., 2008).The residence time of natural gaseous Hg in the atmosphere is about 1 year, allowing at least hemispheric scale Hg mixing before it deposits (Kawai et al., 2020;Semeniuk & Dastoor, 2017).As a result, global atmospheric Hg concentration is homogeneous and potentially influenced by enhanced volcanic degassing on a geological timescale.The atmosphere and the surface ocean actively exchange Hg by large fluxes of wet and dry deposition as well as Hg evasion (Lamborg et al., 2002;Mason et al., 2003), allowing rapid balance between atmospheric and surface seawater Hg concentrations.Sinking particulate organic matter (POM) in surface oceans adsorbs dissolved Hg and transports it to the deep ocean and sediment-water interface (Bowman et al., 2016;Cossa et al., 2018), which is the precursor of geological records.As a result, the sedimentary total Hg concentration (THg) has the potential to record the atmospheric Hg spike caused by volcanic degassing.Box-model study on the Siberian Traps eruption suggests that it leads to orders of magnitude higher Hg content in the atmosphere, which is able to be recorded by marine sediment as a chemostratigraphic marker, and potentially to be toxic to both marine and terrestrial ecosystems (Grasby et al., 2020).
In geological records, Hg should be normalized to its major bounding phase to eliminate the influence of the dilution effect caused by major sediment components such as calcite and clay minerals.POM is the major Hg bounding phase as suggested by sequential extraction experiments (Bloom et al., 2003;Hall et al., 2005;Hall & Pelchat, 2005).In geological records, total organic carbon (TOC) is used to quantify POM and normalize THg (Sanei et al., 2012).Although other Hg bounding phases are put up, for example, sulfide minerals in geologic records with high total sulfur content (Sanei et al., 2012;Shen, Algeo, Chen, et al., 2019;Shen et al., 2020) and secondary minerals in some tropical soils (Shen, Yu, et al., 2019;X. Wang et al., 2021), POM is recognized to be the major Hg bounding phase in most of the cases.However, using THg/TOC ratios as a volcanism proxy has not been tested by marine sediments and may bear uncertainties in two aspects.First, although global large-scale volcanism has the potential to result in Hg enrichment in the geological record, local or regional hydrothermal input may also raise the sedimentary THg/TOC.Second, the transport of Hg from the sea surface to the deep ocean and sediment relies on POM, which is a chemically reactive during sedimentation and early diagenesis.The degradation of POM may have an imprint on the THg/TOC in the geological record.
The total Hg concentration in seawater is about 1 pM and does not vary much globally (Bowman et al., 2020).The hydrothermal input is reported to increase the local total Hg concentration in the seawater up to 5 pM (Bowman et al., 2015;Roberts et al., 2021).THg enrichment is also found in sediments under the influence of hydrothermal fluids (Cox & McMurtry, 1981;Lamborg et al., 2006).Therefore, some of the elevated THg/TOC ratio in geological records are attributed to past hydrothermal input (Jin et al., 2023;Zhu et al., 2021).However, the spatial scale of hydrothermal influence on sedimentary THg/TOC ratios is poorly constrained.
Weathering of sedimentary rock and post-depositional diagenetic degradation of sedimentary POM is observed to change THg/TOC ratios in geological records (Charbonnier et al., 2020), whereas the effect of early diagenesis needs to be tested.Studies on modern marine Hg sedimentation have been conducted around the global (Figure 1), providing the opportunity to investigate the effect of POM degradation on THg/TOC ratios.Sediment under the Arabian oxygen minimum zone shows a decreasing THg/TOC ratio with decreasing bottom water oxygen concentration, which is explained to be the result of preferential Hg loss during early diagenesis (Frieling et al., 2023).Thermo-desorption of marine sediments show that Hg can be bound to POM with different degrees  (Cossa et al., 2021;Fadina et al., 2019;Figueiredo et al., 2020;Frieling et al., 2023;Gehrke et al., 2009;Kim et al., 2020;Kita et al., 2013Kita et al., , 2016;;Munson et al., 2015)).The map was drawn with ODV software (Schlitzer, 2007).For sites with significant change in geographic position during deposition, moving tracks reconstructed by Gplate (R. D. Müller et al., 2018) are indicated by yellow lines with numbers indicating the age in million years.
of degradation (Pérez-Rodríguez et al., 2019).Research on estuary sediments indicates that mature POM is more enriched in Hg, probably due to the increase in complexation of functional groups during degradation (Laurier et al., 2003).These observations and experiments indicate that the diagenetic degradation of POM may change its Hg bounding ability; however, the direction and extent of the change in marine sediment need to be constrained.
Here, we provide a systematic study of marine sediment to determine if distal hydrothermal input and early diagenesis will cause large differences in sedimentary THg/TOC.The core sites used in this study span the globe, including East Equatorial Pacific ODP Sites 677 and 1241, southern Pacific oligotrophic gyre IODP Site U1366, northwestern Pacific ODP Site 1208, Southern Ocean ODP Site 1094, and East China Sea sediment core EC2005.The comparison between sites with and without hydrothermal input under different productivity regimes (Sites 677, 1241, and U1366) indicates that long distance hydrothermal transport under the oxygenated bottom water conditions does not induce sedimentary Hg enrichment.The consistent Hg/TOC ratios (between 50 and 300 ppb/ %) of the sediments with different geographic locations, sedimentation rates, TOC fluxes, POM sources, and early diagenetic environments confirm a strong coupling between POM and Hg in most marine sediments.We observed an abnormally high Hg/TOC ratio at the high TOC variability section at Site 1208, resulting from Hg vertical migration from high to low TOC layers, suggesting that strong POM degradation and Hg migration through neighboring layers will cause THg/TOC enrichment.

