THE EFFECT OF DRY DOWN AND NATURAL FIRES ON MERCURY METHYLATION IN THE FLORIDA EVERGLADES
David P. Krabbenhoft, U.S.
Geological Survey, Middleton, WI 53562 (dpkrabbe@usgs.gov), and Larry E. Fink,
South Florida Water Management District, West Palm Beach, FL 33416
Extensive fires occurred in the
Florida Everglades in May and June 1999 following a La Nińa-driven dry period. In
response, the U.S. Geological Survey and the South Florida Water Management
District conducted a collaborative study on the effect of sediment drying and
fires on mercury (Hg) speciation and bioaccumulation at 13 sites spanning most
of the north-to-south length of the remnant Everglades. At the time of the first
post-burn sampling (July 1999) methylmercury (MeHg) levels in surface water,
porewater, sediments, periphyton, and mosquitofish were about 2x, 18x, 11x,
1.5x, and 0.7x higher, respectively, in the burned areas versus non-burned
locations. Monitoring at these sites showed burdens of MeHg in mosquitofish and
periphyton continued to build throughout the fall of 1999 reaching maximum
observed levels in October and November.
Peat oxidation from burning or intense drying could potentially enhance
methylation of Hg by increasing the availability of sulfate, labile carbon,
and/or Hg (II). Of these three
parameters, only sulfate showed demonstrably higher levels (about 2.4x) in
response to the drying and burning.
Regardless of the precise geochemical mechanism, data collected from
this study suggests that geochemical changes induced by prolonged drying or
burning of peat from the Everglades can favor Hg methylation through increased
availability of necessary constituents such as sulfate.
Mercury
levels in several species of predator sport fish from the Everglades exceed the
Florida action level of 0.5 ppm (Ware et al. 1990). The mean concentration of
Hg in largemouth bass from the central Everglades exceeds 1.5 ppm, which is
high in comparison to any published studies. In March 1989, the Florida
Department of Health and Rehabilitative Services issued fish consumption
advisories for Hg in each of the Water Conservation Areas (WCAs) and the
Everglades National Park. Although
advisories are given for Hg, almost all of mercury in fish tissues is
methylmercury (MeHg), and thus it is necessary to understand the distribution,
production mechanisms, and bioaccumulation pathways of MeHg. The ecological significance of the
occurrence of high concentrations of MeHg in the aquatic environment includes
threats to such endangered species as the Florida panther and fish-eating
birds, such as the wood stork.
It is not yet completely understood
what controls the apparent susceptibility of Everglades to Hg exposure: high
annual inorganic mercuric ion loads; high concentrations of inorganic mercury
in peat soils via historical accumulation and occasional concentration via
plant or muck fires; high bioavailable fraction of Hg available for
methylation; high absolute rates of Hg methylation and corresponding low rates
of demethylation of methylmercury; high biotic uptake rates of produced MeHg;
high biomagnification factors in the aquatic food chain; or any combinations of
the above. Studies by the U.S. Geological Survey (USGS) under the Aquatic
Cycling of Mercury in the Everglades (ACME) Project (Krabbenhoft 1996) of the
underlying transport, transformation, fate, and bioaccumulation processes that
describe these chemical interrelationships have been underway since 1995.
Plant-top and muck fires in the Everglades are part of
the natural biogeochemical cycle of the Everglades. In addition, fires may become more frequent in some areas of the
Everglades under more natural hydrologic conditions that may result from
implementation of the restoration plan.
These extreme chemical oxidation events can change soil and water
chemistries that influence inorganic mercury availability, or other chemical
constituents needed for methylation. A
priori, one might expect that fires could liberate a significant fraction
of otherwise unavailable mercury, sulfate, and/or labile carbon, and thus
stimulate the methylation process.
However, there are no known studies on how fires affect mercury cycling
and bioaccumulation.
