Dissolved gaseous mercury formation and mercury ...

Science of the Total Environment 603–604 (2017) 279–289

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Dissolved gaseous mercury formation and mercury volatilization in intertidal sediments Rute Cesário a,b, Laurier Poissant c,1, Martin Pilote c, Nelson J. O'Driscoll d, Ana M. Mota a, João Canário a,⁎ a

Centro de Química Estrutural, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal IPMA-Instituto Português do Mar e Atmosfera, Rua Alfredo Magalhães Ramalho, 6, 1495-006 Lisboa, Portugal Environment and Climate Change Canada, Aquatic Contaminants Research Division, Water Science and Technology Directorate, Montréal, QC, Canada d Acadia University, Department of Earth and Environmental Science, K.C. Irving Environmental Science Center, Wolfville, NS, Canada b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Tidal flushing increases Hg and MMHg transport in the sediment water interface. • Hg and MMHg export occurs in the first minutes of inundation. • Transported Hg species are easily reduced to Hg° that escape to the atmosphere. • Tidal flushing mechanisms should be considered when estimating regional Hg budgets.

a r t i c l e

i n f o

Article history: Received 25 March 2017 Received in revised form 10 June 2017 Accepted 11 June 2017 Available online xxxx Editor: D. Barcelo Keywords: Mercury speciation Tidal flushing Intertidal sediments Tagus estuary

a b s t r a c t Intertidal sediments of Tagus estuary regularly experiences complex redistribution due to tidal forcing, which affects the cycling of mercury (Hg) between sediments and the water column. This study quantifies total mercury (Hg) and methylmercury (MMHg) concentrations and fluxes in a flooded mudflat as well as the effects on waterlevel fluctuations on the air-surface exchange of mercury. A fast increase in dissolved Hg and MMHg concentrations was observed in overlying water in the first 10 min of inundation and corresponded to a decrease in pore waters, suggesting a rapid export of Hg and MMHg from sediments to the water column. Estimations of daily advective transport exceeded the predicted diffusive fluxes by 5 orders of magnitude. A fast increase in dissolved gaseous mercury (DGM) concentration was also observed in the first 20–30 min of inundation (maximum of 40 pg L−1). Suspended particulate matter (SPM) concentrations were inversely correlated with DGM concentrations. Dissolved Hg variation suggested that biotic DGM production in pore waters is a significant factor in addition to the photochemical reduction of Hg. Mercury volatilization (ranged from 1.1 to 3.3 ng m−2 h−1; average of 2.1 ng m−2 h−1) and DGM production exhibited the same pattern with no significant time-lag suggesting a fast release of the produced DGM. These results indicate that Hg sediment/water exchanges in the physical dominated estuaries can be underestimated when the tidal effect is not considered. © 2017 Elsevier B.V. All rights reserved.

1. Introduction ⁎ Corresponding author. E-mail address: [email protected] (J. Canário). 1 Laurier Poissant is retired.

http://dx.doi.org/10.1016/j.scitotenv.2017.06.093 0048-9697/© 2017 Elsevier B.V. All rights reserved.

Mercury (Hg) is considered a global pollutant and may exist in many chemical and physical forms in natural waters (Hines, Brezonik, &

280

R. Cesário et al. / Science of the Total Environment 603–604 (2017) 279–289

Engstrom, 2004): dissolved divalent mercury species (Hg(II)), dissolved gaseous elemental mercury [Hg(0) (aq)], dissolved monomethylmercury (CH3Hg(I)), and particulate-adsorbed forms. Numerous studies have shown that the cycling of elemental mercury (Hg(0)) is of great importance to Hg transport, residence time, and reactivity in natural waters (Amyot, Gill, & Morel, 1997a; Mason, Lawson, & Sheu, 2001; Mason, Fitzgerald, & Morel, 1994; Wang et al., 2015). Among Hg chemical species, monomethylmercury (herein defined as methylmercury, MMHg) is of most concern because it can be readily biomagnified in food webs (Boening, 2000; Harris, 2003). Methylmercury is produced by the methylation of divalent mercury (Hg(II)) in wetland sediments and its availability is a critical variable determining the production of MMHg (e.g. Ullrich, Tanton, & Abdrashitova, 2001). It is widely accepted that biological methylation is the dominant mechanism of MMHg production in aquatic environments. However, (Celo, Lean, & Scott, 2006) demonstrated that in the presence of chemical methyl donors and under specific environmental conditions (including pH, temperature, and the presence of complexing agents such as chloride), some abiotic pathways for mercury methylation are possible. The presence of MMHg in estuarine and coastal environments is due to several sources, including fluxes from sub-tidal sediment (Cesário et al., 2016; Hollweg, Gilmour, & Mason, 2009) water column methylation (Monperrus et al., 2007), riverine inputs (Mason et al., 1999; Riedel, Williams, Riedel, Gilmour, & Sanders, 2000), groundwater (Black et al., 2009), and coastal wetlands (Bergamaschi et al., 2011; Mitchell, Jordan, Heyes, & Gilmour, 2012). In aquatic ecosystems, MMHg flux from sub-tidal sediment is the most studied pathway and possibly the dominant source of MMHg to estuarine and coastal waters (Hammerschmidt, Fitzgerald, Lamborg, Balcom, & Visscher, 2004; Hollweg, Gilmour, & Mason, 2010), where availability of inorganic Hg appears to be the primary control on its production (Cesário et al., 2016; Hammerschmidt & Fitzgerald, 2004). However, active MMHg production in sediment is not always predictive of water column MMHg concentrations (Balcom, Schartup, Mason, & Chen, 2015). Understanding the variables influencing MMHg production is key to identifying strategies that can be used to reduce MMHg levels in aquatic ecosystems. Elemental Hg in natural waters is measured as part of the operationally defined fraction “dissolved gaseous Hg” (DGM). This fraction includes all purgeable gaseous species of mercury such as dimethylmercury [(CH3)2Hg] and the elemental form [Hg(0)], the latter being responsible for N90% in this pool (Lanzillotta, Ceccarini, & Ferrara, 2002). It is well known that the reduction of dissolved Hg(II) to the gaseous Hg(0) is mostly a photochemical mechanism in sea- and freshwater (Liu et al., 2012). In fact, Rolfhus and Fitzgerald (2004) reported that only 70% of bulk reduction was driven photochemically, being the remaining 30% attributed to biotic reduction and by uncharacterized dark reduction in the dissolved phase. Additionally, (Qureshi, O'Driscoll, Macleod, Neuhold, & Hungerbühler, 2010) found that microbes and colloids did not appreciable influence reduction or oxidation kinetics of mercury in water from the Atlantic Ocean. In water bodies, DGM undergoes bidirectional fluxes trough the water-air interface and this mechanism significantly regulates the elemental mercury pool in these compartments (Lanzillotta & Ferrara, 2001). Several researchers have demonstrated that volatilization is a significant part of the mercury cycle (Rolfhus & Fitzgerald, 2001; Gao et al., 2006). However, the mechanisms and factors that control the DGM production and distribution need to be better constrained. Several factors have been shown to be critical in terms of regulating DGM production in surface waters: (i) the intensity of solar radiation (Amyot, McQueen, Mierle, & Lean, 1994; O'Driscoll, Beauchamp, Siciliano, Rencz, & Lean, 2003a), (ii) radiation wavelengths (O'Driscoll, Poissant, Canário, Ridal, & Lean, 2007), (iii) the availability of a photosensitizer such as DOM (O'Driscoll, Lean, Loseto, Carignan, & Siciliano, 2004; Garcia, Amyot, & Ariya, 2005) and (iv) Hg(0)-suspended particulate matter (SPM) (Wang et al., 2015; Castelle et al., 2009). In fact, a daily cyclic pattern of DGM concentration

in relation with solar intensity is observable in both seawater and freshwater systems (Castelle et al., 2009; Lanzillotta & Ferrara, 2001; O'Driscoll et al., 2003a; O'Driscoll, Poissant, Canário, & Lean, 2008; O'Driscoll et al., 2007). Temperature must also be considered in the DGM flux in water. In fact, Loux (2000) examined diel temperature effects on the water to - air exchange of mercury and observed significant changes in Henry's law constants that are not currently accounted for in mercury flux models. The positive correlation between temperature (from both air and water), DGM production and Hg volatilization reported by O'Driscoll et al. (2003a, 2007) indicated that this factor is a consideration in mercury flux. In Tagus estuary, Portugal, Canário and Vale (2004) reported that intertidal surface sediments exposed to direct solar radiation transferred about 56% of total Hg to the atmosphere, per day, and that salt-marsh plants also played an important role on the emission of Hg(0) to the atmosphere (Canário et al., 2017). Lee, Benoit, and Hu (2000) also found a correlation between Hg flux and sediment temperature, but this relationship is confounded due to the obvious correlation between temperature and solar radiation. Moore and Carpi (2005) suggested that solar radiation mediates Hg(II) reduction to Hg(0) and that temperature mediates the volatilisation of Hg(0) from soils. Furthermore, (Sizmur, McArthur, Risk, Tordon, & O'Driscoll, 2017) supports the hypothesis that photoreduction of Hg(II) to Hg(0) is mediated by both sediment temperature and solar radiation and that photoreduction (rather than volatilisation of Hg(0)) is the rate limiting step in these sediments. Amyot, Mierle, Lean, and Mc Queen (1997b) observed that UVB radiation is particularly important for mercury photoreduction in low dissolved organic carbon (DOC) temperate lakes. Notwithstanding, O'Driscoll et al. (2004) reported that in the surface of freshwaters the increasing of DOC concentration enhanced the photoreduction rate of mercury. A similar pattern was also observed in tropical waters in French Guyana (Beucher et al., 2002; Peretyazhko, Charlet, Muresan, Kazimirov, & Cossa, 2006), where mercury photoreduction occurred in waters with high DOC concentration. Suspended particulate matter concentrations might either enhance or limit DGM production. Indeed, lower DGM levels have been observed in surface waters with higher SPM concentrations, probably due to SPM-bound Hg(0) (Tseng, Lamborg, Fitzgerald, & Engstrom, 2004). The presence of particulate Hg(0) in water should be considered when studying Hg species transformation. A decrease of Hg(0) in water is often attributed to the oxidation of Hg(0) to Hg(II) (Amyot et al., 1997a; Zhang & Lindberg, 2001). However, the inverse reaction (reduction of Hg(II) to Hg(0)) may occur simultaneously, lead to a misleading conclusion. Afterwards, a fraction of Hg(0) could rapidly become non-purgeable upon being adsorbed onto SPM in water. Wang et al. (2015) demonstrated that about 70% of the total Hg(0) was bound to SPM and non-purgeable, suggesting the occurrence of particulate Hg(0) in natural waters. Also Castelle et al. (2009) showed that the presence of high SPM concentrations might strongly modify DGM production by reducing light penetration into estuarine surface waters, suggesting that the hydrodynamic component in the gaseous Hg cycle of turbid estuaries must not be neglected. All these factors give emphasis that the DGM production process can occur under various environments. Although, UV radiation has been indicated as the principal driver of DGM production in freshwater systems (Garcia et al., 2005; O'Driscoll, Siciliano, Lean, & Amyot, 2006a), tidal conditions have also been identified as an important variable controlling the mercury budget in the water column (Castelle et al., 2009; Poissant et al., 2004). Physical (sediment transport, tidal pumping, wave action, bioturbation), chemical (oxidation-reduction, precipitation-dissolution, adsorption), and microbial processes regulate the cycling of elements, including mercury, in intertidal sediments. This type of sediment constitutes a large area of the Tagus estuary. In these areas, the tide alternately inundates and exposes the sediments to the atmosphere creating nonsteady state conditions between solids and pore waters. The flood water arriving with the tide forces a flow through the upper sediments, which may cause a renewal of pore water (Kerner & Wallmann, 1992),


