Mercury wet deposition involves the transfer of mercury from the atmosphere to land or water through precipitation. Several chemical species of mercury exist in ambient air as a result of both natural and man-made emissions and the water-soluble forms may be scrubbed out of the atmosphere by cloud water or rain and snowfall. For many sensitive surface waters, atmospheric wet deposition constitutes the most significant route of mercury input. Dry deposition processes also contribute to the overall rate of atmospheric deposition, but wet deposition is believed to play a larger role in the eastern United States where precipitation rates are higher than in arid western states. Together, these phenomena can contribute to raise methylmercury levels in fish in mercury-sensitive waters.

Wet deposition data is generated through the collection and subsequent analysis of rainfall for total mercury concentration. The DAQ has operated two sites for measurement of mercury in rainfall since 1996. Both deposition monitors are located in the eastern part of the state near mercury-sensitive waters; one at Pettigrew State Park on the shores of Phelps Lake in Washington County (NC42)*, and the other at Waccamaw State Park in Columbus County (NC08). Data collected from these stations are provided to the National Atmospheric Deposition Program Mercury Deposition Network (MDN) to aid in the identification of  geographical and temporal trends in mercury deposition across the U.S. Rainfall is collected weekly in a bucket sampler and sent to a laboratory for quantitative analysis. Mercury levels are measured using EPA Method 1631 for total mercury analysis and undergo full quality assurance/quality control procedures before being reported. Data are presented in two formats: volume-weighted mercury concentration in rainwater (ng/L), and deposition rate (ng/m²). Annualized values are presented in the following table for both North Carolina sites, 1996 - 1999.

Data from 1996-1998 are reported in Table 1 if they met the MDN criteria for valid samples. Data from 1999 are preliminary and have not been fully quality assured by the MDN program office. On a handful of occasions, data were rejected. Reasons for rejecting data included site operation problems and sample collection or processing problems.The net effect of these exclusions is to decrease estimates of cumulative deposition. It cannot be said what impact these missing data have on the calculation of volume-weighted average mercury concentration.

Annual wet deposition rates in eastern North Carolina are similar to other locations in the United States (see figure at right). Wet deposition rates are influenced by rainfall amount and rainwater mercury concentration. Mercury is added to rainwater through cloud-level chemical reactions and/or scrubbing out of reactive gaseous mercury (RGM) or particulate mercury from the air column as the rainwater falls to earth. Local emissions may play a role in this variable, especially under the right meteorological conditions (Dvonch, 1998). Reporting volume-weighted mercury concentrations reduces the relative influence of small precipitation events and provides a better indication of longer-term trends in mercury rainwater concentration. This means of reporting does not, however, capture short-term trends in temporal or spatial variability which may be a better indicator of local-scale impacts.

Both North Carolina deposition monitors are situated in relatively remote locations, at least 20 kilometers away from any urban activity that might be responsible for significant mercury emissions. The Waccamaw site, however, is more likely to be influenced by local industrial activity since it is located approximately 15 miles from a major atmospheric mercury emissions source and about 30 miles from Wilmington, NC. The Phelps site, which is surrounded by parkland and agricultural activities and free of local or regional mercury emissions sources, could be thought of as a good representation of "background" conditions for the eastern coastal plain of North Carolina.

An illustration of week-to-week variability in mercury deposition can be seen in the figure at right, which presents data collected during 1997. The high degree of variability in weekly wet deposition rates is primarily due to differences in precipitation levels. However, mercury concentration also seems to play a role in the observed variability. Variation in rainwater mercury concentration is significant when examined on a week-to-week basis. The weekly values for the Lake Waccamaw site ranged from 3.09 ng/L to 49.04 ng/L during 1997. At Phelps, the range extended from 1.74 ng/L to 26.98 ng/L. An even wider range of values has been reported for precipitation captured in southeastern Florida (Dvonch 1998). In the Florida study, individual rain event samples were collected at 17 sites between August 6 and September 6, 1995. The variability in event mercury concentration for all sampling sites was substantial, with reported values ranging from 4.5 ng/L to 113.2 ng/L. Monthly volume-weighted mean concentrations fluctuated from 13.1 ng/L to 30.5 ng/L, a range similar to those values calculated from North Carolina data. It is interesting to note that in addition to a consistent pattern of lower annual volume-weighted average mercury concentrations there is also a tighter range of weekly mercury precipitation concentrations at the Phelps site. This is consistent with predictions for a rural location unaffected by local mercury sources (EPA).

A pattern of elevated mercury concentrations during summer months can be observed when the data are separated by season (see figure at left). Data from 1996 and 1997 have been split into two groupings: April - September and October - March. During the April - September time period, the average volume-weighted mercury concentration in rainwater is over twice that detected during the remainder of the year. Some of this difference might be explained by seasonal changes in air movement. During summer months winds predominantly arrive from the southwest, while winds from the north or northwest influence wind patterns during winter months. Conceivably, if these sampling sites were downwind of sources during one half of the year and upwind during the remainder, a pattern might emerge from the data. However, higher summertime readings are seen consistently at other MDN sampling sites on the east coast so it seems unlikely that this is the only phenomenon at work. Some have suggested that warmer temperatures will facilitate atmospheric reactions that create RGM from elemental mercury and thus increase the potential for deposition (Mason, 2000). Other potential reasons could include meteorological transport phenomena, precipitation scavenging efficiency (rain vs. snow) or the amount of sunlight. At present the relative impact of local atmospheric mercury sources cannot be determined by examining seasonal data without more complete information on local and mesoscale meteorological conditions during individual rain events.

Currently, data is only available on wet deposition patterns in North Carolina. It is believed that dry deposition also occurs in the absence of precipitation, involving RGM and particulate mercury species (EPA). The net effect of this phenomenon would be to increase the overall rate of deposition relative to that predicted using data for wet deposition only. At this point, it is difficult to speculate what the relative impact of dry deposition is in North Carolina. It is likely that dry deposition occurs to a significant extent near RGM emission points, especially those which emit at ground level where significant scavenging by foliage could take place. This is an important phenomenon to consider, as dry deposition could result in substantial direct and indirect inputs of mercury to waterways located near mercury emission sources. In the near future, the DAQ will acquire more complete information on ambient levels of RGM to assess the potential for dry deposition and source impacts on mercury-sensitive areas. The Lake Waccamaw area is characterized by some of the highest levels of mercury in fish in the state; the identification of atmospheric inputs to this area could aid in ameliorating this situation if proper steps are taken to reduce mercury emissions.

For more information on mercury deposition trends in North Carolina, please contact Jeff Hayward with the Toxics Protection Branch at (919) 733-1475.

* During the summer of 2000, the Phelps site will be relocated to a National Wildlife Refuge area positioned between Phelps Lake and Pungo Lake in Washington County. The new site will be located less than 5 miles from the current site.

References:

Environmental Protection Agency (EPA) (1997). Mercury Study Report to Congress. EPA-452/R-97-003.

Dvonch, et al (1998). An investigation of source-receptor relationships for mercury in south Florida using event precipitation data. The Science of the Total Environment, v 213, pp 95-108.

Mason RP, et al (2000). Annual and Seasonal Trends in Mercury Deposition in Maryland. Atmospheric Environment, v 34, pp 1691-1701.