S1 Measurement and understanding of atmospheric mercury processes

Thursday, 28 July, 2011

RS1-P2 — 11:00-12:00 and 17:30-18:30
Authors: KONOPLEV, Alexei1, PANKRATOV, Fidel1, REIERSEN, Lars-Otto2, STEFFEN, Alexandra3, WILSON, Simon2
(1) Chemical Environmental Centre, konoplev@obninsk.com; (2) AMAP Secretariate, Oslo, Norway; (3) Environment Canada, Toronto, Ontario, Canada;

Long-term continuous monitoring of gaseous elemental mercury (GEM) in the surface air at the polar station Amderma (69,72?N; 61,62oE) using the analyzer Tekran 2537A has been conducted from June 2001 to date. Individual measurements were collected every thirty minutes. It has been shown that during ten years of observations the so-called mercury depletion event (MDE) was observed every year from the end of March to early June. The least variability in the mercury concentration was seen from September to December each year. In September 2009 the mercury concentration was 1.32±0.09 ng/m3. During the monitoring period the mean annual concentration of elemental mercury vapor was decreasing from 1.68±0.29 ng/m3 in 2001 to 1.31±0.31 ng/m3 in 2009. Most likely, this is the reflection of the reduction in European anthropogenic emissions of mercury as a result of adopted bans and restrictions.

Behavior of GEM in the air of Amderma in 2010 was very specific which, in our view, was caused by impact of the Eyjafjallajokull volcano eruption. Prior to April 2010 the mercury concentrations in the air at Amderma were in general slightly lower than in the previous years. This may be a reflection of the decrease in the European mercury levels in the atmosphere due to the undertaken measures to restrict mercury emissions. The eruption of the Eyjafjallajokull volcano began on 20 March 2010 and the massive release of ash and water vapor due to evaporation of snow and glaciers began on 14 April. The mercury concentration in Amderma increased from 13 April to end of May 2010, as compared to the mean many-year values. It may be worth noting that the Eyjafjallajokull eruption coincided with the phenomenon of depletion of atmospheric mercury during the polar sunrise in the spring of 2010.

The analysis of the available calculated maps for the paths of air masses from Iceland suggests that in the middle and second half of April Amderma was in the area affected by the volcano unlike other global stations of air mercury monitoring (Alert, Canada and Ny Alesund, Svalbard, Norway). In this context, the observation of the increased levels of mercury in the ambient air in the vicinity of Amderma appears to be unique.

In summary, the long-term monitoring of GEM in the ambient air in Amderma has made it possible to detect the impact of the Eyjafjallajokull volcano eruption on the atmospheric mercury pollution levels on a global scale, first of all, in the Arctic.

Financial support for the monitoring program was provided by Environment Canada, AMAP Secretariate and Russian Federal Service for Hydrometeorology and Environmental Monitoring.

RS1-P3 — 11:00-12:00 and 17:30-18:30
Authors: WEISS-PENZIAS, Peter1, GUSTIN, Mae S.2, LYMAN, Seth N.3
(1)University California, Santa Cruz, pweiss@ucsc.edu; (2) University of Nevada, Reno; (3) University of Washington, Bothell.