Extensive fires occurred in the
impounded Everglades in May and June 1999 following a La Nińa -driven dry period. In some locations, the fire
burned through exposed peat soil to the underlying rock. Our study hypothesis was that oxidation of
peat from burning or intense drying would enhance availability of sulfate,
labile carbon, and/or Hg (II) for methylation.
To test this hypothesis, the U.S.
Geological Survey and the South Florida Water Management District conducted a
collaborative study on the effect of sediment drying and fires on Hg speciation
and bioaccumulation in the Everglades.
The first post-burn sampling was conducted in July 1999, after allowing for a period of about five weeks of inundation due to the onset of the summer rainy season. Thirteen sites were sampled (Table 1), including ten sites previously studied by the ACME project (one of which, 3A33, was burned) and three additional sites from burned areas. The ten ACME sampling sites span almost the entire north-to-south length of the remaining Everglades, and represent many of the varied sub-ecosystem types. The sampled burn sites were focused in and around northern WCA-3A where the most widespread burns occurred. With this array of sites, we make data comparisons on three levels: (1) 1999 data from burned versus unburned areas, (2) 1995-98 data versus post-burn data collected at site 3A33, and (3) ACME data for 1995-98 from all sites versus the post burn time period. In addition, to track the temporal effects of the burn and dry down, follow-up sampling was conducted at six sites in August, October, and November for surface water, mosquitofish, sediment, and periphyton.
Previously established ultra-clean sampling methods by
the ACME project were used to collect and analyze the samples for this study
(Hurley et al. 1998; Krabbenhoft et al. 1998; Cleckner et al. 1998; Olson et
al. 1997; and Olson and DeWild 1999).
Sites were accessed primarily by helicopter, but in some cases airboats
were used. When helicopter transport
was utilized, field crews walked several hundred feet away from the helicopter
to ensure undisturbed samples were acquired.
At each site, samples of surface water (mid water column, filtered [0.45
mm] and
unfiltered), peat porewater (from 5 cm depth), sediment (top 5 cm), floating
periphyton (may not have reestablished yet at burn sites), and mosquitofish
(composite analysis of 10-15 individuals) were collected for analysis of total
Hg (HgT) and methylmercury.
In addition, aqueous samples were analyzed for temperature, pH, and
dissolved (0.45 mm) sulfate, chloride, sulfide, dissolved organic carbon, specific UV
absorbance (SUVA) at 254 nm, and total suspended solids. Temperature was
measured at two surface water depths (mid water column and 5-10 cm above the
sediments) as well as at 5 cm sediment depth.
Water column depth was measured at 5 locations per site to estimate
average water column depth.
At
the time of the first post-burn sampling (July, 1999) MeHg levels in surface
water, porewater, sediments, periphyton, and mosquitofish were about 2x, 18x,
11x, 1.5x, and 0.7x higher, respectively, in the burned or severely dried areas
versus non-burned locations. The fact that the greatest changes were observed
for porewater and sediment is not surprising because Hg methylation in the
Everglades is primarily facilitated by of sulfate reducing bacteria (SRB) in
the near-surface (top 5 cm) sediments (Gilmour et al., 1998), although
periphyton has also been showed to methylate Hg (Cleckner et al. 1999). Maximum MeHg levels in surface water and
sediment were observed in July and follow-up monitoring showed that surface
water remained significantly elevated at least until November 1999 when
compared to the four-years of study by the ACME project that preceded the
burn. Other the other hand, sedimentary
MeHg concentrations returned to “normal” levels by about October for severely
dried areas, but remained high in burned areas through November.
Trends in MeHg levels in
biological samples showed different trends. Burdens of MeHg in mosquitofish and
periphyton continued to build throughout the fall of 1999 reaching maximum
observed levels in October and November.
This observation suggests an inherent time lag on the order of 90 to 120
days between punctuated methylmercury production in response to peat oxidation
by burning and dry down, and bioaccumulation response in the food web. In addition, though most of the routine
sampling sites were not burned in the spring of 1999, many of these sites
incurred a prolonged period of draw down and oxidation. This may explain why, overall, we observed
about 2x overall higher levels of MeHg in surface water than observed during
four years of sampling from 1995 to 1998.