bringing fresh supplies of dissolved oxygen and changes the concentrations of highly reactive redox sensitive species on short-time scales (Caetano, Falcão, Vale, & Bebianno, 1997). When sediments are covered with tidal water, molecular diffusion is an important but not the only transport mechanism across the sediment-water. Advective processes related with several factors such as: macrofauna (Aller & Yingst, 1985); pressure variation on a permeable bottom sediments generated by gravity waves and bottom currents (Shum & Sundby, 1996); and episodic wind-induced resuspension of sediments (De Jorge & van Beusekom, 1995; Fanning, Carder, & Betzer, 1982) also contribute to the export of solutes from the sediments to the water column. Therefore, advection and other related physical processes play an important role in the transport of solutes from sediments to the water column. However, in intertidal sediments and during the flooding period, advective transport can be considered the most important factor since this induces considerable mass fluxes through the sediment-water interface (Hemond, Nuttle, Burke, & Stolzenbach, 1984; Huettel, Ziebis, Forster, & Luther, 1998). Many studies have shown that in physical-dominated estuaries tidal flushing induces the export of solutes from pore-water to the overlying water (Caetano, Madureira, & Vale, 2007; Santos-Echeandía, Vale, Caetano, Pereira, & Prego, 2010). However, to the best of our knowledge, there is almost no information about the tidal effect on Hg (and MMHg) export and the resultant processes in the water column. In fact, only Bothner, Jahnke, Peterson, and Carpenter (1980) showed a flux of Hg from the sediment to the water column as a consequence of redox processes in the sediment/water interface and later, Guédron et al. (2012) reported that during the tidal flooding both Hg species peaked at the sediment–water interface and a moderate increase of dissolved THg and MMHg was also observed in the water column. These results highlighted the strong influence of tidal flushing on the release of both dissolved THg and MMHg from surface sediments to water column during the tidal flooding. More recently, Zhang, Moffett, WindhamMyers, and Gorelick (2014) reported the effect of tidal flooding on MMHg distribution and sediment-water flux on a tidal saltmarsh in the San Francisco Bay estuary. In spite of these studies the effect of tidal flushing on the formation (and export) of Hg° and its volatilization was never reported. To address this critical knowledge gap, this study reports simultaneously the air–water exchanges of DGM, the Hg volatilization to the atmosphere, and the changes of Hg and MMHg concentrations in both pore waters and overlying waters during 180 min of tidal inundation. This work provides critical science needed to accurately predict mercury movements and impacts in tidal-dominated estuaries and highlights the importance of the advective process in the regional mercury cycle.

281

Teflon dynamic flux chamber was deployed in the sediment and Hg(0) sediment-air flux measurements started. The floating apparatus also contained a small Teflon tube with ¼ inch o.d. that was connected to a peristaltic pump, to pump overlying water when the flooding water arrived at the sampling point (Fig. S2). A multi-parameter Hydrolab probe was also deployed near the sampling point to continuously measure physical and chemical parameters such as temperature, salinity, dissolved oxygen and redox potential. Before the arriving of the tide (time = 0), surface sediments were collected as well as some water present in small tidal ponds. Sediments were stored in aciddecontaminated polyproylene centrifuge tubes while water samples were stored in polypropylene syringes. Mercury flux measurements began when the flooding water was at 1 m distance from the sampling point, and DGM measurements started when the water arrived at this point. These measurements were made continuously from time zero to 180 min in intervals of 10 min (total of 19 discreet measurements). After the arrival of the tide overlying water samples were collected and a small portion of surface sediment was also sampled by hand using the same procedures. This sampling took place at times 0 (immediately after the water reach the sediments), 5, 10, 20, 30, 45, 60, 120 and 180 min after inundation of the chosen sampling site. 2.3. Sample site preparation In a mobile laboratory vehicle, and immediately after sampling, sediment subsamples were centrifuged at 5000 rpm for 15 min at 4 °C to separate solid sediments from pore waters, while the remaining sediment subsamples were stored in decontaminated bags. Centrifuged pore waters and overlying water samples were immediately filtered (0.45 μm HA Millipore) inside an anaerobic glovebox filled with N2. All water samples were stored in Teflon bottles and preserved with HCl (Merck Hg-free) to 0.5% for dissolved MMHg (DMMHg) determinations. Total dissolved mercury samples (DHg) were preserved in 5% HCl (Merck Hg-free), and nutrients determinations preserved with HgCl2 far from the mobile laboratory to avoid Hg contamination of the other samples and after were stored in a special sealed container. Water samples for DOC analyses were preserved with 50% H3PO4 to a pH b 2. All samples were transported to the laboratory at 4 °C, or in a freezer at −8 °C (DOC and nutrients). In the laboratory, sediments and filters were dried at 40 °C while water samples were immediately analysed (maximum 48 h after sampling). After the drying process sediments were powdered and homogenized using an agate mortar. Filters from the overlying waters were acid decontaminated for the analyses of mercury and methylmercury in particulate fraction (PHg and PMMHg).

2. Material and methods 2.4. Analytical methods 2.1. Study site This study was conducted in the Alcochete mudflats (38°45′39 N; 8°56′19 W) of the Tagus Estuary Natural Park (Fig. S1), in May 2012. Alcochete site is an intertidal area located inside a protected area of Tagus estuary, covering salt marsh areas, small river islands and agriculture land. Twice a day this intertidal area is inundated with tidal amplitude ranging from 1.5 to 2.3 m. According to previous works conducted in this site (Canário et al., 2010; Canário, Vale, & Caetano, 2005; Cesário et al., 2016), this area is considered a low to moderate Hg-contaminated area, with concentrations of Hg species ranges between 0.20 and 0.60 μg Hg g−1; 0.20–2.5 ng MMHg g−1 in surface sediments and from 11 to 12 ng Hg L−1; 0.80–1.6 ng MMHg L−1 in overlying waters. 2.2. Experimental setup At the study site one intertidal sampling point was chosen, according to the maximum tidal amplitude. Before the arrival of the tide a floating

2.4.1. Nutrients and dissolved organic carbon (DOC) in overlying and pore water samples Dissolved nutrient concentrations (ammonium-NH+ 4 and silicaSi(OH)4), were determined applying standard colourimetric methods with continuous flow, using a TRAACS 2000 (Bran-Luebbe) autoanalyser (Tréguer & Le Corre, 1974). Sagami CSK Standards were used to control the accuracy of nutrient levels (Ambe, 1978). Detection limit was 0.1 μmol L−1 for both compounds and precision for three replicates was better than 2.0%. Dissolved organic carbon (DOC) analyses were performed by high temperature catalytic oxidation (HTCO), using a Shimadzu TOC-5000A analyser. The system was previously standardized/calibrated in the range of 0–500 μM with potassium hydrogen phthalate in Milli-Q water. The coefficient of variation (CV) for the slope of the seven 5-point calibration curves performed was 1.6%. System blanks were determined before and after analysis by injecting Milli-Q water (Benner & Strom, 1993). The DOC blank was 8 μM. The CV for the 3–5 replicate analyses of each sample was better than 2.0%.