Wet deposition measurements have shown that relative to other sites in the US, the southeastern region has the highest mercury (Hg) inputs. The source of this Hg has been investigated by multiple researchers and is suggested to be derived from local, regional and global sources. Here we focus on trying to understand potential sources of Hg to this area during periods dominated by dry gaseous oxidized mercury (GOM) deposition. Dry deposition of Hg to a surrogate surface was measured in conjunction with speciated atmospheric Hg, and ancillary parameters from September 2007 through September 2008 at two sites located within 25 km of coal-fired power plants (CFPPs): Yorkville, GA, and Pensacola, FL. Mean weekly Hg dry deposition, daily gaseous oxidized mercury (GOM) and daily sulfur dioxide (SO2) concentrations were significantly (P < 0.01) higher at the Yorkville site by factors of 1.5, 2.0, and 1.8, respectively, likely due to the large size of a nearby CFPP. At both sites, GOM and SO2 concentrations were significantly correlated (P < 0.05) in every season on hourly and daily timescales. Wind rose diagrams showed enhanced GOM and SO2 concentrations occurred when air moved to the sites from the direction of the local CFPPs. Most periods of enhanced GOM concentrations ([GOM] > 98th percentile), were also associated with NOy/SO2 ratios that were within 25% of that reported for the local CFPPs (N = 27 of 33 at Yorkville, N = 18 of 26 at Pensacola). During these events, mean GOM/SO2 enhancement ratios were 3.1 ± 1.4 and 3.2 ± 1.4 pg m-3 ppb-1 for Yorkville and Pensacola, respectively (range = 0.5 - 5.5). The remaining events had higher GOM/SO2 ratios due to 3-4 times lower SO2 concentrations. These events occurred primarily in the spring, coinciding with a period of enhanced Hg dry deposition at both sites. The potential sources of GOM during the latter periods were investigated using gridded frequency distributions of 72-h atmospheric back trajectories for both sites. During these periods air transport was from the northnorthwest, with a component from above the mixed boundary layer, and with little precipitation along the trajectory paths. Based on the GOM/SO2 ratios, approximately 30% and 67% of the Hg dry deposited during these events was derived from sources outside of the region.

RS1-P5 — 11:00-12:00 and 17:30-18:30
(1)Instituto de Geología Aplicada, Universidad de Castilla-La Mancha, alba.martinez@ucl,.es

The Almadén mercury district can be regarded as the largest geochemical anomaly of mercury on Earth. The ore deposits have been mined for more than 2000 years, and the main mine of the district (Almadén), has been active from Roman times to present day. Las Cuevas is the fourth largest Hg deposit in the district; it was exploited by the Romans and rediscovered in 1982, its exploitation by underground mining was completed in September 1997. Actually is a mercury handling and storage warehouse.

Mercury was determined by Atomic Fluorescence Spectrophotometry using a Tekran device equipped with speciation modules for determining RGM and PM, with detection limit at 1 pg m-3. Local meteorological parameters were monitored using a station Davis Vantage Pro, and the ozone concentrations were measured with a ML®9810B device, by absorption photometry.

We also present data corresponding to one year of measurements of GEM and RGM from a regional background area: the town of Puertollano, located 80 km away from Almadén and with no atmospheric mercury sources.

Measurements reveal the existence of high and extremely high atmospheric mercury values in this area; as well, the ozone measurements put forward an anomaly with values up to 0.5 mg m-3. We compared the values of the different species of mercury with ozone and the most important meteorological parameters (wind speed, wind direction, rainfall, UV, temperature, solar radiation, atmospheric pressure, humidity, soil temperature and moisture), taking also into account the different seasons and working schedule at the mercury handling warehouse. In average, mercury contents for seasons were: spring (GEM: 280 ng m-3, RGM: 0.348 ng m-3, PM: 0.401 ng m-3), summer (GEM: 223 ng m-3, RGM: 0.08 ng m-3, PM: 0.426 ng m-3), autumn (GEM: 2842 ng m-3, RGM: 0.084 ng m-3, PM: 0.422 ng m-3) and winter (GEM: 127 ng m-3, RGM: 0.033 ng m-3, PM: 0.228 ng m-3), meanwhile in the background area (Puertollano) the average mercury concentrations for the complete year were 1.5 ng m-3 (GEM) and 0.005 ng m-3 (RGM).

RS1-P6 — 11:00-12:00 and 17:30-18:30
Author: PYTA, Halina1
(1) Institute of Environmental Engineering of the Polish Academy of Sciences, pyta@ipis.zabrze.pl

In Poland there are four regional background stations that monitor concentrations of total gaseous mercury (TGM). At one of them the measurements are taken periodically with manual sampling. At three remaining stations they are conducted continuously by applying the automatic method. The study presents the results of TGM monitoring recorded at the last three stations. These stations work within the air monitoring network of Inspection for Environmental Protection. One of the stations is situated in a village Granica in east-central Poland, 40 km to the west of Warsaw. It lies within Kampinos Forest. The second station is located in a village Zielonka in northern Poland, over 90 km to the south-west of Gda?sk, in the area of Tuchola Forest. The third station is situated near a village Zloty Potok within Eagles’ Nests Landscape Park (Czestochowa County) in southern Poland. It lies 40 km to the north of Upper Silesia Agglomeration, the most industrialized region of Poland.