In order to efficiently
carry out their anaerobic metabolic processes, SRBs require sulfate and a
labile carbon substrate. To methylate
Hg, they also need a bioavailable pool of inorganic Hg (Gilmour et al. 1991). Our hypothesis was that one or more of these
three necessary ingredients (sulfate, labile carbon, and bioavailable Hg) for
Hg methylation would increase in abundance due drying and burning of peat, and
yield higher levels of methylmercury.
This could result from oxidation of organic or inorganic sulfide to
yield sulfate; degradation of non-biodegradable carbon pools to yield labile
carbon; and, liberation of mercury from unavailable pools bound to
sediments. Of these three parameters,
our sampling and analysis show that only sulfate in surface water and sulfide
in porewater showed demonstrably higher levels (about 2.4x and 7.0x,
respectively) in response to the drying and burning. Benoit et al (1999)
recently demonstrated that sulfide can have a controlling influence on mercury
speciation in aqueous environments, and at the right concentrations can
potentially increase the availability of mercury for methylation by formation
of zero-charged, mono-sulfide species.
No appreciable changes in HgT were observed in water,
sediment or biota, and DOC levels, and DOC quality showed no change compared to
historical observations. Given that HgT
and DOC quantity and quality were constant for burned sites and we observed 10x
increase in net methylation efficiency (defined here as the percent of HgT
as MeHg in sediment and porewater), we infer that liberation of sulfate from
sediments and secondary stimulation of SRBs was a primary driving factor of
excess methylmercury production in burned/dried areas. Regardless of the
precise geochemical mechanism, data collected from this study suggests that
geochemical changes induced by prolonged drying or burning of Everglades peat
favors substantial Hg methylation through increased availability of important
substrates such as sulfate.
Benoit, J.M., R. P. Mason and C.C. Gilmour, 1999, Estimation of mercury-sulfide speciation and bioavailability in sediment pore waters using octanol-water partitioning, Environ. Toxicol. & Chem., 18: 2138-2141.
Cleckner, LB, Garrison, PJ,
Hurley, JP, Olson, ML, and Krabbenhoft, DP, 1998, Trophic transfer of
methylmercury in the northern Everglades, Biogeochemistry,
40: 347-361.
Cleckner, L.B., C.C Gilmour, J.P. Hurley, and D.P. Krabbenhoft, 1999, Mercury Methylation by Periphyton in the Florida Everglades, Limnology and Oceanography 44, pp. 1815-1825.
Gilmour, CC, Riedel, GS,
Ederington, MC, Bell, JT, Beniot, JM, Gill, GA, and Stordal, MC, 1998,
Methylmercury concentrations and production rates across a trophic gradient in
the Northern Everglades. Biogeochemistry,
40: 326-346.
Gilmour, CC, Henry, EA and Mitchell, R. 1991, Sulfate stimulation of mercury methylation in freshwater sediments. Environ. Sci. Technol. 26: 2281-2287.
Hurley, JP,
Krabbenhoft, DP, Cleckner, LB, Olson, ML, Aiken, GR, and Rawlik, PJ, 1998,
System controls on aqueous mercury distribution in the northern Everglades, Biogeochemistry, 40: 293-310.
Krabbenhoft, DP, 1996, Mercury Studies in the Florida Everglades, 1996, U.S. Geological Survey Fact Sheet, FS-166-96 (4 p).
Krabbenhoft, DP,
Hurley, JP, Olson, ML, and Cleckner, LB, 1998, Diel variability of mercury
phase and species distributions in the Florida Everglades, Biogeochemistry, 40:311-325.
