FEDDS: The Florida Everglades Dry-Deposition Study

Preliminary Results from an Ongoing Investigation of the Dry-Depositional Loading of Mercury to the Florida Everglades Ecosystem

Frank J. Marsik¹, Gerald J. Keeler¹, Elizabeth G. Malcolm¹, J. Timothy Dvonch¹, James A. Barres¹, Steven E. Lindberg², Hong Zhang², Robert K. Stevens³ and Matthew S. Landis⁴

¹University of Michigan Air Quality Laboratory (Ann Arbor, MI, USA); ²Oak Ridge National Laboratory (Oak Ridge, TN, USA); ³Florida Department of Environmental Protection (Tallahassee, FL, USA); ⁴US Environmental Protection Agency (Research Triangle Park, NC, USA)

[E-mail Corresponding Author: marsik@umich.edu]

ABSTRACT

The Florida Everglades Dry-Deposition Study (FEDDS) was designed to test the viability of new and existing measurement techniques for use in the estimation of the dry-depositional loading of speciated mercury to a mixed sawgrass (cladium jamaicense) and cattail (typha domingensis) stand within the Florida Everglades.

Following the February 1999 “dry-season” intensive, results from water-based, surrogate surface measurements indicate that for the period studied, the average daily dry-deposition rate for total mercury to the site was 15.7 ± 9.8 ng/m2/day, with an average daytime (0900 to 1700 LT) rate of 21.8 ± 11.3 ng/m2/day and an average nighttime (1800 to 0800 LT) rate of 10.3 ± 3.4 ng/m2/day. Concurrent measurements of elemental and reactive gaseous mercury, as well as size-segregated particulate mercury, were also made. These data will be used as input into an inferential dry-deposition model for comparison with measured results.

A second measurement intensive was conducted during June 2000 in an effort to obtain summer, or wet-season, estimates of dry-deposition to the same site. Results from both measurement intensives will be presented.

INTRODUCTION

The discovery of elevated levels of methylmercury in freshwater fish in South Florida prompted the State of Florida Department of Health, in March 1989, to issue limited-consumption and no-consumption advisories for some species of fish found within the Florida Everglades. Since these initial advisories were issued, a number of consumption advisories have been issued statewide. The observed elevated levels of methylmercury are a result of the strong tendency of methylmercury to bioaccumulate up the aquatic food chain, with human consumption of aquatic wildlife being a common endpoint. In a study of human consumption of food from the Florida Everglades, Fleming et al. (1995) indicated that the populations that were at particular risk include sport fishermen, Everglades' residents and subsistence fishermen.

The atmosphere has been found to be a significant pathway for the transport and deposition of mercury in South Florida (Guentzel et al., 1995; Dvonch et al., 1998; Dvonch et al., 1999), as well as other parts of North American and Europe (e.g., Sorenson et al., 1994; Hoyer et al., 1995; Munthe et al. 1995; Lorey and Driscoll, 1999). The relative importance of atmospheric deposition pathways (wet- and dry-) varies considerably depending upon the location considered. Studies investigating the deposition of mercury to forested ecosystems in Tennessee (Lindberg et al., 1994), Sweden (Munthe et al., 1995) and Vermont (Rea et al., 1996) have shown both wet- and dry-depositional loading to be of importance. Guentzel et al. (1995) suggest that wet-deposition is an important pathway for mercury loading across South Florida during the summer season due to the high frequency of convective storms. However, the dry-deposition of mercury is also likely to be important across this region during non-precipitation periods, especially during the months of October through May, when precipitation is relatively infrequent.

The previous studies investigating the atmospheric deposition of mercury to South Florida have mainly focused on quantifying the contributions due to wet-depositional fluxes. The results presented in this paper represent the first effort to quantify the dry-depositional fluxes of total mercury to the Florida Everglades.

METHODOLOGY

Two measurement intensives were planned as part of the Florida Everglades Dry-Deposition Study. A “dry season” intensive was performed during the period: 24 February to 04 March 1999. Preliminary results from this study are presented in this paper. A “wet season” intensive was conducted in June 2000.

