Use of Passive Sampling Methods and Models to Understand Sources

Dec 8, 2014 - Western US, data were collected with passive samplers for ambient GOM .... 2012 and 2013 from the Pacific Ocean coast to Great Basin. Na...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/est

Use of Passive Sampling Methods and Models to Understand Sources of Mercury Deposition to High Elevation Sites in the Western United States Jiaoyan Huang* and Mae Sexauer Gustin Department of Natural Resources and Environmental Sciences, University of Nevada, Reno, 1664 N. Virginia Street, Reno, Nevada 89557, United States S Supporting Information *

ABSTRACT: To understand gaseous oxidized mercury (GOM) sources to the Western US, data were collected with passive samplers for ambient GOM concentrations and dry deposition at 10 sites from the coast of the Pacific Ocean to Great Basin National Park. Tests were done to better understand the samplers and the materials used. Measured dry deposition of GOM was significantly higher at sites >2000 m elevation relative to those below due to high GOM concentrations and atmospheric turbulence. At these high elevation sites, GOM dry deposition was higher in spring due to long-range transport from Asia (air parcels from the free troposphere) and some high GOM dry deposition events were related to regional emissions. Dry deposition of GOM at two sites was calculated using the passive sampler data and a multiple-resistance model. A previously developed relationship between the sampling rate of the passive sampler and GOM concentrations was used to estimate dry deposition and a scaling factor of 3 was used to adjust GOM concentrations, due to underestimation by KCl-coated denuder measurements. With the scaling factor of 3, modeled deposition was in the range of results estimated from two different models settings. However, dry deposition did not correlate consistently with either model. The disagreement could be due to uncertainties associated with measurements and/or modeling, or different GOM compounds existing in the atmosphere. If the atmospheric GOM compounds are known, dry deposition velocities could be estimated more accurately. Lastly, we investigated the potential for use of a new sampling material for GOM and checked the efficiency of the passive sampler.

■

atmospheric Hg chemistry, and model results.3,11,13−15 Lyman et al.15 reported GOM could be reduced by ozone on denuder surface; thus, GEM might be overestimated by Tekran 2537 measurement. Passive samplers are considered the next generation of atmospheric Hg measurements due to their low cost and simple deployment.3,16 This paper focuses on presenting research done with two major objectives: (1) furthering our understanding of the use of passive samplers for the measurement of GOM concentration and deposition and (2) describing their use in a network established for understanding Hg concentrations and deposition at rural locations in the Western US. A variety of passive samplers have been developed by multiple groups for atmospheric Hg measurements, including total gaseous Hg (TGM) and GOM concentrations, and dry deposition (ref 16 and references therein). A current concern is that passive samplers for GOM measurements have been calibrated using the Tekran (2357/1130/1135) system,17,18 which has been demonstrated to be biased low for collection of

INTRODUCTION The Minamata Convention on Mercury (Hg) has been signed by 102 nations and the number of nations ratifying is increasing.1 To support this effort, the United Nations Environment Programme (UNEP) published the Global Mercury Assessment that updated annual estimates of anthropogenic Hg emissions to the atmosphere based on emission inventories for 2010, and described Hg chemistry, transport, and fate in the environment.2 Pirrone et al.3 pointed out that the key challenge for Hg research is understanding the links between the atmosphere and fish. Recently, Blum et al.4 reported that the increased anthropogenic Hg emissions enhanced Hg content in marine food webs. Currently, fish and rice consumption are the major path for human exposure for methylHg.2 Atmospheric Hg is currently measured, and operationally defined, based on the Tekran 2537/1130/1135 measurement methods, as gaseous elemental Hg (GEM), gaseous oxidized Hg (GOM), and fine-particulate bound Hg (PBM, particle size 0.5, 4 sets of samples). This indicates that the samplers may be deployed for 2-to-4 weeks with no significant loss of GOM.