282


2.4.2. Mercury and methylmercury in water samples Dissolved mercury (DHg) were analysed by cold vapor-atomic fluorescence spectroscopy (CV-AFS) according EPA method 1631 (USEPA, 2002), in a Tekran 2600 and dissolved methylmercury (DMMHg) according to the EPA method 1630 (USEPA, 2001), by CV-AFS in a Brooks Rand Model III system. Concentrations were determined applying the standard addition method, using a fresh MMHg-Cl solution prepared daily from CH3HgCl (Aldrich) in ethanol (Canário et al., 2008; Canário, Caetano, & Vale, 2006). Recoveries of MMHg ranged from 92 to 103% (MMHg-Cl spikes). Replicate samples were also used to assess variability of the data and was b 5%. Blanks were repeated every 20 samples to evaluate cross contaminations and to ensure that the equipment was operating within the same conditions. Detection limit (3σ) for both dissolved Hg species was b 0.1 ng L−1 (n = 4). 2.4.3. Mercury and methylmercury in particulate fraction In overlying waters Hg and MMHg analysis of the particulate fraction were made using the membranes of the filtered samples. Particulate mercury (PHg) was determined by acid digestion and wet oxidation using the method proposed by Morrison and Watras (1999). Samples were then reduced with NH2OH-HCl followed by SnCl2 and purged onto dual gold traps for analysis by cold-vapor atomic fluorescence spectroscopy (Bloom & Fitzgerald, 1988). Particulate methylmercury (PMMHg) on filters was extracted by placing each filter in 10 mL of water with 200 mL of 20% KCl and 500 mL of 8 M H2SO4 and then distilling as described by Horvat, Bloom, and Liang (1993). MMHg in the distillate was determined by aqueous-phase ethylation followed by purging of volatile methylethyl mercury onto Tenax® and separation by isothermal gas chromatography with detection by CV-AFS (Bloom, 1989; Liang, Horvat, & Bloom, 1994). To assess the source and variability of Hg and MMHg associated with the method, three types of blank analysis were performed as described by Morrison and Watras (1999). A minimum of three reagents and three filter blanks were performed with every sample group. In pore waters, the determinations of Hg and MMHg in particulate fraction were made in the solid surface sediments. Total mercury in solid sediments (PHg) was determined in dried samples by atomic absorption spectrometry (AAS), using a silicon UV diode detector LECO AMA-254 (Costley et al., 2000) and methylmercury (PMMHg) was determined by alkaline digestion (KOH/MeOH), organic extraction with dichloromethane (DCM) pre-concentration in aqueous sulfide solution, back-extraction into DCM, and quantification by atomic fluorescence spectrometry coupled with gaseous chromatography (GC-AFS) using an Agilent Chromatograph and a Millennium Merlin PSA detector (Canário, Antunes, Lavrado, & Vale, 2004). Certified Reference Materials (CRMs) (BCR 580 and IAEA 405, both Estuarine Sediments from Institute for Reference Materials and Measurements (IRMM) and International Atomic Energy Agency (IAEA), respectively) were also analysed in each batch of samples. Results obtained were in good agreement with the certified values from the CRMs used and the precision expressed as relative standard deviation was better than 4% for all metals investigated (p b 0.05). 2.4.4. Dissolved gaseous mercury (DGM) Continuous analysis of dissolved gaseous mercury (DGM) was performed using the technique described by O'Driscoll, Siciliano, and Lean (2003b). The pumped overlying water was transported to the mobile laboratory using a two-channel peristaltic pump with silicone pump tubing and then to the bottom of a 1 L graduated glass sparger. With a flow rate of 50 mL min−1 the volume of sample analysed was 250 mL every 10 min. This flow rate and sparge time has been shown to be comparable with previous discrete analysis methods (O'Driscoll et al., 2003a, 2003b). A Tekran 1100 zero air generator was used to supply mercury-free air to the glass sparger at a rate of 1.0 L min− 1. DGM was carried from the sparger to the sample inlet of the Tekran 2537A and analysed for mercury content. Each Tekran 2537A used for this

study was calibrated prior to this analysis using the internal mercury permeation calibration source. The internal mercury calibration source was checked for accuracy with a standard air injection of elemental mercury using a Hamilton digital syringe and a Tekran 2505 mercury vapor calibration unit. The detection limit for DGM was 20 fmol L−1, and the relative standard deviation (RSD) of duplicates (n = 36) was 4.0 ± 2.6%. More detailed procedure can be found in Supporting Information. 2.4.5. Mercury flux analysis Water-to-air mercury volatilization was measured using a floating Teflon dynamic flux chamber with a Tekran continuous gaseous mercury equipment as described by Poissant and Casimir (1998) and by Poissant et al. (2004, 1999). Flux measurements were made continuously between time zero and time 180 min. The chamber consists of a hemispheric stainless steel bowl coated with Teflon. The open area of the chamber (A) is 0.125 m2 and its volume is 0.010 m3 (chamber is 20 cm in height). This chamber was previously inter-compared with other types of flux chambers during an international study conducted in Reno, Nevada (Gustin et al., 1999; Poissant, Pilote, & Casimir, 1999). Following Xiao, Munthe, Schroeder, and Lindqvist (1991) the mercury flux (ng m−2 h−1) from the dynamic flux chamber is computed as follows: F Hg ¼

½Hgo −½Hg i Q A

ð1Þ

where: [Hg]o is the outlet air concentration; [Hg]i is the inlet air concentration into the chamber; A is the open surface of the chamber (0.125 m2) and Q is the flow rate into the chamber (0.09 m3 h−1). Before deploying the chamber, blanks were determined in the laboratory over a clean Teflon surface. After several flushing's with mercuryfree air the laboratory blanks reached the instrumental lowest blank level (2% precision of the Tekran instrument) ~ 0.04 ng m2 h−1. After blank determinations, the flux chamber was put together with a clean Teflon plate and sealed in a clean plastic bag. More detailed procedure can be found in Supporting Information. 2.4.6. Estimation of sediment-water exchanges The concentration of mercury and methylmercury between pore waters of upper sediment layers and overlying water were related with export/import to/from the water column. Once intertidal sediments are alternatively covered with water and exposed to the atmosphere, two types of fluxes were considered: diffusive fluxes during the submerged period and advective transport associated with the tidal flooding. 2.4.6.1. Molecular diffusion fluxes. The diffusive fluxes across the sediment-water interface (SWI) were estimated according to Cesário et al. (2016) using Fick's first law of diffusion: J = (− ϕDw/θ2)(∂ C/∂ z), where J is the diffusion flux (ng m−2 d−1) of solute with concentration C (ng dm−3) at depth z (cm), ϕ and θ respectively are the porosity and tortuosity, Dw is the ionic/molecular diffusion coefficient in water and (∂ C/∂ z) is the concentration gradient in pore water. For all cores we used z = 2 cm (i.e. the average depth of the uppermost pore water sample). Tortuosity was estimated from porosity using Boudreau's formulation: θ2 = 1 − ln(ϕ2) (Boudreau, 1996). According to Cesário et al. (2016 and references therein), we used Dw, 25 °C = 1.2 × 10−5 cm2 s−1 for DMMHg and 2 × 10−6 cm2 s−1 for DHg (assuming that inorganic Hg is bound to macromolecules in the colloidal size range). 2.4.6.2. Advective flux calculations. The advective daily flux (T, ng m−2 d−1) of dissolved THg and MMHg was calculated by introducing the temporal variation of Hg (and MMHg) concentration during tidal inundation into the following global mass balance equation: T = ∑ [(Ct + 1 × ht + 1) − (Ct × ht)]/dt (Santos-Echeandía et al., 2010), were


Ct + 1 and Ct are concentrations of dissolved Hg and MMHg in the flooding water at times t + 1 and t; ht + 1 and ht are the water depth at the same times and dt is the time interval between samples. Detailed descriptions and calculations of diffusive fluxes and advective transport are in Supporting Information. 3. Results and discussion 3.1. Nutrients in overlying and pore waters In extensive intertidal areas (like the Tagus Estuary) the tide alternatively inundates and exposes the sediment surface to the atmosphere. These periodic covering and uncovering situations induce a nonsteady-state condition in sediments. The way that water floods the intertidal area is complex and within the first minutes of inundation sediment may be resuspended and seawater mixed with pore water. This mixing induces a change in composition and promotes the export of solutes to the overlying water (Caetano et al., 1997; Falcão & Vale, 1998). This natural process has been studied in some estuaries (e.g. Falcão & Vale, 1998; Santos-Echeandía et al., 2010) and researchers suggested that the changes of some nutrient concentrations in the overlying water during inundation are good indicators of this effect. In our work, changes in the concentrations of NH+ 4 and Si(OH)4 in the overlying and pore water measured over the first 180 min of inundation were observed (Fig. 1). Higher levels of ammonium and silica in the flooding water were recorded 5 min after the inundation (24 μM and 112 μM, respectively) decreasing rapidly to values reaching 10 μM of NH+ 4 and 16 μM of Si(OH)4 by the end of the experiment thus indicating that a tidal flushing mechanism was occurring. Simultaneously, pore water concentrations of those nutrients showed a corresponding decrease, suggesting that the observed increases in overlying water were due to an upward transport mechanism related with the flushing effect. After this period, nutrient concentrations decrease gradually until 45 min of inundation, remaining relatively constant until the end of the

Fig. 1. Ammonium (NH+ 4 ) (a) and Silica (Si(OH)4) (b) concentrations (μM) in overlying and pore waters from intertidal sediments of Alcochete during 180 min of inundation.

283

experiment. This gradual decrease may be related with biological processes occurring in the water column (e.g. uptake of nutrients by phytoplankton) and/or by a dilution effect (Kanda, Ziemann, Conquest, & Bienfang, 1990). Our results were in line with those reported by Caetano et al. (2007) and Falcão and Vale (1998) in intertidal sediments from Ria Formosa (Portugal). Thus, we may conclude that in the beginning of the inundation the mixing of sea water with the sediments seems to change the reversible adsorption-desorption equilibrium of ammonium (Mackin & Aller, 1984) and silica in the sediment and causes a twice daily increase in the supply of both nutrients to the overlying water. Thus, our results clearly suggest that the pulse mechanisms (e.g. molecular diffusion and advective transport) induced by the tidal flooding in this intertidal area affect the equilibrium of NH+ 4 and Si(OH)4 in pore waters of surface sediments over a semi-diurnal tidal cycle. 3.2. Dissolved mercury and methylmercury in overlying and pore waters The tidal flushing effect has been proven to be responsible not only for the transport of nutrients but also for other trace elements. Huettel et al. (1998) reported the effect of tidal flushing on the release of Fe and Mn in a laboratory-controlled experiment and later Caetano et al. (2007) observe this process in Ria Formosa, (Portugal). More recently, Santos-Echeandía et al. (2010) observed a similar effect for Zn, Cu, Pb and Cd in salt-marsh sediments of the Tagus estuary. Studies concerning this tidal flushing effect on mercury biogeochemistry in intertidal sediments are scarce. To the best of our knowledge only Zhang et al. (2014) performed a similar study in a saltmarsh in San Francisco, California, and found that MMHg persists in these sediments over the tidal cycle. Therefore, our work is, in our opinion, a significant contribution to better quantify temporal effects on the Hg cycle in permeable intertidal sediments. Our observations in intertidal sediments pointed to dramatically fast changes in dissolved Hg (DHg) and MMHg (DMMHg) concentrations in pore waters of the sediments occurring in the first 10 min of tidal inundation (Fig. 2). In overlying waters, both DHg and DMMHg concentrations, showed an increase in the first 10 min of inundation, reaching values of 46 ng L−1 and 21 ng L−1, respectively. After, DHg concentrations decreased sharply until 30 min of inundation, reaching concentrations of 25 ng L− 1. For DMMHg, the sharp decrease in concentration occurs until 60 min of inundation (from 21 ng L−1 to 6.0 ng L−1), and then remains constant after this period. There are several possible mechanisms that may explain this result. The surface sediments of intertidal zones are physically mixed by the flooding water, which may lead to a renewal of pore water by percolation (Kerner & Wallmann, 1992), and the physical and chemical tide-driven modifications in the oxic and suboxic layers may take place in the first minutes of flooding. Indeed, the wave driven shear stress applied to surface sediment during tidal flushing could enhance oxygen penetration and change the redox state of surface sediments (Precht & Huettel, 2004; Taillefert, Neuhuber, & Bristow, 2007). The high concentrations of DHg and DMMHg in the overlying water, after 10 min of the inundation, coupled with a decrease of exchangeable Hg and MMHg (10 and 30 min after inundation for DHg and DMMHg, respectively) in the pore waters clearly points to the removal of this species from the sediment, indicating that tidal flooding modifies the chemical equilibrium in pore waters, as was earlier observed for nutrients. During inundation, overlying water containing lower metal concentrations mixes with pore waters of the upper sediment layers. Also the alterations associated with precipitation/dissolution and sorption/desorption during this period should not also be excluded (Caetano et al., 1997; Huettel et al., 1998). The release of dissolved Hg species from the sediment during tidal flushing is significant enough to affect the water column concentrations. Similar observations were reported in other studies, which evidenced THg and MMHg exports from salt marsh sediments to the water column during flooding events (Canário, Caetano, Vale, & Cesário, 2007; Choe et