TGM was measured between January and December 2010 using a Tekran 2537A. The hourly average concentration of TGM (as well as minimum and maximum) was 1.53 (0.8-3.9) ngm-3 Granica, 1.69 (0.7-8.4) ngm-3 Zielonka and 2.00 (0.7-18.2) ngm-3 Zloty Potok. An average concentrations at the station in Granica and Zielonka were within the range of the global TGM background concentrations (1.5-1.8 ngm-3). Higher concentrations in Zloty Potok were caused by the advection of mercury from industrial sources of Czestochowa and Upper Silesia. All higher concentrations (>8 ngm-3) were observed in Zloty Potok between 1:00 and 7:00 CET. Seasonal distribution of TGM concentrations was smooth, typical for background conditions, particularly in Granica. In winter the concentrations were the highest, in summer they were the lowest (except Zielonka – the lowest concentrations were noticed in autumn). In spring and autumn the distribution of TGM concentrations was similar.

RS1-P7 — 11:00-12:00 and 17:30-18:30
Authors: PYTA, Halina1, PAWLOWSKI, Marek2
(1) Institute of Environmental Engineering of the Polish Academy of Sciences, pyta@ipis.zabrze.pl; (2) TSI Sp. z o.o..

Mercury exists in the air mainly as a gaseous elemental mercury (Hg0) with minor amounts of divalent reactive gaseous (Hg2+) and particle-bound mercury (Hgp). The different physicochemical properties of atmospheric mercury define its tate in the environment. Measurements of atmospheric mercury species have been made in different parts of the world, but significant gaps in description of mercury transport and chemical conversion still exist, particularly in the complex urban environment.

The study presents the initial results of the measurements of atmospheric mercury speciation (Hg0, Hg2+, Hgp bound with PM2.5) in Zabrze obtained in winter season 2010/2011. Zabrze is a city (ca. 190,000 inhabitants) in Upper Silesia Agglomeration, the most industrialized part of southern Poland. The air was sampled from the roof of the two-storey building of Institute of Environmental Engineering of the Polish Academy of Sciences. The measurements were performed using Tekran Model 2537A together with a Model 1130 Speciation Unit and a Model 1135 Hgp Unit. Hg2+ is captured in the Model 1130 (KCl-coated denuder), while Hgp is trapped onto a regenerable filter inside the Model 1135. Hg0 remaining in the air stream is then directed into the Model 2537A analyzer and detected using Cold Vapour Atomic Fluorescence Spectrometry. After sampling, 1130 and 1135 Units are flushed with zero air and Hg2+ and Hgp are sequentially thermodesorbed and analyzed. The concentrations of Hg0 (5-minutes data) were from 1 to 22 ng m-3, with 3,2 ng m-3 as a mean value. Hg2+ concentrations (1-hour) varied from 0,7 to 700 pg m-3, a mean was 31 pg m-3. The concentrations of Hgp (1-hour) ranged from about 6 pg m-3 to 1,5 ng m-3, on average 90 pg m-3. While the concentrations of Hg0 did not diverge generally from the levels recorded in urban and industrial areas in Western Europe and Northern America, the concentrations of two remaining Hg species, that can be considered as results of hard coal combustion and coking processes, were distinctly higher.

RS1-P8 — 11:00-12:00 and 17:30-18:30
Authors: PETERSON, Christianna1, GUSTIN, Mae1, WEISS, Peter2
(1) University of Nevada, Reno, cpeterson@cabnr.unr.edu; (2) University of California, Santa Cruz.

