Mercury concentration and isotopes composition during last glacial maximum to the Holocene
Deep-sea sediments receive particulate Hg deposited from the upper ocean after transformation by physical, chemical, and biological processes. As shown in Fig. 2, a sharp increase in Hg content was observed in the 215–240 cm layer, reaching a peak of 299 ng/g, which is six-fold higher than the average concentration of 49 ± 20 ng/g (1SD, n = 19) in the profile. In view of Hg/TOC ratio also significantly increased around the same time; therefore, the spike is likely caused by an anomalous Hg influx during sedimentation. At the depth of anomalous high Hg concentration, δ202Hg values presented a large negative shift (decrease to -4.48‰), falling well below the average MDF (-1.79 ± 1.05‰, 1SD, n = 23) of the profile. This variation in isotopic compositions was likely caused by process or source differences during Hg deposition. The Δ199Hg value exhibits initially a descending fluctuation trend from the surface down to a depth of 215 cm (from 0.24‰ to 0.10‰, mean 0.22 ± 0.08‰, 1SD, n = 15), followed by more variable values between 215–325 cm (0.29‰ − 0.21‰, mean 0.29 ± 0.08‰, 1SD, n = 8). It is worth noting that the MIF trend is in line with the turning point of the Hg concentrations at the same depth of 210 cm. We hypothesized that the variation in Δ199Hg values is influenced by photochemical dynamics in the upper atmosphere and water column during its transport and migration 18.
Despite limitations in establishing precise chronological frameworks, an approximate timeline for the cores can be inferred from the MT sedimentation rates and geological core data. It has been estimated that sedimentation rates in the MT were between 0.02 and 0.04 cm yr− 1 based on Pb-210 data from sediment cores retrieved from the southern slope and the axis of the MT 31. This suggests that the sediment core MT03 may contain sedimentary age information spanning the past 5 to 20 thousand years 32. Laminated diatom mats (LDMs) are present at the base of the MT03 sediment core (Fig. 1C), which could be a result of silicon diatom accumulation during the last glacial maximum (LGM) 33. This extensive diatom blooming was likely caused by an increase in nutrients and minerals source from aeolian input during the LGM 34. Thus, our date suggest that the deepest trenches showed anomalous increases Hg input during the LGM due to possible environmental or geologic evolution, which also affected the Hg isotopic composition on this profile. For comparison, the geo-environmental Hg levels reported for Arctic ice cores and shelf peat 35, 36, likely suggests there was global sediment Hg anomaly from the Last Glacial Termination into the early Holocene. The possible reasons for the anomalous increase in Hg level and isotopic variations are discussed further below.
Mercury source variability revealed from Hg isotopic composition
The ratios of Δ199Hg/Δ201Hg in the Mariana Trench sediment, as shown in Fig. 3, was approximately 0.95 ± 0.12 (R2 = 0.76, p < 0.001), closely aligning with the Δ199Hg/Δ201Hg slope (1.00 ± 0.01) typically seen in studies on photochemical reduction of Hg2+ to Hg0 37. In addition, MIF varies very little with depth, implying that the Hg found at different depths was similarly affected by photochemistry. The consistent MIF > 0 across the sedimentary profile suggests that buried Hg was constantly influenced by atmospheric processes 38. However, an atmospheric Hg source alone seems insufficient to explain the large variations in MDF across the sedimentary profile, because existing data on atmospheric Hg (particulate Hg and gaseous elemental Hg) do not exhibit such a substantial negative MDF. Therefore, we hypothesize that the sediment core MT03 may contain Hg from additional sources (sediment rock/aquatic plant) 39, where Hg undergoes photochemical processes similar to Hg in the upper atmosphere or water column before being deposited into the trench sediments.
The average δ202Hg isotopic composition in 240–325 cm profile was − 2.25 ± 0.62 (n = 6, 1 SD), significantly more negative (t-test, p < 0.01) than that in the 0-215 cm profile (-1.26 ± 0.52, 1 SD, n = 15). Although enhanced particle adsorption or microbial methylation of Hg2+ may result in a negative MDF shift in 240–325 cm profile, they would be insufficient to explain the large negative δ202Hg observed for the period 18. Specifically, dissolved Hg2+ is adsorbed by Fe-Mn oxides or thiol ligand particles, with fractionation (enrichment) ranging from 0.30‰ to 0.62‰ 40. Additionally, microbial-controlled methylation of Hg2+ under anoxic conditions can lead to significant isotopic fractionation in produced methylmercury (~ -2‰) 41. However, these processes typically represent the maximum fractionation between methylmercury and total dissolved Hg2+ observed in natural environment matrices (e.g., sediment/porewater). The extent of fractionation is also influenced by the Hg bioavailability and the methylation rate 42. Methylation rates are usually less than 2% 43, suggesting that microbial methylation would not generate sufficient methylmercury to significantly alter the isotope ratio of the total Hg and therefore is not the primary contributor to the significant negative MDF observed in the profile. The isotopic composition of residual Hg, which is enriched in Hg2+ by algae, is heavier than the atmospheric mercury isotopic composition. The positive MDF of the isotopic composition of residual Hg outside of algae is also inconsistent with the isotopic trend of the sediment profile. Consequently, it is reasonable to argue that the main differences in Hg concentration and isotope ratios between the sedimentary profile are caused by a different Hg source.