Study Sites
IODP Site U1366 (26°03´S, 156°54´W, water depth 5,127 m) is located at the western portion of the oligotrophic South Pacific Gyre (Figure 1).The 10 samples used in this study range from 2 to 30 mbsf, with ages of 10-92 Ma and sedimentation rate ranges between 12 and 210 cm/Ma (Dunlea, Murray, Sauvage, Pockalny, et al., 2015).From 92 to 40 Ma, the sediment is characterized by high Fe-Mn oxyhydroxide content (up to 80%), which is suggested to be of hydrothermal origin due to the high Fe/Al and Zn/Al ratios, high Fe-Mn oxide mass accumulation rate and the vicinity to a mid-ocean range (Dunlea, Murray, Sauvage, Spivack, et al., 2015).The content of POM in the sediment is low due to the low export productivity and slow sedimentation rate.Total organic carbon declines from about 0.2% at the seafloor to about 0.03% at 10.75 mbsf (with an age of ∼40 Ma) (D'Hondt et al., 2011).Site U1366 was used to illustrate the THg content in hydrothermal Fe-Mn oxides.
ODP Sites 1241 (5°50.570´N,86°26.676´W,water depth 2,027 m) and 677 (1°12.14´N,83°44.22´W,water depth 3,461 m) are located in the Eastern Equatorial Pacific upwelling system (Figure 1) (Shipboard Scientific Party, 1988, 2003).The samples used in this study range from 20 to 210 mbsf, with ages of 0.5-8.3Ma and sedimentation rate ranges between 1 and 9 cm/ka based on biostratigraphy (Alexandrovich, 1989;Shipboard Scientific Party, 2003).Their sedimentation rates are 1-6 cm/ka and 4-9 cm/ka for Sites 1241 and 677, respectively.Sites 1241 and 677 locate near the East Pacific Rise (EPR) hydrothermal system.The EPR owns one of the fastest spreading rates and the highest magmatic budgets.Its large hydrothermal flux is revealed by the helium plumes in seawater (Lupton, 1998).Based on the Pb-isotope record, the hydrothermal input flux varies during the past 7 Ma, with a high flux between 4.4 and 2.9 Ma.The porewater at Site 677 is influenced by hydrothermal activity based on temperature and chemistry data (Mottl, 1989), whereas there is no known hydrothermal influence on Site 1241 (Shipboard Scientific Party, 2003).These two sites were chosen to investigate whether the Hg deposition is influenced by high primary productivity and distal hydrothermal input.
ODP Site 1208 (36°7.630´N,158°12.095´E,water depth of 3,346 m) is located at the Central High of Shatsky Rise in the North Pacific oligotrophic gyre (Figure 1) (Shipboard Scientific Party, 2002).The samples used in this study have a depth range from 2 to 321 mbsf, with an age of up to 26 Ma (Bown, 2005).From the deepest sample used in this study to the core top, the sedimentation rate gradually increases from 0.02 to 4.24 cm/ka and the color of the sediment changes from yellowish orange to greenish gray, indicating a transition from an oxic to a more reduced early diagenetic environment (Shipboard Scientific Party, 2002).Based on the sedimentation rate and the TOC data measured in this study (Section 3.1), the samples at Site 1208 are divided into two sections.The samples above 310 mbsf (with ages younger than 15.5 Ma) are referred to as the upper section with a high sedimentation rate and moderate TOC content.The samples below 310 mbsf (with ages between 15.5 and 26.5 Ma) are referred as high TOC variability section.At the depth between 310 and 319 mbsf, some of the samples have TOC up to 1.3%, while most of the samples in the high TOC variability section have TOC lower than 0.1%.Site 1208 is used to illustrate the large effect of Hg migration through neighboring layers on THg/TOC during early diagenesis.
ODP Site 1094 (53°10.8149´S,5°7.8259´E, water depth 2,807 m) is located in the Antarctic Zone in the Southern Ocean (Figure 1).The site is located at the north of the winter sea-ice edge, and it was covered by sea-ice during the last glaciation (Shipboard Scientific Party, 1999).The samples used in this study range from core top to 31 mbsf, with ages covering the past 175 ka and sedimentation rate ranges between 2 and 56 cm/ka (Hasenfratz et al., 2019).The sediment is composed of diatoms and calcareous ooze and is characterized by glacialinterglacial cycles (Shipboard Scientific Party, 1999).Site 1094 represents an open ocean site with young and relatively high sedimentation rates.
Site EC2005 (121°20.0036´E,27°25.0036´N,water depth 36 m) is located in the East China Sea (Figure 1).The samples used in this study ranged from core tope to 3.8 mbsf, with ages covering the past 12 ka.Based on the radiocarbon dating, the sedimentation rate varies from 48 to 3,200 cm/ka (Xu et al., 2009).The sediment is mainly composed of terrestrial minerals and half of the TOC is of terrestrial origin (X.Liu, et al., 2020).Porewater sulfidation is observed at this site (X.Liu et al., 2020).Site EC2005 is used as a comparison to the abovementioned open ocean cores, which have no terrestrial POM input.

Methods
For bulk sediment major and trace element analyses, samples were milled and homogenized with an agate mortar.About 50 mg sample was weighed into a 15 mL PFA vial and treated with mixed concentrated acids of HCl, HF, and HNO 3 at a ratio of 3:2:1.The vial is closed and heated at 120°C overnight, after which it is evaporated at 95°C.The precipitate was reacted with a 3:1 mixture of concentrated HNO 3 and HCl at 120°C overnight to further decompose the remaining organic matter (OM).After drying down at 95°C, 0.5 mL of concentrated HNO 3 was added and evaporated twice to destroy all the remaining fluorides.The final precipitation was dissolved in 6 M HCl for preservation.Major and trace elements were measured by Thermo iCap 6000 ICP-OES and an Agilent 7900 Quadrupole ICP-MS, respectively.The sample preparation and the major and trace element analysis were performed at the State Key Laboratory for Mineral Deposit Research and the MOE Key Laboratory of Surficial Geochemistry, Nanjing University.The RSD is less than 10% for all elements based on the certified reference material (CRM) marine sediment GBW07316.