Wet deposition rates reported by the Mercury Deposition Network in Florida are some of the highest in the United States. Little however is known about the magnitude of dry deposition to the region. To address this uncertainty, dry deposition of gaseous oxidized mercury (GOM) was estimated for three Florida TMDL Supersites using surrogate surfaces and a dry deposition model. The latter applied air GOM concentrations measured on-site using a Tekran air speciation system. Passive samplers for monitoring GOM concentrations were also deployed. The TMDL sites were located near industrial and urban sources, with coal fired power plants located near the Pensacola and Tampa sites, while the Davie location was in a large urban area. From July 2009-2010 surrogate and passive samplers were deployed for two week intervals. Air Hg speciation data, trace gas concentrations and meteorological parameters were collected by the SEARCH and TMDL networks. Samplers located in Pensacola had the lowest annual Hg dry deposition rate and GOM uptake (0.2 ng m-2 hr-1 and 0.4 pg h-1; respectively), while the highest rates (0.4 ng m-2 hr-1and 0.6 pg h-1; respectively) were observed at the Davie site. Sampler Hg concentrations were highest across all sites during the spring and lowest in the fall and winter. Tekran derived GOM concentrations followed the same seasonal and spatial patterns, with Pensacola having the lowest mean concentration in the winter and Davie the highest mean in the summer (2 pg m-3 and 7 pg m-3; respectively), while the highest hourly GOM concentration (157 pg m-3) was recorded at Tampa. Hg dry deposition to surrogate surfaces was significantly correlated to mean GOM concentrations measured over the same time period at Pensacola and Tampa, while GOM uptake by the passive samplers was significantly correlated to mean GOM at Pensacola and Davie (p<0.05). Modeled dry deposition followed the same spatial and seasonal trends, and was significantly correlated to dry deposition to the surrogate surfaces and passive samplers (p<0.05). Based on deposition to the surrogate surfaces and modeled values, dry deposition in Florida accounts for 3 to 15% of total annual Hg deposition (wet + dry). Variations between the samplers and real-time Hg concentrations indicate that the samplers are a good method for understanding seasonal and spatial variations in Hg dry deposition, even at low concentration sites. Further use of these methods will allow for greater understanding of Hg cycling across large areas.

RS1-P9 — 11:00-12:00 and 17:30-18:30
Authors: KAWAKAMI, Tomonori1, ISEZAKI, Yukihiro2, KINOSHITA, Hazumu2, HASHIMOTO, Naoki 3, KINOSHITA, Kuriko4
(1) Toyama Prefectural University, kawakami@pu-toyama.ac.jp; (2) Graduate School of siga Prefecture; (3) Graduate Shool of Shiga Prefecture; (4) Toyohashi University of Technology.

UNEP initiated Global Mercury Partnership to protect human health and global environment from the release of mercury and long-range transport issue, especially from anthropogenic mercury such as coal combustion.

To elucidate the vertical distribution of mercury concentration the different altitudes of Mt. Fuji, we have developed a passive sampler to measure the gaseous mercury concentration in the atmosphere.

A passive sampler does not require electric power supply and is suitable for multi points sampling. The body was made of fluorocarbon to prevent mercury from deposition on it. As an adsorbent of mercury in the sampler, a quarts fiber filter coated with gold was prepared in order to make it possible to measure the mercury by the heating atomic absorption spectrophotometry method.

The developed passive sampler was applied to measure the vertical distribution of gaseous mercury concentration in the atmosphere of Mt. Fuji. The sampling was carried out on July and August, 2010.

The passive sampler for the atmospheric mercury was successfully developed to measure the vertical distribution of mercury concentration each altitude of Mt. Fuji.

This research was partially supported by Mitui&Co.,Ltd. Environment Fund, the Environment Research and Technology Development Fund(B-1008)of the Ministry of the Environment, Japan, the Watanabe Memorial Foundation for the Advancement of Technology, and the financial support of Japan Post Service Co.,Ltd. In 2009. This work was performed during the period in which the NPO Valid Utilization of Mt. Fuji Weather Station maintained the facilities.

