G10 (I) Mercury in marine ecosystems

In the framework MEDOCEANOR program atmospheric mercury dynamic processes taking place in the Mediterranean Basin have been studied through several cruises over the Mediterranean Sea during different seasons from 2000 to 2010.

Since almost all our knowledge on the regional distribution of atmospheric mercury is derived from ground-based measurements at single locations for different time periods, the major scientific aim of this contribution is to provide spatially and temporally resolved information on the mercury distribution and speciation in the Mediterranean Marine Boundary Layer (MBL). The overall objective is to provide a quantitative assessment of Hg concentrations in different compartments of the air and water ecosystems and evaluate major chemical and physical mechanisms in the lower atmosphere and sea water, atmospheric transport and deposition as well as gaseous exchanges at the air–water interface with changing meteorological conditions. Mercury evasion derived from measurements of dissolved gaseous mercury (DGM) as well as speciated Hg measurements were performed across the Mediterranean Sea during nine seasonal cruises covering the eastern and western sectors of the Mediterranean basin. The relationship between Hg⁰_(g), RGM and HgP requires careful examination as the relative concentrations of the three species depends on the combined effects of local emissions, chemistry and long-range transport patterns. A diurnal cycle of the RGM concentrations has been observed over the off shore waters investigated during the cruises as result of in situ oxidation processes possibly by as yet unconfirmed daytime oxidants such as BrO. The highly variable RGM concentrations in the MBL have been indeed attributed to intense atmospheric-sea water exchange processes. Results showed, therefore, that the RGM formation in the Marine Boundary Layer is the key to all chemical processes in gaseous, aqueous and heterogeneous phases that may drive the formation of oxidized mercury and transform elemental Hg to oxidized Hg. Measurements performed during summer and early spring seasons consistently demonstrated that maximum solar radiation at midday occurred when DGM concentration was at its highest value. During nights DGM concentrations were generally lower compared to the concentrations measured during daylight. Further researches will be carried out within the ongoing GMOS project in order to better understand the relative contributions of all these mechanisms in the overall mercury cycle on regional and global scale and the relative contribution of natural vs. anthropogenic vs. recycling mechanisms which represents a crucial step in the assessment of the global cycle in the Mediterranean sea basin.

Most ocean surface waters are supersaturated with gaseous elemental Hg (Hg⁰), resulting in net evasion to the atmosphere globally. New global modeling results confirm that Hg evasion from the ocean accounts for approximately one-third of atmospheric emissions and around 80% of Hg deposited to the ocean surface reenters the atmosphere as the result of air-sea exchange. Efforts to evaluate model predictions and to gain insights into future improvements needed are constrained by sparse data availability on air-sea exchange processes. Here we present new continuous measurements of dissolved Hg⁰ in seawater and total atmospheric gaseous mercury (TGHg) from four cruises in Bermuda between 2008-2010 and one off the New England Shelf in 2008. Seawater Hg⁰ was estimated from equilibrium air concentrations determined every 5 minutes with a Tekran 2537A and collected using a continuous sparging set-up with a high water/air flow ratio that ensured gas:water equilibrium (seawater flow was 10-16 L/min and the Hg air flow 1.5 L/min). Water concentrations were calculated from these measured air concentrations using Henry’s law. Atmospheric concentrations were measured every 5 minutes using another Tekran 2537A. Results show atmospheric concentrations are fairly uniform across all cruises ranging from 1.47 ng/m³ off the New England Shelf and 1.41-1.43 ng/m³ in Bermuda. Measured seawater Hg⁰ concentrations were also fairly consistent across cruises and ranged from 0.11-0.14 pM, while the corresponding fluxes of Hg⁰ to the atmosphere predicted from the observational data were between 10-24 pmol/m²/hr, depending on wind speed and water temperature. We compare observations to model results from the GEOS-Chem biogeochemical Hg model. The model includes atmospheric oxidation of Hg⁰ by Br atoms and an updated surface ocean model that includes mechanistic descriptions of redox processes, particle associated Hg fluxes, and subsurface ocean inputs to the mixed layer. Preliminary results suggest the model reasonably reproduces observed atmospheric and oceanic Hg⁰ concentrations in most regions but underestimates seawater concentrations in coastal regions. We present results for a variety of sensitivity studies with the model, including a new parameterization for air-sea exchange that takes into account wave height and wave age for whitecapping.

