Mercury budgets of watersheds and estimation of riverine discharge based on the measurements of major rivers in eastern Japan.

Abstract

Atmospheric mercury is deposited on land and in the ocean through both wet and dry deposition. In recent years, the role of rivers in transporting mercury from terrestrial to marine environments has received increasing attention, with estimates suggesting that annual riverine transport may be comparable to atmospheric deposition (Liu et al., 2021a, b). However, observational data on mercury concentrations in Japanese rivers remain limited to the Hokuriku region (Ohno et al., 2020), leaving many aspects of mercury cycling at the island arc scale unresolved. Furthermore, the mercury speciation in river water is essential for assessing the contribution of terrestrial sources to methylmercury input into the ocean. In this study, we collected samples from major rivers across eastern Japan to quantify concentrations of dissolved mercury (DHg), particulate mercury (PHg), total mercury (THg), and dissolved monomethylmercury (MeHg), and to estimate mercury budgets and riverine fluxes to the ocean.The mean (± standard deviation) concentrations of each mercury species in river water were: DHg, 0.68 ± 0.40 ng/L; PHg, 1.78 ± 2.78 ng/L; THg, 2.48 ± 2.93 ng/L; and MeHg, 6.1 ± 4.4 pg/L. PHg was suggested to be the dominant mercury species in river water (PHg/THg: 57.4 ± 18.5%). The estimated riverine mercury flux was 1.9 ± 2.2 µg/m²/yr, accounting for 12.6 ± 10.9% of atmospheric deposition. The total annual mercury export from the Japanese archipelago via rivers was estimated to be 0.95 ± 0.15 t/yr for THg and 2.0 x 10^-3 t/yr for MeHg. The THg flux accounted for approximately 14% of atmospheric deposition to land (wet deposition: 3.8 ± 0.11 t/yr; dry deposition: 3.0 ± 1.0 t/yr). The discrepancy between atmospheric deposition and riverine export is likely due to the reemission of deposited mercury and retention in vegetation and soils. The THg flux estimated in this study was approximately one order of magnitude lower than the previous estimate of 8.3 ± 6.0 t/yr by Liu et al. (2021b). Liu et al. estimated PHg and the solid-liquid partition coefficient (K_d) using an empirical equation in which suspended solids concentration (SS) was the explanatory variable. DHg was then calculated from PHg and K_d. However, SS in Chinese rivers are generally higher than those in Japanese rivers and K_d values tend to be lower (estimated from Ohno et al., 2020). Therefore, Liu et al.'s method may have led to an overestimation of mercury fluxes from Japanese rivers to the ocean.