RS1-P10 — 11:00-12:00 and 17:30-18:30
Authors: DURNFORD, Dorothy A1, DASTOOR, Ashu P2, STEEN, Anne O3, BERG, Torunn3, RYJKOV, Andrei1, FIGUERAS-NIETO, Daniel2, HOLE, Lars R.4, PFAFFHUBER, Katrine A5, HUNG, Hayley2
(1) independent researcher, dorothy.durnford@ec.gc.ca; (2) Environment Canada; (3) Norwegian University of Science and Technology; (4) Norwegian Meteorological Institute; (5) Norwegian Institute for Air Research;

Mercury deposited onto snow- and ice-covered surfaces is only partially revolatilised. In order to understand the fate of cryospheric mercury, we reviewed the physical and chemical processes that govern its behaviour. These processes are applicable at all locations. Regional variations in the behaviour of cryospheric mercury are caused by differing environmental conditions. This discussion is based on an examination of field observations, laboratory data, modelling results, and theoretical considerations.

We further performed a statistical analysis of 5 types of cryospheric mercury field observations and 20 model environmental variables. This analysis reveals that the presence of cryospheric halogen species has a significant impact on the fate of cryospheric mercury. Physically, we find that the fractional loss of mercury from surface snow within a 24-hour period is governed by mercury deposition through both dry and wet processes, as well as surface-level atmospheric wind speed and stability. In contrast, the concentration of mercury in long-term cryospheric records is affected by latitude, ventilation and the burial of mercury by fresh snowfalls. Considering all five types of cryospheric mercury observations analysed, the dry and wet depositions of oxidized mercury were found to have the strongest impact on cryospheric mercury observations. The burial of mercury by fresh snowfalls and the wind driven ventilation of snowpacks are also important processes.

RS1-P11 — 11:00-12:00 and 17:30-18:30
Author: WETHERBEE, Gregory1
(1) U.S. Geological Survey, Branch of Quality Systems, wetherbe@usgs.gov

A U.S. Geological Survey (USGS) inter-laboratory comparison program provides independent quality assurance of total-mercury (Hg) concentrations in wet-deposition samples for the National Atmospheric Deposition Program (NADP) / Mercury Deposition Network (MDN). The program quantifies Hg-concentration analysis variability and bias among participating laboratories that provide data for global Hg fate and transport modeling. Participating laboratories include: Frontier Global Sciences1 (FGS), USA; official Mercury Analytical Laboratory for MDN; ACZ Laboratories, USA; Atlantic Laboratory of Environmental Testing (ALET), Canada; Flett Research, Ltd. (FRL), Canada; Swedish Environmental Institute (IVL), Sweden; North Shore Analytical (NSA), USA; Northern Lake Service (NLS), USA; Flemish Institute for Technological Research (VITO), Belgium; and USGS Wisconsin Mercury Laboratory (WML), USA. All laboratories analyze Hg in water using methods equivalent to USEPA Method 1631.

Friedman’s test indicated at least one laboratory’s results were significantly different from at least one other laboratory for paried comparisons of Hg-concentrations during 2007-09 (nine laboratories), but not during 2004-06 (six laboratories) at the alpha=0.05 significance level. A non-parametric multiple comparison test indicated significant (alpha=0.05) differences between selected pairs of laboratories, but not for any laboratory paired with FGS. Parametric analysis of variance indicated no significant (alpha=0.05) difference between reported mean concentration values from each laboratory and mean concentrations determined from the remaining laboratories.

Percentages of reported-minus-most probable value (MPV) differences outside +3 f-pseudosigma statistical control limits for each laboratory ranged from 0-24 percent (where: 1 f-pseudosigma=interquartile range =0.7413). Comparison of the percentages indicated relative data variability within laboratories in the following order: FRL<VITO<IVL<WML<ALET<NSA<FGS<NLS<ACZ. Comparison of the ratios of individual laboratory f-pseudosigma to f-pseudosigma for all laboratories combined revealed relative data variability among laboratories in the order: FRL<IVL<VITO<FGS<WML<ALET<NSA<NLS<ACZ.