to estimate air concentrations. Because the Tekran system has been reported to underestimate GOM concentrations by a factor of 2-to-4;11 a scaling factor of 3 was used in this study to adjust the GOM concentrations in air. The number was selected because it is in the middle of the range of GOM underestimation determined by the Tekran KCl-denuder for Nevada11,13 and is within the range of the discrepancy between the Tekran GOM measurement and that collected using an active system and cation exchange membranes (1.6−3.7) in Huang et al.14 In this study, these concentrations were used to calculate Hg dry deposition under different conditions (α = β = 10, α = β = 2, no canopy resistance). The first is similar to HNO3 and the second to HONO. The calculated deposition was then compared with deposition measured using the surrogate surfaces. In addition, we conducted additional field and laboratory tests to better understand the potential for loss of GOM when deployed over 1 month, and tested a new cation-exchange membrane material for collection of GOM, because the cation-exchange material previously used has been phased out by the manufacturer.19,26,31 High GOM concentrations (up to 300 pg m−3) have been observed at high elevation sites in the Western US and in the free troposphere using KCl-coated denuders, the Tekran system, the Detector for Oxidized Hg Species (DOHGS), and passive systems.19,32−37 Most high GOM events were correlated with dry air, high ozone concentrations, and air masses that originated above the planetary boundary.32,33,37,38 However, based on back trajectory analyses, GOM could also be derived from the marine boundary layer.19,37 Generally, GOM deposition and concentrations are higher in summer than other seasons in Nevada;19,39 however, some spring and fall peaks in the GOM uptake rate were observed using passive samplers at high elevation Western US coastal sites and Great Basin National Park.19 In this study, using a passive sampler designed for measurement of GOM (Lyman et al.23 stated cation-exchange membrane can only capture GOM not GEM, so, in this study, GOM is defined as the Hg captured by cation-exchange membranes), passive sampler uptake were investigated using a computable fluid dynamic model, and some modifications of sampling and analytical methods were applied. The passive samplers were deployed at 10 sites in Nevada and California in 2012 and 2013 from the Pacific Ocean coast to Great Basin National Park, located on the border of Nevada and Utah, to understand the spatial and temporal variation of atmospheric Hg concentration and dry deposition. Dry deposition of GOM was then calculated using a multiple-resistance model with different settings and corrected GOM ambient air concentrations; these results were compared to measured GOM dry deposition to understand the uncertainties in the model. Our hypothesis for this work was that dry deposition would be higher at high elevation sites due to inputs from the free troposphere. In addition, a major goal for this work was to understand the level of uncertainty in model values assuming surrogate surface measurements reflect natural surfaces for GOM dry deposition. Furthermore, GOM sources related to dry deposition could be identified using this passive sampler coupled with gridded frequency distribution analysis.

■

METHODS Field Measurements. Site Information. This study involved 10 field sites (Table SI 1, Supporting Information, and Figure 1) located in Nevada and California. These included C

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

D

PRMP

PAVA

CGSP

CHALK

PEAV

GBNP

ECHO

CCRS

BISP

AGPK

0.5 ± 0.6 (21)

1.4 ± 0.6 (8)

1.5

0.7

0.9

0.6

0.4 ± 0.4 (21) 0.8(2)

3.5 ± (6) 0.5 ± (7) 1.2 ± (6) 2.8 ± (9)

0.7 ± 0.2 (15)

2.6 ± 1.2 (10) 1.7 ± 1.5 (8) 4.8 ± 4.4 (3) 2.6 ± 0.3 (23)

fall

2012

0.3 ± 0.2 (12)

0.2(4)

1.5(4)

3.6 ± 0.6 (3)

winter

2.3(2)

1.2 ± 0.4 (3)

3.1 ± 0.2 (15)

2.5 ± 1.6 (9)

1.0 ± 0.8 (9)

0.8 ± 0.4 (6)

3.8 ± 1.5 (18)

2.8 ± 1.9 (21) 2.0 ± 0.8 (18) 1.2 ± 0.6 (9) 3.4(2)

4.4 ± 0.5 (3) 1.1 ± 0.3 (3)

2013 summer

spring

1.0 ± 0.4 (6)

0.5(2)

2.3 ± 0.9 (18)

1.6 ± 0.8 (12) 1.6 ± 0.2 (6) 1.5 ± 0.2 (4) 0.7(2)

fall 2.3 ± 1.7 (36) 2.2 ± 1.0 (43) 1.2 ± 0.9 (28) 2.8 ± 2.2 (25) 2.6 ± 2.7 (32) 2.9 ± 1.5 (49) 0.5 ± 0.4 (36) 0.7 ± 0.4 (17) 0.7 ± 0.6 (41) 1.0 ± 0.8 (17)

overall

1.7 ± 1.0 (9)

0.4 ± 0.3 (12)

5.1 ± 3.1 (9) 1.9 ± 2.3 (9) 12.5 ± 13 (3) 5.1 ± 1.1 (21)

summer

0.4

0.8

0.5

0.6

0.2 ± 0.3 (21)

0.4 ± 0.4 (12)

1.7 ± (6) 0.4 ± (9) 2.8 ± (9) 1.6 ± (6)

fall

2012

0.2 ± 0.2 (12)

0.3(2)

1.5 ± 0.3(6)

10.0 ± 6.4 (6)

winter

2.1 ± 1.0 (3)

1.3 ± 0.3 (3)

3.6 ± 0.3 (15)

3.4 ± 0.3 (9)

5.6 ± 1.4 (3) 7.7 ± 2.2 (3)

spring

2013 summer

3.1 ± 0.6 (9)

1.4 ± 0.2 (6)

5.6 ± 1.6 (18)

4.2 ± 4.3 (21) 3.4 ± 2.0 (18) 0.9 ± 0.5 (12) 5.5 ± 0.9 (3)

uptake rate (pg h−1)

fall

0.8 ± 0.3 (6)

3.5 ± 1.7 (18)

2.3 ± 1.2 (12) 1.1 ± 1.0 (6) 1.5 ± 0.6 (6) 2.0 ± 0.5 (3)

overall 3.7 ± 3.4 (36) 3.9 ± 2.6 (42) 1.1 ± 1.2 (36) 6.2 ± 6.1 (33) 4.4 ± 8.3 (27) 4.1 ± 2.0 (49) 0.4 ± 0.4 (24) 1.1 ± 0.6 (11) 0.6 ± 0.9 (42) 2.2 ± 1.2 (18)