284


Fig. 2. Total dissolved mercury (DHg) (a) and dissolved methylmercury (DMMHg) (b) concentrations (ng L−1) in overlying waters (ow) and pore waters (pw) from the intertidal sediments of Alcochete during the 180 min of inundation.

al., 2004; Langer, Fitzgerald, Visscher, & Vandal, 2001; Santos-Echeandía et al., 2010). The sharp increase of both dissolved Hg species concentrations measured at the overlying water with the tidal flushing might also be caused by shear stress along sediments (Guédron et al., 2012). This physical phenomenon may increase molecular diffusion or advective transport of dissolved species from surface sediments and/or promote in situ production of MMHg or desorption of both Hg species from sediment particles due to the oxidative dissolution of redox sensitive sulfide minerals (Domènech, De Pablo, & Ayora, 2002; Lewis et al., 2007; Moses & Herman, 1991; Schäfer et al., 2010). It is logical to assume that the differences in Hg and MMHg concentrations between pore waters of surface sediment and overlying water induce a diffusive flux estimated by Fick's first law (Eq. 2-Supporting Information). It was assumed that intertidal sediments were covered by water for 10 h per tide, and therefore approximately 20 h per day. Within the first 60 min of flooding the relationship between DHg in the overlying water and the pore water was not in steady-state equilibrium, the diffusive flux was estimated after this period of inundation until the end of the experiment, being multiplied by a factor of 18, once we considered two tides and subtract one hour per tide by the 20 h submerged per day. For DMMHg, this proxy of a steady-state equilibrium is only achieved after 180 min of inundation, and therefore we multiply the diffusive flux by a factor of 14, considering the same procedure from DHg. Thus, the predicted flux (ng m−2 d−1) of Hg and MMHg daily exchanged between sediment pore water and overlying waters was estimated. Efflux values (from pore water to overlying water) obtained for DHg (J = 17 ng Hg m−2 d−1) and DMMHg (J = 8.0 ng MMHg m−2 d−1) indicated a significant export of both Hg species to the water column. Guédron et al. (2012) reported lower diffusive fluxes in sediments tidally affected from the Venice Lagoon (Italy) (J(Hg) = 9.5 ngm−2 day−1; J(MMHg) = 1.1 ng m−2 day−1). Interestingly, a previous study conducted by Cesário et al. (2016) in the same site of Tagus estuary, reported a lower export of Hg and MMHg in the winter season, being the

diffusive fluxes 10 (for Hg) and 6 (for MMHg) times lower than those presented in this study. However, in the summer season, these sediments behave as sink of both mercury species. These results suggested that in addition to the temperature, also the tidal flooding influenced molecular diffusion promoting the export of Hg and MMHg from sediments to the water column. Although diffusion plays an important role in Hg and MMHg fluxes between sediments and overlying water, the advective transport should be also considered, once during the flooding period this transport induces mass fluxes through the sediment/water interface. Therefore, shear stress might play a key role in the advection of dissolved Hg and MMHg from the pore waters to the overlying water. In fact, the increase of DHg and DMMHg in overlying waters coupled with the decrease of these concentrations in pore waters, during the first minutes of inundation, points to the escape of these mercury species from pore waters. These observations were consistent with findings made in intertidal salt marsh sediments which showed that at tidal flushing pore waters advection pushed reduced species (e.g. ferrous iron, sulphur, manganese) toward the overlying water during tidal inundations (Beck, Dellwig, Schnetger, & Brumsack, 2008; Chow, 2007; Taillefert et al., 2007). Based on these one may calculate the advective transport to the water column using Eq. 3 (Supporting Information). The amount of Hg and MMHg on a daily basis, which corresponds to the sum of two pulses associated with the two tidal flooding periods, was estimated. The obtained advective fluxes (T = 35 × 105 ng Hg m− 2 d− 1 and T = 19 × 105 ng MMHg m− 2 d− 1) clearly points to the fact that advection is the primary transport process responsible for the Hg and MMHg export. These values were up to 10 5 times higher than those predicted by Fick's 1st law. Our results were in line with those reported by Precht and Huettel (2004) who found that wave-driven advection was at least 3 orders of magnitude faster than solute transfer by molecular diffusion. Additionally, SantosEcheandía et al. (2010) reported that advective transport of several metals were up to four orders of magnitude greater than diffusive fluxes, in intertidal sediments from saltmarshes of Tagus estuary. An interesting fact is the difference between Hg and MMHg advective transport (T of Hg is 1.9 higher than T of MMHg), whereas diffusive fluxes presented more than a double of Hg export when compared with MMHg diffusive transport from sediments to overlying water. This may be due to the fact that while molecular diffusivity depends of the solute being DHg more mobile than DMMHg, the advection mechanism is just a physical process that occurs independently of the composition of the solute. Previous works on permeable intertidal sediments have shown the importance of pulse mechanisms associated with the tidal flushing on sediment-water exchanges (Caetano et al., 2007; Falcao & Vale, 1990). Considering the magnitude and the periodicity of advective transport, successive tidal inundation would deplete the pool of Hg and MMHg in the intertidal sediments if not balanced by equivalent inputs. Conservation of mass requires an import of Hg and MMHg during the submerged period by settling of suspended particles, advection of water and deposition of particles during the ebb tide. Additionally, Cesário et al. (2016) reported that Alcochete have some anthropogenic influences from Hg contaminated areas of Tagus estuary, namely North Channel and Barreiro (Cesário et al.-submitted), depending on the hydrodynamic conditions of the estuary. Consequently Hg and MMHg may also be supplied by those sources. The increased concentrations of both dissolved Hg species in overlying waters with tidal flushing may also be due to: (i) a rapid desorption from sediment particles (Hintelmann & Harris, 2004), (ii) in situ microbial MMHg production, or (iii) release from organic matter degradation of both Hg species during redox transition events (Bouchet et al., 2011; Muresan, Cossa, Richard, & Dominique, 2008). Moreover, comparing the concentrations of Hg and MMHg between dissolved (Fig. 2) and particulate fractions (Fig. 3) in both overlying and pore waters, it is clear that there is equilibrium between both fractions in the first 20 min of inundation, evidenced by


285

Fig. 4. Concentrations of dissolved mercury (DHg, ng L−1) and dissolved gaseous mercury (DGM, pg L−1) in overlying water from an intertidal sediment area of Alcochete, during 180 min of inundation.

Fig. 3. Concentrations of particulate mercury (PHg, in μg g−1 in ow and in ng g−1 in pw) (a) and particulate methylmercury (PMMHg, ng g−1) (b) in overlying waters (ow) and in pore waters (pw) from intertidal sediments of Alcochete during 180 min of inundation.

DGM production parallels DHg variation in water column until 60 min of inundation suggesting that DHg might be adsorbed to SPM remobilized with the sediments resuspension occurred during this flooding period, and/or inhibition in the mercury photochemical reduction. In fact, the study reported by Wang et al. (2015) in natural waters from Florida Everglades indicated that 70% of Hg(0) is strongly adsorbed to the particulate fraction. Indeed, in our study, the variation of DGM production with SPM concentrations (Fig. 5a), along the flooding period, showed a negative correlation (r = 0.75, p b 0.05, Fig. S3). This suggests that an increase in SPM concentrations may lead to lower amounts of Hg photochemical reduction or that a large percentage of Hg(0) was adsorbed to SPM. Castelle et al. (2009) also observed a clear increase in DGM concentrations attributed to increasing solar radiation coupled with a decrease in SPM in a turbid estuary system. Although our flooding experiment started approximately at the peak of daily solar radiation (1

strong positive correlations (r N 0.7, p b 0.05 for Hg and r N 0.8, p b 0.05 for MMHg) between both fractions for Hg and MMHg. In contrast to DHg concentrations in overlying waters after this period, PHg concentrations (Fig. 3) increase between 30 and 60 min after flooding. This was attributed to a rapid adsorption onto particles, suggesting that Hg retention is primarily associated with particles during this period. In summary, MMHg mobilization that is tidal driven must be considered as an important variable in methylmercury studies in tidal environments since it may considerably influence MMHg availability to aquatic organisms, particularly those feeding on (suspended) particles. 3.3. DGM production Dissolved gaseous mercury (DGM) formation and mercury volatilization are important mechanisms by which estuarine waters may naturally reduce their mercury burden. To the best of our knowledge this is the first time that the effect of tidal flushing on DGM concentrations in intertidal sediments has been quantified and reported. As for total dissolved mercury (DHg), similar results were observed for DGM concentrations in the water column during the flooding period (Fig. 4). Even though in different times of inundation, DGM concentrations (mean = 19.7 pg L−1, SD = 5.9, n = 19) varied between 13.4 pg L−1 at t = 0 min and 24.6 pg L−1 at the end of the experiment (t = 180 min), with a maximum of 40.0 pg L−1 at 20 min of inundation, after the maximum concentration obtained for DHg. The observed DGM concentrations obtained in this study were in the same range of those reported by Castelle et al. (2009) in Gironde estuary (France) (2–150 pg L−1), by O'Driscoll et al. (2007) in St. Lawrence River in Cornwall (Canada) (0–60.4 pg L−1) and also in a freshwater lake in KejimkujiK Park, Nova Scotia (Canada) (Puzzle Lake) (1.3–110 pg L−1), reported by O'Driscoll et al. (2003a). The sharp increase of DGM after the maximum of DHg suggests the mercury photochemical reduction from Hg(II) (DHg) to Hg(0) (DGM) maximum, after 20 min of inundation. After this period,

Fig. 5. Concentrations of dissolved gaseous mercury (DGM, pg L−1) versus suspended particulate matter concentrations (SPM, mg L−1) (a) and dissolved organic carbon concentrations (DOC, mg L−1) (b) in overlying waters of Alcochete intertidal sediments during 180 min of inundation.