Two approaches were employed during the study, with dry-deposition estimates obtained using both direct and inferential techniques. Direct measurements of total mercury dry-depositional flux were obtained using a water-based, surrogate surface technique developed by the University of Michigan Air Quality Laboratory. To make this measurement, aerodynamic frisbee-shaped deposition plates were deployed at a height equal to the average height of the vegetative canopy. Each deposition plate, made of Teflon, contained a removable well that holds 300 ml of ultra-pure water. Both daytime and nighttime samples were collected with daytime samples deployed from 0900 to 1700 local time and nighttime samples deployed from 1800 to 0800 local time. All sample handling was performed using particle-free gloves to insure that sample contamination did not occur. Samples were analyzed for both total mercury and trace metals within a Class 100 cleanroom at the University of Michigan using techniques detailed in Landis and Keeler (1997) and Hoyer et al. (1995).

FEDDS SITE

MIAMI

Figure 1. Location of FEDDS sampling site.

Modeled fluxes were obtained using the inferential dry-deposition model of Hicks et al. (1987). In this technique, measurements of ambient chemical concentrations of the gaseous and particulate species of interest are combined with species-specific deposition velocities to “infer” rates of pollutant dry-deposition to the surface of interest. The model of Hicks et al. (1987) mathematically describes the variations in the dry-deposition velocity of a given species based upon an electrical resistance analogy. Namely, the delivery of a pollutant from the atmosphere to a given surface is described in terms of a series of “resistances to deposition” that arise due to meteorological, chemical and biological processes that control pollutant delivery, adsorption and capture at natural surfaces.

Until very recently, little information has been available regarding the chemical levels of mercury species in the atmosphere and preliminary estimates of the dry-deposition of mercury to vegetation have employed the assumption that reactive gaseous mercury (believed to be the dominant species that is dry-deposited) behaves like gaseous nitric acid. Now, for the first time, measurement methodologies have been developed to characterize the ambient concentrations of the three dominant species of mercury: elemental [Hg(0)], reactive [Hg(II)] and particulate [Hg(p)].

For this study, elemental mercury was measured using Tekran 2537A continuous mercury vapor analyzers. Reactive mercury was collected manually using annular denuders coated with potassium chloride, and with an automated Tekran 1130 Mercury Speciation Unit. Total particulate mercury was collected using 47mm glass fiber filters in Teflon open-faced filter packs. Sample analysis was performed in a Class 100 cleanroom based upon the methods described in Keeler et al. (1995).

The flux of vapor phase mercury over Everglades vegetation has been found to be bi-directional, thus two unique methods are being employed to measure both the microscale (<1 m²), as well as the larger scale (approximately 100 m²) emissive fluxes of elemental mercury from the ecosystem. These methods involve state of the art micrometeorological approaches that have been modified for application to mercury in Florida, and an inert dynamic flux chamber (FC) developed for mercury water surface flux measurements. The modified Bowen ratio (MBR) method is a micrometeorological technique that is now widely used for measuring mercury fluxes over soils and vegetation. The method combines real-time trace gas flux measurements by eddy correlation with high precision time-averaged concentration gradients of these trace gases and mercury to obtain inferred fluxes of mercury from the ecosystem. Recent applications of the MBR approach over wetlands in the Everglades Nutrient Removal (ENR) area by members of our research team have proven highly successful (Lindberg et al., 1999). Through the MBR approach, we are also collecting data on the fluxes of heat, CO₂, and water vapor to allow for the investigation of linkages between the surface fluxes of mercury and plant photosynthetic processes.

The FC method uses a polycarbonate enclosure to directly measure the mercury concentration changes over a known surface area, from which dry deposition and emission rates can be computed directly. The FC approach allows for measurements of spatial flux patterns and for testing of Hg flux mechanisms through surface manipulations.

RESULTS

At the time of this writing, the laboratory analyses has been completed for those samples collected during the 24 February to 04 March 1999 intensive. While data analysis is ongoing, preliminary results from some portions of the study are currently available.

The results from our water-based, surrogate dry-deposition surfaces are presented in Table 1. These values represent the average dry-deposition measured using the two co-located deposition surfaces for each measurement period.

TABLE 1

Average Measured Total Mercury Dry-Deposition (ng/m²/period) during the 1999 FEDDS Measurement Intensive.