The data without standard deviation indicate sample n < 3. Site abbreviations include Angel Peak (AGPK), Berlin Ichthyosaur State Park (BISP), Crooked Creek Research Station (CCRS), Cathedral Gorge State Park (CGSP), Chalk Mountain (CHALK), Echo Peak (ECHO), Great Basin National Park (GBNP), Paradise Valley (PAVA), Peavine Peak (PEAV), and Pahrump (PRMP).

a

low elevation

high elevation

summer

dry deposition (ng m−2 h−1)

Table 1. Seasonal and Overall Hg Dry Deposition and Passive Sampler Uptake Rate (Mean ± σ (Total Number of Samples)) in California and Nevadaa

Environmental Science & Technology Article

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

San Francisco area (SF, −123 to −121°W and 39 to 37°N), the San Joaquin Valley (SJV, −121 to −119°W and 37 to 35°N), the Los Angeles Basin (LA, −120 to −117°W and 35 to 32.5°N), and the Las Vegas area (LV, −116 to −114°W and 37 to 35°N) with the air parcel heights of 3 km. The probability of the air passing through a box was calculated as

At UNR, the analytical method for measuring Hg collected on the cation-exchange membrane was changed in January 2013 from a manual total Hg analytical method18,19,23,24 to an automated system (Tekran 2600, Tekran Corp., Ontario, Canada)40,41 with the same membrane preparation. The automated system showed more stable and lower blanks (0.3 ± 0.3 and 0.3 ± 0.1 ng for aerohead (n = 58) and box sampler (n = 59), respectively) than those analyzed by the manual method in 2012 (1.1 ± 2.1 and 0.5 ± 0.5 ng for aerohead (n = 57) and box sampler (n = 56), respectively). This new analytical method improved the method detection limit. Data Analyses. Ancillary measurements, ozone concentrations, and meteorology are reported in the Supporting Information (Table SI 2). A computational fluid dynamic model (CFD), FLUENT (ANASYS, Inc., Canonsburg, PA), was used to resolve the flow fields within the sampler structure. The parameters used in this study are the same as those used in previous studies (see the Supporting Information for details).20,22 The 2-D geometries and the mesh or grid were created using the workbench, a subprogram under FLUENT (ANASYS, Inc., Canonsburg, PA). A standard k-ω model was used to investigate the influences near the wall with a low Reynolds’ number. The continuity and momentum equations for these simulations were converged to at least 1 × 10−6. A multiple-resistance model modified from Zhang et al.27 was used to calculate Hg dry deposition velocities during the sampling periods at PEAV and AGPK. Modeled Hg dry deposition fluxes were estimated using dry deposition velocities multiplied by the Hg ambient concentrations estimated using the passive sampler uptake and the linear relationship in Lyman et al.18 In previous work, the sampling rate of the box sampler was calibrated for GOM concentration using the Tekran system under different wind conditions (cf. ref 18). Because the underestimation of the Tekran KCl-coated denuder measurements in ambient air in Nevada was reported as being 2-to-4fold biased low,11 a 3 times correction factor was applied to the Tekran derived concentrations calculated from the box sampler uptake rate in this study.11 These two sites were classified as thorn shrubs sites. The canopy resistance of thorn shrubs and zero resistance were applied as defined in the model to investigate the uncertainty resulting from the usage of different values. Similar to Lyman et al.5 and Marsik et al.,28 α and β values were set at 2 and 10 to investigate the effect on calculated dry deposition velocities under constant environmental conditions. The first is similar to HNO3 and the second to HONO. Detailed meteorological data for these calculations were not available for these sites; therefore, data were extracted from the Eta Data Assimilation System (EDAS) 40 km meteorological model using an interpolation method.42 To ensure the extracted data represent actual local conditions, they were compared to the measured air temperature and humidity. The daily mean temperature was highly correlated (r2 = 0.56− 0.87 with slope close to one (0.8−1.2)), and the correlation for RH was slightly lower (r2 = 0.21−0.52) with slope of 0.9 (n = 133−156). Back trajectories were calculated using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLT 4.9) model using EDAS 40 km with a 1000 m above ground level starting height and 72 h duration.43 The starting times selected were Pacific Standard Time 0:00, 4:00, 8:00, 12:00, 16:00, and 20:00 for each day during the sampling periods. The numbers of endpoints were calculated for different air basins: the marine boundary layer (MBL, −134 to −124°W and 46 to 31°N), the

Pi , k =

Ni , k Mi , k

N and M are the number of end points during high dry deposition events (>mean value) and entire sampling period, respectively, for the specific source box and elevation. Global Data Assimilation System (GDAS) 1 × 1° meteorology was used to run back trajectories for 2 overlapping high Hg events in the spring. Back trajectories were initiated daily (11:00 pacific standard time, 120 h duration) during 2 weeks. Overall, the uncertainties of back trajectories calculated from HYSPLIT are 20% of the air parcel traveling distance.44−47 Sigmaplot 14.0 (Systat Software Inc., San Jose, CA) and Minitab 16.0 (Minitab Inc., PA, US) were used to do t tests and correlation analyses.