Total mercury content in bulk sample (THg) is measured by a commercial direct mercury analyzer DMA-80 (Milestone) at the Center for Marine Geochemistry Research, State Key Laboratory for Mineral Deposit Research, Nanjing University.The DMA-80 was calibrated with a liquid Hg standard (High-Purity Standards, HPS) with Hg content ranging from 0.02 to 10 ng.Before measurement, about 30 mg of milled and homogenized sediment was weighed into a sample holder by a six-digit balance.Within the DMA-80 standard measurement cycle, the sediment is thermally decomposed in an oxygen flow and the products are further decomposed in the catalyst tube to eliminate gaseous OM.The resulting gas flow carrying Hg vapor passes through a gold amalgamator, and the Hg is trapped and enriched.The Hg in gold amalgamator is released at a higher temperature and measured by atomic absorption spectrophotometry at 254 nm.Blanks of sample holders are measured before usage.The reliability of measurement was tested by CRMs, including 86 ± 2 ppb (n = 7) for MESS-4 with a reference value of 80 ± 6 ppb, and 465 ± 6 ppb (n = 7) for NIST 2702 with a reference value of 454 ± 18 ppb.The RSD of Hg measurement is less than 5%.
The percentage of Hg leachable by oxidizing solution in the sediment (Hg ox %) is calculated by the difference between samples with and without the sodium hypochlorite treatment.About 40 mg of the sample was weighed into a PE tube and reacted with 40 ml sodium hypochlorite solution (6 wt%) for 12 hr at room temperature in a shaker.After centrifugation, the supernate was decanted and the solid was washed with Milli-Q water for 3 times.The solid was further reacted with 40 ml 1% HNO 3 for 6 hr to release the adsorbed Hg on the mineral surface under an alkalescent environment.After the reaction, the solid was centrifuged and washed again before freeze drying.The control group experienced the same treatment except for replacing the sodium hypochlorite solution by Milli-Q water in the oxidizing step.The Hg contents of the oxidized sediment and the control group are denoted as c(Hg) ox and c(Hg) control , respectively.To correct for the mass loss caused by mineral dissolution and the resulting Hg content enrichment, the percentage of Hg leachable by oxidizing solution is calculated as: Hg ox % = 1 c(Hg) ox /c(Hg) control × 100%.The effectiveness of the sodium hypochlorite solution oxidation was tested using marine sediment certified reference materials (CRMs).The 1% dilute HNO 3 washing, which has been tested and has negligible effect on washing away Hg, is designed to remove various proportions of carbonate in the marine sediments before introducing into the DMA-80.
Total organic carbon (TOC) analysis was performed on a CNHS-O element combustion analyzer (Costech ECS 4024, NC Technologies) at the MOE Key Laboratory of Surficial Geochemistry, Nanjing University.About 0.5-1 g milled sediment sample was weighed into a PE tube and reacted with 1 M HCl overnight to remove carbonate minerals.The remaining solid was washed to neutral pH and freeze dried.About 20 mg freeze dried sample is warped into a tin cup and used to measure TOC.The RSD is less than 2% for regular measurement.For samples with extremely low TOC content, the RSD rises to 5% when the organic carbon input for measurement is 40 μg based on repeated measurement on an external standard.
In order to quantify the authigenic Fe-Mn oxide, sequential extraction is performed on the sediment based on the procedure of Poulton and Canfield (2005).Carbonate was extracted by 1 M sodium acetate (adjusted with acetic acid to pH 4.5) at room temperature for 24 hr.Easily reducible iron (Fe ox1 ) and manganese (Mn ox1 ) was extracted by 1 M hydroxylamine-HCl in 25% v/v acetic acid at room temperature for 48 hr.Reducible iron (Fe ox2 ) was extracted by sodium dithionite solution (50 g/L) buffered to pH 4.8 with 0.35 M acetic acid/0.2M sodium citrate for 2 hr.Between each extraction step, the solid was washed with Milli-Q water for three times.The concentrations of major elements of the resulting extraction solution were measured as described above.
The barium in the barite phase (Ba barite ) is quantified by the DTPA leaching method (House & Norris, 2020).The freeze dried and homogenized sample was first treated with 0.2 M ascorbic acid in 5 M acetic acid for 12 hr at room temperature in order to remove carbonates and Fe-Mn (hydr)oxides bound barium.The sample was centrifuged and the supernate was decanted.After washing with Milli-Q water for 3 times, the sample was treated with 0.2 M DTPA at pH 11.5-12 with sonication for 3 hr and agitation at 60°C for 12 hr.After centrifugation, the supernate was collected and diluted properly for ICP-OES measurement as mentioned above.
Rock-Eval pyrolysis was conducted at the School of Earth Sciences and Engineering, Nanjing University, Nanjing, China.The powdered sample was heated with a Rock-Eval VI pyrolysis instrument.The sample was heated at 300°C for 3 min, and then heated to 600°C; the carbon released are measured as free hydrocarbons (S1) and cracked hydrocarbons (S2), and the temperature of the highest carbon release at stage two (T max ) is recorded.

The TOC, THg, Hg Leachable by Oxidizing Solution (Hg ox ), and THg/TOC Ratio
The TOC measured in this study covers a wide range from less than 0.1%-4.6%.For the low TOC samples, Grasby et al. (2019) suggested that the questionable TOC data make their THg/TOC ratio unreliable, and should be excluded when discussing volcanism signal.There are two possible reasons why the low TOC sample will give false THg/TOC peaks (Grasby et al., 2019): (a) the large error of TOC data makes the THg/TOC unreliable; (b) the decarbonate process before element analyzer measurement causes OM loss during the 1 M HCl treatment, and this portion of OM is non-negligible.