Annual z-values ([laboratory median reported value-MPV]/[interquartile range for all laboratories combined = 0.7413]) indicated less variability among laboratory results for the low-concentration (6 ng/L) solutions and greater variability between laboratories with increasing concentrations in the range 9 - 22 ng/L. Although statistically significant (alpha=0.05) interlaboratory differences exist, median annual reported-minus-MPV Hg concentration differences among the laboratories was 0.92 ng/L over the test-solution concentration range of 6 - 22 ng/L, and the median laboratory-specific f-pseudosigma was 0.68 ng/L, indicating that inter-laboratory bias and variability are low.

1Firm or trade names mentioned herein are for identification purposes only and do not constitute endorsement by the U.S. government.

RS1-P12 — 11:00-12:00 and 17:30-18:30
Authors: MILLER, Matthieu1, GUSTIN, Mae1
(1) UNR, matthieum@unr.edu

Gaseous oxidized mercury (GOM) in air is currently measured using collection on annular denuders and on a variety of surrogate surfaces, with subsequent conversion to elemental Hg and quantification by Cold Vapor Atomic Fluorescence Spectroscopy (CVAFS). The exact chemical composition of GOM is not known, however current thinking is that halogenated forms are important, and possibly sulfur and oxygen containing compounds as well. This project focuses on developing a method to indirectly determine the form of GOM in air through collection on a surface and subsequent temperature-specific desorption. A variety of filter media are being tested for the collection of GOM in ambient air and in air with specific GOM-permeated forms. Thus far, quartz fiber and ion exchange membranes deployed passively, and PTFE and nylon filters deployed in a multiport manifold with forced air flow through the filters, have been investigated. During these exposures, atmospheric Hg concentrations are measured concurrently using an automated air mercury analyzer (Tekran 2537A/1130). After deployment, GOM deposition to filter media has been quantified by thermal desorption in a tube furnace using a sequential heating program, as well as by aqueous digestion and CVAFS analysis (EPA Method 1631). GOM deposition to quartz fiber and PTFE filters was not significantly different from blank values using both quantification methods. Ion exchange membranes and nylon filters showed significant collection of GOM, with nylon filters showing better agreement in total Hg recovery between quantification methods. HgCl2 was released from nylon filters dominantly in the temperature range between 140 and 180°C, whereas HgBr2 was mostly released at temperatures below 140°C. We are continuing to refine our data collection method with laboratory tests and field deployments in the summer.

RS1-P13 — 11:00-12:00 and 17:30-18:30
Authors: MAO, Huiting1, TALBOT, Robert2, HEGARTY, Jennifer3, KOERMER, James4
(1) SUNY-ESF, hmao@esf.edu; (2) University of Houston; (3) AER, Inc.; (4) Plymouth State University.

A comprehensive analysis was conducted using long-term continuous measurements of elemental gaseous mercury (Hg°), reactive mercury (RGM), and particulate phase mercury (HgP) at a coastal (Thompson Farm, denoted as TF), marine (Appledore Island, denoted as AI), and elevated inland (Pac Monadnock, denoted as PM) monitoring sites of the University of New Hampshire AIRMAP Observatory Network. Diurnal, seasonal, annual, and interannual variability in Hg°, RGM, and HgP from the three distinctly different environments were characterized and compared. Diurnal, seasonal, and annual variability in Hg° was most profound at TF and AI whereas was relatively dampened at PM due to its location in the free troposphere. The diurnal cycles of Hg° at TF and AI were of opposite phase in summer indicating Hg° sinks during daytime in the marine boundary layer, which is consistent with Hg° oxidation by halogen radicals in the marine environment reported previously. Annual RGM maximums at TF and AI occurred in spring, while at PM RGM was mostly below the limit of detection. HgP at AI and TF was largely below 1 ppqv; annual maximums occurred in winter and minimums in fall. Correlations between Hg°/RGM/HgP and climate variables were generally obscure although a tendency of higher levels of RGM and HgP was observed in spring and summer under sunny, dry, and warm conditions. To identify source types, correlations between Hg°/RGM/Hgp and tracers of different sources were carefully examined for all seasons. Hg°-CO relationship was well defined for winters 2003 – 2008 at TF and became rather scattered in winters 2009 and 2010. Higher levels of RGM were observed together with enhancement in CO, NOy, and SO2 in plumes at TF and AI, whereas no similar relationships were observed for HgP. Further, higher RGM levels at all three sites were found to be in photochemically aged air masses.

