G7 (II) Environmental Biogeochemistry: Environmental Approaches

Freshwater runoff from rivers to continental shelves has a crucial influence on both the hydrodynamics and biogeochemistry of many coastal zones. Buoyant freshwater coming from the river outflows tends to spread away from the river mouth and forms an estuarine plume.

These areas are thus preferential sites where material of terrestrial origin, including that from the anthropogenic activities, is captured, transformed and eventually stored on the continental margins or exported more or less directly to the open ocean. Along estuarine – coastal continuum, mercury may undergo significant changes in speciation due to changes in salinity regime, properties of the dissolved organic matter and photochemical reactivity. These important changes within a river discharge coastal plume will drastically control the net amount and bioavailability of monomethyl Hg (MeHg) transferred to the coastal zone via both biotic and abiotic methylation, demethylation and redox processes.

The aim of this study is to characterize the behaviour and distribution of Hg species in the Adour Estuary (South Bay of Biscay) and examined the role of estuarine – coastal mixing zone in the reactivity of Hg. The spatial variations of mercury species concentrations following a longitudinal and vertical sampling strategy were performed during two campaigns (RV Côte de la Manche, 04/07-05/10) exhibiting different river discharges and plume extent over the coastal zone. Gaseous (Hg°, Me₂Hg), particulate and dissolved (MeHg, Hg²⁺) mercury species concentration have been measured. Specific water incubation experiments using isotopically labelled species (¹⁹⁹Hg(II) and Me²⁰¹Hg) have been used to investigate transformation mechanisms (i.e. methylation, demethylation and reduction) at the different water masses dilution of the plume and under different experimental conditions in order to assess the relative contribution of photochemical versus biological processes to these mechanisms. A conceptual model of the mercury species fate in this coastal system has been elaborated and demonstrate that while Hg methylation remains low within the estuarine plume (yield range: 0-0.4 %/day), land discharged MeHg is efficiently demethylated via biotic and abiotic pathways (6-50 %/day). In addition, elemental gaseous Hg is increasingly photochemically produced seaward (0.5-43.5 %/day).

We made use of the long-term research site at the Smithsonian Environmental Research Center (SERC) in central Maryland to study the fluxes of mercury (Hg) and methylmercury (MeHg) in three small first-order mid-Atlantic coastal plain watersheds. Our goals include the first assessment of watershed Hg yields in the mid-Atlantic; an assessment of MeHg production in shallow groundwater; and establish a baseline to assess any change following implementation of Hg emissions regulations. While tracking changes in atmospheric deposition is relatively straight forward, observing changes on MeHg concentrations in the various environmental compartments of an ecosystem, including the biological receptors, is a bigger challenge. Our watershed research is part of a larger framework we have been developing at SERC, which includes a Mercury Deposition Network Site, and studies of Hg and MeHg dynamics in the estuarine receiving water body and tidal wetlands.

This presentation will focus on deposition and flux of Hg and MeHg from 2008 through 2010. The watersheds reflect three landuse types – forest, agriculture and mixed development. Deposition of Hg in precipitation prior to implementation of emissions controls on Maryland power plants was between 5.98 and 8.2 ug m^-2, with the highest loadings, driven by rainfall, in the spring and summer. In our watersheds, spring and summer rain were generally released whereas winter precipitation mainly went to storage.

The three studied watersheds produced relatively high yields of Hg (20 to 37%) in relation to deposition, exporting approximately 30 mg of Hg ha^-1 in 2008. We found substantial production of MeHg in all three watersheds which resulted in fluxes ranging from 0.6 to 1.3 mg ha^-1 in 2008. Studies within the watersheds suggest that a substantial fraction of MeHg is produced in stream banks/riparian zones. While the importance of MeHg production in stream banks has been proposed, the variations in productivity based on the seasonal distribution and amount of precipitation have only been hypothesized. Stream water and riparian zone chemistry, namely the production of ammonia, strongly connects the evolution of microbial community within the riparian zone to MeHg export. The location and importance of the Hg methylation sites, both in terms of flux and biological relevance are largely controlled by stream hydrology. We suggest MeHg originating in stream banks are an undocumented yet potentially significant source of MeHg to the Chesapeake Bay.

During the loading phase of the METAALICUS project, Lake 658 at the Experimental Lakes area received annual additions of isotope enriched mercury, which approximately doubled the amount of mercury loading to the lake surface. Loading ceased at the conclusion of the 2007 field season, and since that time the project team has been monitoring the recovery. This paper will describe and discuss the processes controlling the response of the water column to a reduced load, and the fate of the added spike. Labeled mercury concentrations continuously declined in the lake water with inorganic mercury dissipating the fastest. Within one year, concentrations in surface water dropped to less than 0.1 ng/L (>95% of the observed peak concentration) and after only another 12 months until it became undetectable. Initially, concentrations were marginally higher at the bottom of the lake, presumably indicating a sediment source. However, during a period in late Fall of 2009, inorganic spike Hg became undetectable in the water column. Methylmercury (MMHg) on the other hand continued to show its characteristic annual variations, albeit at much lower concentrations compared to those observed during the loading phase. Each year, spike MMHg accumulated in bottom waters during late summer until early fall and gets mixed into the entire water column after fall turnover – a pattern also observed for ambient MMHg. The total mass of spike MMHg in the entire water column declined with each successive year. However, three years into the recovery phase, the MMHg spike is still clearly visible in the water column and contributing MMHg bioaccumulation in biota. This shows that although older inorganic Hg disappears relatively quickly after cessation of the source, spike MMHg will have a longer effect on the study lake. This observation probably has implications for the recovery times of other lakes should widespread emissions be enacted.