286


to 4 pm), the effect of solar radiation intensity must still also be considered since this parameter is directly related with production of DGM (Amyot, Lean, Poissant, & Doyon, 2000; Amyot et al., 1994; Batrakova, Travnikov, & Rozovskaya, 2014; Castelle et al., 2009). Limitation of DGM production due to depletion in dissolved Hg(II) would be expected in the estuary. This suggests that Hg(0) evasion to the atmosphere (due to increased current velocities and wave action) and/or DGM redistribution in the water column by tidal mixing may efficiently control DGM levels in surface water. Nevertheless, one cannot exclude that DGM produced biotically or abiotically under reducing conditions near the sediment/water interface (Tseng et al., 2001) may also be redistributed in the water column. In fact, these results suggest that in estuarine systems DGM production may vary locally with SPM resuspension/sedimentation cycles induced by tidal currents. However, it is not clear yet what factors control binding and release of Hg(0) from SPM (Wang et al., 2015). Several works have demonstrated that both dissolved organic matter (DOM) structure and concentrations are key variables regulating photoreducible mercury in natural waters (O'Driscoll et al., 2006a, 2003a, 2008; O'Driscoll, Siciliano, Peak, Carignan, & Lean, 2006b; Poulain et al., 2004). In fact, Poulain et al. (2004) reported that Hg reduction is thought to be triggered by DOC through its photoexcitation. Our results showed a decrease of DOC concentrations from the beginning of the experiment (9.7 mg L−1 of DOC at t = 0 min) to the end of the inundation (3.6 mg L− 1, at t = 180 min). Although high DOC levels could increase the concentration of Hg photoreductants at the air-water interface, DGM photoproduction maximum yields occurred between 20 and 30 min of inundation, while the higher DOC values were found in the beginning of the flooding experiment (Fig. 5b). This result suggests that, even if reduction occurred faster in waters with higher DOC values, potentially as a result of high concentrations of photoproduced reductants, the pool of photoreducible Hg was smaller, likely because of Hg binding by DOC. This can be due to a mixture of DOC release from the pore waters of sediments in the beginning of the flooding. However, after 20 min of inundation, the decrease of DGM parallels the DOC decrease until the end of flooding, suggesting that mercury photoreduction increases with DOM concentration due to an overall increase in the associated photoreducible mercury, assuming that DOM structure is held constant during this flooding period. Between 70 and 180 min DGM production slightly increased (from 16.6 pg L−1 to 24.6 pg L−1) when DHg remains constant (Fig. 4), suggesting that (i) Hg reduction increased during this period of inundation or (ii) the sediment may be also a continuous source of Hg° to the water column. Although DGM production was related with DHg concentrations in water column, no correlation (p N 0.05) was observed between both mercury species, indicating that photochemical reduction of DHg from the water column was not the only process responsible for the DGM production. In fact, the hypothesis of DGM formation and export in/from pore waters from the intertidal sediments during tidal flooding, and furthermore the contribution of this DGM production for the water column, should be considered. Another possibility is the release of DGM from SPM that has been suspended in the water column. We also should not exclude that the increase of DGM in the surface water may be due to either abiotic photoreduction reaction or microbial reduction (Siciliano, O'Driscoll, & Lean, 2002). However, the increasing trend in DGM production during this flooding period reinforces the hypothesis that sediment diffusion and/or near sediment microbial reduction may be important in this system. This diffusion may be attributed to direct export of Hg(0) from the sediment or by diffusion of inorganic mercury in the sediment-water interface fallowed by reduction of Hg(II) to Hg(0) by microbial reduction or mixotrophs (O'Driscoll et al., 2008; Poulain et al., 2004). In a study conducted in the St. Lawrence River near Cornwall, Poissant et al. (2007) determined that total gaseous mercury (TGM) is a minor fraction of the gases release from sediment to the water column, in the sediment/water interface. Therefore, it can be assumed that in our flooding experiment the increase of DGM in bottom

of the water column after 70 min of inundation results from the reduction of Hg(II) diffusing from the sediments and/or that, in contrast to the Cornwall experiments, more gaseous mercury is escaping from sediments. Since DGM sampling and measurements were also done during the inundation, the accumulation of DHg in the pore waters of surface sediment suggest that the reduction of Hg(II) to Hg(0) or the release of Hg(0) from the sediment particles were very slow processes. After 70 min of inundation the water level reached almost 1 m above surface sediment and in those conditions Hg photoreduction should not play an important role due to the low penetration of UV radiation (O'Driscoll et al., 2007). Thus, the biotic Hg reduction or abiotic desorption from particles should be the main processes producing DGM in the deepest layers of the water column as reported by Siciliano et al. (2002). 3.4. Mercury flux across the air-water interface The results obtained in the flooding experiment also showed that sediment flushing by tide may also play an important role in Hg volatilization. Fig. 6 presents the water-to-air Hg flux (ng m2 h−1) during the flooding experiment. Gaseous elemental Hg fluxes across the air-water interface increased in the first hour of inundation and were positively correlated (r = 0.77, p b 0.05) with DGM concentrations in surface water (Fig. S4a). During the flooding experiment Hg fluxes from water to atmosphere ranged between 1.12 and 3.28 ng m−2 h−1, with an average of 2.11 ng m−2 h−1 (Fig. 6). These values were in the same range of those reported by O'Driscoll et al. (2007) in the Upper St. Lawrence River, near Cornwall (Ontario, Canada), but lower than those reported for the same river seven years before (Poissant, Amyot, Pilote, & Lean, 2000), and for Gironde estuary, France (Castelle et al., 2009). However the Hg water-to-air fluxes observed were slightly higher than those reported by O'Driscoll et al. (2008) in a frozen freshwater fluvial lake and those reported for the surface waters of the Arctic Ocean (Andersson, Sommar, Gårdfeldt, & Lindqvist, 2008). The same pattern exhibited by Hg flux and DGM concentrations, with a sharp peak in this flooding period, suggested that parameters affecting DGM concentration (e.g. solar radiation, SPM and DOC concentrations, abundance of Hg(II), etc.) may strongly influence Hg fluxes across the air-water interface. Values of DGM saturation ranged between 352% and 1077%, indicating that the surface waters of the Alcochete site were supersaturated during the flooding period, with the maximum between 20 and 30 min of inundation. The above-mentioned strong correlations between Hg flux and DGM concentrations (Fig. S4a) and DGM saturation (Fig. S4b) clearly evidenced that the Hg flux water-to-air is strongly dependent on the DGM production and showed no time lag between both processes. These results indicated a fast release of the produced DGM, suggesting that kinetics of DGM production supercedes advective transport of mercury from water to atmosphere. The same pattern was observed in summer studies in the surface water of freshwater sites, which have

Fig. 6. Dissolved gaseous mercury production (DGM, pg L−1) and water-air mercury volatilization (ng m−2, h−1) in overlying water of Alcochete intertidal sediments during 180 min of inundation.


observed increases in DGM photo-production near water/air interface and subsequent volatilization from the water (Castelle et al., 2009; O'Driscoll et al., 2003a; Poissant et al., 1999). In summary our results indicate that intertidal areas act as a source of DHg and DMMHg to the water column and a potential source of Hg(0) to the atmosphere during the flooding period. Accordingly, these results suggest that Hg sediment/water exchanges in the Tagus Estuary and likely other macro- and meso-tidal ecosystems have been underestimated. This study demonstrates the usefulness of one-site continuous analysis for examining trends in DGM and Hg volatilization in intertidal sediments during the tidal flooding. In addition, the role of bottom turbulence in such shallow ecosystems induced by tidal currents on the exchange of mercury species has been addressed for the first time. Therefore, these findings based on field data an insight into the mechanisms controlling DGM formation and Hg volatilization contribute to improve understanding of the exchange of Hg in the waterair interface in a tidal estuarine environment, and provide a mechanistic framework which can be used for integrating coastal wetlands into global Hg flux models. However, the development of a specific model to investigate Hg transfer velocity between sediment-water, water-atmosphere and sediment-atmosphere interfaces, in sediments influenced by tidal currents (and also by wind velocity) would be very useful to improve our knowledge of the global mercury cycle.

Acknowledgments This work was performed under the projects PROFLUX (PTDC/MAR/ 102748/2008), PLANTA (PTDC/AAC-AMB/115798/2009), and UID/QUI/ 00100/2013, all funded by the Portuguese Foundation for Science and Technology (FCT). Rute Cesário would also like to acknowledge FCT for her PhD grant (SFRH/BD/86441/2012). The authors would also like to acknowledge Marta Nogueira, Pedro Brito and Miguel Caetano for the help in the manuscript. This paper greatly benefit with the comments of three anonymous reviewers. Appendix A. Supplementary data Detailed information on analytical methodologies; description and calculation of diffusive fluxes and advective transport; a figure presented the study area; a figure showing the mercury flux chamber apparatus used in the in situ experiment; description of statistical analysis; a figure showing the correlation between DGM and SPM concentrations in overlying waters and a figure showing the correlations between: (a) DGM concentrations and Hg volatilization and (b) DGM saturation and Hg volatilization. Also two tables with physic-chemical parameters of the overlying and pore waters along the inundation period (180 min) were presented. (PDF). Supplementary data associated with this article can be found in the online version, at doi: http://dx.doi.org/10.1016/j. scitotenv.2017.06.093.

References Aller, R.C., Yingst, J.Y., 1985. Effects of the marine deposit-feeders Heteromastus filiformis (Polychaeta), Macoma balthica (Bivalvia), and Tellina texana (Bivalvia) on averaged sedimentary solute transport, reaction rates, and microbial distributions. J. Mar. Res. 43:615–645. http://dx.doi.org/10.1357/002224085788440349. Ambe, M., 1978. Note of the experience in the preparation of CSK standard solutions and the ICES—SCOR intercalibration experiment, 1969–1970. Mar. Chem. 6:171–178. http://dx.doi.org/10.1016/0304-4203(78)90026-9. Amyot, M., McQueen, D.J., Mierle, G., Lean, D.R.S., 1994. Sunlight-induced formation of dissolved gaseous mercury in lake waters. Environ. Sci. Technol. 28:2366–2371. http:// dx.doi.org/10.1021/es00062a022. Amyot, M., Gill, G.A., Morel, F.M.M., 1997a. Production and loss of dissolved gaseous mercury in coastal seawater. Environ. Sci. Technol. 31:3606–3611. http://dx.doi.org/10. 1021/es9703685. Amyot, M., Mierle, G., Lean, D., Mc Queen, D.J., 1997b. Effect of solar radiation on the formation of dissolved gaseous mercury in temperate lakes. Geochim. Cosmochim. Acta 61:975–987. http://dx.doi.org/10.1016/S0016-7037(96)00390-0.