Olson, ML,
Cleckner, LB, Hurley, JP, Krabbenhoft, DP, and Heelan, TW, 1997, Resolution of
matrix effects on analysis of total and methyl mercury in aqueous samples from
the Florida Everglades, Fresenius J.
Analytical Chemistry, 358: 392-396.
Olson, ML, and DeWild, JF, 1999, Low-level techniques for the collection and species-specific analysis of low levels of mercury in water, sediment, and biota, in Morganwalp, D.W., and Buxton, H.T., eds., U.S. Geological Survey Toxic Substances Hydrology Program--Proceedings of the Technical Meeting, Charleston, South Carolina, March 8-12, 1999--Volume 2 of 3--Contamination of Hydrologic Systems and Related Ecosystems: U.S. Geological Survey Water-Resources Investigations Report 99-4018B: 191-200.
Ware, FJ, Royals, H, and Lange, T, 1990, Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies, 44: 5-12.
Table 1.
Sampling results for unfiltered surface water and sediment from the
thirteen study sites, and comparison of historical average values (1995-98) to
July 1999 conditions. Latitude and
longitude in degrees, minutes and hundredths of minutes
|
Site |
Spring 1999 Condition |
Lati- tude |
Longi- tude |
7/99 Surface Water
(ng/L) |
7/99 Sediment (ng/g, dry) |
Water Ratio: 1995-98
Average/7-99 |
Sediment Ratio: 1995-98 Average/7-99 |
||||
|
|
|
|
|
HgT |
MeHg |
HgT |
MeHg |
HgT |
MeHg |
HgT |
MeHg |
|
ENR103 |
Wet |
26 38.24 |
80 25.36 |
0.57 |
0.019 |
105.38 |
0.035 |
|
|
|
|
|
WCA1 |
Wet |
26 30.61 |
80 18.63 |
3.24 |
0.260 |
204.92 |
13.143 |
0.72 |
1.00 |
0.82 |
2.76 |
|
F1 |
Dried |
26 21.58 |
80 22.23 |
2.03 |
0.484 |
116.51 |
8.600 |
0.65 |
1.86 |
1.22 |
28.75 |
|
U3 |
Dried |
26 17.25 |
80 24.68 |
4.39 |
1.618 |
257.96 |
7.124 |
0.95 |
2.74 |
2.32 |
7.14 |
|
2BS |
Wet |
26 09.82 |
80 22.68 |
2.7 |
0.923 |
249.45 |
1.387 |
0.97 |
1.85 |
1.62 |
1.36 |
|
2AB |
Burned |
26 23.48 |
80 27.81 |
3.45 |
2.080 |
99.63 |
19.810 |
new site |
new site |
new site |
new site |
|
3A-1 |
Burned |
26 11.21 |
80 44.41 |
2.33 |
0.463 |
149.32 |
11.074 |
new site |
new site |
new site |
new site |
|
3A-4 |
Burned |
26 19.05 |
80 47.83 |
3.16 |
1.142 |
81.02 |
57.582 |
new site |
new site |
new site |
new site |
|
3A-33 |
Burned |
26 16.16 |
80 36.82 |
2.1 |
0.545 |
110.09 |
35.734 |
1.13 |
1.76 |
1.56 |
7.08 |
|
3A-15 |
Wet |
25 58.45 |
80 40.13 |
1.95 |
0.193 |
350.28 |
0.997 |
1.15 |
0.86 |
0.94 |
2.99 |
|
3A-TH |
Wet |
25 46.87 |
80 41.12 |
1.85 |
0.644 |
257.05 |
1.708 |
0.89 |
1.37 |
0.69 |
27.42 |
|
TS-7 |
Wet |
25 17.23 |
80 38.78 |
2.07 |
0.622 |
41.42 |
1.772 |
0.71 |
3.45 |
1.33 |
18.25 |
|
TS-9 |
Wet |
25 14.85 |
80 40.94 |
1.32 |
0.413 |
77.53 |
4.313 |
0.44 |
1.53 |
2.41 |
0.29 |