■

RESULTS AND DISCUSSION Field Measurement: Spatial/Temporal Patterns. Except for the CCRS site, overall mean GOM dry deposition and uptake rate at high elevation sites for the data periods presented in this paper were 2.2−2.9 ng m−2 h−1 and 3.7−6.2 pg h−1, respectively (Table 1). These varied by season and by year. Other studies have reported high values in the spring and fall in the southwestern United States (0.4−1.0 ng m−2 h−1),31 and in Florida (∼0.1 ng m−2 h−1 and 0.5 pg h−1).48 The higher GOM dry deposition and uptake rates are due to higher GOM ambient air concentration at Western US high elevation sites. Sather et al.,49 who collected data in Oklahoma and Texas, observed no significant seasonal trend. In this study, there were significantly high Hg levels in the summer and winter 2012 at ECHO, and at AGPK in the spring 2013 (Table 1). Uptake rate and deposition measured at >2000 m elevation above sea level (asl) was significantly higher than that at low elevation ( low elevation sites (0.6 to 2.2 pg h−1) > marine boundary layer (0.4 pg h−1). Despite the fact that Hg dry deposition and uptake rates measured using these passive techniques may be influenced by environmental conditions and the chemical forms of GOM, if both measurements are linked with ambient air concentrations, then a correlation between dry deposition and uptake rate should be observed. A moderate correlation between Hg dry deposition and uptake rate was obtained using the entire data set (outlier included) for all sites (Figure SI 3, Supporting Information, top; r2 = 0.43 p < 0.01 n = 109). However, the data measured at GBNP are scattered and the individual correlations at BISP, CHALK, CGSP, and PRMP were low (r2 < 0.25). The reason for weak correlations at CGSP and PRMP may be due different compounds occurring at different times that are not adequately measured by the box sampler, but recorded by the aerohead. At CHALK, dry deposition and uptake rate were low and close to the method detection limit. If the data are parsed out into the high elevation and low elevations sites (see Figure SI 3, Supporting Information), there is no correlation for the former; however, there is for the latter. The different trends and outliers suggest that there are different chemical forms of RM occurring temporally. In addition, based on data collected with a total gaseous Hg passive sampler, Wright et al.19 suggested that there may be forms of GOM not being measured by the cation-exchange membrane. Because of the small air movement in the box sampler, wind speed can only slightly change the uptake rate. Taking PEAV as a specific example, because of the availability of more data, the correlation between dry deposition and uptake rate measured by aerohead and passive samplers was good (Figure SI 3, Supporting Information, bottom; r2= 0.66 p < 0.01 n = 24). This could be due to a form consistently present at this site that has similar collection efficiency on the membranes in the aerohead and box samplers. Measured and Modeled Fluxes. Modeled estimates of GOM dry deposition velocities ranged from 0.3 to 2.8 cm s−1 at PEAV and AGPK depending on the environmental parameters and different canopy resistances. Sampling rates of the GOM

passive samplers were applied here using the equation developed by Lyman et al.23 and associated concentrations were adjusted by a scaling factor of 3 due to the underestimation ratio of GOM by the Tekran KCl-denuder.11 Sampling rates were used to estimate ambient air GOM concentrations that were used in the dry deposition model. Concentrations were then multiplied by the corresponding deposition velocities calculated from the resistance model, and resulting deposition ranged from 0.1 to 22.4 ng m−2 h−1 (Figure 2). As expected, the highest GOM dry deposition velocity was estimated when canopy resistance was excluded; while the lowest velocity was found when α = β = 2 was applied. Measured GOM dry deposition (1.0 to 6.6 ng m−2 h−1) using aerohead samplers was similar to modeled values at these two sites. Previous work compared GOM dry deposition fluxes derived from the surrogate surface measurements and model simulations;5,17,26,28 however, different conclusions were reached due to different sampling methods and model parameters. In this study, the average measured GOM dry deposition fluxes (2.4 to 3.2 ng m−2 h−1) were close to modeled values with α = β = 10 (∼2.9 ng m−2 h−1). Lyman et al.5 found the measured fluxes were 3-to-7 times higher than model results, if the sampling rate was not corrected and α = β = 2 was used. Data presented in Figure 2 suggests that (1) the surrogate surface does not represent a surface of no surface resistance, rather it may be a good simulation for natural surfaces in some cases, and (2) based on the variable agreement between the measured and modeled data, different chemical compounds are present with different deposition velocities. For example, at AGPK, measured fluxes were close to those with α = β = 2 in the first 6 weeks; however, the fluxes were similar to those with α = β = 10 in the following 10 weeks and there was one 2 week period when the measured value was higher than modeled. At PEAV, the biweekly mean fluxes varied over the time. There was one period during which measured flux was lower than modeled flux using α = β = 2; however, most of time values were higher than fluxes using α = β = 10. This suggests a form present with a higher deposition velocity that would facilitate higher rates of deposition. In other words, there are different GOM forms in the atmosphere during different time periods. Alternatively, the presence of high turbulence at this site would facilitate higher rates of deposition due to more contact with the surface (this site can be windy due to its location as a prominent peak in the free troposphere). Potential Source Areas. The probability of air being derived from specific source areas delineated as boxes based on 72 h trajectories was assessed (Table 2). Values were calculated based on the number of end points > mean divided by the total number of endpoints in that box and at a specific height. One must consider how this calculation is done and use caution when interpreting the probability values. For example, for PEAV, the Los Angeles (LA) Basin has the highest probability for all heights, but the numbers of end points are low, whereas the marine boundary layer (MBL) at all heights has the highest number of trajectory points and the air coming into this site is predominantly from the west (cf. Fine et al.51). In addition, the size of the boxes varied. Using the trajectory analyses, at AGPK, major source areas were LA and MBL boxes, when the air parcels were derived from high elevation, and LA, Las Vegas, MBL, and San Joaquin Valley when from low elevation. This implies that GOM sources (either direct sources or sources of oxidants) are lofted pollution from California that subsides into G