At Site 1208, most of the samples in the high TOC variability section have TOC content less than 0.1% (Figure 2a, symbols under gray highlight).For these low TOC content samples, the organic carbon induced by the element combustion analyzer for TOC measurement is 15-30 μg.Repeated measurement of an external standard with 40 μg organic carbon gives a 5% error.As a result, the low TOC samples might have higher but still comparable analytical uncertainties to the external standard with higher TOC content.As a result, the value of the TOC data may bear large uncertainties as Grasby et al. (2019) suggested, but the data do indicate that the organic carbon content is very low compared to other sediment samples in this study.It was also suggested that the OM loss during the decarbonate process may be non-negligible for low TOC samples (Grasby et al., 2019).Based on the organic matter hydrolysis experiment using a stronger reaction condition (stronger acid and higher reaction temperature), the acid soluble organic matter corresponds to less than 50% TOC for most of the water column particles and sediments in both marginal and open ocean sites (Hwang & Druffel, 2006;Hwang et al., 2006;Roland et al., 2008;X. C. Wang & Druffel, 2001).Even if our TOC content was underestimated by a factor of 2, the actual THg/TOC was still abnormally high in the THg-enriched section at Site 1208 (Figure 2a, light blue line).As a result, the potential error in the low TOC data will not change the fact that the THg-enriched section has abnormally high THg/TOC ratios.Except for samples from the high TOC variability section at Site 1208, the THg mimics the TOC pattern in all of the core sites (Figure 2).Although the sediment cores in this study are characterized by different TOC and pyrite content, age, and the proportion of terrestrial OM input, most of the THg/TOC ratios range between 50 and 300 ppb/% (average = 165 ± 102 ppb/%, 1sd, n = 254).Sites 1241 and 677 experienced porewater pyrite formation and their pyrite content varies greatly (Figures 2b and 2c).At Site 677, the THg/TOC decreased where the pyrite content exhibited peaks (about 3 Ma).Lower THg/TOC ratios in sulfidic sediment have been reported (Frieling et al., 2023), and the mechanism will be discussed in Section 4.2.Site 1094 is characterized by its young age and high productivity.Site EC2005 is a continental shelf site with half of its TOC of terrestrial origin (X.Liu et al., 2020).The THg/TOC at EC2005 is lower than other open ocean sites but still within the 50-300 ppb/% range (Figure 2e).The TOC content at Site U1366 is expected to be lower than 0.03% (D'Hondt et al., 2011), which is too low to be measured in this study.The THg of samples from Site U1366 are less than 7 ppb (not shown in figures), which is the lowest in marine sediment measured in this and published studies.
For Site 1208 sediment, Hg ox % is used to distinguish between Hg bound to OM and Hg adsorbed to minerals because no pyrite is observed in the THg-enriched section (Shipboard Scientific Party, 2002).Most of the Hg ox % measured through the depth profile have Hg ox % higher than 80%, even for the samples from the high TOC variability section (symbols with blue edges).This suggests that although the TOC is extremely low, most of the Hg in the sediment is still bound to OM.

Element Concentrations and Sediment Sequential Leaching
The element and sediment component contents of Site 1208 are shown in Figure 3. Titanium (Ti), lithium (Li) and Fe ox2 contents have almost identical trends, which are similar to operational defined eolian dust (ODED, W. Zhang et al. (2016)) representing terrestrial material (Figures 3a and 3b).Almost all of the extractable Mn is leached in the first step as Mn ox1 , which has a similar trend as Fe ox1 , especially in the high TOC variability section.The similar trend of Fe ox1 , Mn ox1 and the authigenic component (W.Zhang et al., 2016) indicated that the Fe-Mn oxide extracted in the first step is a marine authigenic phase.The THg has the same trend with authigenic components in the high TOC variability section below the depth of 319 mbsf (Figures 3c and 3d).Ba barite has a similar trend to TOC with a few discrepancies (Figures 3e and 3f).At the high TOC content layer (310-319 mbsf), both TOC and Ba barite have two peaks, with small differences in timing and duration.Below the high TOC content layer, both TOC and Ba barite are low.

The Characterization of the Nature of Organic Matter
The result of Rock-Eval pyrolysis is shown in Figure 4. S2 evaluates the cracked hydrocarbons in an inert gas atmosphere within the temperature range of 300°C-650°C.In previous studies, S2/TOC is used to evaluate the OM sources and preservation in marine sediments, whose value is between 100 and 350 mg HC/g TOC for open ocean sediment.The S2/TOC value will be higher under oxygen-depleted bottom water and/or with high OM flux and lower in sediment with terrestrial OM input (see Figure 4d, shaded area).In this study, continental shelf site EC2005 has the lowest S2/TOC, which is comparable to published continental margin sediment (Holtvoeth et al., 2005), in agreement with the previous estimation which suggests half of the OM at EC2005 is of terrestrial origin (X.Liu et al., 2020).The high TOC variability section at Site 1208 has a wide S2/TOC range, with low S2/TOC corresponding to the two TOC peaks shown in Figure 3f, suggesting a highly degraded nature of OM caused by the low linear sedimentation rate (Figure 4f).The high TOC variability section, characterized by its extremely low sedimentation rate (Figure 4f) as well as low productivity suggested by Ba barite (Figure 3f), should have a low OM content which is highly degraded.In fact, its S2/TOC is roughly within the open ocean range.This may suggest DOM migration from the overlying high TOC layers (see discussion 4.3).The S2/TOC of the other sites in this study is generally within the published open ocean range, suggesting a consistent OM nature in marine Geochemistry, Geophysics, Geosystems 10.1029/2023GC011201 sediment under well-oxygenated bottom water.The S1 and S1/TOC data (Figures 4a and 4b) in this study are comparable to published marine sediment data (Figure 4b).The same trend between S1 and S2 in the studied sites suggests a similar controlling factor.