Recent works pointed to the elevated production of methylmercury (MeHg) in salt marsh sediments and its accumulation in the root-sediment system. During flooding tide the complex mosaic of channels in the rooting sediment are filled with tidal water. To the best of our knowledge the effect of this pumped mechanism on MeHg produced in salt marshes has not been documented.

In order to gain a better understanding of the distribution and export of MeHg, flooding water and sediment cores (10-cm long) were sampled several times during the first 30 minutes of tidal flooding in two salt-marshes of the Tagus estuary. Immediately after sampling, water was filtered and sediment cores sliced and stored in decontaminated plastic bags. In the laboratory, separations were performed to obtain subsamples of sediments, pore waters, and roots. Reactive mercury, total mercury MeHg_D were quantified in water samples and total mercury (Hg) and methylmercury (MeHg) in solid sediments and belowground biomass. Physico-chemical parameters such as organic carbon, pH, salinity and dissolved sulphate and sulphide were also determined in water and sediments.

Concentrations of dissolved mercury species in flooding water increased up to 10 times during the first four minutes of tidal flooding. Concomitantly MeHg levels decreased in pore waters and solid sediments. Before flooding MeHg in roots was up to 35% of the total Hg, but decreased to 8% and 4% only 2 and 8 minutes after the inundation. These results suggest that MeHg is far mobile that would be expected if the sequestration processes was dominated by the strong affinity to cysteine-rich peptides. The short-time variations registered in this study indicate a shift on the mercury equilibrium between pore waters, solid sediments and root surface. Moreover, as water floods the salt marsh mercury, namely MeHg, escape the sediments being potentially available to biota living in salt marshes.

Estuaries act as repositories for river-borne particulate contaminants such as mercury (Hg). The high degree of variability in geochemical and physical characteristics and infaunal density render estuaries ideal sites for Hg transformation. Porewater and sediment chemical and molecular analyses were performed at two mudflats to study Hg dynamics in the Great Bay estuary, New Hampshire, USA. The site situated at the mouth of a river (Squamscott) was tidally and fluvially influenced with a wide variation in salinity. The site located closer to the mouth of the estuary (Portsmouth) was tidally dominated with higher salinity, a lower advective velocity and insignificant sediment transport. Sediment from both sites was mixed, as indicated by the presence of ⁷Be in the sediment (half-life 53 d). However, mixing at Squamscott was due to sediment erosion and deposition, while mixing at Portsmouth was due to bioturbation. Sediment inorganic Hg (Hg_i) concentrations and profiles were similar at both sites. Porewater Hg_i concentrations, acid-volatile sulfides (AVS), alkalinity and DOC were lower at Portsmouth while sediment Fe(III) was higher due to infaunal burrowing activity. Sediment-water partitioning of Hg_i and lower DOC concentrations at Portsmouth were proposed to be controlled by adsorption to freshly precipitated Fe(III) hydroxides and removal by bioirrigation. Lower AVS at Portsmouth was due to oxidation of FeS(s) with oxygen introduction. Fluorescence analysis of DOC revealed a greater percentage of labile, protein-like DOC in Portsmouth porewater compared to Squamscott. Microbial DNA analysis showed similar concentrations and distributions of sulfate-reducers and iron-reducers at both mudflats, whereas mer-A (a gene responsible for the production of mercuric reductase which reduces Hg(II) to volatile Hg⁰) and methanogen concentrations were significantly higher at Squamscott. Porewater Hg_i concentrations were higher at Squamscott and peak methylmercury (MeHg) concentration was closer to the sediment-water interface. Peak porewater MeHg concentration was higher and at a greater depth at Portsmouth which may be due to the introduction of labile DOC deeper in the sediment, indicating greater methylation efficiency with bioturbation. Porewater and solid-phase MeHg peaks at Squamscott corresponded to one another and occurred within the top 4 cm, similar to other estuarine sediments with no significant degree of bioturbation. A solid-phase MeHg peak was not found in the top 10 cm at Portsmouth. The findings of this study suggest that bioirrigation may affect the methylation efficiency of sediment, and high sediment Fe(III) levels may affect the sediment-water partitioning of Hg_i, controlling the availability of Hg_i for methylation in bioturbated sediments.