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

similar. Variation between modeled and measured values indicates different chemical forms in the air. Data also suggest that the surrogate surface sampler simulated natural surfaces. Although the measured GOM dry deposition were within the range of modeled GOM dry deposition, the ratio of measured/ modeled GOM dry deposition varied with the time at PEAV and AGPK. This also implies different GOM compounds are present in the atmosphere with different dry deposition velocities due to physical and chemical properties. High elevation sites had higher rates of deposition than the low elevation sites, and the MBL site. Higher dry deposition and passive sampler uptake at the low elevation sites relative to the MBL indicated mixing of free troposphere air to the low elevation locations in Nevada. Higher GOM deposition at the high elevation sites may be attributable to regional and longrange transport. GOM could have been derived from a specific source or generated as it interacted with oxidants during transport. Sources for GOM were examined for AGPK and PEAV because of a more robust set of samples. Overall, the major GOM or oxidant source areas contributing to observed GOM at AGPK were from long-range transport and the LA Basin, and high altitude long-range transport at PEAV. A local source of oxidants at AGPK would be Las Vegas and regional sources for PEAV include San Francisco and Sacramento. Trajectories for 2 overlapping high Hg events measured in the spring at PEAV and AGPK were associated with Asia longrange transport.

Table 2. Probability Function of Back Trajectory Analyses for Different Source Boxes (LA-Los Angeles, LV-Las Vegas, MBL-Marine Boundary Layer, SF-San Francisco, and SJVSan Joaquin Valley), and Elevations at AGPK and PEAVa 3 km

total

47(38) 0(59) 60(822) 0(33) 20(15)

40(8404) 22(14951) 36(9016) 8(7472) 27(6154)

77(144) 29(34) 23(1587) 12(33) 30(228)

44(2031) 25(201) 33(21772) 37(7472) 54(3086)

a

Elevation represents that above ground level. Data in parentheses indicate total number of end point for that box and elevation.

Nevada, or subsiding air that traveled from Asia and over the MBL (Table 2). Fine et al.51 pointed out that air associated with some high maximum daily 8 h average (MDA-8) ozone events at GBNP sites included air that was derived from Asian long-range transported through the LA area. Weiss-Penzias et al.33 showed high GOM concentrations in Nevada were correlated with high ozone concentrations and derived from the free troposphere and related to long-range transport. For AGPK, located in southern Nevada, when air traveled over the surface in the boundary layer, Las Vegas and LA, and the San Joaquin Valley were important regional sources. There was no significant influence from San Francisco box, because the predominant wind direction at this location was from the southwest and the San Francisco box is to the northwest and ∼450 km away. At PEAV, despite the high probability associated with the LA box, this area does not have many points, nor does LV or the SJV. The marine boundary layer is an important source area at high and low elevations, and this suggests a free troposphere influence. The SF box, as boundary layer flow, and SJV (at 1 to 3 km) are also source areas for PEAV. Although PEAV is a high elevation site with air derived primarily from the west, air coming from LA was associated with high GOM events. This implies that there are Hg/oxidant sources in the LA area; however, the influence is low due to less air movement from this direction. At both sites, trajectories were from the East Pacific Ocean; however, the general trajectory patterns during 2 weeks with high GOM events in spring, based on 10 day trajectories, were from the West Pacific Ocean (Figure SI 4, Supporting Information). Among these trajectories, 58 and 65% of end points were above 1 km; this implies that these events are from Asia long-range transport.

Comparison of different deployment times (monthly and biweek) and collection surfaces (Mustang S and ICE 450) using passive samplers (PAS) aerohead (top) and box samplers (bottom) at the Peavine (PEAV) site, configuration of the box sampler, overall correlation of Hg dry deposition and uptake rate measured by surrogate surfaces and passive samplers among all sites, back trajectories (HYSPLIT with GDAS meteorology, 120 h, 1000 m starting height) during spring high Hg dry deposition events, summary of sampling site information including site abbreviation, elevation, general description, data collected at each site, the sampling period, and site operator, and summary of ancillary data. This material is available free of charge via the Internet at http://pubs.acs.org.

