S1 (I) Measurement and understanding of atmospheric mercury processes

The atmosphere is a significant pathway by which Hg is distributed and introduced to terrestrial and aquatic ecosystems. Elemental Hg is the dominant form in the atmosphere and exhibits bidirectional exchange with surfaces. Gaseous oxidized mercury may be emitted by point sources and is also thought to be produced in the atmosphere by reaction of elemental mercury with oxidants. The predominant mechanisms of formation and the chemistry of gaseous oxidized mercury in the atmosphere are not known and continue to be debated. Additionally, the gas-particulate partitioning of Hg has had limited investigation, and the role of aerosols in the fate and transport of atmospheric Hg is uncertain. In this session we will focus on new observational methods, use of additional atmospheric chemistry and meteorological information, and modeling that can improve our understanding atmospheric Hg processing, distribution and fate. This includes discussion of:

1. Observations support an improved understanding of the atmospheric oxidation mechanisms and/or the gas/particle conversion processes;
2. Observations that support a better understanding of the chemical forms of Hg(II) in the atmosphere;
3. Observations that support a better understanding of flux of Hg between terrestrial and aquatic surfaces and the atmosphere;
4. Observations that improve our understanding of the distribution of Hg species throughout the atmospheric column;
5. New instrumental methods, analyses of data collected in conjunction with Hg, and modeling that support improved understanding of atmospheric Hg.

The free troposphere has been shown to be a reservoir for oxidized mercury compounds, but free tropospheric measurements of oxidized mercury are extremely limited. Aircraft observations would be valuable to better understand oxidation mechanisms and to validate the array of global models currently being developed. We first deployed a prototype aircraft system for total, elemental, and oxidized mercury in 2008. This system used two different approaches to measure gaseous oxidized mercury, which allowed for an internal comparison of the approaches (Swartzendruber et al. 2009). Based on these measurements and additional tests, we made significant improvements to our system for the recent WAMO/ICARE project (Western Airborne Mercury Observations / International Conference on Airborne Research for the Environment), which flew domestic and intercontinental flights on the NCAR C-130 in the fall of 2010 in a four-week campaign. Prior to the campaign the system was extensively tested in our laboratory with synthetic HgX₂ compounds to verify and characterize its performance. In-flight calibrations included zero checks and additions of span gas to zero and ambient air.

We found that oxidized mercury is common in cloud-free air in the free troposphere. The highest concentrations of oxidized mercury were mostly associated with either air from the upper troposphere/lower stratosphere or aged pollution plumes. Air from the upper troposphere/lower stratosphere (high ozone and potential vorticity) had the most oxidized mercury (up to 600 pg std m^-3) and was associated with reduced total mercury concentrations. In these plumes the ratio of oxidized mercury to total mercury was as high as 63%. We found oxidized mercury concentrations of 100-350 pg m^-3 in aged pollution plumes that had not interacted with clouds. Oxidized mercury in pollution plumes was correlated with CO and ozone and anti-correlated with gaseous elemental mercury and relative humidity. These data are being used in consort with a variety of chemical models to better understand the distribution and sources of elemental and oxidized mercury in the atmosphere.

The National Atmospheric Deposition Program’s (NADP) Mercury Deposition Network (MDN) benefits from the activities of both internal and external quality-assurance (QA) programs. The internal program operates within the NADP’s Program Office (PO) and within its contract laboratory. The external program is operated by the U.S. Geological Survey’s Branch of Quality Systems (USGS/BQS).

MDN samples are analyzed for total mercury (Hg) by the Mercury Analytical Laboratory (HAL) at Frontier Global Sciences, Inc.³ (Seattle, WA). The USGS MDN Inter-laboratory Comparison program quantifies variability and bias in MDN analyses for comparison with laboratories in USA, Canada and Europe. HAL’s analytical bias is -1 nanograms Hg per liter (ng /L), over the analytical range of 6 to 24 ng/L, relative to participating laboratories. HAL ranked fourth out of nine laboratories in interlaboratory variability. Results of a USGS Blind Audit program, which monitors HAL’s performance using double-blind QA samples (disguised as field samples), confirms this bias.

System Blank Program results indicate that average Hg contamination in MDN samples is less than the second percentile of field sample concentrations. Source(s) of this contamination have not yet been identified in the field or laboratory.

Analysis of MDN field data suggests that sample loss through evaporation can impact data quality. Preliminary results suggest that sample evaporation can reduce measured total Hg concentrations by as much as 50 percent.