The Toulon Bay is a semi-closed area (North-western Mediterranean), which receives various anthropogenic inputs from harbors (marina, navy) and industrial activities. It is divided in two parts: the inner (“small bay”) and the outer bay (“large bay”). Hydrodynamic in the “small bay” leads to long water residence time, increasing the risk of sediment and/or water column contamination by remobilization processes. The objectives of this study were to establish Hg contamination in sediment, and understand the main mechanisms of Hg remobilization to the water column.

Surface sediments (0-5cm) were sampled in the entire Bay to map the contamination state and its dispersion. Results have shown a large contamination, especially for the most enclosed parts of the small bay, with Hg concentrations up to 40µg/g. MMHg levels were significantly correlated to Hg, consisting less than 1% of the total Hg.

High resolution interface cores (2cm slices, under nitrogen) were analyzed to study Hg/MMHg profiles in sediments, in relation to geochronology of the settling material and diagenesis processes. Hg profiles showed a marked pollution event at a depth (~11cm) corresponding to the period of the 2^nd World War. In addition to the correlation with Hg profiles, the MMHg profiles presented a surface maximum.

The most probable modes of Hg dispersion within the bay are: (1) transport of Hg contaminated particles through the whole bay due to sediment resuspension events (storm, boat traffic, dredging), (2) Hg remobilisation to the water column during these events, and (3) Hg diffusive fluxes at the water/sediment interface. The first mode is attested by Hg surface cartography and hydrodynamic of the bay. Additional experiments were carried out to investigate the two others. Batch remobilization experiments were carried out by mixing fresh sediment (from contrasted areas, 0-2 cm top slice) with ambient seawater under oxic conditions (1g/L of dry sediment). Time variations (15min to 100h) of dissolved Hg/MMHg concentrations were measured and displayed only a weak Hg mobilisation from solid. In the same contrasted areas, the dynamic of MMHg at the benthic interfaces was studied using Peepers. The MMHg distributions in interstitial waters showed a broad maximum at the top of the anoxic sediments, suggesting an “usual” in situ Hg methylation.

Determination of only total mercury in sediments does not give an accurate estimate of the likely environmental impacts. Speciation study of mercury in sediment provides information on the potential availability of mercury to biota under various environmental conditions. The toxicity of mercury depends especially on its chemical forms rather than its total content. According to regulation model (water levels were projected to fluctuate in a cycle opposite from natural conditions, with lower levels during the summer and higher levels in the winter) for Three Gorges Reservoir, the characteristic of speciation of mercury in sediment of water-level-fluctuating zone of Three Gorge Reservoir Area was investigated. Samples of flooding sediments, semi-flooding sediments, drying sediments, original soil and overlying water were collected from five sites during the flood season (summer) and mercury speciation (operationally defined) and some physicochemical parameters were analyzed. The results indicated that mercury mainly existed in residual fraction for flooding sediments, alkali-soluble fraction and residual fraction for semi-flooding sediments and original soil, Fe-Mn oxides and residual fraction for drying sediments. In flood season, with the increasing waterlogging duration caused the increase of TOC and pH and decrease of ORP. Mercury has the tendency to release to water body. The amount of migration of bioavailable fractions of mercury in five sites were 0.002mg/kg, 0.005 mg/kg, 0.010 mg/kg, 0.027 mg/kg and 0.003 mg/kg respectively. Correspondingly, the migration rates of mercury were 5.5%, 16.90%, 33.50%, 47.90%, and 7.20%, respectively. The total mercury and methyl mercury in overlying water were 29.88±18.78ng/L and 0.19±0.11ng/L. The regulation model for Three Gorges Reservoir increased the ecological risk of mercury. Still, even though the concentrations of Hg in the study areas were low, a long-term water quality monitoring in Three Gorges Reservoir was necessary.

Methylmercury (MeHg), a potent neurotoxin, is among the most widespread contaminants that are posing potential threat to both humans and wildlife. In the Florida Everglades, MeHg is recognized as one of the major water quality concerns. Degradation of MeHg in the water is thought to be one of the most important processes to the cycling of MeHg, but there is a lack of quantitative estimations of its effect on the distribution and cycling of MeHg in this ecosystem. Stable isotope (Me²⁰¹Hg) addition method was implemented to investigate the degradation of MeHg in the Everglades. By combining these results with the field monitoring data, effects of photodegradation on MeHg distribution and its contribution to MeHg cycling were estimated. The results indicate that degradation of MeHg in Everglades water is mediated by sunlight, and that UV-A and UV-B radiations are the principal driver. The spatial pattern of MeHg photodegradation potential (P_PD) generally illustrated an increasing trend from north to south in the Everglades, which was opposite to the distribution of MeHg in water column. Correlation analysis shows that MeHg concentration in the water had a significant negative relation to P_PD, suggesting that photodegradation could play an important role in controlling the distribution of MeHg in the Everglades waters. Furthermore, about 31.4% of MeHg input into the water body was removed by photodegradation, indicating its importance in the biogeochemical cycling of MeHg in the Everglades. This percent reduction is much lower than that reported for other ecosystems, which could be caused by the higher concentration of DOC in the Everglades. The relatively slower degradation of MeHg could be one of the main reasons for the high ratio of MeHg to total mercury (THg) in this ecosystem.