IMPLICATIONS Based on the comparison of the Mustang S and the ICE450 membranes, there are different forms of Hg in the air that are collected differently by these materials. Materials may be deployed for 2 to 4 weeks without loss of GOM. Use of an automated Hg analytical system resulted in improvement of the detection limit. Using passive sampler uptake rates and a 3 times correction factor, modeled and measured dry deposition values were

This study was funded by Electric Power Research Institute (EPRI). We thank undergraduate students (Matthew Peckham, Douglas Yan, Travis Lyman, Musheng Alishahi, and Jennifer Arnold) at UNR, and the site operators (Jeff Morris (BISP), Cody Tingey (CGSP), Tim Forsell (CCRS), Russ Merle (PRMP), Jerry Dries (ECHO), Daren Winkelman (AGPK), and Li Zhang (AGPK), Tony Lesperance (PAVA), Ben Roberts (GBNP), and Peter Weiss-Penzias (CHALK)). The authors thank Rebekka Fine for detailed review of this paper.

■

ASSOCIATED CONTENT

S Supporting Information *

■

AUTHOR INFORMATION

Corresponding Author

*Jiaoyan Huang. E-mail: [email protected]. Tel: 1-775784-4722. Fax: 1-775-784-4789. Notes

The authors declare no competing financial interest.

■

■

H

ACKNOWLEDGMENTS

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

■

Article

Computational modeling and experimental measurements. Atmos. Environ. 2011, 45 (26), 4354−4359. (21) Zhang, X.; Wania, F. Modeling the uptake of semivolatile organic compounds by passive air samplers: Importance of mass transfer processes within the porous sampling media. Environ. Sci. Technol. 2012, 46 (17), 9563−9570. (22) Huang, J.; Liu, Y.; Holsen, T. M. Comparison between knifeedge and frisbee-shaped surrogate surfaces for making dry deposition measurements: Wind tunnel experiments and computational fluid dynamics (CFD) modeling. Atmos. Environ. 2011, 45 (25), 4213− 4219. (23) Lyman, S. N.; Gustin, M. S.; Prestbo, E. M.; Kilner, P. I.; Edgerton, E.; Hartsell, B. Testing and application of surrogate surfaces for understanding potential gaseous oxidized mercury dry deposition. Environ. Sci. Technol. 2009, 43 (16), 6235−6241. (24) Peterson, C.; Alishahi, M.; Gustin, M. S. Testing the use of passive sampling systems for understanding air mercury concentrations and dry deposition across Florida, USA. Sci. Total Environ. 2012, 424 (0), 297−307. (25) Peterson, C.; Gustin, M. Mercury in the air, water and biota at the Great Salt Lake (Utah, USA). Sci. Total Environ. 2008, 405 (1−3), 255−268. (26) Castro, M. S.; Moore, C.; Sherwell, J.; Brooks, S. B. Dry deposition of gaseous oxidized mercury in Western Maryland. Sci. Total Environ. 2012, 417−418 (0), 232−240. (27) Zhang, L.; Brook, J. R.; Vet, R. A revised parameterization for gaseous dry deposition in air-quality models. Atmos. Chem. Phys. 2003, 3 (6), 2067−2082. (28) Marsik, F. J.; Keeler, G. J.; Landis, M. S. The dry-deposition of speciated mercury to the Florida Everglades: Measurements and modeling. Atmos. Environ. 2007, 41 (1), 136−149. (29) Zhang, L.; Wright, L. P.; Blanchard, P. A review of current knowledge concerning dry deposition of atmospheric mercury. Atmos. Environ. 2009, 43 (37), 5853−5864. (30) Zhang, L.; Blanchard, P.; Johnson, D.; Dastoor, A.; Ryzhkov, A.; Lin, C. J.; Vijayaraghavan, K.; Gay, D.; Holsen, T. M.; Huang, J.; Graydon, J. A.; St. Louis, V. L.; Castro, M. S.; Miller, E. K.; Marsik, F.; Lu, J.; Poissant, L.; Pilote, M.; Zhang, K. M. Assessment of modeled mercury dry deposition over the Great Lakes region. Environ. Pollut. 2012, 161 (0), 272−283. (31) Sather, M. E.; Mukerjee, S.; Smith, L.; Mathew, J.; Jackson, C.; Callison, R.; Scrapper, L.; Hathcoat, A.; Adam, J.; Keese, D.; Ketcher, P.; Brunette, R.; Karlstrom, J.; Jagt, G. V. d. Gaseous oxidized mercury dry deposition measurements in the Four Corners area and Eastern Oklahoma, U.S.A. Atmos. Pollut. Res. 2013, 4, 168−180. (32) Weiss-Penzias, P.; Jaffe, D.; Swartzendruber, P.; Hafner, W.; Chand, D.; Prestbo, E. Quantifying Asian and biomass burning sources of mercury using the Hg/CO ratio in pollution plumes observed at the Mount Bachelor observatory. Atmos. Environ. 2007, 41 (21), 4366− 4379. (33) Weiss-Penzias, P.; Gustin, M.; Lyman, S. N. Observations of speciated atmospheric mercury at three sites in Nevada, USA: Evidence for a free tropospheric source of reactive gaseous mercury. J. Geophys. Res. 2009, DOI: 10.1029/2008JD011607. (34) Swartzendruber, P. C.; Jaffe, D. A.; Finley, B. Development and first results of an aircraft-based, high time resolution technique for gaseous elemental and reactive (oxidized) gaseous mercury. Environ. Sci. Technol. 2009, 43 (19), 7484−7489. (35) Obrist, D.; Hallar, A. G.; McCubbin, I.; Stephens, B. B.; Rahn, T. Atmospheric mercury concentrations at Storm Peak Laboratory in the Rocky Mountains: Evidence for long-range transport from Asia, boundary layer contributions, and plant mercury uptake. Atmos. Environ. 2008, 42 (33), 7579−7589. (36) Sheu, G.-R.; Lin, N.-H.; Wang, J.-L.; Lee, C.-T.; Ou Yang, C.-F.; Wang, S.-H. Temporal distribution and potential sources of atmospheric mercury measured at a high-elevation background station in Taiwan. Atmos. Environ. 2010, 44 (20), 2393−2400.