Limited Influence of Distal Hydrothermal Input on Hg Deposition in the Deep Pacific
Proximal hydrothermal activity is observed to increase the ambient seawater and sedimentary Hg concentration (Bowman et al., 2015;Cox & McMurtry, 1981;Lamborg et al., 2006;Roberts et al., 2021); however, distal hydrothermal input with oxygenated bottom water conditions needs to be explored.The lower part of Site U1366 (deposited at 92 to 40 Myr, seven samples are measured in this study) is characterized by a large flux of hydrothermal Fe-Mn oxide deposition (Dunlea, Murray, Sauvage, Spivack, et al., 2015), while its THg is no more than 7 ppb and indistinguishable from the upper part of the core.This THg value is significantly lower than other marine sediments in this study and in published data (Figure 5).At Site U1366, the hydrothermal influenced samples are lack of POM (TOC content less than 0.03%, D 'Hondt et al., 2011).As a result, the THg content is most probably controlled by the adsorption of Hg by Fe-Mn oxide.The adsorption capacity of Hg on minerals can be calculated by seawater Hg concentration and adsorption distribution coefficient (K d ), which is used to describe the distribution of an element between the solid and the solution phase.The K d of Hg adsorption on freshly prepared Fe hydrous oxide gel is about 1500 L/kg  a-c) at the depth between 300 and 321 mbsf.Below 300 mbsf, the Fe ox1 , Mn ox1 , THg and THg/TOC increased downward, covarying with the authigenic component.The RSD for major and trace element measurement is less than 10%.(Kinniburgh & Jackson, 1978), and becomes much lower on crystalized Fe-Mn oxides and clay minerals (Farrah & Pickering, 1978;Forbes et al., 1974;Thanabalasingam & Pickering, 1985).By contrast, the K d of Hg on POM is 10 5 -10 6 L/kg (Morel et al., 1998), more than two orders of magnitude higher.Considering that the Hg concentration in the south Pacific oligotrophic gyre deep water is 0.98 pM (Munson et al., 2015), the THg of Fe-Mn oxide will be less than 5 ppb, consistent with the measured values (less than 7 ppb).It indicates that Fe-Mn oxide in POM depleted sediment is not able to enrich Hg from the long-distance hydrothermal influenced seawater due to the low K d of Hg on minerals.
The ability of POM-rich sediment to record hydrothermal influence was tested by the Eastern Equatorial Pacific Sites 677 and 1241.Site 677 is influenced by distal hydrothermal fluid circulation (Mottl, 1989), while there is no sign of hydrothermal input to Site 1241 (Shipboard Scientific Party, 2003), probably due to the inhibition of the Cocos Ridge.The THg/TOC ratios are almost identical for Sites 677 and 1241 through the past 6 Myr, and comparable to sediments in other sites (Figure 6).By comparing to the Pb-isotope based hydrothermal input record (van de Flierdt et al., 2004), there is a limited seawater and sedimentary Hg signature caused by distal hydrothermal activity.Hydrothermal Hg exists mainly in three phases, including elemental gaseous Hg, sulfide minerals bounding Hg, and dissolved Hg.A proportion of the elemental gaseous Hg will transform into dissolved Hg, and the rest of the elemental gaseous Hg will rise upward and seep into the air (Roberts et al., 2021).The crystallized sulfide minerals will deposit near the hydrothermal vent in meter-scale (Findlay et al., 2019),  d) are ranges of published data.① Sediment under coastal upwelling system with large TOC burial flux (Hatcher et al., 2014).②③④ Sediment under Arabian Sea extended oxygen minimum zone with bottom water being suboxic, dysoxic, and fully oxic, respectively (Nierop et al., 2017).⑤ Open ocean sediment with little terrestrial OM input (Freudenthal et al., 2001).⑥ Marine sediment with part of its OM of terrestrial origin (Hare et al., 2014;Moreno et al., 2008;Tribovillard et al., 2009).⑦ Continental margin sediment (Holtvoeth et al., 2005).(e) The Hg versus TOC flux of our studied samples.(f) The LSR of the core sites used in this study ranges over five orders.
restricting their influence in the proximal area.As a result, the effect of distal hydrothermal Hg input on sedimentary Hg content is negligible, as shown by the Eastern Equatorial Pacific Sites 677 and 1241.

Tight Coupling Between Hg and Organic Burial in Marine Sediment Under Well-Oxygenated Bottom Waters
Published studies are confined to sediment under oxygen-depleted bottom water and a large flux of POM deposition.Under oxygen-depleted bottom water, the THg/TOC decreases with the decrease in bottom water oxygen content, probably due to microbial activity (Frieling et al., 2023).In geological records rich in POM and pyrite, the THg/TOC is enriched due to the sulfide mineral associated with Hg in the sediment (Shen et al., 2020).There is still a lack of research on Hg burial under well-oxygenated bottom water.In this study, we report data on Hg deposition in open ocean sediment under well-oxygenated bottom water, and compare it with published data on marine sediment and geological record (Figure 6), in order to complete the understanding of marine Hg burial.
In open ocean sediments, the degradation of POM is significantly influenced by oxygen exposure time, which is controlled by the sedimentation rate and POM content (P.J. Müller & Suess, 1979;Hartnett et al., 1998).The upper section of Site 1208, Sites 677,1241, and 1094 have large variations in linear sedimentation rate, TOC flux and age, while their THg/TOC ratios are confined in the 50-300 ppb/% range (Figure 6), suggesting that POM degradation has little influence on THg/TOC under well-oxygenated bottom water.This suggests that even though the sedimentary environment is different, after early diagenesis under well-oxygenated bottom water, the nature of POM tends to be similar, which is also supported by the Rock-Eval pyrolysis result (Figure 4).The continental shelf site EC2005 has half of its POM of terrestrial origin (X.Liu et al., 2020).Its nature of POM is slightly different from the open ocean sites as indicated by the lower S1/TOC and S2/TOC values.As a result, the average THg/TOC at EC2005 was the lowest in this study.However, it is still within the range of other open ocean sites (Figure 6), indicating a small influence of terrestrial POM input on the THg/TOC ratio.The THg and TOC data of marine sediment under well-oxygenated bottom water in published studies, which are Black Sea (Shen et al., 2020), Caribbean Sea (Kita et al., 2013), Kuril area (Kita et al., 2013), northeast (NE) of Brazil (Fadina et al., 2019), north Atlantic ice-rafted debris (IRD) belt (Kita et al., 2016), and Southwestern (SW) Atlantic (Figueiredo et al., 2020).