Co-located sampler configurations at operating MDN sites are used to estimate overall variability in MDN data. Annual precipitation-weighted mean Hg concentrations have a statistically significant (a=0.05) measurement resolution of +2 ng Hg/L over the entire range of field concentrations.

Results from each of these programs are available on the USGS/BQS website. Without these programs, the quality of MDN data would be unknown, and potential environmental spatial and temporal patterns could not be discerned from measurement variability and bias. ³Firm or trade names mentioned herein are for identification purposes only and do not constitute endorsement by the U.S. government.

The gas-phase reaction of halogen species with gaseous elemental mercury (GEM) is a source for gaseous oxidized mercury (GOM). It has been established that these processes occur at high latitudes, but it has only recently been demonstrated that they might also occur in mid-latitude regions (Peleg et al. 2007). Here we present measurements of reactive halogen species bromine oxide (BrO) and iodine oxide (IO) along with gaseous oxidized mercury (GOM), gaseous elemental mercury (GEM), and particulate mercury (Hg_p) at a coastal location in Gulf Breeze, Fl. The University of Colorado has deployed a research grade Multi Axis Differential Optical Absorption Spectroscopy (MAX-DOAS) instrument to measure BrO, IO, as well as formaldehyde (HCHO), glyoxal (CHOCHO), nitrogen dioxide (NO₂) and oxygen dimers (O₄). MAX-DOAS is based on the well established DOAS technique that uses Beer-Lambert’s Law and characteristic absorption structures to derive differential slant column densities for atmospheric trace gases. MAX-DOAS makes use of light collected from multiple elevation angles (the angle between the horizon and the pointing direction of the telescope) as well as a zenith measurement from the same solar zenith angle. Here we present the compilation of the data collected by this instrument over the time period from May 2009 to November 2010, which include the first measurements of BrO, IO, and CHOCHO over the Gulf of Mexico. We also present several case studies for days where significant amounts of reactive halogens were measured and explore the sources and history of the air masses carrying these compounds. We also take an in depth look at these case studies to investigate their implications on the reactions converting GOM to GEM in the coastal marine boundary layer over the Gulf of Mexico.

Current measurement techniques for atmospheric elemental mercury (Hg⁰) require several liters of sample air and several minutes for each analysis. Measurements with improved temporal and spatial resolution will enhance our ability to understand and track chemical cycling of mercury in the atmosphere such as high frequency Hg⁰ fluctuations that may correspond to atmospheric sources, sinks, and chemical transformation processes. We are developing a new sensor using cavity ring-down spectroscopy (CRDS) to measure Hg⁰ mass concentration. CRDS is a direct absorption technique that uses path lengths up to multiple kilometers in a compact absorption cell and has much higher sensitivity, temporal resolution, and reduced sample volume requirements (<0.5 l of sample air) than conventional absorption spectroscopy. Our current prototype includes a frequency-doubled, dye-laser emitting pulses tunable from 215 to 280 nm with a pulse repetition frequency of 50 Hz. CRDS measurements of Hg⁰ are linearly related to Hg⁰ concentrations determined by a Tekran 2537B analyzer over an Hg⁰ concentration range of 0.2 ng m^-3 to 573 ng m^-3, suggesting a linear response for both instruments.

Recent results using the first ambient air measurements of atmospheric mercury with this sensor show a CRD sensor detection limit of 0.09 ng Hg⁰ m^-3 at a temporal resolution of 10 sec, which is similar in range to the detection limit estimated using laboratory tests. We present a new method to control the laser wavelength by locking it to the hyperfine structure of the Hg isotope Hg²⁰² using an external low-pressure mercury cell and differential extinction measurements by two photodiodes. We found that the best method for calibration of the sensor is by quantification of the system’s baseline losses using ambient air filtered from ambient mercury (using a charcoal filter) and from interfering constituents such as fine particulate matter (using a HEPA filter) and ozone (by means of an ozone denuder). We present measurements of the CRD system in ambient air of the urban atmosphere of Reno, NV, and discuss the value and new opportunities of high-resolution measurements for better air mass characterization and source quantification of mercury. Potential future applications for our system include use with micrometeorological techniques such as Eddy Covariance, in situ and laboratory chemical and physical transformation and speciation studies of mercury, and airborne measurements if the sensor can be operated from an aircraft platforms.