REFERENCES

(1) UNEP. Minamata Convention on Mercury. http://www. mercuryconvention.org/. (2) UNEP. Global Merucry Assessment 2013-Sources, Emissions, Releases, and Environmental Transport; UNEP Chemicals Branch: Geneva, Switzerland, 2013. (3) Pirrone, N.; Aas, W.; Cinnirella, S.; Ebinghaus, R.; Hedgecock, I. M.; Pacyna, J.; Sprovieri, F.; Sunderland, E. M. Toward the next generation of air quality monitoring: Mercury. Atmos. Environ. 2013, 80 (0), 599−611. (4) Blum, J. D.; Popp, B. N.; Drazen, J. C.; Anela Choy, C.; Johnson, M. W. Methylmercury production below the mixed layer in the North Pacific Ocean. Nat. Geosci. 2013, 6 (10), 879−884. (5) Lyman, S. N.; Gustin, M. S.; Prestbo, E. M.; Marsik, F. J. Estimation of dry deposition of atmospheric mercury in Nevada by direct and indirect methods. Environ. Sci. Technol. 2007, 41 (6), 1970− 1976. (6) Schroeder, W. H.; Munthe, J. Atmospheric mercuryAn overview. Atmos. Environ. 1998, 32 (5), 809−822. (7) Lindberg, S. E.; Bullock, R.; Ebinghaus, R.; Engstrom, D.; Feng, X.; FItzgerald, W.; Pirrone, N.; Prestbo, E.; Seigneur, C. A synthesis of progress and uncertainites in attributing the sources of mercury in deposition. Ambio 2007, 36, 19−32. (8) Gay, D. A.; Schmeltz, D.; Prestbo, E.; Olson, M. L.; Sharac, T.; Tordon, R. The Atmospheric Mercury Network: Measurement and initial examination of an ongoing atmospheric mercury record across North America. Atmos. Chem. Phys. Discuss. 2013, 13, 10521−10546. (9) Steffen, A.; Scherz, T.; Olson, M.; Gay, D.; Blanchard, P. A comparison of data quality control protocols for atmospheric mercury speciation measurements. J. Environ. Monit. 2012, 14 (3), 752−765. (10) Gustin, M.; Jaffe, D. Reducing the uncertainty in measurement and understanding of mercury in the atmosphere. Environ. Sci. Technol. 2010, 44 (7), 2222−2227. (11) Gustin, M. S.; Huang, J.; Miller, M. B.; Peterson, C.; Jaffe, D. A.; Ambrose, J.; Finley, B. D.; Lyman, S. N.; Call, K.; Talbot, R.; Feddersen, D.; Mao, H.; Lindberg, S. E. Do we understand what the mercury speciation instruments are actually measuring? Results of RAMIX. Environ. Sci. Technol. 2013, 47 (13), 7295−7306. (12) Talbot, R.; Mao, H.; Feddersen, D.; Smith, M.; Kim, S. Y.; Sive, B.; Haase, K.; Ambrose, J.; Zhou, Y.; Russo, R. Assessment of particulate mercury measured with the manual and automated methods. Atmos. 2010, 2, 1−20. (13) Ambrose, J. L.; Lyman, S. N.; Huang, J.; Gustin, M. S.; Jaffe, D. A. Fast time resolution oxidized mercury measurements during the Reno Atmospheric Mercury Intercomparison Experiment (RAMIX). Environ. Sci. Technol. 2013, 7 (13), 7285−7294. (14) Huang, J.; Miller, M. B.; Weiss-Penzias, P.; Gustin, M. S. Comparison of gaseous oxidized Hg measured by KCl-coated denuders, and nylon and cation exchange membranes. Environ. Sci. Technol. 2013, 47 (13), 7307−7316. (15) Lyman, S. N.; Jaffe, D. A.; Gustin, M. S. Release of mercury halides from KCl denuders in the presence of ozone. Atmos. Chem. Phys. 2010, 10, 8197−8204. (16) Huang, J.; Lyman, S. N.; Hartman, J. S.; Gustin, M. S. A review of passive sampling systems for ambient air mercury measurements. Environ. Sci.: Processes Impacts 2014, 16 (3), 374−392. (17) Huang, J.; Choi, H.-D.; Landis, M. S.; Holsen, T. M. An application of passive samplers to understand atmospheric mercury concentration and dry deposition spatial distributions. J. Environ. Monit. 2012, 14 (11), 2976−2982. (18) Lyman, S. N.; Gustin, M. S.; Prestbo, E. M. A passive sampler for ambient gaseous oxidized mercury concentrations. Atmos. Environ. 2010, 44 (2), 246−252. (19) Wright, G.; Gustin, M. S.; Weiss-Penzias, P.; Miller, M. B. Investigation of mercury deposition and potential sources at six sites from the Pacific Coast to the Great Basin, USA. Sci. Total Environ. 2014, 470−471 (0), 1099−1113. (20) May, A. A.; Ashman, P.; Huang, J.; Dhaniyala, S.; Holsen, T. M. Evaluation of the polyurethane foam (PUF) disk passive air sampler: I