Using a newly developed method (continuous flow isotope dilution-pyrolysis-inductively coupled plasma mass spectrometry) we have determined the mercury concentration in annual layers of calcium carbonate laid down by massive, colonial Schleractinian hard corals. Results from at least two sites will be presented, including impacted and relatively pristine sites in Bermuda. The mercury to calcium ratio from contemporary layers is similar to that observed in seawater at present, suggesting little selective retention or rejection from calcifying fluids. These results demonstrate the utility of this archive in reconstructing seawater mercury concentrations over time.

The Paleocene-Eocene Thermal Maximum (PETM) (55.8 Ma) marks a global climate excursion during which surface and bottom water temperatures increased by more than 5°C, the carbonate compensation depth shoaled to <2000 m, and the marine oxygen minimum zones expanded. Here, we investigate the effects of this global warm period on marine Hg deposition. We compared the Hg concentrations in continental margin and open ocean sediments deposited prior to, during, and through the recovery period following the PETM climate excursion. During the warming period, Hg concentrations in continental margin ODP Leg 174AX Bass River core sediments increased from 14 to 30 ng/g correlated with published d¹⁸O (r² = 0.82) and d¹³C (r² = 0.59) values. Additionally, since published sedimentation rates at Bass River quadrupled through the warming interval, we estimate that the Hg mass accumulation rate increased from ~50 to ~400 ng/cm²/ky as the climate excursion intensified. Open ocean Walvis Ridge (ODP Leg 208 site 1262) sediment Hg concentrations also increased, but remained extremely low. Sediments deposited before and after the PETM had Hg concentrations below the ~0.25 ng/g detection limit (combustion CV-AAS). Clay-rich PETM sediments in the Walvis Ridge core had measurable Hg concentrations ranging from 1.0 to 2.0 ng/g. Unlike the continental margin however, studies have determined that sedimentation rates at Walvis Ridge decreased through the climate excursion. In the peak of the PETM, we estimate a Hg mass accumulation rate of only < 3 ng/cm²/ky at Walvis Ridge. Studies have shown that both open ocean and continental margins experienced suboxic conditions during the PETM, but that mid-latitude continental margins were important organic C sinks while open ocean sediments had very low organic matter deposition. Elevated Hg deposition in continental margin sediments through the PETM is consistent with enhanced Hg delivery by organic matter. By virtue of being sinks for organic C, continental margins also were likely important sinks for Hg through the PETM. We assert that the organic matter cycle has a larger influence on the Hg cycle than do ambient redox conditions and we suggest that any effects of global climate change on the Hg biogeochemical cycle will be associated with changes in the organic carbon cycle.

Marine fish harvested from the South Pacific and Indian Oceans and sold globally contribute substantially to human methylmercury exposures. We present new data for total mercury and methylated mercury concentrations in 1000 m vertical profiles from the Indian Ocean collected in 2009 on the CLIVAR I5 cruise (32°S transect between 32E-105E). Total mercury in the surface mixed layer ranged between 0.50-0.71 pM. Vertical profiles generally exhibit nutrient like distributions, with lowest concentrations in the surface mixed layer and increasing to a maximum of 1.47 pM between 700-1000 m depths. We present the results of coupled oceanic and atmospheric modeling analyses to project trends in seawater concentrations given current atmospheric loading rates. We also show source attribution of present-day and potential future inputs based on emissions growth scenarios. Observed methylated mercury concentrations (51-158 fM) in the Indian Ocean are much lower than observed in our previous work in the North Pacific (average 266 fM in mesopelagic waters). Lower methylated mercury concentrations in the Indian ocean correspond to lower overall productivity in this region. Accordingly, we find that the empirical relationship between methylated mercury and organic carbon remineralization rates observed in the North Pacific can be extended to include the Indian Ocean seawater data, reinforcing the robustness of this relationship.