RS1-P14 — 11:00-12:00 and 17:30-18:30
Authors: CORBETT HAINS, Hamish1, VAN HEYST, Bill1
(1) School of Engineering, University of Guelph, hcorbett@uoguelph.ca

Mercury flux from frozen soils has been shown to increase during periods of temperature change in past field-experiments. Due to the nature of these experiments, the influence of individual environmental parameters on the flux of mercury during periods of changing temperature has not previously been identified. This research isolates both soil temperature and soil moisture content at sub-zero temperatures in order to determine the effect of both on mercury flux.

Mercury flux from frozen soils was observed in a dynamic flux chamber and scientific grade freezer using a Tekran 2537a mercury analyzer. Additionally, soil temperature, air temperature and relative humidity were recorded over the duration of the experiment. It was found that mercury flux from soils was suppressed significantly over most temperature ranges below 10oC. Periods of rapid positive change in soil temperature were seen to result in a significant increase in mercury flux from frozen soils. This effect, which had measured spikes in mercury flux approach levels seen at room temperature, is proposed to be a result of physical changes in the soil during temperature change. This effect was exaggerated at lower temperatures and was found to be most prevalent at soil moisture contents around 60% of field capacity. At a soil moisture content of 75% of field capacity, the flux was almost fully suppressed.

RS1-P15 — 11:00-12:00 and 17:30-18:30
Authors: HOROWITZ, Hannah M.1, CORBITT, Elizabeth S.2, TALBOT, Robert W.3, MAO, Huiting4, JACOB, Daniel J.2, AMOS, Helen M.2, SUNDERLAND, Elsie M.2
(1)Harvard College, hmhorow@fas.harvard.edu; (2) Harvard University; (3) University of Houston; (4) SUNY-ESF;

We use a global 3-D chemical transport model (GEOS-Chem CTM) to interpret long-term observations of Hg(0), RGM, and particulate Hg at a mixed forest site in New Hampshire (Thompson Farm). The observations show large diurnal variations for Hg(0), particularly during the growing season, with midday maxima and nighttime minima (Mao et al., 2008). We use these observational constraints to improve the GEOS-Chem representation of mercury uptake by vegetation, including the roles of leaf cuticles, dew formation, and in-leaf Hg(0) compensation points. We compare the diurnal variations of mercury species at Thompson Farm to those observed and simulated at other North American sites and relate differences to the underlying vegetation types. We also analyze other features of the observations of mercury species at Thompson Farm, including seasonal and synoptic variations, and show that these provide important constraints for mercury chemistry and sources in GEOS-Chem. Sensitivity analyses on model processes are conducted and implications for the atmospheric simulation of mercury in GEOS- Chem will be discussed.

RS1-P16 — 11:00-12:00 and 17:30-18:30
Author: HOLMES, Christopher1
(1)University of California, Irvine, cdholmes@uci.edu

Atmospheric processes controlling the fate of mercury--including transport, chemistry, scavenging, and deposition--are thought to depend linearly on mercury emissions and atmospheric abundance. For steady-state conditions, the rate coefficients for these processes are constants. The fluxes between mercury reservoirs and oxidation states can therefore be written as a single matrix equation with constant coefficients and solved with linear algebraic methods. This analytic approach has been previously used to estimate hemispheric gradients and mercury isotope fractionation in simple models with 10 or fewer reservoirs representing the global mercury cycle (Lamborg et al., Geochim et Cosmochim Acta 2002; Sonke, Geochim et Cosmochim Acta 2010).