287

Amyot, M., Lean, D.R.S., Poissant, L., Doyon, M.R., 2000. Distribution and transformation of elemental mercury in the St. Lawrence River and Lake Ontario. Can. J. Fish. Aquat. Sci. 57:155–163. http://dx.doi.org/10.1139/f99-248. Andersson, M.E., Sommar, J., Gårdfeldt, K., Lindqvist, O., 2008. Enhanced concentrations of dissolved gaseous mercury in the surface waters of the Arctic Ocean. Mar. Chem. 110: 190–194. http://dx.doi.org/10.1016/j.marchem.2008.04.002. Balcom, P.H., Schartup, A.T., Mason, R.P., Chen, C.Y., 2015. Sources of water column methylmercury across multiple estuaries in the Northeast U.S. Mar. Chem. 177:721–730. http://dx.doi.org/10.1016/j.marchem.2015.10.012. Batrakova, N., Travnikov, O., Rozovskaya, O., 2014. Chemical and physical transformations of mercury in the ocean: a review. Ocean Sci. 10:1047–1063. http://dx.doi.org/10. 5194/os-10-1047-2014. Beck, M., Dellwig, O., Schnetger, B., Brumsack, H.J., 2008. Cycling of trace metals (Mn, Fe, Mo, U, V, Cr) in deep pore waters of intertidal flat sediments. Geochim. Cosmochim. Acta 72:2822–2840. http://dx.doi.org/10.1016/j.gca.2008.04.013. Benner, R., Strom, M., 1993. A critical evaluation of the analytical blank associated with DOC measurements by high-temperature catalytic oxidation. Mar. Chem. 41: 153–160. http://dx.doi.org/10.1016/0304-4203(93)90113-3. Bergamaschi, B.A., Fleck, J.A., Downing, B.D., Boss, E., Pellerin, B., Ganju, N.K., Schoellhamer, D.H., Byington, A.A., Heim, W.A., Stephenson, M., Fujii, R., 2011. Methyl mercury dynamics in a tidal wetland quantified using in situ optical measurements. Limnol. Oceanogr. 56:1355–1371. http://dx.doi.org/10.4319/lo. 2011.56.4.1355. Beucher, C., Wong-Wah-Chung, P., Richard, C., Mailhot, G., Bolte, M., Cossa, D., 2002. Dissolved gaseous mercury formation under UV irradiation of unamended tropical waters from French Guyana. Sci. Total Environ. 290:131–138. http://dx.doi.org/10. 1016/S0048-9697(01)01078-6. Black, F.J., Paytan, A., Knee, K.L., De Sieyes, N.R., Ganguli, P.M., Gary, E., Flegal, A.R., 2009. Submarine groundwater discharge of total mercury and monomethylmercury to central California coastal waters. Environ. Sci. Technol. 43:5652–5659. http://dx.doi.org/ 10.1021/es900539c. Bloom, N., 1989. Determination of Picogram levels of methylmercury by aqueous phase Ethylation, followed by cryogenic gas chromatography with cold vapour atomic fluorescence detection. Can. J. Fish. Aquat. Sci. 46:1131–1140. http://dx.doi.org/10.1139/ f89-147. Bloom, N., Fitzgerald, W.F., 1988. Determination of volatile mercury species at the picogram level by low-temperature gas chromatography with cold-vapour atomic fluorescence detection. Anal. Chim. Acta 208:151–161. http://dx.doi.org/10.1016/ S0003-2670(00)80743-6. Boening, D.W., 2000. Ecological effects, transport, and fate of mercury: a general review. Chemosphere 40:1335–1351. http://dx.doi.org/10.1016/S0045-6535(99)00283-0. Bothner, M.H., Jahnke, R.A., Peterson, M.L., Carpenter, R., 1980. Rate of mercury loss from contaminated estuarine sediments. Geochim. Cosmochim. Acta 44:273–285. http:// dx.doi.org/10.1016/0016-7037(80)90137-4. Bouchet, S., Bridou, R., Tessier, E., Rodriguez-Gonzalez, P., Monperrus, M., Abril, G., Amouroux, D., 2011. An experimental approach to investigate mercury species transformations under redox oscillations in coastal sediments. Mar. Environ. Res. 71:1–9. http://dx.doi.org/10.1016/j.marenvres.2010.09.001. Boudreau, B.P., 1996. The diffusive tortuosity of fine-grained unlithified sediments. Geochim. Cosmochim. Acta 60:3139–3142. http://dx.doi.org/10.1016/00167037(96)00158-5. Caetano, M., Falcão, M., Vale, C., Bebianno, M.J., 1997. Tidal flushing of ammonium, iron and manganese from inter-tidal sediment pore waters. Mar. Chem. 58:203–211. http://dx.doi.org/10.1016/S0304-4203(97)00035-2. Caetano, M., Madureira, M.J., Vale, C., 2007. Exchange of Cu and Cd across the sedimentwater interface in intertidal mud flats from Ria Formosa (Portugal). Hydrobiologia 587:147–155. http://dx.doi.org/10.1007/s10750-007-0673-y. Canário, J., Vale, C., 2004. Rapid release of mercury from intertidal sediments exposed to solar radiation: a field experiment. Environ. Sci. Technol. 38:3901–3907. http://dx. doi.org/10.1021/es035429f. Canário, J., Antunes, P., Lavrado, J., Vale, C., 2004. Simple method for monomethylmercury determination in estuarine sediments. TrAC Trends Anal. Chem. 23:799–806. http:// dx.doi.org/10.1016/j.trac.2004.08.009. Canário, J., Vale, C., Caetano, M., 2005. Distribution of monomethylmercury and mercury in surface sediments of the Tagus Estuary (Portugal). Mar. Pollut. Bull. 50: 1142–1145. http://dx.doi.org/10.1016/j.marpolbul.2005.06.052. Canário, J., Caetano, M., Vale, C., 2006. Validation and application of an analytical method for monomethylmercury quantification in aquatic plant tissues. Anal. Chim. Acta 580: 258–262. http://dx.doi.org/10.1016/j.aca.2006.07.055. Canário, J., Caetano, M., Vale, C., Cesário, R., 2007. Evidence for elevated production of methylmercury in salt marshes. Environ. Sci. Technol. 41:7376–7382. http://dx.doi. org/10.1021/es071078j. Canário, J., Poissant, L., O'Driscoll, N., Ridal, J., Delongchamp, T., Pilote, M., Constant, P., Blais, J., Lean, D., 2008. Mercury partitioning in surface sediments of the upper St. Lawrence River (Canada): evidence of the importance of the sulphur chemistry. Water Air Soil Pollut. 187:219–231. http://dx.doi.org/10.1007/ s11270-007-9510-1. Canário, J., Vale, C., Poissant, L., Nogueira, M., Pilote, M., Branco, V., 2010. Mercury in sediments and vegetation in a moderately contaminated salt marsh (Tagus Estuary, Portugal). J. Environ. Sci. 22:1151–1157. http://dx.doi.org/10.1016/S1001-0742(09)60231-X. Canário, J., Poissant, L., Pilote, M., Caetano, M., Hintelmann, H., O'Driscoll, N.J., 2017. Saltmarsh plants as potential sources of Hg0 into the atmosphere. Atmos. Environ. 152: 458–464. http://dx.doi.org/10.1016/j.atmosenv.2017.01.011. Castelle, S., Schäfer, J., Blanc, G., Dabrin, A., Lanceleur, L., Masson, M., 2009. Gaseous mercury at the air–water interface of a highly turbid estuary (Gironde Estuary, France). Mar. Chem. 117:42–51. http://dx.doi.org/10.1016/j.marchem.2009.01.005.