(c), (d) Published THg and TOC data of marine sediment either in restricted ocean basin or under the influence of oxygen minimum zone, which makes the sediment-water interface probably being oxygen-depleted.The site locations are Japan Sea (Shen et al., 2020), Saanich Inlet (Shen et al., 2020), Peru Margin (Shen et al., 2020), Mediterranean (Cossa et al., 2021;Frieling et al., 2023;Gehrke et al., 2009), Arabian Sea (Frieling et al., 2023), Baltic Sea (Frieling et al., 2023), and Arctic Fjords (Kim et al., 2020).The hollow marks represent the sediment layers that deposited in a relatively oxygenated environment based on TOC and redox sensitive trace metal concentrations.The water column particles (Cossa et al., 2021;Munson et al., 2015), represented by blue stars, are also located at the low THg/TOC area after normalizing its TOC to 100% for convenience.Comparing data from this and published study, most of the sediment under well-oxygenated bottom water has THg/TOC ratios ranging between 50 and 300 ppb/% (Figures 5a and 5b, yellow highlighted area), whereas the sediments under oxygen-depleted bottom water and water column particles have lower THg/TOC ratios (Figures 5c and 5d, blue highlighted area).The difference can be explained by (a) preferential Hg release under oxygen depleted bottom water and/or (b) relative Hg enrichment during OM degradation.Frieling et al. (2023) reported THg/TOC depletion in oxygen-depleted sediment and suggested that microbial Hg methylation may preferentially release Hg from the sediment, which is supported by large benthic flux of methylmercury release from the sediment (Covelli et al., 1999;Hammerschmidt & Fitzgerald, 2008;Tomiyasu et al., 2008).The relative Hg enrichment during POM degradation may result from the change of OM characteristics.The one order of magnitude change in K d of Hg on POM (10 5 -10 6 L/kg (Morel et al., 1998)) indicates a large difference in the Hg bounding ability of POM.Organic matter with high aromaticity and oxygen-containing functional groups are observed to have high affinity for Hg (Li et al., 2022;Ramalhosa et al., 2003), which is selectively preserved during POM degradation (Duan, 2000;Nierop et al., 2017;Yoshimura & Hama, 2012).The elevated Hg bounding ability during POM degradation is also supported by the increased THg/TOC ratio of the water column particle increase with water depth (Cossa et al., 2021;Munson et al., 2015), and the systematically higher THg/TOC ratio of sediments compared to water column particles.The remarkable higher THg/TOC ratio in marine sediment than the water column particles should be considered when evaluating the Hg burial flux in marine sediment, which is always calculated by POM burial flux and the THg/TOC ratio (e.g., Y. Zhang et al., 2014).The THg/TOC data reported in this study indicate that well-oxygenated bottom water will result in higher THg/TOC ratios in sediment than oxygen depleted bottom water.
Despite the alteration of THg/TOC by the above-mentioned two pathways, the THg/TOC ratio of sediment under well-oxygenated bottom water is constraint in a narrower range (50-300 ppb/%) than geological records (Figure 6, ranging from 10 1 to 10 5 ppb/%).The relatively stable THg/TOC ratio in sediment with large variations in TOC burial flux (about five orders of magnitude), age (from several thousand years to several million years), and terrestrial POM input (from open ocean sediment with negligible terrestrial input to continental shelf sediment with half of the POM of terrestrial origin) suggests tight coupling between TOC and THg.Although narrower than the geological record, the six-fold variation of THg/TOC in marine sediment under well-oxygenated bottom water is non-negligible.Two main factors are suggested to control the THg/TOC ratio in marine sediments.First, the changing Hg bounding ability of POM due to degradation will cause THg/TOC difference in water column particle and sediment.During the POM sinking from the surface ocean to the deep sea, its Hg bounding ability increases with depth due to the degradation of POM (Lamborg et al., 2016).During early diagenesis, THg/TOC changes as a result of further TOC decomposition in different bottom water redox condition (Frieling et al., 2023).Besides, the Hg content in seawater blow the mixed layer is also variable, ranging from 0.59 to 1.4 pM globally (Bowman et al., 2020), resulting in variable Hg content in POM.Second, it is also possible that large variation in TOC flux and the associated Hg release into the porewater due to POM degradation might result in Hg diffusion to the neighboring layer (see Section 4.3).Unfortunately, there is no single environment factor that could be quantitatively linked to the six-fold variation in THg/TOC in open ocean sediment, because it is driven by multiple environmental factors which jointly control the bounding ability of TOC and the porewater Hg concentrations.
In geological records, the Hg enrichment is always identified by the THg/TOC peak value compared to the background value in the same deposition profile (Grasby et al., 2013;Sanei et al., 2012).Grasby et al. (2019) suggested that only samples with TOC > 0.2% and THg/TOC peaks higher than 72 ppb/% should be considered as a potential volcanism signal (Figure 6, red diamond).The marine sediment data produced and compiled in this study (Figure 6) further suggest that the variation in THg/TOC of marine sediment under well-oxygenated bottom water has a range between 50 and 300 ppb/%.We suggest that THg/TOC exceeding ∼300 could be considered as distinct Hg enrichment in sedimentary records.However, before the enrichment can be interpreted as reflecting enhanced Hg input, the diagenetic imprint should be excluded (see Section 4.3).Other evidence is further needed to test whether the Hg enrichment is linked to volcanism, such as Hg-isotope data (Grasby et al., 2017;Shen, Algeo, Planavsky, et al., 2019) and the nature of POM (Grasby et al., 2019).

Diagenetic Migration of Hg in Sedimentary Sequence Associated With Abrupt Changes in TOC Burial
The TOC content in most of the samples in the high TOC variability section of Site 1208 is extremely low, whereas its THg (average = 107 ± 40 ppb, 1sd, n = 33) is relatively high.Although Grasby et al. (2019) suggested that TOC data lower than 0.2% will bear large error caused by measurement and decarbonate, after considering these two aspects, the POM content in these samples is still negligible.The abnormal THg enrichment in the high TOC variability section indicates another Hg bounding phase than POM, which is supposed to be the major Hg bounding phase in most of the sediments with no pyrite.
At Site 1208, the high TOC variability section is lack of sulfide minerals and enriched in authigenic components (including zeolite minerals and manganese micronodules) (Shipboard Scientific Party, 2002;W. Zhang et al., 2016).Based on the discussion in Section 4.1, the Hg content in authigenic minerals equilibrated with seawater is too low (<7 ppb) to explain the THg/TOC enrichment.Volcanic and hydrothermal Hg inputs are also ruled out based on the content of high field strength elements (W.Zhang et al., 2016;Yang et al., 2022).Based on the results of sequential leaching, the Hg in the high TOC variability section has a trend similar to Fe ox1 , and about 80% of the Hg is bound to OM related phase, making the metal oxide-OM association being the most probable Hg bounding phase in THg-enriched section.Compared to minerals whose Hg bounding ability is very low, experimental studies show that Fe-and Mn-oxides coated/co-precipitated with OM (i.e., metal oxide-OM association) have Hg adsorption ability greatly enhanced (Zhong & Wang, 2008;Y. Liu et al., 2020).