Standard Operating Procedures (SOPs) and QA/QC protocols for monitoring ambient air concentrations of the operationally defined mercury fractions TGM, RGM, TPM and GEM, and mercury concentrations in precipitation are needed in order to assure a full comparability of site specific observational datasets with that obtained inside and outside the above mentioned networks. SOPs and QA/QC protocols should be in accordance with measurement practice adopted in existing networks and based on the most recent literature. Therefore, mercury species measurements were carried out at the German EMEP station and background measurement site of the German Federal Environmental Agency “Waldhof” from January to December 2009 and have been extended to this day. A complete setup for the measurements of RGM, TPM and GEM was installed at Waldhof station in 2008. A second analyser to determine TGM in ambient air have been in operation since 2002. An annual time series for the four Hg species was achieved for 2009, including ancillary data like particulate matter, NOx, Ozone and meteorological parameter. Furthermore a continuous set of QA/QC data and maintenance intervals was recorded and will be presented. A pre-period of more than 6 month was used to prepare a standard operation procedure and to answer the question if the system was fit for purpose at low level background concentrations? Within this scope the complete measurement procedure, including a 3 hours sampling interval, the amount of desorption and clean cycles, the QA/QC cylces for zero checks and functional test and the frequent exchange of glassware was evaluated to be suitable for the application at background levels at Waldhof station. We will present quality control data and ambient air concentrations, showing that the measurement procedure turned out to be a good compromise to detect RGM and TPM levels in the range of 0.4 to 21 pg m^-3. We will also show that daily and seasonal variability at different sites for RGM and TPM can be extremely high, making it difficult to choose the best sampling and desorption intervals for the individual application. However, a sensitive and precise method with high time resolution is needed, in order to measure atmospheric processes and/or use the automated method for long-term monitoring. We will discuss a maintenance procedures and service intervals which can be the key to achieve this goal.

Annual mercury (Hg) deposition and concentration isopleths maps from the Mercury Deposition Network (MDN) for the United States (U.S.) continue to show a south-to-north gradient. From a source-receptor point-of-view this seems counterintuitive as the largest anthropogenic sources; coal-fired power plants (CFPPs) are located primarily east of the Mississippi river, whereas natural sources are more widely distributed. Hence, factors other than proximity to sources must be considered to explain this consistent pattern. Questions of importance to this issue are: A) How much do U.S. based CFFPs contribute to this pattern? B) how much do local vs. non-local sources contribute? C) How much do natural sources and processes contribute?

EPRI is undertaking an integrated study to answer these questions that will increase understanding of the elevated Hg deposition in the southern U.S. Focal points of this study are:

1. assessing reduction of divalent Hg to elemental Hg in CFPP plumes both in the field and in the laboratory
2. assessing the importance of atmospheric halogens on observed divalent Hg concentrations
3. measuring Hg and trace metals in event-based deposition around a CFPP to establish near-field source receptor relationships
4. measuring Hg isotopes in wet deposition and associated with atmospheric aerosols to determine source types
5. modeling the importance of thunderstorm activity and morphology on observed MDN concentrations and the measured Hg levels around the CFPP
6. sampling of Hg dry deposition to establish the dry deposition component to total deposition and potential source-receptor relationships
7. determining the chemical speciation of divalent Hg in a CFPP plume as well as the atmosphere; crucial to Hg’s atmospheric chemical reactions, fate and transport
8. measurement of speciated atmospheric Hg along with meteorological parameters, other gases, and particles to characterize temporal behavior, atmospheric processes, and sources

The GEOS-Chem global mercury simulation is compared with measurements of reactive gaseous mercury (RGM) to constrain the atmospheric oxidation mechanism and budget of mercury in the atmosphere. Data is used from sites where both RGM and wet deposition information are available, and examine the site-specific budget of RGM in both measurements and model. To explore the influence of reaction mechanism, two simulations are compared: one with an Hg(0) oxidation mechanism including bromine chemistry, and a previous simulation using OH chemistry. Data is identified from the continental U.S. and differences between the western and eastern U.S. to isolate the strongest influence of chemistry. Findings from this analysis include: the model on average overestimates RGM concentrations over continental sites; latitudinal and longitudinal comparisons can provide some insight into oxidation and source processes for RGM; and that RGM measurements in comparison with modeling may help better constrain the Hg cycle than Hg(0) data. An uncertainty analysis focusing on using RGM and Hg(0) measurements to constrain oxidation and the mercury budget in comparison with modeling is also presented.