Date	Period	Measured Hg Dry-Deposition
25-Feb	Day	7.1
25-Feb	Night	7.9
26-Feb	Day	7.9
26-Feb	Night	4.8
27-Feb	Day	14.2
27-Feb	Night	2.9
28-Feb	Day	5.1
28-Feb	Night	6.1
1-Mar	Day	8.6
1-Mar	Night	9.4
2-Mar	Day	8.7
2-Mar	Night	6.9
3-Mar	Day	1.9
3-Mar	Night	3.9
4-Mar	Day	5.5
4-Mar	Night	5.7

Considerable sample-to-sample variability can be observed. When viewed on a “per period” basis, the average daytime total mercury dry-deposition for the period was 7.4 ± 3.6 ng/m², while the average nighttime total mercury dry-deposition for the period was 5.9 ± 2.1 ng/m². Despite the fact that the nighttime period encompassed a longer period of time (14 hours versus 8 hours), the average daytime dry-deposition total exceeds that of the nighttime total. This observation is likely the result of the fact that the generally unstable atmospheric conditions found during the daytime periods resulted in a greater rate of delivery of mercury to the surface. When scaled to a 24-hour deposition rate, the average daytime rate was 21.8 ± 11.3 ng/m²/day and the average nighttime rate was 10.3 ± 3.4 ng/m²/day.

As noted earlier, previous MBR measurements of the flux of vapor phase mercury over Everglades vegetation found these fluxes to be bi-directional. Preliminary results from the vapor phase mercury flux measurements made during the 24 February to 04 March 1999 intensive are now available and concur with the previous observations. Using the modified Bowen ratio approach described earlier, measurements suggest that for the period studied the overall average vapor phase mercury flux was 12 ± 20 ng/m²/hr. The large range in observed values corresponds to the fact that net fluxes are generally positive during the day (away from vegetation) and slightly negative at night (toward the vegetation). On average, daytime fluxes were found to be approximately 32 ± 27 ng/m²/hr. These observed values were found to be in the same range as those measured in the Everglades Nutrient Removal area by team members Steve Lindberg and Tilden Meyers (NOAA-ATDD, Oak Ridge, TN) [Personal Communication].

ACKNOWLEDGEMENTS

We would like to thank our sponsors for this project, the US Environmental Protection Agency and the State of Florida Department of Environmental Protection. Additionally, we would like to thank the South Florida Water Management District, without whose cooperation this project would not have been possible.

REFERENCES

Dvonch, J.T., J.R. Graney, F.J. Marsik, G.J. Keeler and R.K. Stevens (1998). Sci. Total Environ., 213: 95-108.

Dvonch, J.T., J.R. Graney, G.J. Keeler and R.K. Stevens (1999). Environ. Sci. Technol., 33: 4522-4527.

Guentzel, J.L., W.M. Landing, G.A. Gill and C.D. Pollman, 1995: Atmospheric deposition of mercury in Florida: The FAMS Project (1992-1994). Water, Air and Soil Pollut., 80: 393-402.

Hicks, B.B., D.D. Baldocchi, T.P. Meyers, R.P. Hosker, Jr., and D.R. Matt (1987). Atmos. Environ., 32: 1317-1322.

Hoyer, M., J. Burke and G.J. Keeler (1995). Water, Air and Soil Pollut., 80: 199-208.

Keeler, G., G. Glinsorn, N. Pirrone (1995). Water, Air and Soil Pollut., 80: 159-168.

Landis, M.S. and G.J. Keeler (1997). Environ. Sci. Technol., 31: 2610-2615.

Lindberg, S.E., J.G. Owens, and W. Stratton (1994): In: Mercury as A Global Pollutant (J. Huckabee and C. Watras, Eds.), Lewis Publishers, pp. 261-272.

Lindberg, S. E., H. Zhang, and T. P. Meyers (1999): Final Report Everglades Mercury Air/Surface Exchange Study (E-MASE). South Florida Water Management District, West Palm Beach, FL.

Lorey, P. and C.T. Driscoll (1999). Environ. Sci. Technol., 33: 718-722.

Munthe, J., H. Hultberg, A. Iverfeldt (1995). Water, Air and Soil Poll., 80: 363-371.

Rea, A.W., G.J. Keeler and T. Scherbatskoy (1996). Atmos. Environ., 30: 3257-3263.

Sorensen, J.A., G.E. Glass and K.W. Schmidt (1994). Environ. Sci. Technol., 28: 2025-2032.