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

(37) Timonen, H.; Ambrose, J. L.; Jaffe, D. A. Oxidation of elemental Hg in anthropogenic and marine airmasses. Atmos. Chem. Phys. 2013, 13, 2827−2836. (38) Weiss-Penzias, P.; Amos, H. M.; Selin, N. E.; Gustin, M. S.; Jaffe, D. A.; Obrist, D.; Sheu, G.-R.; Giang, A. Use of a global model to understand speciated atmospheric mercury observations at five highelevation sites. Atmos. Chem. Phys. Discuss. 2014, 14, 22763−22792. (39) Lyman, S. N.; Gustin, M. S. Determinants of atmospheric mercury concentrations in Reno, Nevada, U.S.A. Sci. Total Environ. 2009, 408 (2), 431−438. (40) Castro, M. S.; McLaughlin, E. N.; Davis, S. L.; Morgan, I. R. P. Total mercury concentrations in lakes and fish of Western Maryland, USA. Arch. Environ. Contam. Toxicol. 2002, 42 (4), 454−462. (41) Demers, J. D.; Driscoll, C. T.; Fahey, T. J.; Yavitt, J. B. Mercury cycling in litter and soil in different forest types in the Adirondack Region, New York, USA. Ecol. Appl. 2007, 17 (5), 1341−1351. (42) NOAA Eta Data Assimilation System (EDAS40) archive information. https://ready.arl.noaa.gov/edas40.php. (43) Draxler, R. R.; Hess, G. D. HYSPLIT 4 User’s guide; NOAA Technical Memorandum ERL ARL-230; NOAA: Washington, DC, 2005. (44) Stohl, A. Computation, accuracy and applications of trajectories - A review and bibliography. Atmos. Environ. 1998, 32 (6), 947−966. (45) Gebhart, K. A.; Schichtel, B. A.; Barma, M. G. Directional biases in back trajectories caused by model and input data. J. Air Waste Manage. Assoc. 2005, 55, 1649−1662. (46) Draxler, R. What are the levels of uncertainty associated with back trajectory calculations in HYSPLIT. http://www.arl.noaa.gov/ faq_hg17.php (accessed 10/17/2011). (47) Stohl, A.; Forster, C.; Eckhardt, S.; Spichtinger, N.; Huntrieser, H.; Heland, J.; Schlager, H.; Wilhelm, S.; Arnold, F.; Cooper, O. A backward modeling study of intercontinental pollution transport using aircraft measurements. J. Geophys. Res. 2003, 108, 4370. (48) Gustin, M.; Weiss-Penzias, P.; Peterson, C. Investigating sources of gaseous oxidized mercury in dry deposition at three sites across Florida, USA. Atmos. Chem. Phys. 2012, 12, 9201−9219. (49) Sather, M. E.; Mukerjee, S.; Allen, K. L.; Sminth, L.; Mathew, J.; Jackson, C.; Callison, R.; Scrapper, L.; Hathcoat, A.; Adam, J.; Keese, D.; Ketcher, P.; Brunette, R.; Karlstrom, J.; Jagt, G. V. d. Gaseous oxidized mercury dry deposition measurements in the Southwestern USA: A comparison between Texas, Eastern Oklahoma, and the Four Corners Area. Sci. World J. 2014, 2014. (50) Burley, J. Professor St. Mary’s College of California. Personal communication, 2014. (51) Fine, R.; Miller, M.; Burley, J.; Jaffe, D. A.; Pierce, R. B.; Gustin, M. S. MDA8 ozone values at rural surface sites in Nevada, USA: Results from two years of the Nevada rural ozone initiative (NVROI). Sci. Total Environ., accepted. (52) Brooks, S.; Ren, X.; Cohen, M.; Luke, W.; Kelley, P.; Artz, R.; Hynes, A.; Landing, W.; Martos, B. Airborne vertical profiling of mercury speciation near Tullahoma, TN, USA. Atmos. 2014, 5 (3), 557−574. (53) Huang, J.; Gustin, M. S. Evidence for a free troposphere source of mercury in wet deposition in the Western United States. Environ. Sci. Technol. 2011, 46 (12), 6621−6629. (54) Weiss-Penzias, P. Professor University of California Santa Cruz. Personal communication, 2014.

J

dx.doi.org/10.1021/es502836w | Environ. Sci. Technol. XXXX, XXX, XXX−XXX