Here I derive the steady-state transport, chemistry and deposition fluxes of mercury using the same approach, but with 4?x5? spatial resolution globally. The model thus calculates the full 3-D steady state distribution and fluxes of mercury in the atmosphere. Transfer coefficients are derived from the GEOS-Chem global 3-D chemistry and transport model (Holmes et al., Atmos Chem Phys 2010). The matrix version solves the steady-state problem much faster than temporal integration of the 3-D model code. I compare transport and deposition estimates from the matrix model against the time-integrated model and discuss some applications.

RS1-P17 — 11:00-12:00 and 17:30-18:30
Authors: BAYA, Anabelle Pascale1, HINTELMANN, Holger1
(1) Trent University, pascalebaya@trentu.ca

A good knowledge of the processes and Hg composition in the lower atmosphere is important to understand the cycle and bio- accumulation of momenethylmercury (MMHg) in the ecosystem. The direct atmospheric MMHg contribution to the aquatic ecosystem is unclear due to the lack of known reliable measurement methods. An analytical method to determine gaseous organic mercury species concentrations namely monomethylmercury (MMHg)and dimethylmercury (DMHg) in air was developed and successfully used to measure organic Hg in the arctic troposphere during the CCGS Amundsen expedition in 2010. The method is based on species specific Hg isotopic dilution and consists of collecting DMHg and MMHg (after inline ethylation to MeEtHg with sodium tetraethylborate) from air samples on Tenax traps. Up to 40 L of air are sampled and collected Hg species from up to 4 field traps are desorbed onto an analytical Tenax trap to form a composite sample. Isotope enriched gaseous MMHg (MM201Hg) and DMHg (DM198Hg) standards from a permeation tube are mixed in-line with the air sample as a yield tracer to assess the overall method performance. The analytical traps are stored and later analyzed in the lab at Trent University by thermodesorption and introduced into a gas chromatograph connected to an inductively coupled plasma mass spectrometer (GC-ICP/MS) for separation and detection. Since the GC-ICP/MS method can detect as little as 0.1 pg of Hg absolute, a 150 L sample volume translates to an overall method detection limit of 0.7 pg/ m3.

Using the method described above, MMHg and DMHg concentrations were successfully measured and are presented for the first time. It is observed that the organic species composition varies among arctic regions, with a higher proportion of MMHg in the Hudson Bay (mean = 5.5 ± 2.0, n=5 pg/m3), while concentrations of DMHg were 2.8 ± 3.6 pg/m3. In the high Arctic however, DMHg concentrations are highest (mean = 4.1 ± 2.3, n = 13 pg/m3). MMHg levels were significantly lower than those measured in Hudson Bay (mean = 2.3 ± 1.8, n = 8 pg/m3).

RS1-P18 — 11:00-12:00 and 17:30-18:30
Authors: SOMMAR, Jonas1, ZHU, Wei2, SHANG, Lihai1, FU, Xuewu1, LIN, Jerry3, FENG, Xinbin1
(1) Institute of Geochemistry, Chinese Academy of Sciences, Jonas@mails.gyig.ac.cn; (2) (1)Institute of Geochemistry, Chinese Academy of Sciences; (2)Graduate School of the Chinese Academy of Sciences; (3) Lamar University, USA;

Two sampling approaches to measure Hg0 gas exchange have been compared during a bi-weekly campaign over an open field setting in Guiyang, China:

  1. A relaxed eddy accumulation (REA) method relying on turbulent transport in the planetary boundary layer. In this method, the air in the case of positive and negative maximal values of the vertical wind velocity are collected into two separate reservoirs; for fluctuations around zero the air is rejected,
  2. Dynamic flux enclosures (DFC) designed to attain a steady, uniform friction velocity (u*) within the DFC.

The magnitude and diel cycles of the Hg0 vapour fluxes measured with the REA and DFC techniques respectively will be reported. Moreover, open field u* determined by the micro-meteorological apparatus will be compared with the measured u* of DFC. The discrepancy in results between the methods will be discussed.

Thursday, 28 July, 2011