288


Celo, V., Lean, D.R.S., Scott, S.L., 2006. Abiotic methylation of mercury in the aquatic environment. Sci. Total Environ. 368:126–137. http://dx.doi.org/10.1016/j.scitotenv.2005. 09.043. Cesário, R., Monteiro, C.E., Nogueira, M., O'Driscoll, N.J., Caetano, M., Hintelmann, H., Mota, A.M., Canário, J., 2016. Mercury and methylmercury dynamics in sediments on a protected area of Tagus Estuary (Portugal). Water Air Soil Pollut. 227:475. http:// dx.doi.org/10.1007/s11270-016-3179-2. Choe, K.-Y., Gill, G.a., Lehman, R.D., Han, S., Heim, W.a., Coale, K.H., 2004. Sediment-water exchange of total mercury and monomethyl mercury in the San Francisco Bay Delta. Limnol. Oceanogr. 49:1512–1527. http://dx.doi.org/10.4319/lo.2004.49.5.1512. Chow, S.S., 2007. Effects of Biogeochemical and Physical Processes on the Transformation of Trace Metals at Oxic-anoxic Interfaces in Aquatic Systems. (PhD thesis). Georgia Institute of Technology. Costley, C.T., Mossop, K.F., Dean, J.R., Garden, L.M., Marshall, J., Carroll, J., 2000. Determination of mercury in environmental and biological samples using pyrolysis atomic absorption spectrometry with gold amalgamation. Anal. Chim. Acta 179–183. De Jorge, V.N., van Beusekom, J.E.E., 1995. Wind- and tide-induced resuspension of sediment and microphytobenthos from tidal flats in the Ems estuary. Limnol. Oceanogr. 40:776–778. http://dx.doi.org/10.4319/lo.1995.40.4.0776. Domènech, C., De Pablo, J., Ayora, C., 2002. Oxidative dissolution of pyritic sludge from the Aznalcóllar mine (SW Spain). Chem. Geol. 190:339–353. http://dx.doi.org/10.1016/ S0009-2541(02)00124-9. Falcao, M., Vale, C., 1990. Study of the ria Formosa ecosystem: benthic nutrient remineralization and tidal variability of nutrients in the water. Hydrobiologia 207: 137–146. http://dx.doi.org/10.1007/BF00041450. Falcão, M., Vale, C., 1998. Sediment–water exchanges of ammonium and phosphate in intertidal and subtidal areas of a mesotidal coastal lagoon (Ria Formosa). Hydrobiologia 373:193–201. http://dx.doi.org/10.1023/A:1017083724636. Fanning, K.A., Carder, K.L., Betzer, P.R., 1982. Sediment resuspension by coastal waters: a potential mechanism for nutrient re-cycling on the ocean's margins. Deep Sea Res. Part A 29:953–965. http://dx.doi.org/10.1016/0198-0149(82)90020-6. Gao, N., Armatas, N.G., Shanley, J.B., Kamman, N.C., Miller, E.K., Keeler, G.J., Scherbatskoy, T., Holsen, T.M., Young, T., McIlroy, L., Drake, S., Olsen, B., Cady, C., 2006. Mass balance assessment for mercury in Lake Champlain. Environ. Sci. Technol. 40:82–89. http:// dx.doi.org/10.1021/es050513b. Garcia, E., Amyot, M., Ariya, P.A., 2005. Relationship between DOC photochemistry and mercury redox transformations in temperate lakes and wetlands. Geochim. Cosmochim. Acta 69:1917–1924. http://dx.doi.org/10.1016/j.gca.2004.10.026. Guédron, S., Huguet, L., Vignati, D.A.L., Liu, B., Gimbert, F., Ferrari, B.J.D., Zonta, R., Dominik, J., 2012. Tidal cycling of mercury and methylmercury between sediments and water column in the Venice Lagoon (Italy). Mar. Chem. 130–131:1–11. http://dx.doi.org/10. 1016/j.marchem.2011.12.003. Gustin, M.S., Lindberg, S., Marsik, F., Casimir, A., Ebinghaus, R., Edwards, G., HubbleFitzgerald, C., Kemp, R., Kock, H., Leonard, T., London, J., Majewski, M., Montecinos, C., Owens, J., Pilote, M., Poissant, L., Rasmussen, P., Schaedlich, F., Schneeberger, D., Schroeder, W., Sommar, J., Turner, R., Vette, A., Wallschlaeger, D., Xiao, Z., Zhang, H., 1999. Nevada STORMS project: measurement of mercury emissions from naturally enriched surfaces. J. Geophys. Res. Atmos. 104:21831–21844. http://dx.doi.org/10. 1029/1999JD900351. Hammerschmidt, C.R., Fitzgerald, W.F., 2004. Geochemical controls on the production and distribution of methylmercury in near-shore marine sediments. Environ. Sci. Technol. 38:1487–1495. http://dx.doi.org/10.1021/es034528q. Hammerschmidt, C.R., Fitzgerald, W.F., Lamborg, C.H., Balcom, P.H., Visscher, P.T., 2004. Biogeochemistry of methylmercury in sediments of Long Island Sound. Mar. Chem. 90:31–52. http://dx.doi.org/10.1016/j.marchem.2004.02.024. Harris, H.H., 2003. The chemical form of mercury in fish. Science 301 (80):1203. http://dx. doi.org/10.1126/science.1085941. Hemond, H.F., Nuttle, W.K., Burke, R.W., Stolzenbach, K.D., 1984. Surface infiltration in salt marshes: theory, measurement, and biogeochemical implications. Water Resour. Res. 20:591–600. http://dx.doi.org/10.1029/WR020i005p00591. Hines, N.A., Brezonik, P.L., Engstrom, D.R., 2004. Sediment and porewater profiles and fluxes of mercury and methylmercury in a small seepage lake in northern Minnesota. Environ. Sci. Technol. 38:6610–6617. http://dx.doi.org/10.1021/es0496672. Hintelmann, H., Harris, R., 2004. Application of multiple stable mercury isotopes to determine the adsorption and desorption dynamics of Hg(II) and MeHg to sediments. Mar. Chem. 90:165–173. http://dx.doi.org/10.1016/j.marchem.2004.03.015. Hollweg, T.A., Gilmour, C.C., Mason, R.P., 2009. Methylmercury production in sediments of Chesapeake Bay and the mid-Atlantic continental margin. Mar. Chem. 114:86–101. http://dx.doi.org/10.1016/j.marchem.2009.04.004. Hollweg, T., Gilmour, C., Mason, R., 2010. Mercury and methylmercury cycling in sediments of the mid-Atlantic continental shelf and slope. Limnol. Oceanogr. 55: 2703–2722. http://dx.doi.org/10.4319/lo.2010.55.6.2703. Horvat, M., Bloom, N.S., Liang, L., 1993. Comparison of distillation with other current isolation methods for the determination of methyl mercury compounds in low level environmental samples. Anal. Chim. Acta 281:135–152. http://dx.doi.org/10.1016/ 0003-2670(93)85348-N. Huettel, M., Ziebis, W., Forster, S., Luther III, G.W., 1998. Advective transport affecting metal and nutrient distributions and interfacial fluxes in permeable sediments. Geochim. Cosmochim. Acta 62:613–631. http://dx.doi.org/10.1016/S00167037(97)00371-2. Kanda, J., Ziemann, D.a., Conquest, L.D., Bienfang, P.K., 1990. Nitrate and ammonium uptake by phytoplankton populations during the spring bloom in Auke Bay, Alaska. Estuar. Coast. Shelf Sci. 30:509–524. http://dx.doi.org/10.1016/0272-7714(90)90070-8. Kerner, M., Wallmann, K., 1992. Remobilization events involving Cd and Zn from intertidal flat sediments in the elbe estuary during the tidal cycle. Estuar. Coast. Shelf Sci. 35: 371–393. http://dx.doi.org/10.1016/S0272-7714(05)80034-4.

Langer, C.S., Fitzgerald, W.F., Visscher, P.T., Vandal, G.M., 2001. Biogeochemical cycling of methylmercury at Barn Island Salt Marsh, Stonington, CT, USA. Wetl. Ecol. Manag. 9: 295–310. http://dx.doi.org/10.1023/A:1011816819369. Lanzillotta, E., Ferrara, R., 2001. Daily trend of dissolved gaseous mercury concentration in coastal seawater of the Mediterranean basin. Chemosphere 45:935–940. http://dx. doi.org/10.1016/S0045-6535(01)00021-2. Lanzillotta, E., Ceccarini, C., Ferrara, R., 2002. Photo-induced formation of dissolved gaseous mercury in coastal and offshore seawater of the Mediterranean basin. Sci. Total Environ. 300:179–187. http://dx.doi.org/10.1016/S0048-9697(02)00223-1. Lee, X., Benoit, G., Hu, X., 2000. Total gaseous mercury concentration and flux over a coastal saltmarsh vegetation in Connecticut, USA. Atmos. Environ. 34:4205–4213. http://dx.doi.org/10.1016/S1352-2310(99)00487-2. Lewis, B.L., Glazer, B.T., Montbriand, P.J., Luther, G.W., Nuzzio, D.B., Deering, T., Ma, S., Theberge, S., 2007. Short-term and interannual variability of redox-sensitive chemical parameters in hypoxic/anoxic bottom waters of the Chesapeake Bay. Mar. Chem. 105: 296–308. http://dx.doi.org/10.1016/j.marchem.2007.03.001. Liang, L., Horvat, M., Bloom, N.S., 1994. An improved speciation method for mercury by GC/CVAFS after aqueous phase ethylation and room temperature precollection. Talanta 41:371–379. http://dx.doi.org/10.1016/0039-9140(94)80141-X. Liu, G., Cai, Y., O'Driscoll, N., 2012. In: Liu, G., Cai, Y., O'Driscoll, N. (Eds.), Environmnental Chemistry and Toxicology of Mercury. John Wiley & Sons, Inc., New Jersey (596 pp.). http://dx.doi.org/10.1002/9781118146644.ch5. Loux, N.T., 2000. Diel temperature effects on the exchange of elemental mercury between the atmosphere and underlying waters. Environ. Toxicol. Chem. 19:1191–1198. http://dx.doi.org/10.1002/etc.5620190453. Mackin, J.E., Aller, R.C., 1984. Ammonium adsorption in marine sediments. Limnol. Oceanogr. 29:250–257. http://dx.doi.org/10.4319/lo.1984.29.2.0250. Mason, R.P., Fitzgerald, W.F., Morel, F.M.M., 1994. The biogeochemical cycling of elemental mercury: anthropogenic influences. Geochim. Cosmochim. Acta 58:3191–3198. http://dx.doi.org/10.1016/0016-7037(94)90046-9. Mason, R.P., Lawson, N.M., Lawrence, A.L., Leaner, J.J., Lee, J.G., Sheu, G.R., 1999. Mercury in the Chesapeake Bay. Mar. Chem. 65:77–96. http://dx.doi.org/10.1016/S03044203(99)00012-2. Mason, R.P., Lawson, N.M., Sheu, G.-R., 2001. Mercury in the Atlantic Ocean: Factors controlling air–sea exchange of mercury and its distribution in the upper waters. DeepSea Res. II Top. Stud. Oceanogr. 48, 2829–2853. doi:http://dx.doi.org/10.1016/S09670645(01)00020-0. Mitchell, C.P.J., Jordan, T.E., Heyes, A., Gilmour, C.C., 2012. Tidal exchange of total mercury and methylmercury between a salt marsh and a Chesapeake Bay sub-estuary. Biogeochemistry 111:583–600. http://dx.doi.org/10.1007/s10533-011-9691-y. Monperrus, M., Tessier, E., Amouroux, D., Leynaert, A., Huonnic, P., Donard, O.F.X., 2007. Mercury methylation, demethylation and reduction rates in coastal and marine surface waters of the Mediterranean Sea. Mar. Chem. 107:49–63. http://dx.doi.org/10. 1016/j.marchem.2007.01.018. Moore, C., Carpi, A., 2005. Mechanisms of the emission of mercury from soil: role of UV radiation. J. Geophys. Res. Atmos. 110:1–9. http://dx.doi.org/10.1029/2004JD005567. Morrison, K.A., Watras, C.J., 1999. Mercury and methyl mercury in freshwater seston: direct determination at picogram per litre levels by dual filtration. Can. J. Fish. Aquat. Sci. 56:760–766. http://dx.doi.org/10.1139/f99-029. Moses, C.O., Herman, J.S., 1991. Pyrite oxidation at circumneutral pH. Geochim. Cosmochim. Acta 55:471–482. http://dx.doi.org/10.1016/0016-7037(91)90005-P. Muresan, B., Cossa, D., Richard, S., Dominique, Y., 2008. Monomethylmercury sources in a tropical artificial reservoir. Appl. Geochem. 23:1101–1126. http://dx.doi.org/10.1016/ j.apgeochem.2007.11.006. O'Driscoll, N.J., Beauchamp, S., Siciliano, S.D., Rencz, A.N., Lean, D.R.S., 2003a. Continuous analysis of dissolved gaseous mercury (DGM) and mercury flux in two freshwater lakes in Kejimkujik Park, Nova Scotia: evaluating mercury flux models with quantitative data. Environ. Sci. Technol. 37:2226–2235. http://dx.doi.org/10.1021/es025944y. O'Driscoll, N.J., Siciliano, S.D., Lean, D.R.S., 2003b. Continuous analysis of dissolved gaseous mercury in freshwater lakes. Sci. Total Environ. 304:285–294. http://dx.doi.org/10. 1016/S0048-9697(02)00575-2. O'Driscoll, N.J., Lean, D.R.S., Loseto, L.L., Carignan, R., Siciliano, S.D., 2004. Effect of dissolved organic carbon on the Photoproduction of dissolved gaseous mercury in lakes: potential impacts of forestry. Environ. Sci. Technol. 38:2664–2672. http://dx. doi.org/10.1021/es034702a. O'Driscoll, N.J., Siciliano, S.D., Lean, D.R.S., Amyot, M., 2006a. Gross photoreduction kinetics of mercury in temperate freshwater lakes and rivers: application to a general model of DGM dynamics. Environ. Sci. Technol. 40:837–843. http://dx.doi.org/10. 1021/es051062y. O'Driscoll, N.J., Siciliano, S.D., Peak, D., Carignan, R., Lean, D.R.S., 2006b. The influence of forestry activity on the structure of dissolved organic matter in lakes: Implications for mercury photoreactions. Sci. Total Environ. 366:880–893. http://dx.doi.org/10. 1016/j.scitotenv.2005.09.067. O'Driscoll, N.J., Poissant, L., Canário, J., Ridal, J., Lean, D.R.S., 2007. Continuous analysis of dissolved gaseous mercury and mercury volatilization in the upper St. Lawrence River: exploring temporal relationships and UV attenuation. Environ. Sci. Technol. 41:5342–5348. http://dx.doi.org/10.1021/es070147r. O'Driscoll, N.J., Poissant, L., Canário, J., Lean, D.R.S., 2008. Dissolved gaseous mercury concentrations and mercury volatilization in a frozen freshwater fluvial lake. Environ. Sci. Technol. 42:5125–5130. http://dx.doi.org/10.1021/es800216q. Peretyazhko, T., Charlet, L., Muresan, B., Kazimirov, V., Cossa, D., 2006. Formation of dissolved gaseous mercury in a tropical lake (Petit-Saut reservoir, French Guiana). Sci. Total Environ. 364:260–271. http://dx.doi.org/10.1016/j.scitotenv.2005.06.016. Poissant, L., Casimir, A., 1998. Water-air and soil-air exchange rate of total gaseous mercury measured at background sites. Atmos. Environ. 32:883–893. http://dx.doi.org/ 10.1016/S1352-2310(97)00132-5.