Knowledge of sources and mechanisms for the production and destruction of methylated mercury species in the oceans is of broad scientific, environmental, and human-health related significance. Thus, and given the limited information and knowledge regarding Hg distributions and biogeochemistry between near-shore and pelagic regions, we have been investigating processes and reactions affecting the cycling of monomethylmercury (MMHg) in sediments, biota, and waters over a broad portion of continental shelf and slope of the northwestern Atlantic Ocean. Here, principal and unifying findings from three oceanographic cruises (2008–10) will be presented. Gradients are evident for MMHg and dimethylmercury (DMHg) on the shelf with substantial enhancements near the sediments. These distributions suggest that both MMHg and uniquely, DMHg, are produced and mobilized from deposits on the continental margin. These inputs represent a potentially large source of methylmercury to the marine environment including the open ocean.

Upper ocean (< 1000 m) maxima in MMHg, DMHg, and total Hg correlate with the oxygen distribution (e.g., minimum zone) at remote stations on the slope. There are three potential and complementary sources for these enhancements: (1) production and release of methylmercury from shallow slope sediments (< 1000 m), (2) bacterially mediated Hg methylation in low O₂ zones of the water column, and (3) release of bioaccumulated Hg from sinking particles upon remineralization in the O₂ minimum. By either mechanism, Hg substrates and methylmercury species will have an anthropogenic component because methylation is occurring at depths less than 1000 m. Deep water profiles show no evidence for either sedimentary or hydrothermal sources of MMHg and DMHg. Complementary investigations are examining biological and geochemical controls on benthic MMHg production and efflux to overlying water as well as the transport, bioaccumulation, and transformation of MMHg in the water column.

Humans are exposed to toxic monomethylmercury (MMHg) principally by consumption of fish from marine environments. Major sources of MMHg to the ocean are suggested to include continental margins, deep-sea hydrothermal vents and sediments, and production in low-oxygen regions of the marine water column; however, inability to measure MMHg in seawater at levels less than about 50 fM has impeded our understanding of its oceanic distributions and cycling. Here, through use of a new analytical technique during the 2009 U.S. GEOTRACES Intercalibration in the North Pacific, we report the first fully resolved (i.e., all species detected), high-resolution vertical profile of both MMHg and dimethylmercury (DMHg) in seawater of a major ocean basin. Distributions of MMHg and DMHg, which are biogeochemically and oceanographically consistent, suggest that both are synthesized in low-oxygen and oxic strata of the water column and that the deep sea is not an important source of MMHg to surface-dwelling organisms. Our results imply that MMHg in seafood is derived from depths of the ocean that are impacted by anthropogenic mercury inputs and thus may be affected by future changes in mercury emissions to the atmosphere.

Methylated mercury (MeHg) in the marine waters is primarily from internal sources. Freshwater from rivers, groundwater and atmosphere, and hydrothermal inputs represent a minor part of the oceanic loading. The open ocean hypoxic waters, where microbial activity is driven by the decomposition of organic matter, are the main site of net MeHg formation. Recently numerous high resolution MeHg profiles in the open ocean waters corroborate this assessment. However, because of the heterogeneity of the oceanic biomass, Hg methylation in particular milieus of the Ocean may be of important environmental concern. These milieus include deep-sea nepheloide layers, continental margins (shelves, slopes and continental rises), and sea-ice. We present examples from deep Mediterranean waters, Northwestern Mediterranean shelf, and Antarctic sea-ice. High resolution pictures near the liquid-solid interfaces of these environments allow a better estimation of the importance of these particular regions in the global oceanic MeHg budget.