Metal oxide-OM association has been reported in both terrestrial and marine sediments.It is first recognized in soil and found to increase the OM long-term preservation (Hemingway et al., 2019).Lalonde et al. (2012) suggested that about 21% of the sedimentary organic carbon is in the metal oxide-OM association phase in marine sediment.In tephra layers of marine sediment, where reactive metals are not limited, the organic carbon bound with reactive metals accounts for ∼80% of the TOC of the sediment (Longman et al., 2021).Metal oxide-OM association is formed by metal oxides (mainly Fe-and Mn-oxides) coprecipitating with or adsorbing dissolved organic matter (DOM).During the formation of metal oxide-OM association, the DOM experienced molecular fractionation, resulting in the metal oxide bound DOM enriched aromatic moieties, polyphenols, and carboxylic compounds (Eusterhues et al., 2011;Han et al., 2021;Lv et al., 2016).These functional groups have higher affinity to Hg compared to saturated carbon chains (Li et al., 2022;Ramalhosa et al., 2003), making the metal oxide-OM association potential to be able to adsorb large amounts of Hg with low organic carbon content.
To understand the formation of the high THg/TOC ratio in the high TOC variability section, the source of the DOM and Hg in the metal oxide-OM association need to be constrained.The sediment at Site U1366 is rich in reactive Fe-Mn oxides but low in THg (less than 7 ppb), indicating that Fe-Mn oxides are not able to enrich DOM and Hg from seawater, probably due to their low concentrations of DOM (Alperin et al., 1999;Nebbioso & Piccolo, 2013) and Hg (Bowman et al., 2020;Cao et al., 2023;Covelli et al., 1999) compared to porewater.As a result, the DOM and Hg most probably migrated from neighboring sediment layers.In the case of Site 1208, the extremely low TOC layer is covered by several high TOC layers (Figure 3f), which is the most probable source of the DOM and Hg.In high TOC layers, the low S1/TOC and S2/TOC (Figure 4) indicate that the POM is highly degraded.Ba barite content is a productivity proxy in marine sediment (Martinez-Ruiz et al., 2019;Paytan & Griffith, 2007;van Beek et al., 2007).In the TOC-enriched section, Ba barite content has two peaks wider than TOC, indicating a poor preservation of POM in the sediment.The reactive metal oxide, which is represented by sequential leachable Fe ox1 in this study, is able to fix the DOM and further adsorb Hg from the porewater.In the high TOC variability section, the Fe ox1 and THg increase with depth due to the higher Hg bounding capacity associated with the high reactive metal oxide content.
The vertical migration of Hg during early diagenesis is widely observed.In OM-rich sediments experiencing sudden oxidation, the Hg released by the underlying layer due to POM degradation is supposed to be scavenged by the Fe-Mn oxide in the redox front on top of the OM-rich layer (Frieling et al., 2023;Gobeil et al., 1999;Matty & Long, 1995).As Site 1208, the THg/TOC enrichment in the high TOC variability section is envisaged to form in a similar way, but the OM-rich sediment is on top of the metal oxide enriched layer, and the major Hg bounding phase is more likely to be metal oxide-OM association.Another example of abnormal sedimentary Hg/TOC that is not linked to global volcanic activities comes from the Caribbean Sea sediment record during the Quaternary (Figure 5b, green triangles in gray highlight).Here, the THg/TOC ratios of Caribbean Sea sediment range from 68 to 876 ppb/%, with the high values corresponding to glacial periods whose TOC is up to an order of magnitude lower than neighboring layers (Kita et al., 2013).The potential vertical migration of Hg to low TOC layers may not mask the positive correlation between THg and TOC in cross plot, but it is able to induce large variations in the THg/TOC ratios, especially for the low TOC layers.As a result, we suggest that sediments with drastic changes in TOC content, especially those with very low TOC content (Figure 5, gray shaded area), are probably not a reliable geological archive to use the THg/TOC volcanism proxy, not only due to the large uncertainty of TOC measurement (Grasby et al., 2019) but also the potential impact by Hg vertical migration.
In geological records, large environmental perturbations are able to cause variations in primary productivity and POM preservation, resulting in drastic changes in sedimentary TOC content.For example, the Late Ordovician mass extinction and End Devonian mass extinction are both accompanied by ice sheet building up, expended marine anoxia, perturbations in carbon, nitrogen, and phosphorus cycles, as well as enhanced organic carbon burial (Murphy et al., 2000;Shen et al., 2018).The hyperwarming during Permian-Triassic boundary crisis causes ocean stratification, deep water anoxia, and productivity collapse in marine ecosystems (Dal Corso et al., 2022), resulting in fluctuations in seawater redox conditions during this period (Wignall & Twitchett, 1996).Ocean anoxic events are characterized by episodically intensified silicate weathering and nutrient runoff, bio-limiting nutrient cycle changes, and seafloor anoxia (Danise et al., 2015;Westermann et al., 2013).During these geological events, the TOC flux in neighboring sedimentary layers is expected to vary abruptly.The Hg and DOM released from the high TOC layer during early diagenesis might cause abnormal Hg enrichment in the low TOC layer.Under these circumstances, it is crucial to take other proxies, such as Hg-isotope data and the characterization of the nature of OM (Grasby et al., 2017(Grasby et al., , 2019;;Shen, Algeo, Planavsky, et al., 2019), into consideration and test if the Hg enrichment is caused by vertical migration of Hg before linking it to enhanced volcanic activity.