R. Cesário et al. / Science of the Total Environment 603–604 (2017) 279–289 Poissant, L., Pilote, M., Casimir, A., 1999. Mercury flux measurements in a naturally enriched area: correlation with environmental conditions during the Nevada Study and Tests of the Release of Mercury from Soils (STORMS). J. Geophys. Res. 104: 21845. http://dx.doi.org/10.1029/1999JD900092. Poissant, L., Amyot, M., Pilote, M., Lean, D., 2000. Mercury water−air exchange over the upper St. Lawrence River and Lake Ontario. Environ. Sci. Technol. 34:3069–3078. http://dx.doi.org/10.1021/es990719a. Poissant, L., Pilote, M., Constant, P., Beauvais, C., Zhang, H.H., Xu, X., 2004. Mercury gas exchanges over selected bare soil and flooded sites in the bay St. François wetlands (Québec, Canada). Atmos. Environ. 38:4205–4214. http://dx.doi.org/10.1016/j. atmosenv.2004.03.068. Poissant, L., Constant, P., Pilote, M., Canário, J., O'Driscoll, N., Ridal, J., Lean, D., 2007. The ebullition of hydrogen, carbon monoxide, methane, carbon dioxide and total gaseous mercury from the Cornwall area of concern. Sci. Total Environ. 381:256–262. http:// dx.doi.org/10.1016/j.scitotenv.2007.03.029. Poulain, A.J., Amyot, M., Findlay, D., Telor, S., Barkay, T., Hintelmann, H., 2004. Biological and photochemical production of dissolved gaseous mercury in a boreal lake. Limnol. Oceanogr. 49:2265–2275. http://dx.doi.org/10.4319/lo.2004.49.6.2265. Precht, E., Huettel, M., 2004. Rapid wave-driven advective pore water exchange in a permeable coastal sediment. J. Sea Res. 51:93–107. http://dx.doi.org/10.1016/j.seares. 2003.07.003. Qureshi, A., O'Driscoll, N.J., Macleod, M., Neuhold, Y.M., Hungerbühler, K., 2010. Photoreactions of mercury in surface ocean water: gross reaction kinetics and possible pathways. Environ. Sci. Technol. 44:644–649. http://dx.doi.org/10.1021/es9012728. Riedel, G.F., Williams, S.A., Riedel, G.S., Gilmour, C.C., Sanders, J.G., 2000. Temporal and spatial patterns of trace elements in the Patuxent River: a whole watershed approach. Estuaries 23:521–535. http://dx.doi.org/10.1007/BF02694950. Rolfhus, K.R., Fitzgerald, W.F., 2001. The evasion and spatial/temporal distribution of mercury species in Long Island Sound, CT-NY. Geochim. Cosmochim. Acta 65, 407–418. Rolfhus, K.R., Fitzgerald, W.F., 2004. Mechanisms and temporal variability of dissolved gaseous mercury production in coastal seawater. Mar. Chem. 90:125–136. http:// dx.doi.org/10.1016/j.marchem.2004.03.012. Santos-Echeandía, J., Vale, C., Caetano, M., Pereira, P., Prego, R., 2010. Effect of tidal flooding on metal distribution in pore waters of marsh sediments and its transport to water column (Tagus estuary, Portugal). Mar. Environ. Res. 70:358–367. http:// dx.doi.org/10.1016/j.marenvres.2010.07.003. Schäfer, J., Castelle, S., Blanc, G., Dabrin, A., Masson, M., Lanceleur, L., Bossy, C., 2010. Mercury methylation in the sediments of a macrotidal estuary (Gironde Estuary, southwest France). Estuar. Coast. Shelf Sci. 90:80–92. http://dx.doi.org/10.1016/j.ecss. 2010.07.007.

289

Shum, K.T., Sundby, B., 1996. Organic matter processing in continental shelf sediments—the subtidal pump revisited. Mar. Chem. 53:81–87. http://dx.doi.org/10. 1016/0304-4203(96)00014-X. Siciliano, S.D., O'Driscoll, N.J., Lean, D.R.S., 2002. Microbial reduction and oxidation of mercury in freshwater lakes. Environ. Sci. Technol. 36:3064–3068. http://dx.doi.org/10. 1021/es010774v. Sizmur, T., McArthur, G., Risk, D., Tordon, R., O'Driscoll, N.J., 2017. Gaseous mercury flux from salt marshes is mediated by solar radiation and temperature. Atmos. Environ. 153:117–125. http://dx.doi.org/10.1016/j.atmosenv.2017.01.024. Taillefert, M., Neuhuber, S., Bristow, G., 2007. The effect of tidal forcing on biogeochemical processes in intertidal salt marsh sediments. Geochem. Trans. 8:6. http://dx.doi.org/ 10.1186/1467-4866-8-6. Tréguer, P., Le Corre, P., 1974. Manuel d'analyse des sels nutritifs dans l'eau de mer (utilisation de l'autoanalyzer II Technicon R), 2nd ed. Tseng, C.M., Amouroux, D., Abril, G., Tessier, E., Etcheber, H., Donard, O.F.X., 2001. Speciation of mercury in a fluid mud profile of a highly turbid macrotidal estuary (Gironde, France). Environ. Sci. Technol. 35:2627–2633. http://dx.doi.org/10.1021/es001750b. Tseng, C.M., Lamborg, C., Fitzgerald, W.F., Engstrom, D.R., 2004. Cycling of dissolved elemental mercury in Arctic Alaskan lakes. Geochim. Cosmochim. Acta 68:1173–1184. http://dx.doi.org/10.1016/j.gca.2003.07.023. Ullrich, S.M., Tanton, T.W., Abdrashitova, S.a., 2001. Mercury in the aquatic environment: a review of factors affecting methylation. Crit. Rev. Environ. Sci. Technol. 31:241–293. http://dx.doi.org/10.1080/20016491089226. USEPA, 2001. Method 1630: methyl mercury in water by distillation, aqueous ethylation, purge and trap, and cold vapor atomic fluorescence spectrometry (CV-AFS). United States, EPA-821-R-01-020, January, 1–49. USEPA, 2002. Method 1631, Revision E: Mercury in water by oxidation, purge and trap, and cold vapor atomic fluorescence spectrometry (CV-AFS). Unitet States, EPA-821R-02-019, August, 1–46. Wang, Y., Li, Y., Liu, G., Wang, D., Jiang, G., Cai, Y., 2015. Elemental mercury in natural waters: occurrence and determination of particulate hg(0). Environ. Sci. Technol. 49: 9742–9749. http://dx.doi.org/10.1021/acs.est.5b01940. Xiao, Z.F., Munthe, J., Schroeder, W.H., Lindqvist, O., 1991. Vertical fluxes of volatile mercury over forest soil and lake surfaces in Sweden. Tellus 43, 267–279. Zhang, H., Lindberg, S.E., 2001. Sunlight and iron(III)-induced photochemical production of dissolved gaseous mercury in freshwater. Environ. Sci. Technol. 35:928–935. http://dx.doi.org/10.1021/es001521p. Zhang, H., Moffett, K.B., Windham-Myers, L., Gorelick, S.M., 2014. Hydrological controls on methylmercury distribution and flux in a tidal marsh. Environ. Sci. Technol. 48: 6795–6804. http://dx.doi.org/10.1021/es500781g.