Previous reports on short sediment cores (MT1) from the MT indicate an OC/TN ratio of 7.0 ± 0.93 and δ13C of -19.9 ± 0.40‰ (1 SD, n = 13) 29, suggesting that the MT is primarily influenced by atmospheric Hg input from marine sources (average THg concentration of 45 ± 6.7 ng/g). The MT03 core at the depth of 0-215 cm is consistent with the Hg concentrations and stable isotopic compositions (Δ199Hg and δ202Hg) of MT sediments influenced by atmospheric Hg inputs (t-tests, p > > 0.05) 2. This finding suggests that the 0-215 cm sediment profile of the deposited Hg source is associated with atmospheric Hg. Mercury from atmospheric dry and wet deposition enters the upper ocean and subsequently undergoes gravitational settling into the trench sediment 16.
The main Hg sources in marine sediments, including background rock weathering, volcanoes, hydrothermal fluids, biological tissue decomposition, and the transport of particle deposition from the atmosphere, could be potential explanations for the abnormal increase in Hg concentrations between 220–240 cm depth 8. This coincides with a significant negative shift in δ202Hg, reaching values as low as -4.48‰. This value is considerably lower than the isotopic MDF composition of potential source for Hg inputs, such as gaseous elemental Hg/ particulate bound Hg and Hg transported by hydrothermal vents or river (t-test, p < 0.01). Comparable values are only recorded in sedimentary igneous rocks 44 (Fig. 4). Furthermore, the reported TOC content is greater than 0.2% in MT03 sediment core 30, and the proxy of Hg/TOC for volcanic emissions in the geologic record exhibits a significant increase within the MT03 sediment core in 220–240 cm profile (Fig. 2B) 45, 46, corroborating the input of volcanic Hg during the period of LGM 47. Non-zero MIF values at this stage are attributed to particles released by volcanoes adsorbing gaseous oxide Hg from the atmosphere to form particle Hg2+ (positive MIF) 48 or emitted gaseous elemental Hg undergoing photochemical oxidation-reduction transformation during atmospheric transport 49. During the Late Glacial and Holocene, climatic and volcanic signals were evident in the Hg record in ombrotrophic peat bog 36, suggesting that the accumulation of Hg during this period may have been caused by a global volcanic event.
Role of diatom in Hg cycling in the LGM period
The suggested sources for Hg found at depths ranging from 0-240 cm (Holocene to the LGM) cannot fully explain the differences in δ202Hg observed at deeper depths of 240–325 cm. The pronounced global environmental shifts during the LGM, which is manifested at depths of 215–325 cm and associated with an enhanced flux of Asian eolian dust, contributed to a marked increase in marine primary productivity in the western Pacific Ocean 34, 55. Recent research highlights the role of algae and algal-derived organic matter in aqueous-phase mercury removal, underscoring the significance of diatom laminations in Hg sequestration within marine environments 56, 57. The continuous existence of LDMs in the 215–325 cm sediment layer indicates that diatoms may be involved in the Hg cycle. Despite the absence of stable isotopic data for marine diatom laminae, available evidence from marine macroalgae (δ202Hg: -3.22 ± 0.81‰, Δ199Hg: 0.16 ± 0.08‰) indicates a distinct isotopic signature 58. It is reported that isotopic compositions of vegetation Hg suggest moderately negative MDF values (δ202Hg: -2.56 ± 0.64‰) 19. This finding aligns with our observation of similar δ202Hg values in selected sediment layers (δ202Hg values of -2.53‰, -3.15‰, and − 2.64‰ at 250, 295, and 310 cm depth, respectively). Algae may continuously enrich lighter Hg isotopes and deplete odd mass isotopes during the Hg0 re-emission process, resulting in the retention of lighter isotopes (negative MDF) and positive MIF within their structure 59, 60. This phenomenon is consistent with the observation that isotopic compositions in the 215–325 cm layer (-2.25 ± 0.57‰, 1 SD, n = 6) are notably lighter than those atmospheric sources contributing to Hg deposited in the 0-215 cm range (-1.26 ± 0.52‰, 1 SD, n = 14). In addition, Vandal et al. 61 inferred that the oceanic productivity may have been higher during the period of LGM (18,000 years ago) based on changes in Hg concentrations in Antarctica ice cores. Our findings advance the understanding of the hadal Hg cycle in the process of geological historical evolution (Fig. 5), suggesting that during the LGM period, diatoms functioned as the conveyor for the sequestration of Hg, facilitating the Hg transportation from the atmosphere to hadal trenches.
In this study, we reconstructed the geochemical cycle of Hg in the trench environment during the LGM period. Our findings reveal variations in the geochemical cycling of trench Hg, shedding light on the source and environmental drivers of Hg cycling prior to human influence. Geological activities (e.g. volcanism and shelf weathering) transported the sources of Hg input to the trench. The primary source shifted from atmospheric and algal deposits during the LGM to predominantly atmospheric deposition in the Holocene. Furthermore, volcanic activities during the transition may have been a principal cause of sudden increases in trench Hg deposition, potentially indicative of a global phenomenon.