Datasets on the spatial distribution of mercury and its controlling factors in the Yellow Sea

Large amount of anthropogenic mercury (Hg) emitted from China has been transported and deposited in the northwestern Pacific marginal seas; in particular, the Yellow Sea adjacent to China is immediately affected by Chinese-high Hg emissions [1,2]. This article presents the comprehensive baseline dataset on the mercury concentrations and their controlling factors in surface sediments from the entire Yellow Sea shelf, including Korean and Chinese rivers and coastal zones. These data supported the research article entitled “Sedimentary mercury (Hg) in the marginal seas adjacent to Chinese High-Hg emissions: source-to-sink, mass inventory, and accumulation history” Kim et al. [1]. Some of the data was used in Kim et al.’s research paper [3] with the reference [1]. A total of 492 surface sediments were collected from the Yellow Sea shelf and coastal zones, and the rivers around the Sea. All sediment samples were freeze-dried and ground by agate mortar for analyzing total mercury (THg) and related elemental components (total nitrogen, total carbon, total inorganic carbon, total organic carbon, and aluminum). Most previous studies on the sedimentary Hg were conducted locally, mainly in the river-dominated coastal and inner shelf zones of the Yellow Sea, which are associated with riverine Hg inputs. Thus, the quality and quantity of available sedimentary Hg data, on which we rely for mass inventories of Hg in the Sea, are limited. In this respect, our large dataset may contribute significantly to a better understanding of the behaviors of riverine and atmospheric Hg from Chinese sources and will help to further refine global estimates of Hg discharge to ocean margins and open oceans in East Asia. Additionally, the dataset will be essential for improving numerical model for global budget calculation and prediction.

large dataset may contribute significantly to a better understanding of the behaviors of riverine and atmospheric Hg from Chinese sources and will help to further refine global estimates of Hg discharge to ocean margins and open oceans in East Asia. Additionally, the dataset will be essential for improving numerical model for global budget calculation and prediction.
© • This baseline dataset of THg-TOC-Al in the Yellow Sea will be invaluable for the researchers studying the global Hg inventory (especially, numerical modeling study) as it yields crucial and essential information for calculating the quantitative contribution (i.e., depositional flux and budget) of Chinese-emitted Hg in the northwestern Pacific marginal seas as well as other open oceans. • Our dataset can also be used to accurately estimate the natural background level of Hg to better assess the anthropogenic Hg impacts on marine environments and ecosystems in the marginal seas.
Surface sediments of the study area exhibit a broad variation in Al concentration, ranging from 1.6% to 10.4% (average 6.4 ± 1.5%), with a relatively high level of Al ( > 7%) in three finegrained shelf muddy sediments, and coastal and riverine sediments ( Fig. 2 e). In general, Al concentration highly dependent on grain size: Al concentration increases as the contents of finegrained particles in the sediments increase. As shown in Fig. 2 , Al shows a similar spatial distribution pattern with TOC and TN, showing these elements are constrained primarily by the sediment grain size.
THg concentrations in the study area ranged from 1.1 to 215.7 ng/g (average 25.8 ± 27.9 ng/g) with significant spatial variation ( Fig. 2 f). Unlike to Al concentration, the highest concentrations are observed in the rivers (Korean rivers: 12.2-134.6 ng/g; Changjiang: 56.5-215.7 ng/g; Huanghe: 7.9-71.0 ng/g) and their estuarine and coastal zones ( > 50 ng/g). Characteristically, the concentrations in the ocean waste-disposal site (23.4-145.0 ng/g) are as high as those in the coastal sediments, followed by the shelf mud zones (25-40 ng/g) of the central Yellow Sea. Except for these areas, most of the shelf sediments are less than 20 ng/g in concentration. Overall, the THg concentrations in the rivers and their coastal zones, and the ocean waste-disposal site are higher compared to the upper continental crust (50 ng/g, [5] ) and the natural values ( < 50 ng/g, [6] ). In the Yellow Sea shelf area, the concentration is relatively lower than in the East/Japan Sea (30-164 ng/g), Okhotsk Basin (36-55 ng/g), and Western Arctic Ocean (62-68 ng/g) [3] .