Conclusions
In this study, we have systematically measured THg and TOC of the marine sediments under a variety of sediment environments, for example, different hydrothermal influence, terrestrial POM input, sedimentation rate and TOC flux.Hydrothermal oxide deposition and terrestrial POM input do not cause abnormal Hg enrichment in marine sediments.Based on the data compilation from the literature and this study, the THg/TOC ratio in marine sediment under well-oxygenated bottom water ranges between 50 and 300 ppb/%, which is systematically higher than the THg/TOC ratios of sediment under oxygen-depleted sediment and water column particles.The large natural variation of THg/TOC in marine sediment under well-oxygenated bottom water suggests that sediment with THg/TOC significantly higher than 300 ppb/% could be considered as distinct Hg enrichment.However, before the THg/TOC enrichment is interpreted as reflecting enhanced Hg input (e.g., increased volcanic activities), the diagenetic mobilization of Hg should be first excluded.Overall, our study provides a large set of Hg concentration data in sediments under oxygenated bottom waters, which would help understand the Hg cycle in the modern ocean and utilize Hg enrichment as a volcanism proxy.

Figure 2 .
Figure 2. The TOC, THg, and Hg leachable by oxidizing solution of the sites in this study.(a-e) The THg, TOC, and THg/TOC of Sites 1208, 1241, 677, 1094, and EC2005.The pyrite content in Sites 1241 and 677 is based on Guo (2021).The yellow shade indicates the THg/TOC range between 50 and 300 ppb/%.The gray shade indicates the low TOC samples whose analytical uncertainty may be higher.The light blue line indicates the THg/TOC ratio if the TOC content is underestimated by a factor of 2 in the high TOC variability section.(f) The result of Site 1208 samples leached by 6 wt% sodium hypochlorite solution.The x-axis indicates the Hg content of the sediment after the sodium hypochlorite solution leaching, and the y-axis indicates the Hg content of the controlling group.The Hg ox % shown by diagonal lines indicates the percentage of the Hg leachable by oxidizing solution.The symbols with blue edges indicate the samples from the high TOC variability section.(g), (h) Boxplots of the THg and TOC data of this study.

Figure 3 .
Figure 3.The variabilities of sediment components with core depth in Site 1208.(a) The Fe ox2 , Li and Ti content in comparison with operational defined eolian dust (ODED, brown shading, W. Zhang et al. (2016)).(b) The Fe ox1 , Mn ox1 and THg content in comparison with the authigenic component (orange shading, W. Zhang et al. (2016)).(c) TOC and THg/TOC in comparison with Ba barite (purple shading).Both TOC and Ba barite have two corresponding peaks, with the Ba barite peaks wider than the TOC ones.(d-f)Detailed profiles for (a-c) at the depth between 300 and 321 mbsf.Below 300 mbsf, the Fe ox1 , Mn ox1 , THg and THg/TOC increased downward, covarying with the authigenic component.The RSD for major and trace element measurement is less than 10%.

Figure 4 .
Figure 4.The result of Rock-Eval pyrolysis, deposition flux, and linear sedimentation rate (LSR).(a-d) The Rock-Eval pyrolysis results of S1, S1/TOC, S2 and S2/TOC.The shaded areas in (b) and (d) are ranges of published data.① Sediment under coastal upwelling system with large TOC burial flux(Hatcher et al., 2014).②③④ Sediment under Arabian Sea extended oxygen minimum zone with bottom water being suboxic, dysoxic, and fully oxic, respectively(Nierop et al., 2017).⑤ Open ocean sediment with little terrestrial OM input(Freudenthal et al., 2001).⑥ Marine sediment with part of its OM of terrestrial origin(Hare et al., 2014;Moreno et al., 2008;Tribovillard et al., 2009).⑦ Continental margin sediment(Holtvoeth et al., 2005).(e) The Hg versus TOC flux of our studied samples.(f) The LSR of the core sites used in this study ranges over five orders.

Figure 5 .
Figure 5.A compilation of marine sediment THg and TOC data of published studies and this work.The yellow and blue highlights indicate the THg/TOC ratio of 50-300 ppb/% and less than 50 ppb/%, respectively.The gray shade emphasizes data with TOC less than 0.1%.(a) The THg and TOC data in this study.The error bar is smaller than the symbol.(b)The THg and TOC data of marine sediment under well-oxygenated bottom water in published studies, which are Black Sea(Shen et al., 2020), Caribbean Sea(Kita et al., 2013), Kuril area(Kita et al., 2013), northeast (NE) of Brazil(Fadina et al., 2019), north Atlantic ice-rafted debris (IRD) belt(Kita et al., 2016), and Southwestern (SW) Atlantic(Figueiredo et al., 2020).(c), (d) Published THg and TOC data of marine sediment either in restricted ocean basin or under the influence of oxygen minimum zone, which makes the sediment-water interface probably being oxygen-depleted.The site locations are Japan Sea(Shen et al., 2020), Saanich Inlet(Shen et al., 2020), Peru Margin(Shen et al., 2020), Mediterranean(Cossa et al., 2021;Frieling et al., 2023;Gehrke et al., 2009), Arabian Sea(Frieling et al., 2023), Baltic Sea(Frieling et al., 2023), and Arctic Fjords(Kim et al., 2020).The hollow marks represent the sediment layers that deposited in a relatively oxygenated environment based on TOC and redox sensitive trace metal concentrations.The water column particles(Cossa et al., 2021;Munson et al., 2015), represented by blue stars, are also located at the low THg/TOC area after normalizing its TOC to 100% for convenience.

Figure 6 .
Figure 6.Compilation of THg/TOC of water column sediment, marine sediment, and geological record.(a) THg/TOC of core sites in this study.(b) THg/TOC of water column particles and marine sediment.The data sources are the same as Figure 5.The boxes filled by gradient ramps indicate water column particles and sediments under oxygen-depleted bottom water.The faded hollow boxes indicate sediments deposited in relatively oxic interlayers under the oxygen-depleted bottom water.The boxes filled with solid colors are sediments under well-oxygenated bottom water.(c) THg/TOC of the high TOC variability section at Site 1208.(d) THg/TOC of geological records modified from Grasby et al. (2019).The gray bars indicate the range of the THg/TOC ratio, and the red diamonds indicate THg/TOC peak values after eliminating data with TOC less than 0.2%.The faint yellow highlight indicates the THg/TOC ratios between 50 and 300 ppb/%.