Experimental Design, Materials and Methods
East Asia (mainly China and India) is one of the largest Hg emission source regions in the world; for example, China contributes about 30 and 50% of global and Asian anthropogenic Hg emissions, respectively [7 , 8] . In this respect, the study of sedimentary Hg in Chinese marginal seas (e.g., Yellow Sea, East China Sea, and East/Japan Sea) may yield valuable information about regional and global Hg cycles as well as the behavior of anthropogenic Hg. In particular, the Yellow Sea might be vulnerable to the impact of anthropogenic Hg, and their sediments are probably important reservoirs for Chinese land-based Hg via atmospheric and riverine transport. However, there is no available sedimentary Hg dataset covering the entire Yellow Sea because many works of sedimentary Hg are restricted to the local areas, such as estuarine and coastal areas, and severely polluted bays [e.g., [9][10][11]. In addition, such limited availability of data may provide relatively rough calculations and predictions in the numerical model. Thus, a large comprehensive dataset of sedimentary Hg with its controlling components in the entire Yellow Sea has been required for better understanding the behaviors of riverine and atmospheric Hg from Chinese sources.
To achieve large THg data, a total of 492 surface sediment samples were collected in the entire Yellow Sea shelf, including Korean and Chinese rivers, estuaries, and coastal zones ( Fig. 1 ). Simultaneously, data on controlling factors affecting Hg deposition were gathered; it is well known that organic-enriched fine particles likely accelerate the deposition of Hg to the sediments through a scavenging process in the water column and consequently control the Hg distribution in the sediments [12 , 13] . Thus, a comprehensive THg-Al-TOC dataset (including TN and TIC) for 492 surface sediment samples covering the entire Yellow Sea has been establised.
Surface sediment samples were collected by grab sampler and box corer between 2001 and 2010 to achieve extensive spatial coverage of the entire Yellow Sea. All sediment samples were stored in an onboard freezer, then were freeze-dried for 1-2 days to be ground to a powder with a mortar and pestle (model pulverisette 6, FRITSCH, Germany). About 50 mg of powdered sample was analyzed for THg concentrations using an automatic Hg analyzer, involved thermal decomposition followed by catalytic reduction, amalgamation, desorption, and the use of an atomic absorption module (model Hydra C II Direct Hg analyzer; Teledyne Leeman Labs, Hudson, NH, USA), which is based on US EPA 7473 [14] . The powdered sample in the nickel boat is pyrolyzed at about 850 °C and oxidized Hg is transformed into elemental mercury (Hg 0 ) through a catalyst tube. Then, Hg 0 gasses are detected using 820 nm wavelength. THg concentrations were calibrated against marine sediment reference material (MESS-3; total Hg = 91 ±9 ng/g certified by the National Research Council of Canada) for trace metals and other constituents. The detection limit was obtained less than 0.02 ng. In measurements of the Hg concentration, the analytical accuracy was determined to be < 5%, based on replicate analysis of the certified reference material (MESS-3). The analytical precision was determined to be < 10% based on replicate measurements of standard materials and sediment samples.
For TN and TC contents, the about 15 mg powdered sediment samples with reference material (L-cystine; 11.66% N, 29.99% C) were analyzed using a CHN elemental analyzer (FLASH 20 0 0, Thermo Fisher Scientific, USA). The encapsulated powder samples by tin are combusted in silvered cobaltous-cobaltic oxide and chromium oxide filled tube at 900 °C then reduced in copper oxide filled tube at 550 °C. On the basis of replicated measurements of a standard material and sediment samples, the accuracy and precision in TN and TC measurements were within 5%. TIC contents were measured using a CO 2 coulometer (CM5014, UIC, USA) by reacting approximately 15 mg powdered samples with 40% perchloric acid. The precision and accuracy in TIC analyses were < 5% in replicated reference material (calcium carbonate having 12% C) and sediment samples measurements. TOC contents were calculated by subtracting TIC from TC contents.
For Al concentrations, about 0.2 g powdered samples, including an international standard reference material (MAG-1; Al 2 O 3 = 16.4, 0.3%; certified by U.S Geological Survey), were dissolved in a mixture of hydrofluoric-perchloric-nitric acids in an airtight Teflon bomb. These solutions were then analyzed to determine Al concentration using inductively coupled plasma atomic emission spectroscopy (ICP-AES, model Spectro Flame Modula EOP; SPECTRO Analytical Instruments Inc., Germany). The precision and accuracy of concentrations were monitored by repeated analyses of the standard MAG-1. Differences between the determined and recommended values were less than 5%, and the leaching efficiency checked by MAG-1 was about 90%.